From mknappe@juniper.net Fri Oct 1 08:52:03 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 9C53D3A6ECB for ; Fri, 1 Oct 2010 08:52:03 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -105.98 X-Spam-Level: X-Spam-Status: No, score=-105.98 tagged_above=-999 required=5 tests=[AWL=0.619, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6ultuA1ds6ey for ; Fri, 1 Oct 2010 08:52:01 -0700 (PDT) Received: from exprod7og115.obsmtp.com (exprod7og115.obsmtp.com [64.18.2.217]) by core3.amsl.com (Postfix) with ESMTP id 188823A6EC2 for ; Fri, 1 Oct 2010 08:52:01 -0700 (PDT) Received: from source ([66.129.224.36]) (using TLSv1) by exprod7ob115.postini.com ([64.18.6.12]) with SMTP ID DSNKTKYD0RosM5nVMSgJ4D/AOyz80Zp7WmR2@postini.com; Fri, 01 Oct 2010 08:52:49 PDT Received: from EMBX02-HQ.jnpr.net ([fe80::18fe:d666:b43e:f97e]) by P-EMHUB02-HQ.jnpr.net ([fe80::88f9:77fd:dfc:4d51%11]) with mapi; Fri, 1 Oct 2010 08:52:31 -0700 From: Michael Knappe To: "codec@ietf.org" Date: Fri, 1 Oct 2010 08:52:26 -0700 Thread-Topic: [codec] Draft codec WG agenda IETF79 Thread-Index: ActhgKTf54MfUwxY+EmzL8IVmXnmUw== Message-ID: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-Entourage/13.3.0.091002 acceptlanguage: en-US Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: [codec] Draft codec WG agenda IETF79 X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 01 Oct 2010 15:52:03 -0000 Here=92s the proposed codec WG agenda for IETF79 in Beijing, we=92ve put in= a request for a 2 hour timeslot. Please reply with any comments and sugges= tions, thanks. 1. Administrivia [10 min] 2. Guidelines and requirements doc progress [25 min] 3. Codec development status [15 min] 4. Codec design walkthrough (a brief tutorial on the design to-date) [25 mi= n] 5. Codec testing status [20 min] 6. Timelines, actions and wrap-up [20 min] Cheers, Mike From fluffy@cisco.com Fri Oct 1 09:40:20 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 6F4B23A6DAA for ; Fri, 1 Oct 2010 09:40:20 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -110.403 X-Spam-Level: X-Spam-Status: No, score=-110.403 tagged_above=-999 required=5 tests=[AWL=0.196, BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id mGFf7EDFDWDl for ; Fri, 1 Oct 2010 09:40:18 -0700 (PDT) Received: from sj-iport-4.cisco.com (sj-iport-4.cisco.com [171.68.10.86]) by core3.amsl.com (Postfix) with ESMTP id 1BC003A6F8A for ; Fri, 1 Oct 2010 09:38:47 -0700 (PDT) Authentication-Results: sj-iport-4.cisco.com; dkim=neutral (message not signed) header.i=none X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AgkGAAqrpUyrRN+K/2dsb2JhbACUMI4HcasqnD6FRASEUYVrgwI X-IronPort-AV: E=Sophos;i="4.57,267,1283731200"; d="scan'208";a="194680316" Received: from sj-core-4.cisco.com ([171.68.223.138]) by sj-iport-4.cisco.com with ESMTP; 01 Oct 2010 16:39:32 +0000 Received: from [192.168.4.2] (rcdn-fluffy-8711.cisco.com [10.99.9.18]) by sj-core-4.cisco.com (8.13.8/8.14.3) with ESMTP id o91GdVem029247 for ; Fri, 1 Oct 2010 16:39:32 GMT From: Cullen Jennings Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Date: Fri, 1 Oct 2010 10:39:31 -0600 References: <20101001161224.A4F2D3A6CEC@core3.amsl.com> To: codec@ietf.org Message-Id: <163C949F-A3AB-4683-BDD8-4714E9E26B9F@cisco.com> Mime-Version: 1.0 (Apple Message framework v1081) X-Mailer: Apple Mail (2.1081) Subject: [codec] Tentative scheduled for IETF 79 X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 01 Oct 2010 16:40:20 -0000 This is just the tentative time and it could change but as FYI ... CODEC Session 1 (2 hours) Tuesday, Afternoon Session I 1300-1500 Room Name: Valley Ballroom C From fluffy@cisco.com Fri Oct 1 10:26:28 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id A90473A6C09 for ; Fri, 1 Oct 2010 10:26:28 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -110.409 X-Spam-Level: X-Spam-Status: No, score=-110.409 tagged_above=-999 required=5 tests=[AWL=0.190, BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 2J9dZnq4c3ZS for ; Fri, 1 Oct 2010 10:26:27 -0700 (PDT) Received: from sj-iport-5.cisco.com (sj-iport-5.cisco.com [171.68.10.87]) by core3.amsl.com (Postfix) with ESMTP id C78313A6918 for ; Fri, 1 Oct 2010 10:26:27 -0700 (PDT) Authentication-Results: sj-iport-5.cisco.com; dkim=neutral (message not signed) header.i=none X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AgkGAEu2pUyrRN+J/2dsb2JhbACUMI4HcatdnD2FRASEUYVrgwI X-IronPort-AV: E=Sophos;i="4.57,267,1283731200"; d="scan'208";a="263420413" Received: from sj-core-3.cisco.com ([171.68.223.137]) by sj-iport-5.cisco.com with ESMTP; 01 Oct 2010 17:27:08 +0000 Received: from [192.168.4.2] (rcdn-fluffy-8711.cisco.com [10.99.9.18]) by sj-core-3.cisco.com (8.13.8/8.14.3) with ESMTP id o91HR7fR005320; Fri, 1 Oct 2010 17:27:07 GMT From: Cullen Jennings Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Date: Fri, 1 Oct 2010 11:27:06 -0600 Message-Id: <1E7430DC-ED0D-44A3-826B-B1F344CB918E@cisco.com> To: codec@ietf.org Mime-Version: 1.0 (Apple Message framework v1081) X-Mailer: Apple Mail (2.1081) Cc: Jonathan Rosenberg Subject: [codec] Summary of codec specification X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 01 Oct 2010 17:26:28 -0000 With my chair hat on, here are thing thing I believe we have consensus = on to move forward on. 1) The code will be in the appendix of the draft base64 encoded. We can = sort out later if it is gziped tar file of whatever but it will be some = easy to extract format. This has been done on other documents. I have = talked to the AD about this before. I have not talked to RFC Ed but do = not foresee any issues there.=20 2) The code will be under the simplified BSD license as specified by = current IETF process documents and most importantly the IETF Trust = Licensee Policy. (As a side note, this is not a topic that is debatable = in this WG - if you want to change IETF process, go somewhere else) 3) We have rough consensus that where the code and the text version = description of the algorithm disagree, the code version is the one that = is "correct" 4) The chairs have not yet made any decision about if we have consensus = on bit-exactness. I'm not 100% sure where we stand on this but given we = are in doubt on this, there is no consensus yet.=20 5) The chairs have not made any decisions about if we have consensus on = characterization gating. See logic around #4.=20 Once we have a WG guidelines document, we as a WG can make sure it = represents what is and is not consensus - something we can't do for an = individual draft. Given the discussion over the last week, we are = probably going to adopt draft-valin-codec-guidelines as the basis for = the WG document some time real soon now. The chairs need to have a few = more conversations about this first. It's would be fine at this point in = time for a WG draft on guidelines to have proposed text in it that did = not yet have consensus but it should clearly state they did not have = consensus and was simply proposed text.=20 Cullen - That's a lot of "C"s From fluffy@cisco.com Fri Oct 1 10:36:59 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 9B48B3A6CC0 for ; Fri, 1 Oct 2010 10:36:59 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -110.415 X-Spam-Level: X-Spam-Status: No, score=-110.415 tagged_above=-999 required=5 tests=[AWL=0.184, BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id zh6787ejHGFx for ; Fri, 1 Oct 2010 10:36:58 -0700 (PDT) Received: from sj-iport-1.cisco.com (sj-iport-1.cisco.com [171.71.176.70]) by core3.amsl.com (Postfix) with ESMTP id 0F1453A6918 for ; Fri, 1 Oct 2010 10:36:58 -0700 (PDT) Authentication-Results: sj-iport-1.cisco.com; dkim=neutral (message not signed) header.i=none X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AvsEAGS5pUyrRN+K/2dsb2JhbACiN3GrVZw/hUQEhFGFa4MC X-IronPort-AV: E=Sophos;i="4.57,267,1283731200"; d="scan'208";a="367274784" Received: from sj-core-4.cisco.com ([171.68.223.138]) by sj-iport-1.cisco.com with ESMTP; 01 Oct 2010 17:37:46 +0000 Received: from [192.168.4.2] (rcdn-fluffy-8711.cisco.com [10.99.9.18]) by sj-core-4.cisco.com (8.13.8/8.14.3) with ESMTP id o91HbjAi001687; Fri, 1 Oct 2010 17:37:45 GMT Mime-Version: 1.0 (Apple Message framework v1081) Content-Type: text/plain; charset=us-ascii From: Cullen Jennings In-Reply-To: <20100925105523.15884gktkx78i41n@webmail.uniud.it> Date: Fri, 1 Oct 2010 11:37:45 -0600 Content-Transfer-Encoding: quoted-printable Message-Id: <8F3BC3D2-E3CB-49FE-A31C-1B4BDE88142E@cisco.com> References: <4C9D0C24.5080302@usherbrooke.ca> <4c9d216e.1021cc0a.658d.51bf@mx.google.com> <4C9D3288.1000608@usherbrooke.ca> <20100925105523.15884gktkx78i41n@webmail.uniud.it> To: Riccardo Bernardini X-Mailer: Apple Mail (2.1081) Cc: codec@ietf.org Subject: Re: [codec] Format for the codec specification X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 01 Oct 2010 17:36:59 -0000 These are all valid concerns, but I think people have often found that = the providing code path is the "least bad way" of specifying codecs. I = did want to point out that bugs are found in specs all the time = regardless if they are C code, english text, or something else. We have = an errata process for updating RFCs as well as a ways to publish new = versions of RFCs. I think that is about the best we can do. It's also = easy to accidentally write ambiguous text like your C example below. = Again, this happens in both english and C - we just need to do our best = to keep an eye open for those and fix them. That's one of the advantages = of having "running code" - it helps discover the corner cases that are = under specified. The code tends to be "Reference" in that it is written = to help illustrate the algorithm and is often not the most optimal way = to implement the algorithm, may not be thread safe, may need changes to = be ported to some platforms and so on. C may not be the best for this = but it has the advantage of behind widely understood - particularly in = DSP codec circles. We could have a huge argument about what language for = a long time. I have seen that argument before and it's not fun. In the = end, the odds we would choose C are extremely high.=20 Cullen On Sep 25, 2010, at 2:55 AM, Riccardo Bernardini wrote: > Quoting Jean-Marc Valin : >=20 >> Hi Roni, >>=20 >> Thanks for the correction on the case of G.719. Despite that, my = point still stands: there *are* codecs for which the decoder is = standardized without a bit-exact definition. As long as we define the = amount of deviation permitted, then there is no problem. >>=20 >> Oh, and I do agree that the C code should take precedence over the = mathematical description. >=20 > A question just came to my mind: what about bugs in the C code? = Suppose, for example, that the algorithm requires the computation of a = DCT of a block of samples and that there is some typo in the constants = used in the C code, so that what the C code computes is not the DCT, but = something similar and the difference is such that it really does not = matter in most cases, but causes very bad performances (when compared = with theoretical description) in others. In this case, the precedence = would go to the "wrong" description. >=20 > I agree that the example above is somehow weak (I just invented it = "on the spot"), but I hope that spirit is clear: it could happen that = despite of all attention we could pay, some subtle but important bugs = could find their way to the C code. Also I agree that we can always = publish an errata or an RFC that obsolete the old one. >=20 > Another doubt about having a C code as a normative reference: could = portability issues make the code "ambiguous"? For example, a code could = use the assumption that int is _exactly_ 32 bits long and short is = _exactly_ 16 bits. If the same code is run on a processor that uses, = say, 64 bits for int and 32 for shorts, maybe the results could be = different. Of course, you can solve this _specific_ issue by including = stdint.h (not available everywhere, though) and using types like = int16_t, but what I want to point out is that we must pay attention to = this type of portability issues, if we want the meaning of the C code = being non-ambiguous. Another example, much more subtle, could be >=20 > int foo[3]; > int goo, foo=3D3; > int *pt; >=20 > pt =3D (int*) foo; > goo =3D (int) pt; /* Is really pt=3D=3Dfoo? */ >=20 > As far as I know (but I am not a C language-lawyer), the standard = grants that conversion pointer-to-integer-to-pointer will give the = original pointer back, but nothing is said about the case above (imagine = an architecture where addresses can be only even). Although the code = above would work on a typical x86 architecture, I would not use it in a = code that should play the role of a reference. >=20 > By the way, I agree that the code above is quite silly, but I seen = things like this used, for example, in multi-thread applications where = the programmer wanted to pass an integer by using a function that = expected a pointer. >=20 > What about using some other language with less portability issues, = such as Ada or Java? Better yet, some language with a semantic formally = defined? (I do not know any, but maybe someone does) >=20 > [Oh, by the way, I *do not want* to start a flame war about which = language to use. Rather than fighting over it, I would go with C. I = just wanted to share with you some doubts of mine.] >=20 > Although I agree that errors and ambiguities could creep in the = mathematical description, maybe it is easier to write a correct and = unambiguous mathematical description rather than a program code (in = whatever language is written). >=20 >=20 >=20 >> Cheers, >>=20 >> Jean-Marc >>=20 >> On 10-09-24 06:06 PM, Roni Even wrote: >>> 6.5 Description of the codec >>> The description of the coding algorithm of this Recommendation is = made in >>> terms of bit-exact >>> fixed-point mathematical operations. The ANSI-C code indicated in = clause 10, >>> which constitutes an >>> integral part of this Recommendation, reflects this bit-exact, = fixed-point >>> descriptive approach. The >>> mathematical descriptions of the encoder and decoder can be = implemented in >>> other fashions, >>> possibly leading to a codec implementation which does not comply = with this >>> Recommendation. >>> Therefore, the algorithm description of the ANSI-C code of clause 10 = shall >>> take precedence over the >>> mathematical descriptions whenever discrepancies are found. A set of = test >>> signals, which can be >>> used together with the ANSI-C code in order to verify bit-exactness, = is >>> available as an electronic >>> attachment to this Recommendation. >>>=20 >>>=20 >>> Roni Even >>>=20 >>>=20 >>>=20 >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >>=20 >=20 >=20 >=20 > --=20 > Riccardo Bernardini > DIEGM -- University of Udine > via delle Scienze 208 > 33100 Udine > Tel: +39-0432-55-8271 > Fax: +39-0432-55-8251 >=20 > ---------------------------------------------------------------------- > SEMEL (SErvizio di Messaging ELettronico) - CSIT -Universita' di Udine >=20 > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From stephen.botzko@gmail.com Fri Oct 1 12:08:16 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 0F4DE3A6D97 for ; Fri, 1 Oct 2010 12:08:16 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.664 X-Spam-Level: X-Spam-Status: No, score=-2.664 tagged_above=-999 required=5 tests=[AWL=-0.066, BAYES_00=-2.599, HTML_MESSAGE=0.001] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id UnmOnmC4aSbn for ; Fri, 1 Oct 2010 12:08:14 -0700 (PDT) Received: from mail-qy0-f172.google.com (mail-qy0-f172.google.com [209.85.216.172]) by core3.amsl.com (Postfix) with ESMTP id E4ED33A6D51 for ; Fri, 1 Oct 2010 12:08:13 -0700 (PDT) Received: by qyk7 with SMTP id 7so1178640qyk.10 for ; Fri, 01 Oct 2010 12:09:02 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type; bh=qyQApLOXtyQVy7JTusCL6yst4sTsdYsCUePBCP9EDn0=; b=PcQWvzdSpOEqvSJ0fbZAnpiqS1I7Ynm64QlX8A+R7fr87CKKDSdvpefSsL+x22FIOF sSZUukAjMcm78Z5j9RwgBG6NfCQlXpkQ+c7Ne3yalph3Ko27pEcg+smAwyHdZhoLxM+g sywrzo2JdIH5QwJiP5hE6RJYwudKMLAAYvmXg= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=fxnwEUmErPTFU0iF/bEAu6N6vPbB2dT6wJlBRiU1hNfLdOjhY8lYCHnt/oSoexVgEh VcoRaw6MMyHHJ8g7IdTLY7cMbeQbOfAK299Jddkl/r5EWNmfERHTEFuKPm6TFpvM90M8 RH56nN/OJTXg8IR8VBh+NAEb1kvq44j2NCAfY= MIME-Version: 1.0 Received: by 10.229.37.70 with SMTP id w6mr4091855qcd.193.1285960141538; Fri, 01 Oct 2010 12:09:01 -0700 (PDT) Received: by 10.229.63.14 with HTTP; Fri, 1 Oct 2010 12:09:01 -0700 (PDT) In-Reply-To: <8F3BC3D2-E3CB-49FE-A31C-1B4BDE88142E@cisco.com> References: <4C9D0C24.5080302@usherbrooke.ca> <4c9d216e.1021cc0a.658d.51bf@mx.google.com> <4C9D3288.1000608@usherbrooke.ca> <20100925105523.15884gktkx78i41n@webmail.uniud.it> <8F3BC3D2-E3CB-49FE-A31C-1B4BDE88142E@cisco.com> Date: Fri, 1 Oct 2010 15:09:01 -0400 Message-ID: From: Stephen Botzko To: Cullen Jennings Content-Type: multipart/alternative; boundary=001636426b2d6e973e049192ed3e Cc: codec@ietf.org Subject: Re: [codec] Format for the codec specification X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 01 Oct 2010 19:08:16 -0000 --001636426b2d6e973e049192ed3e Content-Type: text/plain; charset=ISO-8859-1 I agree. Also, as a practical matter it is quite difficult to write the mathematical description in an RFC, since you are restricted to plain ASCII text. Stephen Botzko On Fri, Oct 1, 2010 at 1:37 PM, Cullen Jennings wrote: > > These are all valid concerns, but I think people have often found that the > providing code path is the "least bad way" of specifying codecs. I did want > to point out that bugs are found in specs all the time regardless if they > are C code, english text, or something else. We have an errata process for > updating RFCs as well as a ways to publish new versions of RFCs. I think > that is about the best we can do. It's also easy to accidentally write > ambiguous text like your C example below. Again, this happens in both > english and C - we just need to do our best to keep an eye open for those > and fix them. That's one of the advantages of having "running code" - it > helps discover the corner cases that are under specified. The code tends to > be "Reference" in that it is written to help illustrate the algorithm and is > often not the most optimal way to implement the algorithm, may not be thread > safe, may need changes to be ported to some platforms and so on. C may not > be the best for this b > ut it has the advantage of behind widely understood - particularly in DSP > codec circles. We could have a huge argument about what language for a long > time. I have seen that argument before and it's not fun. In the end, the > odds we would choose C are extremely high. > > Cullen > > > On Sep 25, 2010, at 2:55 AM, Riccardo Bernardini wrote: > > > Quoting Jean-Marc Valin : > > > >> Hi Roni, > >> > >> Thanks for the correction on the case of G.719. Despite that, my point > still stands: there *are* codecs for which the decoder is standardized > without a bit-exact definition. As long as we define the amount of deviation > permitted, then there is no problem. > >> > >> Oh, and I do agree that the C code should take precedence over the > mathematical description. > > > > A question just came to my mind: what about bugs in the C code? Suppose, > for example, that the algorithm requires the computation of a DCT of a block > of samples and that there is some typo in the constants used in the C code, > so that what the C code computes is not the DCT, but something similar and > the difference is such that it really does not matter in most cases, but > causes very bad performances (when compared with theoretical description) in > others. In this case, the precedence would go to the "wrong" description. > > > > I agree that the example above is somehow weak (I just invented it "on > the spot"), but I hope that spirit is clear: it could happen that despite of > all attention we could pay, some subtle but important bugs could find their > way to the C code. Also I agree that we can always publish an errata or an > RFC that obsolete the old one. > > > > Another doubt about having a C code as a normative reference: could > portability issues make the code "ambiguous"? For example, a code could use > the assumption that int is _exactly_ 32 bits long and short is _exactly_ 16 > bits. If the same code is run on a processor that uses, say, 64 bits for > int and 32 for shorts, maybe the results could be different. Of course, you > can solve this _specific_ issue by including stdint.h (not available > everywhere, though) and using types like int16_t, but what I want to point > out is that we must pay attention to this type of portability issues, if we > want the meaning of the C code being non-ambiguous. Another example, much > more subtle, could be > > > > int foo[3]; > > int goo, foo=3; > > int *pt; > > > > pt = (int*) foo; > > goo = (int) pt; /* Is really pt==foo? */ > > > > As far as I know (but I am not a C language-lawyer), the standard grants > that conversion pointer-to-integer-to-pointer will give the original pointer > back, but nothing is said about the case above (imagine an architecture > where addresses can be only even). Although the code above would work on a > typical x86 architecture, I would not use it in a code that should play the > role of a reference. > > > > By the way, I agree that the code above is quite silly, but I seen things > like this used, for example, in multi-thread applications where the > programmer wanted to pass an integer by using a function that expected a > pointer. > > > > What about using some other language with less portability issues, such > as Ada or Java? Better yet, some language with a semantic formally defined? > (I do not know any, but maybe someone does) > > > > [Oh, by the way, I *do not want* to start a flame war about which > language to use. Rather than fighting over it, I would go with C. I just > wanted to share with you some doubts of mine.] > > > > Although I agree that errors and ambiguities could creep in the > mathematical description, maybe it is easier to write a correct and > unambiguous mathematical description rather than a program code (in whatever > language is written). > > > > > > > >> Cheers, > >> > >> Jean-Marc > >> > >> On 10-09-24 06:06 PM, Roni Even wrote: > >>> 6.5 Description of the codec > >>> The description of the coding algorithm of this Recommendation is made > in > >>> terms of bit-exact > >>> fixed-point mathematical operations. The ANSI-C code indicated in > clause 10, > >>> which constitutes an > >>> integral part of this Recommendation, reflects this bit-exact, > fixed-point > >>> descriptive approach. The > >>> mathematical descriptions of the encoder and decoder can be implemented > in > >>> other fashions, > >>> possibly leading to a codec implementation which does not comply with > this > >>> Recommendation. > >>> Therefore, the algorithm description of the ANSI-C code of clause 10 > shall > >>> take precedence over the > >>> mathematical descriptions whenever discrepancies are found. A set of > test > >>> signals, which can be > >>> used together with the ANSI-C code in order to verify bit-exactness, is > >>> available as an electronic > >>> attachment to this Recommendation. > >>> > >>> > >>> Roni Even > >>> > >>> > >>> > >> _______________________________________________ > >> codec mailing list > >> codec@ietf.org > >> https://www.ietf.org/mailman/listinfo/codec > >> > > > > > > > > -- > > Riccardo Bernardini > > DIEGM -- University of Udine > > via delle Scienze 208 > > 33100 Udine > > Tel: +39-0432-55-8271 > > Fax: +39-0432-55-8251 > > > > ---------------------------------------------------------------------- > > SEMEL (SErvizio di Messaging ELettronico) - CSIT -Universita' di Udine > > > > _______________________________________________ > > codec mailing list > > codec@ietf.org > > https://www.ietf.org/mailman/listinfo/codec > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > --001636426b2d6e973e049192ed3e Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable I agree.=A0 Also, as a practical matter it is quite difficult to write the = mathematical description in an RFC, since you are restricted to plain ASCII= text.

Stephen Botzko

On Fri, Oct = 1, 2010 at 1:37 PM, Cullen Jennings <fluffy@cisco.com> wrote:

These are all valid concerns, but I think people have often found that the = providing code path is the "least bad way" of specifying codecs. = =A0I did want to point out that bugs are found in specs all the time regard= less if they are C code, english text, or something else. We have an errata= process for updating RFCs as well as a ways to publish new versions of RFC= s. I think that is about the best we can do. It's also easy to accident= ally write ambiguous text like your C example below. Again, this happens in= both english and C - we just need to do our best to keep an eye open for t= hose and fix them. That's one of the advantages of having "running= code" - it helps discover the corner cases that are under specified. = The code tends to be "Reference" in that it is written to help il= lustrate the algorithm and is often not the most optimal way to implement t= he algorithm, may not be thread safe, may need changes to be ported to some= platforms and so on. =A0C may not be the best for this b
=A0ut it has the advantage of behind widely understood - particularly in DS= P codec circles. We could have a huge argument about what language for a lo= ng time. I have seen that argument before and it's not fun. In the end,= the odds we would choose C are extremely high.

Cullen


On Sep 25, 2010, at 2:55 AM, Riccardo Bernardini wrote:

> Quoting Jean-Marc Valin <jean-marc.valin@usherbrooke.ca>:
>
>> Hi Roni,
>>
>> Thanks for the correction on the case of G.719. Despite that, my p= oint still stands: there *are* codecs for which the decoder is standardized= without a bit-exact definition. As long as we define the amount of deviati= on permitted, then there is no problem.
>>
>> Oh, and I do agree that the C code should take precedence over the= mathematical description.
>
> A question just came to my mind: what about bugs in the C code? =A0Sup= pose, for example, that the algorithm requires the computation of a DCT of = a block of samples and that there is some typo in the constants used in the= C code, so that what the C code computes is not the DCT, but something sim= ilar and the difference is such that it really does not matter in most case= s, but causes very bad performances (when compared with theoretical descrip= tion) in others. =A0In this case, the precedence would go to the "wron= g" description.
>
> I agree that the example above is somehow =A0weak (I just invented it = "on the spot"), but I hope that spirit is clear: it could happen = that despite of all attention we could pay, some subtle but important bugs = could find their way to the C code. =A0Also I agree that we can always publ= ish an errata or an RFC that obsolete the old one.
>
> Another doubt about having a C code as a normative reference: could po= rtability issues make the code "ambiguous"? =A0For example, a cod= e could use the assumption that int is _exactly_ 32 bits long and short is = _exactly_ 16 bits. =A0If the same code is run on a processor that uses, say= , 64 bits for int and 32 for shorts, maybe the results could be different. = =A0Of course, you can solve this _specific_ issue by including stdint.h (no= t available everywhere, though) and using types like int16_t, but what I wa= nt to point out is that we must pay attention to this type of portability i= ssues, if we want the meaning of the C code being non-ambiguous. =A0Another= example, much more subtle, could be
>
> =A0int foo[3];
> =A0int goo, foo=3D3;
> =A0int *pt;
>
> =A0pt =A0=3D (int*) foo;
> =A0goo =3D (int) pt; =A0/* Is really pt=3D=3Dfoo? */
>
> As far as I know (but I am not a C language-lawyer), the standard gran= ts that conversion pointer-to-integer-to-pointer will give the original poi= nter back, but nothing is said about the case above (imagine an architectur= e where addresses can be only even). =A0Although the code above would work = on a typical x86 architecture, I would not use it in a code that should pla= y the role of a reference.
>
> By the way, I agree that the code above is quite silly, but I seen thi= ngs like this used, for example, in multi-thread applications where the pro= grammer wanted to pass an integer by using a function that expected a point= er.
>
> What about using some other language with less portability issues, suc= h as Ada or Java? =A0Better yet, some language with a semantic formally def= ined? (I do not know any, but maybe someone does)
>
> [Oh, by the way, I *do not want* to start a flame war about which lang= uage to use. =A0Rather than fighting over it, I would go with C. =A0I just = wanted to share with you some doubts of mine.]
>
> Although I agree that errors and ambiguities could creep in the mathem= atical description, maybe it is easier to write a correct and unambiguous m= athematical description rather than a program code (in whatever language is= written).
>
>
>
>> Cheers,
>>
>> =A0 =A0 =A0Jean-Marc
>>
>> On 10-09-24 06:06 PM, Roni Even wrote:
>>> 6.5 Description of the codec
>>> The description of the coding algorithm of this Recommendation= is made in
>>> terms of bit-exact
>>> fixed-point mathematical operations. The ANSI-C code indicated= in clause 10,
>>> which constitutes an
>>> integral part of this Recommendation, reflects this bit-exact,= fixed-point
>>> descriptive approach. The
>>> mathematical descriptions of the encoder and decoder can be im= plemented in
>>> other fashions,
>>> possibly leading to a codec implementation which does not comp= ly with this
>>> Recommendation.
>>> Therefore, the algorithm description of the ANSI-C code of cla= use 10 shall
>>> take precedence over the
>>> mathematical descriptions whenever discrepancies are found. A = set of test
>>> signals, which can be
>>> used together with the ANSI-C code in order to verify bit-exac= tness, is
>>> available as an electronic
>>> attachment to this Recommendation.
>>>
>>>
>>> Roni Even
>>>
>>>
>>>
>> _______________________________________________
>> codec mailing list
>> codec@ietf.org
>> https://www.ietf.org/mailman/listinfo/codec
>>
>
>
>
> --
> Riccardo Bernardini
> DIEGM -- University of Udine
> via delle Scienze 208
> 33100 Udine
> Tel: +39-0432-55-8271
> Fax: +39-0432-55-8251
>
> ----------------------------------------------------------------------=
> SEMEL (SErvizio di Messaging ELettronico) - CSIT -Universita' di U= dine
>
> _______________________________________________
> codec mailing list
> codec@ietf.org
> https://www.ietf.org/mailman/listinfo/codec

_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec

--001636426b2d6e973e049192ed3e-- From fluffy@cisco.com Sat Oct 2 07:27:46 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id AF8553A6CD8 for ; Sat, 2 Oct 2010 07:27:46 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -110.514 X-Spam-Level: X-Spam-Status: No, score=-110.514 tagged_above=-999 required=5 tests=[AWL=0.085, BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id XKVUn09ltvt1 for ; Sat, 2 Oct 2010 07:27:45 -0700 (PDT) Received: from sj-iport-6.cisco.com (sj-iport-6.cisco.com [171.71.176.117]) by core3.amsl.com (Postfix) with ESMTP id D5BDE3A6C65 for ; Sat, 2 Oct 2010 07:27:45 -0700 (PDT) Authentication-Results: sj-iport-6.cisco.com; dkim=neutral (message not signed) header.i=none X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AgkGAIvepkyrR7H+/2dsb2JhbACUN44HcaVWm3KFRASEUYVrgwI X-IronPort-AV: E=Sophos;i="4.57,271,1283731200"; d="scan'208";a="598164472" Received: from sj-core-2.cisco.com ([171.71.177.254]) by sj-iport-6.cisco.com with ESMTP; 02 Oct 2010 14:28:36 +0000 Received: from [192.168.4.2] (rcdn-fluffy-8711.cisco.com [10.99.9.18]) by sj-core-2.cisco.com (8.13.8/8.14.3) with ESMTP id o92ESaae018178 for ; Sat, 2 Oct 2010 14:28:36 GMT Content-Type: text/plain; charset=us-ascii Mime-Version: 1.0 (Apple Message framework v1081) From: Cullen Jennings In-Reply-To: <1E7430DC-ED0D-44A3-826B-B1F344CB918E@cisco.com> Date: Sat, 2 Oct 2010 08:28:35 -0600 Content-Transfer-Encoding: quoted-printable Message-Id: <8BBCF750-0DDB-466F-811E-17AF1E1D5524@cisco.com> References: <1E7430DC-ED0D-44A3-826B-B1F344CB918E@cisco.com> To: codec@ietf.org X-Mailer: Apple Mail (2.1081) Subject: Re: [codec] Summary of codec specification - Code in RFC X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 02 Oct 2010 14:27:47 -0000 On Oct 1, 2010, at 11:27 AM, Cullen Jennings wrote: > 1) The code will be in the appendix of the draft base64 encoded. We = can sort out later if it is gziped tar file of whatever but it will be = some easy to extract format. This has been done on other documents. I = have talked to the AD about this before. I have not talked to RFC Ed but = do not foresee any issues there.=20 I'd like to encourage people interested in this point to look at = Appendix B of RFC 4474.=20 http://tools.ietf.org/rfc/rfc4474.txt From fluffy@cisco.com Sat Oct 2 13:21:12 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id EF76B3A6D24 for ; Sat, 2 Oct 2010 13:21:12 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -110.515 X-Spam-Level: X-Spam-Status: No, score=-110.515 tagged_above=-999 required=5 tests=[AWL=0.084, BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3QLCzV4NXo4c for ; Sat, 2 Oct 2010 13:21:07 -0700 (PDT) Received: from sj-iport-4.cisco.com (sj-iport-4.cisco.com [171.68.10.86]) by core3.amsl.com (Postfix) with ESMTP id 9A9A73A6DAA for ; Sat, 2 Oct 2010 13:21:07 -0700 (PDT) Authentication-Results: sj-iport-4.cisco.com; dkim=neutral (message not signed) header.i=none X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AgkGAEcxp0yrR7H+/2dsb2JhbACUOY4Hcadpm2GFRASEUYVrgwI X-IronPort-AV: E=Sophos;i="4.57,272,1283731200"; d="scan'208";a="195016896" Received: from sj-core-2.cisco.com ([171.71.177.254]) by sj-iport-4.cisco.com with ESMTP; 02 Oct 2010 20:21:59 +0000 Received: from [192.168.4.2] (rcdn-fluffy-8711.cisco.com [10.99.9.18]) by sj-core-2.cisco.com (8.13.8/8.14.3) with ESMTP id o92KLwDv010040 for ; Sat, 2 Oct 2010 20:21:58 GMT Content-Type: text/plain; charset=us-ascii Mime-Version: 1.0 (Apple Message framework v1081) From: Cullen Jennings In-Reply-To: <8BBCF750-0DDB-466F-811E-17AF1E1D5524@cisco.com> Date: Sat, 2 Oct 2010 14:21:57 -0600 Content-Transfer-Encoding: quoted-printable Message-Id: <070D11DD-137C-42F7-B5E0-21DB8704A4E1@cisco.com> References: <1E7430DC-ED0D-44A3-826B-B1F344CB918E@cisco.com> <8BBCF750-0DDB-466F-811E-17AF1E1D5524@cisco.com> To: codec@ietf.org X-Mailer: Apple Mail (2.1081) Subject: Re: [codec] Summary of codec specification - Code in RFC X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 02 Oct 2010 20:21:13 -0000 There been a bit of discussion suggesting that we should put into the = draft both the machine readable base 64 version in that is easy to = extract, and a version in human readable text that is easy for people to = read. This seems like a reasonable idea to me. Just to get a rough idea = of what this would look like, I took the current GIT repository and had = a look at the code in it. It' a bit over 50k lines including both = encoder and decoder. You get 48 lines of real text per page of RFC so = the human readable version would be over a 1000 pages. If you take the = code, tar it, gzip it, and base64 encode it, you get a bit over 180 = pages. This sounds big but given it's some bits on a web server, I don't = particularly see any problem with it. I also have no idea how the = repository compares to what would be needed to specify the normative = parts of the decoder. I may have including in the 50k lines of code all = kinds of stuff that was not needed.=20 Cullen On Oct 2, 2010, at 8:28 AM, Cullen Jennings wrote: >=20 > On Oct 1, 2010, at 11:27 AM, Cullen Jennings wrote: >=20 >> 1) The code will be in the appendix of the draft base64 encoded. We = can sort out later if it is gziped tar file of whatever but it will be = some easy to extract format. This has been done on other documents. I = have talked to the AD about this before. I have not talked to RFC Ed but = do not foresee any issues there.=20 >=20 >=20 > I'd like to encourage people interested in this point to look at = Appendix B of RFC 4474.=20 >=20 > http://tools.ietf.org/rfc/rfc4474.txt >=20 > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From stewe@stewe.org Sat Oct 2 15:32:01 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 927683A6D8B for ; Sat, 2 Oct 2010 15:32:01 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.655 X-Spam-Level: X-Spam-Status: No, score=-1.655 tagged_above=-999 required=5 tests=[AWL=0.944, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id XJCz-agpSDj3 for ; Sat, 2 Oct 2010 15:32:00 -0700 (PDT) Received: from stewe.org (stewe.org [85.214.122.234]) by core3.amsl.com (Postfix) with ESMTP id 989B83A6C73 for ; Sat, 2 Oct 2010 15:31:59 -0700 (PDT) Received: from [192.168.1.105] (unverified [24.5.132.232]) by stewe.org (SurgeMail 3.9e) with ESMTP id 803318-1743317 for multiple; Sun, 03 Oct 2010 00:32:48 +0200 User-Agent: Microsoft-Entourage/12.26.0.100708 Date: Sat, 02 Oct 2010 15:32:29 -0700 From: Stephan Wenger To: "fluffy@cisco.com" , Message-ID: Thread-Topic: [codec] Summary of codec specification - Code in RFC Thread-Index: ActigbIwpTZPpnVQuUuLt3zCCYQomg== In-Reply-To: <070D11DD-137C-42F7-B5E0-21DB8704A4E1@cisco.com> Mime-version: 1.0 Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit X-Originating-IP: 24.5.132.232 X-Authenticated-User: stewe@stewe.org X-ORBS-Stamp: Your IP (24.5.132.232) was found in the spamhaus database. http://www.spamhaus.net Subject: Re: [codec] Summary of codec specification - Code in RFC X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 02 Oct 2010 22:32:01 -0000 As said before, putting base64 code into an RFC makes sense to me. However, I would definitely put it into its own RFC (which would be standards track) and not into an appendix of the same document. The versioning problem seems to be less an issue than the trees that will fall if people "print all" the RFC in question. So please leave the textual description in another document. Logic suggests that this "human readable" document would be informational, as the code RFC prevails in case of errors. However, I don't mind being slightly illogical here and call it standards track as well. The document could also certain code pieces in such cases where a textual description in English with ASCII fixed width fonts may be to cumbersome. Someone mentioned the "errata" process for fixing errors. That process is not going to work overly well with an base64 encoded zip file... But I don't consider that a major problem--rather an incentive to get our stuff right the first time. Stephan On 10.2.2010 13:21 , "fluffy@cisco.com" wrote: > > There been a bit of discussion suggesting that we should put into the draft > both the machine readable base 64 version in that is easy to extract, and a > version in human readable text that is easy for people to read. This seems > like a reasonable idea to me. Just to get a rough idea of what this would look > like, I took the current GIT repository and had a look at the code in it. It' > a bit over 50k lines including both encoder and decoder. You get 48 lines of > real text per page of RFC so the human readable version would be over a 1000 > pages. If you take the code, tar it, gzip it, and base64 encode it, you get a > bit over 180 pages. This sounds big but given it's some bits on a web server, > I don't particularly see any problem with it. I also have no idea how the > repository compares to what would be needed to specify the normative parts of > the decoder. I may have including in the 50k lines of code all kinds of stuff > that was not needed. > > Cullen > > > On Oct 2, 2010, at 8:28 AM, Cullen Jennings wrote: > >> >> On Oct 1, 2010, at 11:27 AM, Cullen Jennings wrote: >> >>> 1) The code will be in the appendix of the draft base64 encoded. We can sort >>> out later if it is gziped tar file of whatever but it will be some easy to >>> extract format. This has been done on other documents. I have talked to the >>> AD about this before. I have not talked to RFC Ed but do not foresee any >>> issues there. >> >> >> I'd like to encourage people interested in this point to look at Appendix B >> of RFC 4474. >> >> http://tools.ietf.org/rfc/rfc4474.txt >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From jmh@joelhalpern.com Sat Oct 2 15:44:05 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id C53333A6BDD for ; Sat, 2 Oct 2010 15:44:05 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.374 X-Spam-Level: X-Spam-Status: No, score=-102.374 tagged_above=-999 required=5 tests=[AWL=0.225, BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RWOtRedKgWbD for ; Sat, 2 Oct 2010 15:44:04 -0700 (PDT) Received: from hgblob.mail.tigertech.net (hgblob.mail.tigertech.net [64.62.209.71]) by core3.amsl.com (Postfix) with ESMTP id 88D573A6B53 for ; Sat, 2 Oct 2010 15:44:04 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by hgblob.tigertech.net (Postfix) with ESMTP id 520C93236DB1; Sat, 2 Oct 2010 15:44:56 -0700 (PDT) X-Virus-Scanned: Debian amavisd-new at hgblob.tigertech.net Received: from [10.10.10.102] (pool-71-161-50-77.clppva.btas.verizon.net [71.161.50.77]) (using TLSv1 with cipher AES256-SHA (256/256 bits)) (No client certificate requested) by hgblob.tigertech.net (Postfix) with ESMTPSA id B98C73228449; Sat, 2 Oct 2010 15:44:50 -0700 (PDT) Message-ID: <4CA7B5E0.4000508@joelhalpern.com> Date: Sat, 02 Oct 2010 18:44:48 -0400 From: "Joel M. Halpern" User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.9) Gecko/20100915 Lightning/1.0b2 Thunderbird/3.1.4 MIME-Version: 1.0 To: Stephan Wenger References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: codec@ietf.org Subject: Re: [codec] Summary of codec specification - Code in RFC X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 02 Oct 2010 22:44:05 -0000 I have serious problems with a standards track RFC whose normative content can not be read by a human being without additional tools. I also have serious conceptual problems with an RFC whose normative content, in human readable form, is 1,000 pages. I am not objecting to publishing code in RFCs. That is something we do frequently. I am not even objecting to including base64 encoded gzip code, as informational appendices in an RFC. However, it seems to me that treating base64 gzip code as normative in an RFC is distinctly bending the rules for RFC publication. That text can not be editted by the RFC Editors. It can not be subject to erratta. And it can not be read as meaningful plain ascii according to the spirit of our publication rules. Yes, we have once in a great while published RFCs with tables on the order of this long. It was difficult, but in fact the content was carefully reviewed. Yours, Joel M. Halpern PS: If the code were a reasonable length, publishing the text of the code as normative, and the base64 gzip as informative would seem to me to be a very good way to have clear normative code in an RFC while making it easy to use the code. On 10/2/2010 6:32 PM, Stephan Wenger wrote: > As said before, putting base64 code into an RFC makes sense to me. However, > I would definitely put it into its own RFC (which would be standards track) > and not into an appendix of the same document. The versioning problem seems > to be less an issue than the trees that will fall if people "print all" the > RFC in question. So please leave the textual description in another > document. > > Logic suggests that this "human readable" document would be informational, > as the code RFC prevails in case of errors. However, I don't mind being > slightly illogical here and call it standards track as well. The document > could also certain code pieces in such cases where a textual description in > English with ASCII fixed width fonts may be to cumbersome. > > Someone mentioned the "errata" process for fixing errors. That process is > not going to work overly well with an base64 encoded zip file... But I don't > consider that a major problem--rather an incentive to get our stuff right > the first time. > > Stephan > > > > On 10.2.2010 13:21 , "fluffy@cisco.com" wrote: > >> >> There been a bit of discussion suggesting that we should put into the draft >> both the machine readable base 64 version in that is easy to extract, and a >> version in human readable text that is easy for people to read. This seems >> like a reasonable idea to me. Just to get a rough idea of what this would look >> like, I took the current GIT repository and had a look at the code in it. It' >> a bit over 50k lines including both encoder and decoder. You get 48 lines of >> real text per page of RFC so the human readable version would be over a 1000 >> pages. If you take the code, tar it, gzip it, and base64 encode it, you get a >> bit over 180 pages. This sounds big but given it's some bits on a web server, >> I don't particularly see any problem with it. I also have no idea how the >> repository compares to what would be needed to specify the normative parts of >> the decoder. I may have including in the 50k lines of code all kinds of stuff >> that was not needed. >> >> Cullen >> >> >> On Oct 2, 2010, at 8:28 AM, Cullen Jennings wrote: >> >>> >>> On Oct 1, 2010, at 11:27 AM, Cullen Jennings wrote: >>> >>>> 1) The code will be in the appendix of the draft base64 encoded. We can sort >>>> out later if it is gziped tar file of whatever but it will be some easy to >>>> extract format. This has been done on other documents. I have talked to the >>>> AD about this before. I have not talked to RFC Ed but do not foresee any >>>> issues there. >>> >>> >>> I'd like to encourage people interested in this point to look at Appendix B >>> of RFC 4474. >>> >>> http://tools.ietf.org/rfc/rfc4474.txt >>> >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > From lars.eggert@nokia.com Tue Oct 5 02:48:59 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 94B863A6F14 for ; Tue, 5 Oct 2010 02:48:59 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -103.213 X-Spam-Level: X-Spam-Status: No, score=-103.213 tagged_above=-999 required=5 tests=[AWL=-0.614, BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id AUbQIJ0EVWaC for ; Tue, 5 Oct 2010 02:48:58 -0700 (PDT) Received: from mgw-sa01.nokia.com (smtp.nokia.com [147.243.1.47]) by core3.amsl.com (Postfix) with ESMTP id 3D35F3A6F08 for ; Tue, 5 Oct 2010 02:48:58 -0700 (PDT) Received: from mail.fit.nokia.com (esdhcp030222.research.nokia.com [172.21.30.222]) by mgw-sa01.nokia.com (Switch-3.4.3/Switch-3.4.3) with ESMTP id o959npQG018056 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Tue, 5 Oct 2010 12:49:51 +0300 X-Virus-Status: Clean X-Virus-Scanned: clamav-milter 0.96.2 at fit.nokia.com Mime-Version: 1.0 (Apple Message framework v1081) Content-Type: multipart/signed; boundary=Apple-Mail-136-874821813; protocol="application/pkcs7-signature"; micalg=sha1 From: Lars Eggert In-Reply-To: Date: Tue, 5 Oct 2010 11:49:43 +0200 Message-Id: References: To: Stephan Wenger X-Mailer: Apple Mail (2.1081) X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.2.6 (mail.fit.nokia.com); Tue, 05 Oct 2010 12:49:45 +0300 (EEST) X-Nokia-AV: Clean Cc: "codec@ietf.org" Subject: Re: [codec] Summary of codec specification - Code in RFC X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 05 Oct 2010 09:48:59 -0000 --Apple-Mail-136-874821813 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=us-ascii On 2010-10-3, at 0:32, Stephan Wenger wrote: > As said before, putting base64 code into an RFC makes sense to me. = However, > I would definitely put it into its own RFC (which would be standards = track) > and not into an appendix of the same document. The versioning problem = seems > to be less an issue than the trees that will fall if people "print = all" the > RFC in question. So please leave the textual description in another > document. =20 FWIW, this was the approach we followed with the XDR code for NFS 4.1: = RFC 5661 has the spec text and RFC 5662 has the XDR (in this case, as = plain text). Lars= --Apple-Mail-136-874821813 Content-Disposition: attachment; filename=smime.p7s Content-Type: application/pkcs7-signature; name=smime.p7s Content-Transfer-Encoding: base64 MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIKHjCCBMww ggQ1oAMCAQICEByunWua9OYvIoqj2nRhbB4wDQYJKoZIhvcNAQEFBQAwXzELMAkGA1UEBhMCVVMx FzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMTcwNQYDVQQLEy5DbGFzcyAxIFB1YmxpYyBQcmltYXJ5 IENlcnRpZmljYXRpb24gQXV0aG9yaXR5MB4XDTA1MTAyODAwMDAwMFoXDTE1MTAyNzIzNTk1OVow gd0xCzAJBgNVBAYTAlVTMRcwFQYDVQQKEw5WZXJpU2lnbiwgSW5jLjEfMB0GA1UECxMWVmVyaVNp Z24gVHJ1c3QgTmV0d29yazE7MDkGA1UECxMyVGVybXMgb2YgdXNlIGF0IGh0dHBzOi8vd3d3LnZl cmlzaWduLmNvbS9ycGEgKGMpMDUxHjAcBgNVBAsTFVBlcnNvbmEgTm90IFZhbGlkYXRlZDE3MDUG A1UEAxMuVmVyaVNpZ24gQ2xhc3MgMSBJbmRpdmlkdWFsIFN1YnNjcmliZXIgQ0EgLSBHMjCCASIw DQYJKoZIhvcNAQEBBQADggEPADCCAQoCggEBAMnfrOfq+PgDFMQAktXBfjbCPO98chXLwKuMPRyV zm8eECw/AO2XJua2x+atQx0/pIdHR0w+VPhs+Mf8sZ69MHC8l7EDBeqV8a1AxUR6SwWi8mD81zpl Yu//EHuiVrvFTnAt1qIfPO2wQuhejVchrKaZ2RHp0hoHwHRHQgv8xTTq/ea6JNEdCBU3otdzzwFB L2OyOj++pRpu9MlKWz2VphW7NQIZ+dTvvI8OcXZZu0u2Ptb8Whb01g6J8kn+bAztFenZiHWcec5g J925rXXOL3OVekA6hXVJsLjfaLyrzROChRFQo+A8C67AClPN1zBvhTJGG+RJEMJs4q8fef/btLUC AwEAAaOCAYQwggGAMBIGA1UdEwEB/wQIMAYBAf8CAQAwRAYDVR0gBD0wOzA5BgtghkgBhvhFAQcX ATAqMCgGCCsGAQUFBwIBFhxodHRwczovL3d3dy52ZXJpc2lnbi5jb20vcnBhMAsGA1UdDwQEAwIB BjARBglghkgBhvhCAQEEBAMCAQYwLgYDVR0RBCcwJaQjMCExHzAdBgNVBAMTFlByaXZhdGVMYWJl bDMtMjA0OC0xNTUwHQYDVR0OBBYEFBF9Xhl9PATfamzWoooaPzHYO5RSMDEGA1UdHwQqMCgwJqAk oCKGIGh0dHA6Ly9jcmwudmVyaXNpZ24uY29tL3BjYTEuY3JsMIGBBgNVHSMEejB4oWOkYTBfMQsw CQYDVQQGEwJVUzEXMBUGA1UEChMOVmVyaVNpZ24sIEluYy4xNzA1BgNVBAsTLkNsYXNzIDEgUHVi bGljIFByaW1hcnkgQ2VydGlmaWNhdGlvbiBBdXRob3JpdHmCEQDNun9W8N/kvFT+IqyzcqpVMA0G CSqGSIb3DQEBBQUAA4GBALEv2ZbhkqLugWDlyCog++FnLNYAmFOjAhvpkEv4GESfD0b3+qD+0x0Y o9K/HOzWGZ9KTUP4yru+E4BJBd0hczNXwkJavvoAk7LmBDGRTl088HMFN2Prv4NZmP1m3umGMpqS KTw6rlTaphJRsY/IytNHeObbpR6HBuPRFMDCIfa6MIIFSjCCBDKgAwIBAgIQM/3DDmRp5mxHvrbX tRvSsDANBgkqhkiG9w0BAQUFADCB3TELMAkGA1UEBhMCVVMxFzAVBgNVBAoTDlZlcmlTaWduLCBJ bmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMTswOQYDVQQLEzJUZXJtcyBvZiB1 c2UgYXQgaHR0cHM6Ly93d3cudmVyaXNpZ24uY29tL3JwYSAoYykwNTEeMBwGA1UECxMVUGVyc29u YSBOb3QgVmFsaWRhdGVkMTcwNQYDVQQDEy5WZXJpU2lnbiBDbGFzcyAxIEluZGl2aWR1YWwgU3Vi c2NyaWJlciBDQSAtIEcyMB4XDTA5MTAxNDAwMDAwMFoXDTEwMTAxNDIzNTk1OVowggETMRcwFQYD VQQKEw5WZXJpU2lnbiwgSW5jLjEfMB0GA1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29yazFGMEQG A1UECxM9d3d3LnZlcmlzaWduLmNvbS9yZXBvc2l0b3J5L1JQQSBJbmNvcnAuIGJ5IFJlZi4sTElB Qi5MVEQoYyk5ODEeMBwGA1UECxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTMwMQYDVQQLEypEaWdp dGFsIElEIENsYXNzIDEgLSBOZXRzY2FwZSBGdWxsIFNlcnZpY2UxFDASBgNVBAMUC0xhcnMgRWdn ZXJ0MSQwIgYJKoZIhvcNAQkBFhVsYXJzLmVnZ2VydEBub2tpYS5jb20wggEiMA0GCSqGSIb3DQEB AQUAA4IBDwAwggEKAoIBAQCrTp9sa04QO3gKan1sQGFAZ91BKjFdECaw29ZXHYOZSFW7JC/oyuAO x7pucJp9qDLYlv60I8zLJ1eocDn3bsjBWwAS65GGVT9NLiblzCd5DeWTkn9UD7OMvLL+Z1Z17sE0 jlmluQSCLjv0P7d6ZCpwVD2hWvuZu37fIQresIEYQLbZHXogwoCMhqwo8k8/WuwDbNcB9RiLq3JW FRjZILMmf5qFdExz1AZyG3OdFXe1F0vcxfHhbrGU5cVklV9XI3TfxVZciFNqnTuWsaHFxU0ZrwI0 njVtpLko1TW1vC5qS2ZIpWq7iDkIke+XL4IYTHlOdH7sM502qEE8JqadXjuTAgMBAAGjgcwwgckw CQYDVR0TBAIwADBEBgNVHSAEPTA7MDkGC2CGSAGG+EUBBxcBMCowKAYIKwYBBQUHAgEWHGh0dHBz Oi8vd3d3LnZlcmlzaWduLmNvbS9ycGEwCwYDVR0PBAQDAgWgMB0GA1UdJQQWMBQGCCsGAQUFBwME BggrBgEFBQcDAjBKBgNVHR8EQzBBMD+gPaA7hjlodHRwOi8vSW5kQzFEaWdpdGFsSUQtY3JsLnZl cmlzaWduLmNvbS9JbmRDMURpZ2l0YWxJRC5jcmwwDQYJKoZIhvcNAQEFBQADggEBAJraU8e1Ss8a Uj33lweT8ibOcmNLTNn2pnSi7p5BwNCFtrLNp9AdhcGYJfEQMOlanxc7dpQ7OWUbpub4jAYKyP47 hPQayKMJ73KcKmiN4Czj9g/uCTRwW3kiyhTo22Yl2MWGTM7WI39xTEQo4DOGs9EK6Ldb9zZU4oRl ZbaD1G/Lo33LYmuI8MfCLzTBtmT5MybyEnK17457+8bMLXoSFJfFGk5LPCn38SWdiAG/YH2A4had 3wvpd9Mpx38fxw/RZqlifRSTT8zlOwkq/u/+il9D4o2ruwMEMC1ZwPN3/w/u+QCw4o78wm24BOwU KyA6H4H1r9BoUZ1To1zXxwvuI+sxggSLMIIEhwIBATCB8jCB3TELMAkGA1UEBhMCVVMxFzAVBgNV BAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMTswOQYD VQQLEzJUZXJtcyBvZiB1c2UgYXQgaHR0cHM6Ly93d3cudmVyaXNpZ24uY29tL3JwYSAoYykwNTEe MBwGA1UECxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTcwNQYDVQQDEy5WZXJpU2lnbiBDbGFzcyAx IEluZGl2aWR1YWwgU3Vic2NyaWJlciBDQSAtIEcyAhAz/cMOZGnmbEe+tte1G9KwMAkGBSsOAwIa BQCgggJtMBgGCSqGSIb3DQEJAzELBgkqhkiG9w0BBwEwHAYJKoZIhvcNAQkFMQ8XDTEwMTAwNTA5 NDk0M1owIwYJKoZIhvcNAQkEMRYEFFbd0cSlPsMlxL//cV1JhfKFodDtMIIBAwYJKwYBBAGCNxAE MYH1MIHyMIHdMQswCQYDVQQGEwJVUzEXMBUGA1UEChMOVmVyaVNpZ24sIEluYy4xHzAdBgNVBAsT FlZlcmlTaWduIFRydXN0IE5ldHdvcmsxOzA5BgNVBAsTMlRlcm1zIG9mIHVzZSBhdCBodHRwczov L3d3dy52ZXJpc2lnbi5jb20vcnBhIChjKTA1MR4wHAYDVQQLExVQZXJzb25hIE5vdCBWYWxpZGF0 ZWQxNzA1BgNVBAMTLlZlcmlTaWduIENsYXNzIDEgSW5kaXZpZHVhbCBTdWJzY3JpYmVyIENBIC0g RzICEDP9ww5kaeZsR76217Ub0rAwggEFBgsqhkiG9w0BCRACCzGB9aCB8jCB3TELMAkGA1UEBhMC VVMxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3 b3JrMTswOQYDVQQLEzJUZXJtcyBvZiB1c2UgYXQgaHR0cHM6Ly93d3cudmVyaXNpZ24uY29tL3Jw YSAoYykwNTEeMBwGA1UECxMVUGVyc29uYSBOb3QgVmFsaWRhdGVkMTcwNQYDVQQDEy5WZXJpU2ln biBDbGFzcyAxIEluZGl2aWR1YWwgU3Vic2NyaWJlciBDQSAtIEcyAhAz/cMOZGnmbEe+tte1G9Kw MA0GCSqGSIb3DQEBAQUABIIBABmP0ape4sH+pssKfmp8rZ29qnH7jxiElTFRomtQdcMbSPgB52Bi rkr1zRmpLjTex849xkiHUeJA/OHmnM4VEfLrIHd2fU7gJN45NDSKnmGhdT1HO6lHx58hd3DjfHWa hFfvZU21bGKtZkedBo3ja9XtwRBlL+lhvqgKhjOCrRpQVSEr+TemIicb7DpDol22BqxVvz6xiwCr agDxmsoUA22LbyiDUmTaq08ZY4rD0cnoSKqSI+llNIxolteS2rsftBPjWM/JWfo7ja4yogja/39Q k/yYo4dGVnSpXO6oCFJ74WOAE/pK/fksFFGlRTGZSJGsI64A1RmVyrJOGrktVikAAAAAAAA= --Apple-Mail-136-874821813-- From erik.norvell@ericsson.com Fri Oct 8 04:40:16 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 426863A6870 for ; Fri, 8 Oct 2010 04:40:14 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.67 X-Spam-Level: X-Spam-Status: No, score=-3.67 tagged_above=-999 required=5 tests=[AWL=2.929, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id zH+GlpkNZysT for ; Fri, 8 Oct 2010 04:39:40 -0700 (PDT) Received: from mailgw9.se.ericsson.net (mailgw9.se.ericsson.net [193.180.251.57]) by core3.amsl.com (Postfix) with ESMTP id D8BF83A6889 for ; Fri, 8 Oct 2010 04:39:39 -0700 (PDT) X-AuditID: c1b4fb39-b7c6dae000006ad7-61-4caf033b4d80 Received: from esessmw0237.eemea.ericsson.se (Unknown_Domain [153.88.253.124]) by mailgw9.se.ericsson.net (Symantec Mail Security) with SMTP id 8E.21.27351.B330FAC4; Fri, 8 Oct 2010 13:40:43 +0200 (CEST) Received: from ESESSCMS0351.eemea.ericsson.se ([169.254.1.175]) by esessmw0237.eemea.ericsson.se ([153.88.115.90]) with mapi; Fri, 8 Oct 2010 13:40:43 +0200 From: Erik Norvell To: "codec@ietf.org" Date: Fri, 8 Oct 2010 13:40:42 +0200 Thread-Topic: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item Thread-Index: ActcFoPbwa6TwrS6SiGNzLa8Hji6hwKxun3w Message-ID: <027A93CE4A670242BD91A44E37105AEF0CACA09122@ESESSCMS0351.eemea.ericsson.se> References: <4C9CEE29.1090600@jdrosen.net> In-Reply-To: <4C9CEE29.1090600@jdrosen.net> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Brightmail-Tracker: AAAAAA== Subject: Re: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 08 Oct 2010 11:40:16 -0000 Hi all, Reading the codec guidelines document it still includes arguable and unnece= ssary generic statements on the problems associated with encumbered technol= ogy. The present guidelines do not at all safeguard that the codec can be i= mplemented by anybody as freely as was the intention when the WG was establ= ished. Royalty-free conditions do not at all mean free implementation. The charter implies that the goal of the WG is to produce a codec without a= ny usage restrictions. http://tools.ietf.org/wg/codec/charters =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D There exist codecs that are standardized, but that cannot be widely implemented and easily distributed; according to reports, the presence of various usage restrictions (e.g., in the form of requirements to pay royalty fees, obtain a license, enter into a business agreement, or meet other special conditions imposed by a patent holder) has hindered adoptions of such codecs in interactive Internet applications. =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D Considering these exceptional intentions with the codec WG, the guidelines = should also be exceptional compared with the customary IETF guidelines. Pre= ferably, the guidelines should clearly state that the intention is that the= codec should be freely available to any implementer. The IETF Patent Discl= osure and Licensing Declaration template, part VI Licensing Declaration, ha= s an option a) reading "No Licenses required for Implementers" (http://tool= s.ietf.org/html/rfc3905). I suggest the guidelines should state that all co= ntributors should comply with this option. As a further clarification of th= is option a) to meet the intentions behind the establishment of the WG the = guidelines should expressly request that "no implementer shall be obliged t= o make a compensation of any kind to the holder of any patent covering the = codec or be obliged to make any undertaking towards to the holder of any pa= tent covering the codec". Admittedly the suggestion really means an exceptional requirement on the pa= tent holders which would not at all be adequate in general for all IETF WGs= and would provide an exceptional freedom for the implementers, but conside= ring the exceptional intentions with the WG I believe exceptional requireme= nts can be justified. If not, the codec may end up adding to the list of en= cumbered codecs which the WG set out to avoid. Best regards, Erik=20 -----Original Message----- From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of J= onathan Rosenberg Sent: den 24 september 2010 20:30 To: codec@ietf.org Subject: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item At the last IETF meeting, we discussed adopting the codec guidelines docume= nt as a working group item. This did not pass, due to concerns over whether= it was in the right direction. We put out a call for alternative documents= over the next 5 week period. Some text was proposed by Stephan for inclusion, which was incorporated int= o the document. Stephan also contributed some comments, including a few ope= n issues which still require some discussion. However, the chairs feel that it is not necessary for all open items to be = closed prior to adopting a document as a working group item. Indeed, discus= sion on the content of the document is a good sign that it is a reasonable = foundation for the working group item. Given the lack of alternative docume= nts to use as a starting point, the chairs plan on adopting this as a worki= ng group item in two weeks time. If you disagree, please speak up - and even better - submit an alternative = document. Thanks, Jonathan R. --=20 Jonathan D. Rosenberg, Ph.D. SkypeID: jdrosen Chief Technology Strategist Mobile: +1 (732) 766-2496 Skype SkypeIn: +1 (408) 465-0361 jdrosen@skype.net http://www.skype.com jdrosen@jdrosen.net http://www.jdrosen.net _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec From ron.even.tlv@gmail.com Fri Oct 8 06:41:21 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 092FC3A689D for ; Fri, 8 Oct 2010 06:41:21 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.599 X-Spam-Level: X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id KU4W7TBs-HfB for ; Fri, 8 Oct 2010 06:41:19 -0700 (PDT) Received: from mail-wy0-f172.google.com (mail-wy0-f172.google.com [74.125.82.172]) by core3.amsl.com (Postfix) with ESMTP id 76D1A3A6835 for ; Fri, 8 Oct 2010 06:41:19 -0700 (PDT) Received: by wyi11 with SMTP id 11so1280797wyi.31 for ; Fri, 08 Oct 2010 06:42:23 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:references :in-reply-to:subject:date:message-id:mime-version:content-type :content-transfer-encoding:x-mailer:content-language:thread-index; bh=TFCe2VASRV8A95/zYbA3/4MyqZwV6gQHcRIxRnDCZvY=; b=Pp+Fjxp2nL31pIr2rpx0T3CppXvsig8kxC4Fdg1/eomBfGLpng3zuvHAbn6VZNXNpE bOZP+8HLudL4QlN9IhQHjauBse6UDi/oeDNiRqPVCJRxfuC4fn6d8tG9gk6DYf5jphEw A3eHmoMQbn6/ZFcfFNlWnwkHe82IT0ESqTM14= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:references:in-reply-to:subject:date:message-id:mime-version :content-type:content-transfer-encoding:x-mailer:content-language :thread-index; b=hop8JaloeD9MaLaxa6+/GklcbmO2/NJDu0lighNV5oRv1aN+1M5cGKqVND8I6mewPY AiqhQXfMmvvu4nHg3eriqhq2u+qG22nJP447VOrRmj25gCr0hOKkpCiZb2y68cu+0F+C hKmMqMT29Y6HPwv9HzqPOpbxupg6Guzm2Mad4= Received: by 10.227.148.2 with SMTP id n2mr2243614wbv.216.1286545343464; Fri, 08 Oct 2010 06:42:23 -0700 (PDT) Received: from windows8d787f9 ([109.67.2.53]) by mx.google.com with ESMTPS id x3sm354136weq.8.2010.10.08.06.42.17 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 08 Oct 2010 06:42:22 -0700 (PDT) From: "Roni Even" To: "'Stephan Wenger'" , "'Jonathan Rosenberg'" , References: <4C9CEE29.1090600@jdrosen.net> In-Reply-To: Date: Fri, 8 Oct 2010 15:39:36 +0200 Message-ID: <4caf1fbe.03edd80a.57fa.1a10@mx.google.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Mailer: Microsoft Office Outlook 12.0 Content-language: en-us Thread-index: ActcUBDZ9dguzCaxcEKDVnYvpVxdOwKnYjFg Subject: Re: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 08 Oct 2010 13:41:21 -0000 Hi, I agree with Stephan's comments and have one other question The text currently says: "Once IETF participants agree that the codec being developed meets the requirements (e.g., via a working group last call), IETF participants can begin the task of characterizing the codec. The characterization process is described under Section 3." Does it mean that the codec may be published without doing characterization test before. Reading the draft it looks like these are the tests that will verify that the codec is within the requirements. Maybe you should add some text saying that the there will be a working group last call after characterization. Roni Even > -----Original Message----- > From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf > Of Stephan Wenger > Sent: Saturday, September 25, 2010 3:22 AM > To: Jonathan Rosenberg; codec@ietf.org > Subject: Re: [codec] Adopting draft-valin-codec-guidelines-06 as a WG > item > > > > > If you disagree, please speak up - and even better - submit an > > alternative document. > > Hi all, > > My preference would be to resolve strategic issues such as text/code > being > normative (where it seems we are very close to consensus: code it > should > be), bit-exact or not (and what are the conformance points if its not > bit-exactness), and characterization gating, before accepting the draft > as a > WG item. That said, I'm willing to support the adoption of the draft > as WG > item even in the absence of this consensus, but only under the > following two > conditions: > > 1. The draft clearly states that there is no consensus yet on the > subject of > bit-exactness. > 2. The draft clearly states that there is no consensus yet on the > subject of > characterization gating (in the sense of Jonathan's email sent a couple > of > hours ago). > > These conditions represent my points 11 and 15 of my email sent to this > list > on August 21, on neither of which I'm in agreement with the current > language > (or the current intention), nor I have seen any movement on the > proponents > on the other side towards a compromise/consensus position. I don't > have the > bandwidth this week or next to suggest alternative language, but will > try to > do so within the two week time window Jonathan proposed. > > In order to move things forward: in both cases, I have no problems if > the > current text stays in, as long as it is made unambiguously clear that > this > is not a consensus position, but rather one possible option for a > resolution > of the question, and that other options are expressedly solicited. > > The reason for these requests are hopefully obvious: a WG draft should > represent the consensus of the WG, and should not contain non-consensus > positions without disclaimers. > > Stephan > > > > On 9.24.2010 11:30 , "Jonathan Rosenberg" wrote: > > > At the last IETF meeting, we discussed adopting the codec guidelines > > document as a working group item. This did not pass, due to concerns > > over whether it was in the right direction. We put out a call for > > alternative documents over the next 5 week period. > > > > Some text was proposed by Stephan for inclusion, which was > incorporated > > into the document. Stephan also contributed some comments, including > a > > few open issues which still require some discussion. > > > > However, the chairs feel that it is not necessary for all open items > to > > be closed prior to adopting a document as a working group item. > Indeed, > > discussion on the content of the document is a good sign that it is a > > reasonable foundation for the working group item. Given the lack of > > alternative documents to use as a starting point, the chairs plan on > > adopting this as a working group item in two weeks time. > > > > If you disagree, please speak up - and even better - submit an > > alternative document. > > > > Thanks, > > Jonathan R. > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From anisse.taleb@huawei.com Fri Oct 8 08:21:34 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 1E15C3A68A7 for ; Fri, 8 Oct 2010 08:21:34 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: 1.364 X-Spam-Level: * X-Spam-Status: No, score=1.364 tagged_above=-999 required=5 tests=[BAYES_20=-0.74, FH_RELAY_NODNS=1.451, HELO_MISMATCH_COM=0.553, RDNS_NONE=0.1] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id XYDrZZcRzSeL for ; Fri, 8 Oct 2010 08:21:33 -0700 (PDT) Received: from szxga02-in.huawei.com (unknown [119.145.14.65]) by core3.amsl.com (Postfix) with ESMTP id 13E8F3A68D4 for ; Fri, 8 Oct 2010 08:21:33 -0700 (PDT) Received: from huawei.com (szxga02-in [172.24.2.6]) by szxga02-in.huawei.com (iPlanet Messaging Server 5.2 HotFix 2.14 (built Aug 8 2006)) with ESMTP id <0L9Z00LWW9CVDX@szxga02-in.huawei.com> for codec@ietf.org; Fri, 08 Oct 2010 23:22:07 +0800 (CST) Received: from szxeml201-edg.china.huawei.com ([172.24.2.119]) by szxga02-in.huawei.com (iPlanet Messaging Server 5.2 HotFix 2.14 (built Aug 8 2006)) with ESMTP id <0L9Z00EW39CVW8@szxga02-in.huawei.com> for codec@ietf.org; Fri, 08 Oct 2010 23:22:07 +0800 (CST) Received: from SZXEML402-HUB.china.huawei.com (10.82.67.32) by szxeml201-edg.china.huawei.com (172.24.2.39) with Microsoft SMTP Server (TLS) id 14.1.218.12; Fri, 08 Oct 2010 23:22:06 +0800 Received: from szxeml502-mbx.china.huawei.com ([fe80::1c80:305a:245:1afe]) by szxeml402-hub.china.huawei.com ([fe80::952d:2bf3:cb35:cafa%21]) with mapi id 14.01.0218.012; Fri, 08 Oct 2010 23:22:06 +0800 Date: Fri, 08 Oct 2010 15:22:06 +0000 From: Anisse Taleb In-reply-to: X-Originating-IP: [10.200.70.166] To: Michael Knappe , "codec@ietf.org" Message-id: <31A06A2BB2D2AB4F9EF23A695455C007943DCB@szxeml502-mbx.china.huawei.com> MIME-version: 1.0 Content-type: text/plain; charset=us-ascii Content-language: en-US Content-transfer-encoding: 7BIT Accept-Language: en-GB, zh-CN, en-US Thread-topic: [codec] McGill university speech database Thread-index: ActgIQL1rV2vEHjq5UuWF93xCyy/iQG1c4xQ X-MS-Has-Attach: X-MS-TNEF-Correlator: References: Subject: Re: [codec] McGill university speech database X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 08 Oct 2010 15:21:34 -0000 Dear Michael, Just a few comments. The TSP database is indeed quite extensive, however, if I am not mistaken, it contains English language only (?). Regarding the balancing of items, one of the main use cases of this codec are conversational applications, as such, the testing should include as much speech items as possible in different conditions, including noisy, reverberant rooms, error conditions/packet losses and more. Furthermore, It is well known that codecs performance may depend on the language of the items used in testing and it is not unheard-of to have a codec pass a quality performance requirement in a certain language and utterly fails in some others. The need for language diversity in testing is even more important given the intended wide distribution of the codec. Kind regards, /Anisse -----Original Message----- From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Michael Knappe Sent: Wednesday, September 29, 2010 11:55 PM To: codec@ietf.org Subject: [codec] McGill university speech database Just received permission from Dr. Peter Kabal at McGill University to use their 1400 utterance, 24 different talker (12 male, 12 female) Harvard sentence speech database for our codec testing efforts in the IETF codec WG. Includes the original 48 kHz files. My preference is just to put the McGill link to the 539 MB CD ISO image file up on the codec wiki, if that sounds ok with everyone I will get final permission to post the link from Dr. Kabal and get them up on the wiki asap. Next step is to work on getting rights to representative music content. A capella vocals (e.g. Tom's Diner) , orchestral crescendo's, castanets, solo violin, jazz trumpet/ensembles, rock/electronica etc would all be good to include for testing, please reply with any suggestions. Thanks to Jean-Marc for the pointer to Dr. Kabal and the speech database! Cheers, Mike _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec From mknappe@juniper.net Fri Oct 8 08:35:56 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 4EA063A6887 for ; Fri, 8 Oct 2010 08:35:56 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -106.013 X-Spam-Level: X-Spam-Status: No, score=-106.013 tagged_above=-999 required=5 tests=[AWL=0.587, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id DBN9qqykzaFi for ; Fri, 8 Oct 2010 08:35:54 -0700 (PDT) Received: from exprod7og115.obsmtp.com (exprod7og115.obsmtp.com [64.18.2.217]) by core3.amsl.com (Postfix) with ESMTP id BAC063A68CB for ; Fri, 8 Oct 2010 08:35:53 -0700 (PDT) Received: from source ([66.129.224.36]) (using TLSv1) by exprod7ob115.postini.com ([64.18.6.12]) with SMTP ID DSNKTK86mpBy4B+OqFGW5o30UjuHuDgZjfK6@postini.com; Fri, 08 Oct 2010 08:36:59 PDT Received: from EMBX02-HQ.jnpr.net ([fe80::18fe:d666:b43e:f97e]) by P-EMHUB03-HQ.jnpr.net ([::1]) with mapi; Fri, 8 Oct 2010 08:30:36 -0700 From: Michael Knappe To: "codec@ietf.org" Date: Fri, 8 Oct 2010 08:30:34 -0700 Thread-Topic: [codec] IETF78 codec WG session audio archive Thread-Index: Actm/b/Ao2HwKMn/6UakMaSI7HQ6wg== Message-ID: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-Entourage/13.3.0.091002 acceptlanguage: en-US Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: [codec] IETF78 codec WG session audio archive X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 08 Oct 2010 15:35:56 -0000 For everyone=92s reference, here is the link to the IETF78 codec WG session= audio archive: http://nagasaki.bogus.com/ietf78/ietf78-ch8-mon-noon.mp3 Cheers, Mike From mknappe@juniper.net Fri Oct 8 08:54:00 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id C5ADE3A686B for ; Fri, 8 Oct 2010 08:54:00 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -106.042 X-Spam-Level: X-Spam-Status: No, score=-106.042 tagged_above=-999 required=5 tests=[AWL=0.557, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id jda+YnnrnisK for ; Fri, 8 Oct 2010 08:53:59 -0700 (PDT) Received: from exprod7og103.obsmtp.com (exprod7og103.obsmtp.com [64.18.2.159]) by core3.amsl.com (Postfix) with ESMTP id 835B03A686A for ; Fri, 8 Oct 2010 08:53:59 -0700 (PDT) Received: from source ([66.129.224.36]) (using TLSv1) by exprod7ob103.postini.com ([64.18.6.12]) with SMTP ID DSNKTK8+2CB2LGshqGS/dvkSGpBKg44WGcok@postini.com; Fri, 08 Oct 2010 08:55:04 PDT Received: from EMBX02-HQ.jnpr.net ([fe80::18fe:d666:b43e:f97e]) by P-EMHUB03-HQ.jnpr.net ([::1]) with mapi; Fri, 8 Oct 2010 08:49:26 -0700 From: Michael Knappe To: "codec@ietf.org" Date: Fri, 8 Oct 2010 08:49:21 -0700 Thread-Topic: [codec] Codec requirements discussion summary from IETF78 Thread-Index: ActnAF9+Yn5EMnBCXUKMo+pFICRvFw== Message-ID: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-Entourage/13.3.0.091002 acceptlanguage: en-US Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: [codec] Codec requirements discussion summary from IETF78 X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 08 Oct 2010 15:54:00 -0000 Below is a brief summary of the discussion Q&A around the requirements doc = held at IETF78. Please refer to the audio ( http://nagasaki.bogus.com/ietf7= 8/ietf78-ch8-mon-noon.mp3) and jabber archives for details. Jean-Marc and Koen, can you update the requirements doc to reflect consensu= s achieved on the 4 points below? Thanks, Mike IETF 78 requirements discussion summary =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D 1. Q: should we have SDP selectable audio bandwidths? Discussion: RTP pay= load formats (sample rates and channel ordering) should be discussed in AVT= . Need to come up with specific proposal(s) for the negotiation of preferre= d bitrate / internal sample rate. One suggestion was simply provisioning in= vites with the highest preferred rate for the codec in the SDP offer as fir= st preference, with lower bit rates in succession as desired. 2. Q: what is a reasonable upper bit rate for the codec? Discussion: 128 = kbps as an upper bit rate limit seems to be reasonable (further clarificati= on needed =96 is that 128 kbps per channel?), will continue codec developme= nt and see what bit rates are offered before any decisions are made on allo= wable bit rate ranges 3. Q: shall we implement sample rate layering? Discussion: prototype code= c in discussion already implements a two layer scheme 4. Q: shall we provide a bit exact implementation? Discussion: consensus = to be optional with no requirement for bit exactness, where interest lay in= the optional specification of a bit exact implementation as one possible i= mplementation among other bit compatible versions 5. Q: shall we provide a joint stereo mode? Discussion: consensus as high= ly desirable where it provides a bit-rate reduction benefit (or conversely = a quality improvement at the same bit-rate) at reasonable computational cos= t. Some anecdotal evidence of 35% or better bit-rate savings noted, further= study needed. 6. Q: shall we provide a low-delay mode? Discussion: consensus as desirab= le, already implemented as a mode in prototype codec as 5 ms total (2.5 ms = look-ahead + 2.5 ms frame size) 7. Q: shall we provide a basic packet-loss concealment mechanism? Discuss= ion: consensus that an informative version of the PLC mechanism be included= in the codec specification, with the ability to replace that mechanism wit= h another implementation. From mknappe@juniper.net Fri Oct 8 09:06:15 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 8B6153A68D6 for ; Fri, 8 Oct 2010 09:06:15 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -106.068 X-Spam-Level: X-Spam-Status: No, score=-106.068 tagged_above=-999 required=5 tests=[AWL=0.531, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id JD4lsbz6pSFE for ; Fri, 8 Oct 2010 09:06:10 -0700 (PDT) Received: from exprod7og115.obsmtp.com (exprod7og115.obsmtp.com [64.18.2.217]) by core3.amsl.com (Postfix) with ESMTP id 4FB893A686B for ; Fri, 8 Oct 2010 09:06:10 -0700 (PDT) Received: from source ([66.129.224.36]) (using TLSv1) by exprod7ob115.postini.com ([64.18.6.12]) with SMTP ID DSNKTK9BsZxyUUP169dVLrzjrqcGSi67WJ9D@postini.com; Fri, 08 Oct 2010 09:07:15 PDT Received: from EMBX02-HQ.jnpr.net ([fe80::18fe:d666:b43e:f97e]) by P-EMHUB03-HQ.jnpr.net ([::1]) with mapi; Fri, 8 Oct 2010 09:01:07 -0700 From: Michael Knappe To: Anisse Taleb , "codec@ietf.org" Date: Fri, 8 Oct 2010 09:01:05 -0700 Thread-Topic: [codec] McGill university speech database Thread-Index: ActgIQL1rV2vEHjq5UuWF93xCyy/iQG1c4xQAALMfR0= Message-ID: In-Reply-To: <31A06A2BB2D2AB4F9EF23A695455C007943DCB@szxeml502-mbx.china.huawei.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-Entourage/13.3.0.091002 acceptlanguage: en-US Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [codec] McGill university speech database X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 08 Oct 2010 16:06:15 -0000 Anisse, Agreed that language diversity in the source material is a necessity, I am working on obtaining additional language samples. Environmental simulation is also highly desirable. There will certainly be environmental diversity i= n the listening environment in the uncontrolled, widescale mushra testing tha= t we are working on setting up, but let's look at a few representative conditions in the samples themselves. It would be nice to provide representative noise / reverb additive in the semi-formal controlled testing, although I am concerned about the rapid increase in test case permutations and listener fatigue with the limited subject pool size we'll have available. Your suggestions here are certainly welcome. Also agree tha= t PLC stressing packet loss conditions do need to be included. Cheers, Mike On 10/8/10 8:22 AM, "Anisse Taleb" wrote: > Dear Michael, >=20 > Just a few comments. >=20 > The TSP database is indeed quite extensive, however, if I am not mistaken= , it > contains English language only (?). >=20 > Regarding the balancing of items, one of the main use cases of this codec= are > conversational applications, as such, the testing should include as much > speech items as possible in different conditions, including noisy, reverb= erant > rooms, error conditions/packet losses and more. Furthermore, It is well k= nown > that codecs performance may depend on the language of the items used in > testing and it is not unheard-of to have a codec pass a quality performan= ce > requirement in a certain language and utterly fails in some others. The n= eed > for language diversity in testing is even more important given the intend= ed > wide distribution of the codec. >=20 > Kind regards, > /Anisse >=20 > -----Original Message----- > From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of > Michael Knappe > Sent: Wednesday, September 29, 2010 11:55 PM > To: codec@ietf.org > Subject: [codec] McGill university speech database >=20 > Just received permission from Dr. Peter Kabal at McGill University to use > their 1400 utterance, 24 different talker (12 male, 12 female) Harvard > sentence speech database for our codec testing efforts in the IETF codec = WG. > Includes the original 48 kHz files. My preference is just to put the McGi= ll > link to the 539 MB CD ISO image file up on the codec wiki, if that sounds= ok > with everyone I will get final permission to post the link from Dr. Kabal= and > get them up on the wiki asap. >=20 > Next step is to work on getting rights to representative music content. A > capella vocals (e.g. Tom's Diner) , orchestral crescendo's, castanets, so= lo > violin, jazz trumpet/ensembles, rock/electronica etc would all be good to > include for testing, please reply with any suggestions. >=20 > Thanks to Jean-Marc for the pointer to Dr. Kabal and the speech database! >=20 > Cheers, >=20 > Mike > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From jean-marc.valin@usherbrooke.ca Sun Oct 10 17:53:27 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 4EA473A672E for ; Sun, 10 Oct 2010 17:53:27 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.16 X-Spam-Level: X-Spam-Status: No, score=-2.16 tagged_above=-999 required=5 tests=[AWL=0.439, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wjBPZ91Wo6ZZ for ; Sun, 10 Oct 2010 17:53:24 -0700 (PDT) Received: from relais.videotron.ca (relais.videotron.ca [24.201.245.36]) by core3.amsl.com (Postfix) with ESMTP id 897983A67B8 for ; Sun, 10 Oct 2010 17:53:19 -0700 (PDT) MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; charset=ISO-8859-1 Received: from [192.168.1.14] ([70.81.109.112]) by VL-MR-MRZ20.ip.videotron.ca (Sun Java(tm) System Messaging Server 6.3-8.01 (built Dec 16 2008; 32bit)) with ESMTP id <0LA300C1CP6N9150@VL-MR-MRZ20.ip.videotron.ca> for codec@ietf.org; Sun, 10 Oct 2010 20:54:24 -0400 (EDT) Message-id: <4CB26045.1070804@usherbrooke.ca> Date: Sun, 10 Oct 2010 20:54:29 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.12) Gecko/20100915 Thunderbird/3.0.8 To: Erik Norvell References: <4C9CEE29.1090600@jdrosen.net> <027A93CE4A670242BD91A44E37105AEF0CACA09122@ESESSCMS0351.eemea.ericsson.se> In-reply-to: <027A93CE4A670242BD91A44E37105AEF0CACA09122@ESESSCMS0351.eemea.ericsson.se> X-Enigmail-Version: 1.0.1 Cc: "codec@ietf.org" Subject: Re: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 11 Oct 2010 00:53:27 -0000 Hi, The current version of the guidelines document mainly cites the charter on IPR issues. This makes sense because this draft can change neither the charter, nor any IETF IPR policies. Also, judging from the discussion on the earlier versions of the draft that differed from the charter, it seems like the current text is still the one that generates the least objection. So unless there is wide consensus on an alternate text that doesn't contradict the current charter, I think we can live with the current text. Cheers, Jean-Marc On 10-10-08 07:40 AM, Erik Norvell wrote: > Hi all, > > Reading the codec guidelines document it still includes arguable and unnecessary generic statements on the problems associated with encumbered technology. The present guidelines do not at all safeguard that the codec can be implemented by anybody as freely as was the intention when the WG was established. Royalty-free conditions do not at all mean free implementation. > > The charter implies that the goal of the WG is to produce a codec without any usage restrictions. > http://tools.ietf.org/wg/codec/charters > ============== > There exist codecs that are standardized, but that cannot be widely > implemented and easily distributed; according to reports, the presence > of various usage restrictions (e.g., in the form of requirements to pay > royalty fees, obtain a license, enter into a business agreement, or meet > other special conditions imposed by a patent holder) has hindered > adoptions of such codecs in interactive Internet applications. > ============== > > Considering these exceptional intentions with the codec WG, the guidelines should also be exceptional compared with the customary IETF guidelines. Preferably, the guidelines should clearly state that the intention is that the codec should be freely available to any implementer. The IETF Patent Disclosure and Licensing Declaration template, part VI Licensing Declaration, has an option a) reading "No Licenses required for Implementers" (http://tools.ietf.org/html/rfc3905). I suggest the guidelines should state that all contributors should comply with this option. As a further clarification of this option a) to meet the intentions behind the establishment of the WG the guidelines should expressly request that "no implementer shall be obliged to make a compensation of any kind to the holder of any patent covering the codec or be obliged to make any undertaking towards to the holder of any patent covering the codec". > > Admittedly the suggestion really means an exceptional requirement on the patent holders which would not at all be adequate in general for all IETF WGs and would provide an exceptional freedom for the implementers, but considering the exceptional intentions with the WG I believe exceptional requirements can be justified. If not, the codec may end up adding to the list of encumbered codecs which the WG set out to avoid. > > Best regards, > Erik > > > -----Original Message----- > From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Jonathan Rosenberg > Sent: den 24 september 2010 20:30 > To: codec@ietf.org > Subject: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item > > At the last IETF meeting, we discussed adopting the codec guidelines document as a working group item. This did not pass, due to concerns over whether it was in the right direction. We put out a call for alternative documents over the next 5 week period. > > Some text was proposed by Stephan for inclusion, which was incorporated into the document. Stephan also contributed some comments, including a few open issues which still require some discussion. > > However, the chairs feel that it is not necessary for all open items to be closed prior to adopting a document as a working group item. Indeed, discussion on the content of the document is a good sign that it is a reasonable foundation for the working group item. Given the lack of alternative documents to use as a starting point, the chairs plan on adopting this as a working group item in two weeks time. > > If you disagree, please speak up - and even better - submit an alternative document. > > Thanks, > Jonathan R. From stewe@stewe.org Mon Oct 11 08:34:09 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 0B90F3A68FF for ; Mon, 11 Oct 2010 08:34:08 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.599 X-Spam-Level: X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id uV1v8cyv3X2w for ; Mon, 11 Oct 2010 08:34:02 -0700 (PDT) Received: from stewe.org (stewe.org [85.214.122.234]) by core3.amsl.com (Postfix) with ESMTP id 4F5573A682F for ; Mon, 11 Oct 2010 08:33:31 -0700 (PDT) Received: from [172.16.7.245] (unverified [160.79.219.114]) by stewe.org (SurgeMail 3.9e) with ESMTP id 810854-1743317 for multiple; Mon, 11 Oct 2010 17:34:42 +0200 User-Agent: Microsoft-Entourage/12.26.0.100708 Date: Mon, 11 Oct 2010 11:34:37 -0700 From: Stephan Wenger To: Jean-Marc Valin , Erik Norvell Message-ID: Thread-Topic: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item Thread-Index: ActpWc/UU0Cj2BOVjkeKwEQcVNYc2g== In-Reply-To: <4CB26045.1070804@usherbrooke.ca> Mime-version: 1.0 Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit X-Originating-IP: 160.79.219.114 X-Authenticated-User: stewe@stewe.org Cc: "codec@ietf.org" Subject: Re: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 11 Oct 2010 15:34:09 -0000 +1. Beyond that, Erik, my feeling is that parts of your proposal, if implemented, would have antitrust issues. I'm willing to discuss those in private if need be. Stephan On 10.10.2010 17:54 , "Jean-Marc Valin" wrote: > Hi, > > The current version of the guidelines document mainly cites the charter > on IPR issues. This makes sense because this draft can change neither > the charter, nor any IETF IPR policies. Also, judging from the > discussion on the earlier versions of the draft that differed from the > charter, it seems like the current text is still the one that generates > the least objection. So unless there is wide consensus on an alternate > text that doesn't contradict the current charter, I think we can live > with the current text. > > Cheers, > > Jean-Marc > > > > On 10-10-08 07:40 AM, Erik Norvell wrote: >> Hi all, >> >> Reading the codec guidelines document it still includes arguable and >> unnecessary generic statements on the problems associated with encumbered >> technology. The present guidelines do not at all safeguard that the codec can >> be implemented by anybody as freely as was the intention when the WG was >> established. Royalty-free conditions do not at all mean free implementation. >> >> The charter implies that the goal of the WG is to produce a codec without any >> usage restrictions. >> http://tools.ietf.org/wg/codec/charters >> ============== >> There exist codecs that are standardized, but that cannot be widely >> implemented and easily distributed; according to reports, the presence >> of various usage restrictions (e.g., in the form of requirements to pay >> royalty fees, obtain a license, enter into a business agreement, or meet >> other special conditions imposed by a patent holder) has hindered >> adoptions of such codecs in interactive Internet applications. >> ============== >> >> Considering these exceptional intentions with the codec WG, the guidelines >> should also be exceptional compared with the customary IETF guidelines. >> Preferably, the guidelines should clearly state that the intention is that >> the codec should be freely available to any implementer. The IETF Patent >> Disclosure and Licensing Declaration template, part VI Licensing Declaration, >> has an option a) reading "No Licenses required for Implementers" >> (http://tools.ietf.org/html/rfc3905). I suggest the guidelines should state >> that all contributors should comply with this option. As a further >> clarification of this option a) to meet the intentions behind the >> establishment of the WG the guidelines should expressly request that "no >> implementer shall be obliged to make a compensation of any kind to the holder >> of any patent covering the codec or be obliged to make any undertaking >> towards to the holder of any patent covering the codec". >> >> Admittedly the suggestion really means an exceptional requirement on the >> patent holders which would not at all be adequate in general for all IETF WGs >> and would provide an exceptional freedom for the implementers, but >> considering the exceptional intentions with the WG I believe exceptional >> requirements can be justified. If not, the codec may end up adding to the >> list of encumbered codecs which the WG set out to avoid. >> >> Best regards, >> Erik >> >> >> -----Original Message----- >> From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of >> Jonathan Rosenberg >> Sent: den 24 september 2010 20:30 >> To: codec@ietf.org >> Subject: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item >> >> At the last IETF meeting, we discussed adopting the codec guidelines document >> as a working group item. This did not pass, due to concerns over whether it >> was in the right direction. We put out a call for alternative documents over >> the next 5 week period. >> >> Some text was proposed by Stephan for inclusion, which was incorporated into >> the document. Stephan also contributed some comments, including a few open >> issues which still require some discussion. >> >> However, the chairs feel that it is not necessary for all open items to be >> closed prior to adopting a document as a working group item. Indeed, >> discussion on the content of the document is a good sign that it is a >> reasonable foundation for the working group item. Given the lack of >> alternative documents to use as a starting point, the chairs plan on adopting >> this as a working group item in two weeks time. >> >> If you disagree, please speak up - and even better - submit an alternative >> document. >> >> Thanks, >> Jonathan R. > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From stephen.botzko@gmail.com Mon Oct 11 08:38:39 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 4522D3A68D6 for ; Mon, 11 Oct 2010 08:38:39 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.563 X-Spam-Level: X-Spam-Status: No, score=-2.563 tagged_above=-999 required=5 tests=[AWL=0.035, BAYES_00=-2.599, HTML_MESSAGE=0.001] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id M6-vWc-OVaTD for ; Mon, 11 Oct 2010 08:38:36 -0700 (PDT) Received: from mail-bw0-f44.google.com (mail-bw0-f44.google.com [209.85.214.44]) by core3.amsl.com (Postfix) with ESMTP id 0F02F3A682A for ; Mon, 11 Oct 2010 08:38:35 -0700 (PDT) Received: by bwz14 with SMTP id 14so1410374bwz.31 for ; Mon, 11 Oct 2010 08:39:47 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type; bh=r8z4w15ClQas5L423pnmW5Es8u4ojuzlEMI81heSWs4=; b=EpFhkjGb441Q+blVrzwWHhxLFWQyoskjCjOnQxoSeovONKytkwQrsDM+2Uqkfe4XNx Ed5vP9W9cdmt0+aqykoXe3FSyAg9/usx5W4alMkToYb9wYVZQ1g/R0I0PelPrBGkhezJ l9YnrK72NfVjZtRr2Cv/7TiVsEsHatWj6pleE= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=bCY+73B+iRkFCUdx8VB8bFLVT4KD9cwOOGC9N5LNdZahjFVyUjkor1U0mr2K+JyFMm G5OQpSJqfM0+hObv0c1Gp01Hc1mX/v21xfK6oTZMkJwNz8QwpLQz9TJ90QWZ5w8SWaho I6D4vrwOiTrVNif53WcIDadn0gC1RFJqmJP/8= MIME-Version: 1.0 Received: by 10.204.115.133 with SMTP id i5mr2624964bkq.27.1286811586264; Mon, 11 Oct 2010 08:39:46 -0700 (PDT) Received: by 10.204.121.204 with HTTP; Mon, 11 Oct 2010 08:39:46 -0700 (PDT) In-Reply-To: <4CB26045.1070804@usherbrooke.ca> References: <4C9CEE29.1090600@jdrosen.net> <027A93CE4A670242BD91A44E37105AEF0CACA09122@ESESSCMS0351.eemea.ericsson.se> <4CB26045.1070804@usherbrooke.ca> Date: Mon, 11 Oct 2010 11:39:46 -0400 Message-ID: From: Stephen Botzko To: Jean-Marc Valin Content-Type: multipart/alternative; boundary=0016e6d625ae7dffb80492592b86 Cc: "codec@ietf.org" Subject: Re: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 11 Oct 2010 15:38:39 -0000 --0016e6d625ae7dffb80492592b86 Content-Type: text/plain; charset=ISO-8859-1 +1 I think the IPR stuff has already been discussed at length, and was worked out adequately in the charter. It'd be best to move on to actually completing the project... Stephen Botzko On Sun, Oct 10, 2010 at 8:54 PM, Jean-Marc Valin < jean-marc.valin@usherbrooke.ca> wrote: > Hi, > > The current version of the guidelines document mainly cites the charter > on IPR issues. This makes sense because this draft can change neither > the charter, nor any IETF IPR policies. Also, judging from the > discussion on the earlier versions of the draft that differed from the > charter, it seems like the current text is still the one that generates > the least objection. So unless there is wide consensus on an alternate > text that doesn't contradict the current charter, I think we can live > with the current text. > > Cheers, > > Jean-Marc > > > > On 10-10-08 07:40 AM, Erik Norvell wrote: > > Hi all, > > > > Reading the codec guidelines document it still includes arguable and > unnecessary generic statements on the problems associated with encumbered > technology. The present guidelines do not at all safeguard that the codec > can be implemented by anybody as freely as was the intention when the WG was > established. Royalty-free conditions do not at all mean free implementation. > > > > The charter implies that the goal of the WG is to produce a codec without > any usage restrictions. > > http://tools.ietf.org/wg/codec/charters > > ============== > > There exist codecs that are standardized, but that cannot be widely > > implemented and easily distributed; according to reports, the presence > > of various usage restrictions (e.g., in the form of requirements to pay > > royalty fees, obtain a license, enter into a business agreement, or > meet > > other special conditions imposed by a patent holder) has hindered > > adoptions of such codecs in interactive Internet applications. > > ============== > > > > Considering these exceptional intentions with the codec WG, the > guidelines should also be exceptional compared with the customary IETF > guidelines. Preferably, the guidelines should clearly state that the > intention is that the codec should be freely available to any implementer. > The IETF Patent Disclosure and Licensing Declaration template, part VI > Licensing Declaration, has an option a) reading "No Licenses required for > Implementers" (http://tools.ietf.org/html/rfc3905). I suggest the > guidelines should state that all contributors should comply with this > option. As a further clarification of this option a) to meet the intentions > behind the establishment of the WG the guidelines should expressly request > that "no implementer shall be obliged to make a compensation of any kind to > the holder of any patent covering the codec or be obliged to make any > undertaking towards to the holder of any patent covering the codec". > > > > Admittedly the suggestion really means an exceptional requirement on the > patent holders which would not at all be adequate in general for all IETF > WGs and would provide an exceptional freedom for the implementers, but > considering the exceptional intentions with the WG I believe exceptional > requirements can be justified. If not, the codec may end up adding to the > list of encumbered codecs which the WG set out to avoid. > > > > Best regards, > > Erik > > > > > > -----Original Message----- > > From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf > Of Jonathan Rosenberg > > Sent: den 24 september 2010 20:30 > > To: codec@ietf.org > > Subject: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item > > > > At the last IETF meeting, we discussed adopting the codec guidelines > document as a working group item. This did not pass, due to concerns over > whether it was in the right direction. We put out a call for alternative > documents over the next 5 week period. > > > > Some text was proposed by Stephan for inclusion, which was incorporated > into the document. Stephan also contributed some comments, including a few > open issues which still require some discussion. > > > > However, the chairs feel that it is not necessary for all open items to > be closed prior to adopting a document as a working group item. Indeed, > discussion on the content of the document is a good sign that it is a > reasonable foundation for the working group item. Given the lack of > alternative documents to use as a starting point, the chairs plan on > adopting this as a working group item in two weeks time. > > > > If you disagree, please speak up - and even better - submit an > alternative document. > > > > Thanks, > > Jonathan R. > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > --0016e6d625ae7dffb80492592b86 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable +1

I think the IPR stuff has already been discussed at length, and w= as worked out adequately in the charter.=A0 It'd be best to move on to = actually completing the project...

Stephen Botzko

On Sun, Oct 10, 2010 at 8:54 PM, Jean-Marc Valin <jean-marc.valin@usherbrooke.c= a> wrote:
Hi,

The current version of the guidelines document mainly cites the charter
on IPR issues. This makes sense because this draft can change neither
the charter, nor any IETF IPR policies. Also, judging from the
discussion on the earlier versions of the draft that differed from the
charter, it seems like the current text is still the one that generates
the least objection. So unless there is wide consensus on an alternate
text that doesn't contradict the current charter, I think we can live with the current text.

Cheers,

=A0 =A0 =A0 =A0Jean-Marc



On 10-10-08 07:40 AM, Erik Norvell wrote:
> Hi all,
>
> Reading the codec guidelines document it still includes arguable and u= nnecessary generic statements on the problems associated with encumbered te= chnology. The present guidelines do not at all safeguard that the codec can= be implemented by anybody as freely as was the intention when the WG was e= stablished. Royalty-free conditions do not at all mean free implementation.=
>
> The charter implies that the goal of the WG is to produce a codec with= out any usage restrictions.
> = http://tools.ietf.org/wg/codec/charters
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
> =A0 There exist codecs that are standardized, but that cannot be widel= y
> =A0 implemented and easily distributed; according to reports, the pres= ence
> =A0 of various usage restrictions (e.g., in the form of requirements t= o pay
> =A0 royalty fees, obtain a license, enter into a business agreement, o= r meet
> =A0 other special conditions imposed by a patent holder) has hindered<= br> > =A0 adoptions of such codecs in interactive Internet applications.
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>
> Considering these exceptional intentions with the codec WG, the guidel= ines should also be exceptional compared with the customary IETF guidelines= . Preferably, the guidelines should clearly state that the intention is tha= t the codec should be freely available to any implementer. The IETF Patent = Disclosure and Licensing Declaration template, part VI Licensing Declaratio= n, has an option a) reading "No Licenses required for Implementers&quo= t; (http:/= /tools.ietf.org/html/rfc3905). I suggest the guidelines should state th= at all contributors should comply with this option. As a further clarificat= ion of this option a) to meet the intentions behind the establishment of th= e WG the guidelines should expressly request that "no implementer shal= l be obliged to make a compensation of any kind to the holder of any patent= covering the codec or be obliged to make any undertaking towards to the ho= lder of any patent covering the codec".
>
> Admittedly the suggestion really means an exceptional requirement on t= he patent holders which would not at all be adequate in general for all IET= F WGs and would provide an exceptional freedom for the implementers, but co= nsidering the exceptional intentions with the WG I believe exceptional requ= irements can be justified. If not, the codec may end up adding to the list = of encumbered codecs which the WG set out to avoid.
>
> Best regards,
> Erik
>
>
> -----Original Message-----
> From: codec-bounces@ietf.org= [mailto:codec-bounces@ietf.o= rg] On Behalf Of Jonathan Rosenberg
> Sent: den 24 september 2010 20:30
> To: codec@ietf.org
> Subject: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item=
>
> At the last IETF meeting, we discussed adopting the codec guidelines d= ocument as a working group item. This did not pass, due to concerns over wh= ether it was in the right direction. We put out a call for alternative docu= ments over the next 5 week period.
>
> Some text was proposed by Stephan for inclusion, which was incorporate= d into the document. Stephan also contributed some comments, including a fe= w open issues which still require some discussion.
>
> However, the chairs feel that it is not necessary for all open items t= o be closed prior to adopting a document as a working group item. Indeed, d= iscussion on the content of the document is a good sign that it is a reason= able foundation for the working group item. Given the lack of alternative d= ocuments to use as a starting point, the chairs plan on adopting this as a = working group item in two weeks time.
>
> If you disagree, please speak up - and even better - submit an alterna= tive document.
>
> Thanks,
> Jonathan R.
_____________________________= __________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec

--0016e6d625ae7dffb80492592b86-- From koen.vos@skype.net Mon Oct 11 13:19:27 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 3EEB73A6B6F for ; Mon, 11 Oct 2010 13:19:27 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.273 X-Spam-Level: X-Spam-Status: No, score=-2.273 tagged_above=-999 required=5 tests=[AWL=0.325, BAYES_00=-2.599, HTML_MESSAGE=0.001] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1DPFNgX4DHGp for ; Mon, 11 Oct 2010 13:19:25 -0700 (PDT) Received: from mx.skype.net (mx.skype.net [78.141.177.88]) by core3.amsl.com (Postfix) with ESMTP id E13273A6B3B for ; Mon, 11 Oct 2010 13:19:24 -0700 (PDT) Received: from mx.skype.net (localhost [127.0.0.1]) by mx.skype.net (Postfix) with ESMTP id 70ED8171D; Mon, 11 Oct 2010 22:20:36 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=skype.net; h=date:from:to :cc:message-id:in-reply-to:subject:mime-version:content-type; s= mx; bh=A+sk54R/5Ybnbc6p+tby4Q0L6YU=; b=uk76iN9q7X0XJxn5SWo86DC3I sKG7/tOGHi/KKCXiRuOwTZSrrjtaPfpDRx4QvpfEU3kQKx96oyGD8JYR4QQxDhkH /SSZETw9AA13tLW5/AxW77IbZtNJFzS1kgimbXAoBJs5IoE/shWjyh4vonF/VUIv fkIFSs2ICgdOoAP71g= DomainKey-Signature: a=rsa-sha1; c=nofws; d=skype.net; h=date:from:to:cc :message-id:in-reply-to:subject:mime-version:content-type; q=dns ; s=mx; b=G2PEPVZlWYkLd1XOUZr8kYX/oR5mmieGgi5/sDssga5O2LnAxvcLTC SPdXLbi31So6nP6t73npGgLUq22wrkHOy1rSuIGDwTs10+CaNwcVo8NqtwWyC/WD aJBrViBUAYPpoHWBHgxHMTl9MaAPTauE/WZAq24UpnMEKPvehGL1Q= Received: from zimbra.skype.net (zimbra.skype.net [78.141.177.82]) by mx.skype.net (Postfix) with ESMTP id 65C677F3; Mon, 11 Oct 2010 22:20:36 +0200 (CEST) Received: from localhost (localhost [127.0.0.1]) by zimbra.skype.net (Postfix) with ESMTP id 21A26350823E; Mon, 11 Oct 2010 22:20:36 +0200 (CEST) X-Virus-Scanned: amavisd-new at lu2-zimbra.skype.net Received: from zimbra.skype.net ([127.0.0.1]) by localhost (zimbra.skype.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id A2MtHvJjizsl; Mon, 11 Oct 2010 22:20:35 +0200 (CEST) Received: from zimbra.skype.net (lu2-zimbra.skype.net [78.141.177.82]) by zimbra.skype.net (Postfix) with ESMTP id 4B7BF3508235; Mon, 11 Oct 2010 22:20:35 +0200 (CEST) Date: Mon, 11 Oct 2010 22:20:35 +0200 (CEST) From: Koen Vos To: Stephen Botzko Message-ID: <926899051.714652.1286828435112.JavaMail.root@lu2-zimbra> In-Reply-To: MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_714651_323704586.1286828435111" X-Originating-IP: [75.28.48.186] X-Mailer: Zimbra 6.0.6_GA_2330.UBUNTU8_64 (ZimbraWebClient - FF3.0 (Win)/6.0.6_GA_2330.UBUNTU8_64) Cc: codec@ietf.org Subject: Re: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 11 Oct 2010 20:19:27 -0000 ------=_Part_714651_323704586.1286828435111 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit +1 :) koen. From: "Stephen Botzko" To: "Jean-Marc Valin" Cc: codec@ietf.org Sent: Monday, October 11, 2010 8:39:46 AM Subject: Re: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item +1 I think the IPR stuff has already been discussed at length, and was worked out adequately in the charter. It'd be best to move on to actually completing the project... Stephen Botzko On Sun, Oct 10, 2010 at 8:54 PM, Jean-Marc Valin < jean-marc.valin@usherbrooke.ca > wrote: Hi, The current version of the guidelines document mainly cites the charter on IPR issues. This makes sense because this draft can change neither the charter, nor any IETF IPR policies. Also, judging from the discussion on the earlier versions of the draft that differed from the charter, it seems like the current text is still the one that generates the least objection. So unless there is wide consensus on an alternate text that doesn't contradict the current charter, I think we can live with the current text. Cheers, Jean-Marc On 10-10-08 07:40 AM, Erik Norvell wrote: > Hi all, > > Reading the codec guidelines document it still includes arguable and unnecessary generic statements on the problems associated with encumbered technology. The present guidelines do not at all safeguard that the codec can be implemented by anybody as freely as was the intention when the WG was established. Royalty-free conditions do not at all mean free implementation. > > The charter implies that the goal of the WG is to produce a codec without any usage restrictions. > http://tools.ietf.org/wg/codec/charters > ============== > There exist codecs that are standardized, but that cannot be widely > implemented and easily distributed; according to reports, the presence > of various usage restrictions (e.g., in the form of requirements to pay > royalty fees, obtain a license, enter into a business agreement, or meet > other special conditions imposed by a patent holder) has hindered > adoptions of such codecs in interactive Internet applications. > ============== > > Considering these exceptional intentions with the codec WG, the guidelines should also be exceptional compared with the customary IETF guidelines. Preferably, the guidelines should clearly state that the intention is that the codec should be freely available to any implementer. The IETF Patent Disclosure and Licensing Declaration template, part VI Licensing Declaration, has an option a) reading "No Licenses required for Implementers" ( http://tools.ietf.org/html/rfc3905 ). I suggest the guidelines should state that all contributors should comply with this option. As a further clarification of this option a) to meet the intentions behind the establishment of the WG the guidelines should expressly request that "no implementer shall be obliged to make a compensation of any kind to the holder of any patent covering the codec or be obliged to make any undertaking towards to the holder of any patent covering the codec". > > Admittedly the suggestion really means an exceptional requirement on the patent holders which would not at all be adequate in general for all IETF WGs and would provide an exceptional freedom for the implementers, but considering the exceptional intentions with the WG I believe exceptional requirements can be justified. If not, the codec may end up adding to the list of encumbered codecs which the WG set out to avoid. > > Best regards, > Erik > > > -----Original Message----- > From: codec-bounces@ietf.org [mailto: codec-bounces@ietf.org ] On Behalf Of Jonathan Rosenberg > Sent: den 24 september 2010 20:30 > To: codec@ietf.org > Subject: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item > > At the last IETF meeting, we discussed adopting the codec guidelines document as a working group item. This did not pass, due to concerns over whether it was in the right direction. We put out a call for alternative documents over the next 5 week period. > > Some text was proposed by Stephan for inclusion, which was incorporated into the document. Stephan also contributed some comments, including a few open issues which still require some discussion. > > However, the chairs feel that it is not necessary for all open items to be closed prior to adopting a document as a working group item. Indeed, discussion on the content of the document is a good sign that it is a reasonable foundation for the working group item. Given the lack of alternative documents to use as a starting point, the chairs plan on adopting this as a working group item in two weeks time. > > If you disagree, please speak up - and even better - submit an alternative document. > > Thanks, > Jonathan R. _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec ------=_Part_714651_323704586.1286828435111 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable <= div style=3D'font-family: Times New Roman; font-size: 12pt; color: #000000'= >+1
:)
koen.


From: "Stephen Botzko" <stephe= n.botzko@gmail.com>
To: "Jean-Marc Valin" <jean-marc.valin@= usherbrooke.ca>
Cc: codec@ietf.org
Sent: Monday, Oct= ober 11, 2010 8:39:46 AM
Subject: Re: [codec] Adopting draft-vali= n-codec-guidelines-06 as a WG item

+1

I think the IPR stuff h= as already been discussed at length, and was worked out adequately in the c= harter.  It'd be best to move on to actually completing the project...=

Stephen Botzko

On Sun, Oct 10, 2010 at 8:54 PM, Jean-Marc Valin <jean-marc.v= alin@usherbrooke.ca> wrote:
Hi,

The current version of the guidelines document mainly cites the charter
on IPR issues. This makes sense because this draft can change neither
the charter, nor any IETF IPR policies. Also, judging from the
discussion on the earlier versions of the draft that differed from the
charter, it seems like the current text is still the one that generates
the least objection. So unless there is wide consensus on an alternate
text that doesn't contradict the current charter, I think we can live
with the current text.

Cheers,

       Jean-Marc



On 10-10-08 07:40 AM, Erik Norvell wrote:
> Hi all,
>
> Reading the codec guidelines document it still includes arguable and u= nnecessary generic statements on the problems associated with encumbered te= chnology. The present guidelines do not at all safeguard that the codec can= be implemented by anybody as freely as was the intention when the WG was e= stablished. Royalty-free conditions do not at all mean free implementation.=
>
> The charter implies that the goal of the WG is to produce a codec with= out any usage restrictions.
> = http://tools.ietf.org/wg/codec/charters
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>   There exist codecs that are standardized, but that cannot be wi= dely
>   implemented and easily distributed; according to reports, the p= resence
>   of various usage restrictions (e.g., in the form of requirement= s to pay
>   royalty fees, obtain a license, enter into a business agreement= , or meet
>   other special conditions imposed by a patent holder) has hinder= ed
>   adoptions of such codecs in interactive Internet applications.<= br> > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>
> Considering these exceptional intentions with the codec WG, the guidel= ines should also be exceptional compared with the customary IETF guidelines= . Preferably, the guidelines should clearly state that the intention is tha= t the codec should be freely available to any implementer. The IETF Patent = Disclosure and Licensing Declaration template, part VI Licensing Declaratio= n, has an option a) reading "No Licenses required for Implementers" (http://tools.iet= f.org/html/rfc3905). I suggest the guidelines should state that all con= tributors should comply with this option. As a further clarification of thi= s option a) to meet the intentions behind the establishment of the WG the g= uidelines should expressly request that "no implementer shall be obliged to= make a compensation of any kind to the holder of any patent covering the c= odec or be obliged to make any undertaking towards to the holder of any pat= ent covering the codec".
>
> Admittedly the suggestion really means an exceptional requirement on t= he patent holders which would not at all be adequate in general for all IET= F WGs and would provide an exceptional freedom for the implementers, but co= nsidering the exceptional intentions with the WG I believe exceptional requ= irements can be justified. If not, the codec may end up adding to the list = of encumbered codecs which the WG set out to avoid.
>
> Best regards,
> Erik
>
>
> -----Original Message-----
> From: code= c-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Jonathan Rosenberg=
> Sent: den 24 september 2010 20:30
> To: codec@ietf.org=
> Subject: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item=
>
> At the last IETF meeting, we discussed adopting the codec guidelines d= ocument as a working group item. This did not pass, due to concerns over wh= ether it was in the right direction. We put out a call for alternative docu= ments over the next 5 week period.
>
> Some text was proposed by Stephan for inclusion, which was incorporate= d into the document. Stephan also contributed some comments, including a fe= w open issues which still require some discussion.
>
> However, the chairs feel that it is not necessary for all open items t= o be closed prior to adopting a document as a working group item. Indeed, d= iscussion on the content of the document is a good sign that it is a reason= able foundation for the working group item. Given the lack of alternative d= ocuments to use as a starting point, the chairs plan on adopting this as a = working group item in two weeks time.
>
> If you disagree, please speak up - and even better - submit an alterna= tive document.
>
> Thanks,
> Jonathan R.
_____________________________= __________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec


_______________________________________________
codec mailing listcodec@ietf.org
https://www.ietf.org/mailman/listinfo/codec
------=_Part_714651_323704586.1286828435111-- From root@core3.amsl.com Tue Oct 12 15:45:01 2010 Return-Path: X-Original-To: codec@ietf.org Delivered-To: codec@core3.amsl.com Received: by core3.amsl.com (Postfix, from userid 0) id B93503A6A22; Tue, 12 Oct 2010 15:45:01 -0700 (PDT) From: Internet-Drafts@ietf.org To: i-d-announce@ietf.org Content-Type: Multipart/Mixed; Boundary="NextPart" Mime-Version: 1.0 Message-Id: <20101012224501.B93503A6A22@core3.amsl.com> Date: Tue, 12 Oct 2010 15:45:01 -0700 (PDT) Cc: codec@ietf.org Subject: [codec] I-D Action:draft-ietf-codec-requirements-02.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 12 Oct 2010 22:45:01 -0000 --NextPart A New Internet-Draft is available from the on-line Internet-Drafts directories. This draft is a work item of the Internet Wideband Audio Codec Working Group of the IETF. Title : Codec Requirements Author(s) : J. Valin, K. Vos Filename : draft-ietf-codec-requirements-02.txt Pages : 21 Date : 2010-10-12 This document provides specific requirements for an Internet audio codec. These requirements address quality, sampling rate, bit-rate, and packet loss robustness, as well as other desirable properties. A URL for this Internet-Draft is: http://www.ietf.org/internet-drafts/draft-ietf-codec-requirements-02.txt Internet-Drafts are also available by anonymous FTP at: ftp://ftp.ietf.org/internet-drafts/ Below is the data which will enable a MIME compliant mail reader implementation to automatically retrieve the ASCII version of the Internet-Draft. --NextPart Content-Type: Message/External-body; name="draft-ietf-codec-requirements-02.txt"; site="ftp.ietf.org"; access-type="anon-ftp"; directory="internet-drafts" Content-Type: text/plain Content-ID: <2010-10-12153156.I-D@ietf.org> --NextPart-- From jean-marc.valin@usherbrooke.ca Tue Oct 12 19:07:50 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 421AC3A6877 for ; Tue, 12 Oct 2010 19:07:50 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.194 X-Spam-Level: X-Spam-Status: No, score=-2.194 tagged_above=-999 required=5 tests=[AWL=0.405, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id NbeG3e9+tWWo for ; Tue, 12 Oct 2010 19:07:48 -0700 (PDT) Received: from relais.videotron.ca (relais.videotron.ca [24.201.245.36]) by core3.amsl.com (Postfix) with ESMTP id D2B1D3A686E for ; Tue, 12 Oct 2010 19:07:47 -0700 (PDT) MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; charset=ISO-8859-1 Received: from [192.168.1.14] ([70.81.109.112]) by VL-MH-MR002.ip.videotron.ca (Sun Java(tm) System Messaging Server 6.3-4.01 (built Aug 3 2007; 32bit)) with ESMTP id <0LA70045YHZ2C1A0@VL-MH-MR002.ip.videotron.ca> for codec@ietf.org; Tue, 12 Oct 2010 22:09:02 -0400 (EDT) Message-id: <4CB514BE.6080104@usherbrooke.ca> Date: Tue, 12 Oct 2010 22:09:02 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.12) Gecko/20100915 Thunderbird/3.0.8 To: codec@ietf.org References: <20101012224501.B93503A6A22@core3.amsl.com> In-reply-to: <20101012224501.B93503A6A22@core3.amsl.com> X-Enigmail-Version: 1.0.1 Subject: Re: [codec] I-D Action:draft-ietf-codec-requirements-02.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 13 Oct 2010 02:07:50 -0000 There's also an updated version of the guidelines draft. Cheers, Jean-Marc On 10-10-12 06:45 PM, Internet-Drafts@ietf.org wrote: > A New Internet-Draft is available from the on-line Internet-Drafts directories. > This draft is a work item of the Internet Wideband Audio Codec Working Group of the IETF. > > > Title : Codec Requirements > Author(s) : J. Valin, K. Vos > Filename : draft-ietf-codec-requirements-02.txt > Pages : 21 > Date : 2010-10-12 > > This document provides specific requirements for an Internet audio > codec. These requirements address quality, sampling rate, bit-rate, > and packet loss robustness, as well as other desirable properties. > > A URL for this Internet-Draft is: > http://www.ietf.org/internet-drafts/draft-ietf-codec-requirements-02.txt > > Internet-Drafts are also available by anonymous FTP at: > ftp://ftp.ietf.org/internet-drafts/ > > Below is the data which will enable a MIME compliant mail reader > implementation to automatically retrieve the ASCII version of the > Internet-Draft. > > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From fluffy@cisco.com Wed Oct 13 10:58:53 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id D89C43A69F5 for ; Wed, 13 Oct 2010 10:58:53 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -110.521 X-Spam-Level: X-Spam-Status: No, score=-110.521 tagged_above=-999 required=5 tests=[AWL=0.078, BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0TOGWBpov1e3 for ; Wed, 13 Oct 2010 10:58:52 -0700 (PDT) Received: from sj-iport-4.cisco.com (sj-iport-4.cisco.com [171.68.10.86]) by core3.amsl.com (Postfix) with ESMTP id 51F6B3A69A0 for ; Wed, 13 Oct 2010 10:58:52 -0700 (PDT) Authentication-Results: sj-iport-4.cisco.com; dkim=neutral (message not signed) header.i=none X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AvsEAE2QtUyrR7Ht/2dsb2JhbAChUHGiP5x2hUgEhFKFb4MD X-IronPort-AV: E=Sophos;i="4.57,326,1283731200"; d="scan'208";a="200247372" Received: from sj-core-1.cisco.com ([171.71.177.237]) by sj-iport-4.cisco.com with ESMTP; 13 Oct 2010 18:00:09 +0000 Received: from [192.168.4.2] (rcdn-fluffy-8711.cisco.com [10.99.9.18]) by sj-core-1.cisco.com (8.13.8/8.14.3) with ESMTP id o9DI07C5008568; Wed, 13 Oct 2010 18:00:07 GMT Mime-Version: 1.0 (Apple Message framework v1081) Content-Type: text/plain; charset=us-ascii From: Cullen Jennings In-Reply-To: <4CA7B5E0.4000508@joelhalpern.com> Date: Wed, 13 Oct 2010 12:00:06 -0600 Content-Transfer-Encoding: quoted-printable Message-Id: References: <4CA7B5E0.4000508@joelhalpern.com> To: "Joel M. Halpern" X-Mailer: Apple Mail (2.1081) Cc: codec@ietf.org Subject: Re: [codec] Summary of codec specification - Code in RFC X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 13 Oct 2010 17:58:54 -0000 These all seem like good points. Let's see if we can figure out a way to = do plain text split across two drafts and follow the XDR model.=20 On Oct 2, 2010, at 4:44 PM, Joel M. Halpern wrote: > I have serious problems with a standards track RFC whose normative = content can not be read by a human being without additional tools. > I also have serious conceptual problems with an RFC whose normative = content, in human readable form, is 1,000 pages. >=20 > I am not objecting to publishing code in RFCs. That is something we = do frequently. I am not even objecting to including base64 encoded gzip = code, as informational appendices in an RFC. > However, it seems to me that treating base64 gzip code as normative in = an RFC is distinctly bending the rules for RFC publication. That text = can not be editted by the RFC Editors. It can not be subject to = erratta. And it can not be read as meaningful plain ascii according to = the spirit of our publication rules. >=20 > Yes, we have once in a great while published RFCs with tables on the = order of this long. It was difficult, but in fact the content was = carefully reviewed. >=20 > Yours, > Joel M. Halpern >=20 > PS: If the code were a reasonable length, publishing the text of the = code as normative, and the base64 gzip as informative would seem to me = to be a very good way to have clear normative code in an RFC while = making it easy to use the code. >=20 > On 10/2/2010 6:32 PM, Stephan Wenger wrote: >> As said before, putting base64 code into an RFC makes sense to me. = However, >> I would definitely put it into its own RFC (which would be standards = track) >> and not into an appendix of the same document. The versioning = problem seems >> to be less an issue than the trees that will fall if people "print = all" the >> RFC in question. So please leave the textual description in another >> document. >>=20 >> Logic suggests that this "human readable" document would be = informational, >> as the code RFC prevails in case of errors. However, I don't mind = being >> slightly illogical here and call it standards track as well. The = document >> could also certain code pieces in such cases where a textual = description in >> English with ASCII fixed width fonts may be to cumbersome. >>=20 >> Someone mentioned the "errata" process for fixing errors. That = process is >> not going to work overly well with an base64 encoded zip file... But = I don't >> consider that a major problem--rather an incentive to get our stuff = right >> the first time. >>=20 >> Stephan >>=20 >>=20 >>=20 >> On 10.2.2010 13:21 , "fluffy@cisco.com" wrote: >>=20 >>>=20 >>> There been a bit of discussion suggesting that we should put into = the draft >>> both the machine readable base 64 version in that is easy to = extract, and a >>> version in human readable text that is easy for people to read. This = seems >>> like a reasonable idea to me. Just to get a rough idea of what this = would look >>> like, I took the current GIT repository and had a look at the code = in it. It' >>> a bit over 50k lines including both encoder and decoder. You get 48 = lines of >>> real text per page of RFC so the human readable version would be = over a 1000 >>> pages. If you take the code, tar it, gzip it, and base64 encode it, = you get a >>> bit over 180 pages. This sounds big but given it's some bits on a = web server, >>> I don't particularly see any problem with it. I also have no idea = how the >>> repository compares to what would be needed to specify the normative = parts of >>> the decoder. I may have including in the 50k lines of code all kinds = of stuff >>> that was not needed. >>>=20 >>> Cullen >>>=20 >>>=20 >>> On Oct 2, 2010, at 8:28 AM, Cullen Jennings wrote: >>>=20 >>>>=20 >>>> On Oct 1, 2010, at 11:27 AM, Cullen Jennings wrote: >>>>=20 >>>>> 1) The code will be in the appendix of the draft base64 encoded. = We can sort >>>>> out later if it is gziped tar file of whatever but it will be some = easy to >>>>> extract format. This has been done on other documents. I have = talked to the >>>>> AD about this before. I have not talked to RFC Ed but do not = foresee any >>>>> issues there. >>>>=20 >>>>=20 >>>> I'd like to encourage people interested in this point to look at = Appendix B >>>> of RFC 4474. >>>>=20 >>>> http://tools.ietf.org/rfc/rfc4474.txt >>>>=20 >>>> _______________________________________________ >>>> codec mailing list >>>> codec@ietf.org >>>> https://www.ietf.org/mailman/listinfo/codec >>>=20 >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >>=20 >>=20 >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >>=20 From erik.norvell@ericsson.com Fri Oct 15 07:41:30 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id EADC53A6CAF for ; Fri, 15 Oct 2010 07:41:30 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -5.134 X-Spam-Level: X-Spam-Status: No, score=-5.134 tagged_above=-999 required=5 tests=[AWL=1.465, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0nZi+m4aWiCu for ; Fri, 15 Oct 2010 07:41:29 -0700 (PDT) Received: from mailgw9.se.ericsson.net (mailgw9.se.ericsson.net [193.180.251.57]) by core3.amsl.com (Postfix) with ESMTP id 62A553A6CB2 for ; Fri, 15 Oct 2010 07:41:27 -0700 (PDT) X-AuditID: c1b4fb39-b7bbbae000007e67-38-4cb86867adcf Received: from esessmw0237.eemea.ericsson.se (Unknown_Domain [153.88.253.125]) by mailgw9.se.ericsson.net (Symantec Mail Security) with SMTP id 2F.FF.32359.76868BC4; Fri, 15 Oct 2010 16:42:47 +0200 (CEST) Received: from ESESSCMS0351.eemea.ericsson.se ([169.254.1.175]) by esessmw0237.eemea.ericsson.se ([153.88.115.90]) with mapi; Fri, 15 Oct 2010 16:42:46 +0200 From: Erik Norvell To: Jean-Marc Valin Date: Fri, 15 Oct 2010 16:42:45 +0200 Thread-Topic: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item Thread-Index: Acto3tzJyxSZgFCYS4+KFEpS7RdSZADmDDMw Message-ID: <027A93CE4A670242BD91A44E37105AEF0CACA603ED@ESESSCMS0351.eemea.ericsson.se> References: <4C9CEE29.1090600@jdrosen.net> <027A93CE4A670242BD91A44E37105AEF0CACA09122@ESESSCMS0351.eemea.ericsson.se> <4CB26045.1070804@usherbrooke.ca> In-Reply-To: <4CB26045.1070804@usherbrooke.ca> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Brightmail-Tracker: AAAAAA== Cc: "codec@ietf.org" Subject: Re: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 15 Oct 2010 14:41:31 -0000 Hi, My suggestion was not to debate the charter nor the IETF policies, but rath= er to make the guidelines more aligned with them. I realize that parts of m= y suggestion was strong-worded, such as having contributors complying with = a "No License option". Still, considering what the charter says about encum= bered codecs, the guidelines should also reflect this in some way. I suggest that the following text is added to the guidelines in page 10: ..that require royalties or other encumbrances that would prevent such tech= nologies from being easy to redistribute and use.[Added text] For the codec= to be truly unencumbered, the contributors should adhere to the 'No licens= e required' option, as expressed in BCP 79 section 6.5: "..c) without the n= eed to obtain a license from the IPR holder."[\Added text] Best, Erik=20 -----Original Message----- From: Jean-Marc Valin [mailto:jean-marc.valin@usherbrooke.ca]=20 Sent: den 11 oktober 2010 02:54 To: Erik Norvell Cc: codec@ietf.org Subject: Re: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item Hi, The current version of the guidelines document mainly cites the charter on = IPR issues. This makes sense because this draft can change neither the char= ter, nor any IETF IPR policies. Also, judging from the discussion on the ea= rlier versions of the draft that differed from the charter, it seems like t= he current text is still the one that generates the least objection. So unl= ess there is wide consensus on an alternate text that doesn't contradict th= e current charter, I think we can live with the current text. Cheers, Jean-Marc On 10-10-08 07:40 AM, Erik Norvell wrote: > Hi all, >=20 > Reading the codec guidelines document it still includes arguable and unne= cessary generic statements on the problems associated with encumbered techn= ology. The present guidelines do not at all safeguard that the codec can be= implemented by anybody as freely as was the intention when the WG was esta= blished. Royalty-free conditions do not at all mean free implementation. >=20 > The charter implies that the goal of the WG is to produce a codec without= any usage restrictions. > http://tools.ietf.org/wg/codec/charters > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > There exist codecs that are standardized, but that cannot be widely > implemented and easily distributed; according to reports, the presence > of various usage restrictions (e.g., in the form of requirements to pay > royalty fees, obtain a license, enter into a business agreement, or mee= t > other special conditions imposed by a patent holder) has hindered > adoptions of such codecs in interactive Internet applications. > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D >=20 > Considering these exceptional intentions with the codec WG, the guideline= s should also be exceptional compared with the customary IETF guidelines. P= referably, the guidelines should clearly state that the intention is that t= he codec should be freely available to any implementer. The IETF Patent Dis= closure and Licensing Declaration template, part VI Licensing Declaration, = has an option a) reading "No Licenses required for Implementers" (http://to= ols.ietf.org/html/rfc3905). I suggest the guidelines should state that all = contributors should comply with this option. As a further clarification of = this option a) to meet the intentions behind the establishment of the WG th= e guidelines should expressly request that "no implementer shall be obliged= to make a compensation of any kind to the holder of any patent covering th= e codec or be obliged to make any undertaking towards to the holder of any = patent covering the codec". >=20 > Admittedly the suggestion really means an exceptional requirement on the = patent holders which would not at all be adequate in general for all IETF W= Gs and would provide an exceptional freedom for the implementers, but consi= dering the exceptional intentions with the WG I believe exceptional require= ments can be justified. If not, the codec may end up adding to the list of = encumbered codecs which the WG set out to avoid. >=20 > Best regards, > Erik >=20 >=20 > -----Original Message----- > From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf=20 > Of Jonathan Rosenberg > Sent: den 24 september 2010 20:30 > To: codec@ietf.org > Subject: [codec] Adopting draft-valin-codec-guidelines-06 as a WG item >=20 > At the last IETF meeting, we discussed adopting the codec guidelines docu= ment as a working group item. This did not pass, due to concerns over wheth= er it was in the right direction. We put out a call for alternative documen= ts over the next 5 week period. >=20 > Some text was proposed by Stephan for inclusion, which was incorporated i= nto the document. Stephan also contributed some comments, including a few o= pen issues which still require some discussion. >=20 > However, the chairs feel that it is not necessary for all open items to b= e closed prior to adopting a document as a working group item. Indeed, disc= ussion on the content of the document is a good sign that it is a reasonabl= e foundation for the working group item. Given the lack of alternative docu= ments to use as a starting point, the chairs plan on adopting this as a wor= king group item in two weeks time. >=20 > If you disagree, please speak up - and even better - submit an alternativ= e document. >=20 > Thanks, > Jonathan R. From root@core3.amsl.com Fri Oct 15 13:30:03 2010 Return-Path: X-Original-To: codec@ietf.org Delivered-To: codec@core3.amsl.com Received: by core3.amsl.com (Postfix, from userid 0) id C18F03A6B0D; Fri, 15 Oct 2010 13:30:02 -0700 (PDT) From: Internet-Drafts@ietf.org To: i-d-announce@ietf.org Content-Type: Multipart/Mixed; Boundary="NextPart" Mime-Version: 1.0 Message-Id: <20101015203002.C18F03A6B0D@core3.amsl.com> Date: Fri, 15 Oct 2010 13:30:02 -0700 (PDT) Cc: codec@ietf.org Subject: [codec] I-D Action:draft-ietf-codec-opus-00.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 15 Oct 2010 20:30:04 -0000 --NextPart A New Internet-Draft is available from the on-line Internet-Drafts directories. This draft is a work item of the Internet Wideband Audio Codec Working Group of the IETF. Title : Definition of the Opus Audio Codec Author(s) : J. Valin, K. Vos Filename : draft-ietf-codec-opus-00.txt Pages : 12 Date : 2010-10-15 This document describes the Opus codec, designed for interactive speech and audio transmission over the Internet. A URL for this Internet-Draft is: http://www.ietf.org/internet-drafts/draft-ietf-codec-opus-00.txt Internet-Drafts are also available by anonymous FTP at: ftp://ftp.ietf.org/internet-drafts/ Below is the data which will enable a MIME compliant mail reader implementation to automatically retrieve the ASCII version of the Internet-Draft. --NextPart Content-Type: Message/External-body; name="draft-ietf-codec-opus-00.txt"; site="ftp.ietf.org"; access-type="anon-ftp"; directory="internet-drafts" Content-Type: text/plain Content-ID: <2010-10-15131844.I-D@ietf.org> --NextPart-- From koen.vos@skype.net Fri Oct 15 13:37:25 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 36C463A6CE2 for ; Fri, 15 Oct 2010 13:37:25 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.594 X-Spam-Level: X-Spam-Status: No, score=-1.594 tagged_above=-999 required=5 tests=[AWL=-0.484, BAYES_05=-1.11] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id XpCQifVwRe4d for ; Fri, 15 Oct 2010 13:37:23 -0700 (PDT) Received: from mx.skype.net (mx.skype.net [78.141.177.88]) by core3.amsl.com (Postfix) with ESMTP id 45BD83A6CD4 for ; Fri, 15 Oct 2010 13:37:23 -0700 (PDT) Received: from mx.skype.net (localhost [127.0.0.1]) by mx.skype.net (Postfix) with ESMTP id 61458170E for ; Fri, 15 Oct 2010 22:38:44 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=skype.net; h=date:from:to :message-id:in-reply-to:subject:mime-version:content-type: content-transfer-encoding; s=mx; bh=lMK1sn6EgwG1jnxZbemP8RrIHy4= ; b=Gcz19MH+s7KdqdjiBb+PmzYwZ46jI4lFSBz7InEPK2UCvnyyDuKsZAL7N3Cy GC1AcpIzY/ljtrXW2SilIvYDz8tw/gsNGJbJKaAeuHIK3uI363oalKo9OK7JDbeY RpCjoWv/fzqQcpBjU6MKV6M6DIKoii77lPlZizWYEkFFJn0= DomainKey-Signature: a=rsa-sha1; c=nofws; d=skype.net; h=date:from:to :message-id:in-reply-to:subject:mime-version:content-type: content-transfer-encoding; q=dns; s=mx; b=qzAtAh71o9r0EQ1e5KlS8G oPLapFUc55XKNcm36Ca9rTRTJpW215hD5TjsBhQ9gKj13mJyO9Wtg8lGockvRW4H gdYPZfB7RPFHLKh14uIk8bR0DzrYldQB0TpSDYclRGKdocM6gc63CJvHQ0TOB07b Mv172O2PG6mRqXYyoRDSg= Received: from zimbra.skype.net (zimbra.skype.net [78.141.177.82]) by mx.skype.net (Postfix) with ESMTP id 5FA097FE for ; Fri, 15 Oct 2010 22:38:44 +0200 (CEST) Received: from localhost (localhost [127.0.0.1]) by zimbra.skype.net (Postfix) with ESMTP id 3F1381672687 for ; Fri, 15 Oct 2010 22:38:44 +0200 (CEST) X-Virus-Scanned: amavisd-new at lu2-zimbra.skype.net Received: from zimbra.skype.net ([127.0.0.1]) by localhost (zimbra.skype.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id AJ+Nna3PA9ey for ; Fri, 15 Oct 2010 22:38:43 +0200 (CEST) Received: from zimbra.skype.net (lu2-zimbra.skype.net [78.141.177.82]) by zimbra.skype.net (Postfix) with ESMTP id D40341672684 for ; Fri, 15 Oct 2010 22:38:43 +0200 (CEST) Date: Fri, 15 Oct 2010 22:38:43 +0200 (CEST) From: Koen Vos To: codec@ietf.org Message-ID: <564940776.851879.1287175123702.JavaMail.root@lu2-zimbra> In-Reply-To: <762896280.851866.1287175064601.JavaMail.root@lu2-zimbra> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [69.181.192.115] X-Mailer: Zimbra 6.0.6_GA_2330.UBUNTU8_64 (ZimbraWebClient - FF3.0 (Win)/6.0.6_GA_2330.UBUNTU8_64) Subject: [codec] Harmony became Opus X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 15 Oct 2010 20:37:25 -0000 Hi Folks, We've now chosen the final name for the SILK/CELT hybrid codec: Opus. It was temporarily called Harmony, but we decided against that in the end - sorry for the confusion :) A draft was submitted to reflect the new name. best, koen. From hoene@uni-tuebingen.de Mon Oct 18 15:20:08 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 07F073A6AB6 for ; Mon, 18 Oct 2010 15:20:08 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.739 X-Spam-Level: X-Spam-Status: No, score=-4.739 tagged_above=-999 required=5 tests=[AWL=1.510, BAYES_00=-2.599, HELO_EQ_DE=0.35, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id LVfw1+pSPQb7 for ; Mon, 18 Oct 2010 15:20:06 -0700 (PDT) Received: from mx06.uni-tuebingen.de (mx06.uni-tuebingen.de [134.2.3.3]) by core3.amsl.com (Postfix) with ESMTP id 7F7AE3A6B8F for ; Mon, 18 Oct 2010 15:20:05 -0700 (PDT) Received: from hoeneT60 ([178.2.218.235]) (authenticated bits=0) by mx06.uni-tuebingen.de (8.13.6/8.13.6) with ESMTP id o9IMLQGa028619 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Tue, 19 Oct 2010 00:21:33 +0200 From: "Christian Hoene" To: "'Jean-Marc Valin'" , References: <20101012224501.B93503A6A22@core3.amsl.com> <4CB514BE.6080104@usherbrooke.ca> In-Reply-To: <4CB514BE.6080104@usherbrooke.ca> Date: Tue, 19 Oct 2010 00:21:28 +0200 Organization: =?UTF-8?Q?Universit=C3=A4t_T=C3=BCbingen?= Message-ID: <001001cb6f12$d3705540$7a50ffc0$@de> MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Mailer: Microsoft Office Outlook 12.0 Thread-Index: Actqe6TizuGd7nwIRPSKcbEN9X81uQElW0gQ Content-Language: de X-AntiVirus: NOT checked by Avira MailGate (version: 3.0.0-4; host: mx06) Subject: Re: [codec] I-D Action:draft-ietf-codec-requirements-02.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 18 Oct 2010 22:20:08 -0000 Hi, just a few comments from regarding the guidelines draft (see = attachment). I made some minor improvements to increase the clarity. I would suggest to split Section 4 into two. One section on how to = describe the codec and its source code and a second on suggestion what = to specify and what not to specify. Also, I added a section on how to collaborate with other WGs and on what = topics the collaboration shall take place. It might be useful to plan = when go for a bar Bof with those guys... With best regards, Christian=20 --------------------------------------------------------------- Dr.-Ing. Christian Hoene Interactive Communication Systems (ICS), University of T=C3=BCbingen=20 Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532=20 http://www.net.uni-tuebingen.de/ -----Original Message----- From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf = Of Jean-Marc Valin Sent: Wednesday, October 13, 2010 4:09 AM To: codec@ietf.org Subject: Re: [codec] I-D Action:draft-ietf-codec-requirements-02.txt There's also an updated version of the guidelines draft. Cheers, Jean-Marc On 10-10-12 06:45 PM, Internet-Drafts@ietf.org wrote: > A New Internet-Draft is available from the on-line Internet-Drafts = directories. > This draft is a work item of the Internet Wideband Audio Codec Working = Group of the IETF. >=20 >=20 > Title : Codec Requirements > Author(s) : J. Valin, K. Vos > Filename : draft-ietf-codec-requirements-02.txt > Pages : 21 > Date : 2010-10-12 >=20 > This document provides specific requirements for an Internet audio > codec. These requirements address quality, sampling rate, bit-rate, > and packet loss robustness, as well as other desirable properties. >=20 > A URL for this Internet-Draft is: > = http://www.ietf.org/internet-drafts/draft-ietf-codec-requirements-02.txt >=20 > Internet-Drafts are also available by anonymous FTP at: > ftp://ftp.ietf.org/internet-drafts/ >=20 > Below is the data which will enable a MIME compliant mail reader > implementation to automatically retrieve the ASCII version of the > Internet-Draft. >=20 >=20 >=20 > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec From hoene@uni-tuebingen.de Mon Oct 18 15:22:42 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 9EAAC3A6BEC for ; Mon, 18 Oct 2010 15:22:42 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -5.117 X-Spam-Level: X-Spam-Status: No, score=-5.117 tagged_above=-999 required=5 tests=[AWL=1.132, BAYES_00=-2.599, HELO_EQ_DE=0.35, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ylU2sWEVD-Tw for ; Mon, 18 Oct 2010 15:22:41 -0700 (PDT) Received: from mx06.uni-tuebingen.de (mx06.uni-tuebingen.de [134.2.3.3]) by core3.amsl.com (Postfix) with ESMTP id 2130D3A6B8F for ; Mon, 18 Oct 2010 15:22:39 -0700 (PDT) Received: from hoeneT60 ([178.2.218.235]) (authenticated bits=0) by mx06.uni-tuebingen.de (8.13.6/8.13.6) with ESMTP id o9IMNq55028753 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Tue, 19 Oct 2010 00:23:58 +0200 From: "Christian Hoene" To: "'Jean-Marc Valin'" , References: <20101012224501.B93503A6A22@core3.amsl.com> <4CB514BE.6080104@usherbrooke.ca> In-Reply-To: <4CB514BE.6080104@usherbrooke.ca> Date: Tue, 19 Oct 2010 00:23:55 +0200 Organization: =?UTF-8?Q?Universit=C3=A4t_T=C3=BCbingen?= Message-ID: <001201cb6f13$2a24fd00$7e6ef700$@de> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_NextPart_000_0013_01CB6F23.EDADCD00" X-Mailer: Microsoft Office Outlook 12.0 Thread-Index: Actqe6TizuGd7nwIRPSKcbEN9X81uQElW0gQ Content-Language: de X-AntiVirus: NOT checked by Avira MailGate (version: 3.0.0-4; host: mx06) Subject: Re: [codec] I-D Action:draft-ietf-codec-requirements-02.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 18 Oct 2010 22:22:42 -0000 This is a multi-part message in MIME format. ------=_NextPart_000_0013_01CB6F23.EDADCD00 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Hi, just a few comments from regarding the guidelines draft (see = attachment). I made some minor improvements to increase the clarity. I would suggest to split Section 4 into two. One section on how to = describe the codec and its source code and a second on suggestion what = to specify and what not to specify. Also, I added a section on how to collaborate with other WGs and on what = topics the collaboration shall take place. It might be useful to plan = when go for a bar Bof with those guys... With best regards, Christian=20 --------------------------------------------------------------- Dr.-Ing. Christian Hoene Interactive Communication Systems (ICS), University of T=C3=BCbingen=20 Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532=20 http://www.net.uni-tuebingen.de/ -----Original Message----- From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf = Of Jean-Marc Valin Sent: Wednesday, October 13, 2010 4:09 AM To: codec@ietf.org Subject: Re: [codec] I-D Action:draft-ietf-codec-requirements-02.txt There's also an updated version of the guidelines draft. Cheers, Jean-Marc On 10-10-12 06:45 PM, Internet-Drafts@ietf.org wrote: > A New Internet-Draft is available from the on-line Internet-Drafts = directories. > This draft is a work item of the Internet Wideband Audio Codec Working = Group of the IETF. >=20 >=20 > Title : Codec Requirements > Author(s) : J. Valin, K. Vos > Filename : draft-ietf-codec-requirements-02.txt > Pages : 21 > Date : 2010-10-12 >=20 > This document provides specific requirements for an Internet audio > codec. These requirements address quality, sampling rate, bit-rate, > and packet loss robustness, as well as other desirable properties. >=20 > A URL for this Internet-Draft is: > = http://www.ietf.org/internet-drafts/draft-ietf-codec-requirements-02.txt >=20 > Internet-Drafts are also available by anonymous FTP at: > ftp://ftp.ietf.org/internet-drafts/ >=20 > Below is the data which will enable a MIME compliant mail reader > implementation to automatically retrieve the ASCII version of the > Internet-Draft. >=20 >=20 >=20 > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec ------=_NextPart_000_0013_01CB6F23.EDADCD00 Content-Type: application/msword; name="draft-valin-codec-guidelines-07-ch.doc" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="draft-valin-codec-guidelines-07-ch.doc" 0M8R4KGxGuEAAAAAAAAAAAAAAAAAAAAAPgADAP7/CQAGAAAAAAAAAAAAAAACAAAA2AAAAAAAAAAA EAAA2gAAAAEAAAD+////AAAAANYAAADXAAAA//////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// ///////////////////////////////////////////////////////////////////////////s pcEAA4AJBAAA8BK/AAAAAAAAEAAAAAAACAAAp5UAAA4AYmpiarRWtFYAAAAAAAAAAAAAAAAAAAAA AAAHBBYANAABANY8AQDWPAEA6IoAAAAAAAAAAAAAAAAAAL4CAAAAAAAAAAAAAAAAAAD//w8AAAAA AAAAAAD//w8AAAAAAAAAAAD//w8AAAAAAAAAAAAAAAAAAAAAALcAAAAAACQJAAAAAAAAJAkAAGgW AAAAAAAAaBYAAAAAAAB0FgAA8gAAAOQXAAAkAAAACBgAABQAAAAAAAAAAAAAAP////8AAAAAHBgA AAAAAAAcGAAAAAAAABwYAAAAAAAAHBgAAHQAAACQGAAAPAEAABwYAAAAAAAA+IIAABwCAADMGQAA AAAAAMwZAAAAAAAAzBkAAAAAAADMGQAAAAAAAMwZAAAAAAAApxoAAAAAAACnGgAAAAAAAKcaAAAA AAAAg4IAAAIAAACFggAAAAAAAIWCAAAAAAAAhYIAAAAAAACFggAAAAAAAIWCAAAAAAAAhYIAAAAA AAAUhQAAogIAALaHAAA+AAAAhYIAABUAAAAAAAAAAAAAAAAAAAAAAAAAaBYAAAwAAACnGgAAWgAA AAAAAAAAAAAAAAAAAAAAAACnGgAAAAAAAKcaAAAAAAAAARsAADwAAAA9GwAAIAAAAIWCAAAAAAAA AAAAAAAAAABoFgAAAAAAAGgWAAAAAAAAzBkAAAAAAAAAAAAAAAAAAMwZAADbAAAAmoIAACIAAAD1 fgAAAAAAAPV+AAAAAAAA9X4AAAAAAABdGwAANGMAAGgWAAAAAAAAzBkAAAAAAABoFgAAAAAAAMwZ AAAAAAAAg4IAAAAAAAAAAAAAAAAAAPV+AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAApxoAAAAAAACDggAAAAAAAAAAAAAAAAAA9X4AAAAAAAD1fgAA HgAAAJeBAAAYAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA44EAAAAAAADMGQAAAAAAAP////8AAAAAIINhXBJv ywEAAAAAAAAAABwYAAAAAAAAkX4AAGQAAACvgQAACAAAAAAAAAAAAAAAb4IAABQAAAC8ggAAPAAA APiCAAAAAAAAt4EAACwAAAD0hwAAAAAAAPV+AAAAAAAA9IcAABAAAADjgQAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADj gQAAFAAAAPSHAAAAAAAAAAAAAAAAAABmFwAAfgAAAPeBAAB4AAAApxoAAAAAAACnGgAAAAAAAPV+ AAAAAAAApxoAAAAAAACnGgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAApxoA AAAAAACnGgAAAAAAAKcaAAAAAAAAhYIAAAAAAACFggAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAA9X4AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAKcaAAAA AAAApxoAAAAAAACnGgAAAAAAAPiCAAAAAAAApxoAAAAAAACnGgAAAAAAAKcaAAAAAAAApxoAAAAA AAAAAAAAAAAAAP////8AAAAA/////wAAAAD/////AAAAAAAAAAAAAAAA/////wAAAAD/////AAAA AP////8AAAAA/////wAAAAD/////AAAAAP////8AAAAA/////wAAAAD/////AAAAAP////8AAAAA /////wAAAAD/////AAAAAP////8AAAAA/////wAAAAD/////AAAAAPSHAAAAAAAApxoAAAAAAACn GgAAAAAAAKcaAAAAAAAApxoAAAAAAACnGgAAAAAAAKcaAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACnGgAAAAAAAKcaAAAAAAAApxoA AAAAAAAkCQAACgwAAC4VAAA6AQAABQASAQAABwQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA0NDU5l dHdvcmsgV29ya2luZyBHcm91cCAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgIEpNLiBWYWxpbg1JbnRlcm5ldC1EcmFmdCAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgICBPY3Rhc2ljIEluYy4NSW50ZW5kZWQgc3RhdHVzOiBTdGFuZGFyZHMg VHJhY2sgICAgICAgICAgICAgICAgICAgICAgICAgICAgICBTLiBCb3JpbGluDUV4cGlyZXM6IEFw cmlsIDE1LCAyMDExICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgU1BJUklU IERTUA0gICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICBLLiBWb3MNICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgIFNreXBlIFRlY2hub2xvZ2llcyBTLkEuDSAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgQy4gTW9udGdvbWVyeQ0gICAg ICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIFhpcGguT3Jn IEZvdW5kYXRpb24NICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICBSLiBDaGVuDSAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICAgICBCcm9hZGNvbSBDb3Jwb3JhdGlvbg0gICAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIE9jdG9iZXIgMTIsIDIw MTANDQ0gICAgICAgICAgR3VpZGVsaW5lcyBmb3IgdGhlIENvZGVjIERldmVsb3BtZW50IFdpdGhp biB0aGUgSUVURg0gICAgICAgICAgICAgICAgICAgIGRyYWZ0LXZhbGluLWNvZGVjLWd1aWRlbGlu ZXMtMDcNDUFic3RyYWN0DQ0gICBUaGlzIGRvY3VtZW50IHByb3ZpZGVzIGdlbmVyYWwgZ3VpZGVs aW5lcyBmb3Igd29yayBvbiBkZXZlbG9waW5nIGFuZA0gICBzcGVjaWZ5aW5nIGEgcHJvYmFibHkg b25lBSBjb2RlYyB3aXRoaW4gdGhlIElFVEYuICBUaGVzZSBndWlkZWxpbmVzIGNvdmVyIHRoZQ0g ICBkZXZlbG9wbWVudCBwcm9jZXNzLCBldmFsdWF0aW9uLCByZXF1aXJlbWVudHMgY29uZm9ybWFu Y2UsIGFuZA0gICBpbnRlbGxlY3R1YWwgcHJvcGVydHkgaXNzdWVzLg0NU3RhdHVzIG9mIHRoaXMg TWVtbw0NICAgVGhpcyBJbnRlcm5ldC1EcmFmdCBpcyBzdWJtaXR0ZWQgaW4gZnVsbCBjb25mb3Jt YW5jZSB3aXRoIHRoZQ0gICBwcm92aXNpb25zIG9mIEJDUCA3OCBhbmQgQkNQIDc5Lg0NICAgSW50 ZXJuZXQtRHJhZnRzIGFyZSB3b3JraW5nIGRvY3VtZW50cyBvZiB0aGUgSW50ZXJuZXQgRW5naW5l ZXJpbmcNICAgVGFzayBGb3JjZSAoSUVURikuICBOb3RlIHRoYXQgb3RoZXIgZ3JvdXBzIG1heSBh bHNvIGRpc3RyaWJ1dGUNICAgd29ya2luZyBkb2N1bWVudHMgYXMgSW50ZXJuZXQtRHJhZnRzLiAg VGhlIGxpc3Qgb2YgY3VycmVudCBJbnRlcm5ldC0NICAgRHJhZnRzIGlzIGF0IGh0dHA6Ly9kYXRh dHJhY2tlci5pZXRmLm9yZy9kcmFmdHMvY3VycmVudC8uDQ0gICBJbnRlcm5ldC1EcmFmdHMgYXJl IGRyYWZ0IGRvY3VtZW50cyB2YWxpZCBmb3IgYSBtYXhpbXVtIG9mIHNpeCBtb250aHMNICAgYW5k IG1heSBiZSB1cGRhdGVkLCByZXBsYWNlZCwgb3Igb2Jzb2xldGVkIGJ5IG90aGVyIGRvY3VtZW50 cyBhdCBhbnkNICAgdGltZS4gIEl0IGlzIGluYXBwcm9wcmlhdGUgdG8gdXNlIEludGVybmV0LURy YWZ0cyBhcyByZWZlcmVuY2UNICAgbWF0ZXJpYWwgb3IgdG8gY2l0ZSB0aGVtIG90aGVyIHRoYW4g YXMgIndvcmsgaW4gcHJvZ3Jlc3MuIg0NICAgVGhpcyBJbnRlcm5ldC1EcmFmdCB3aWxsIGV4cGly ZSBvbiBBcHJpbCAxNSwgMjAxMS4NDUNvcHlyaWdodCBOb3RpY2UNDSAgIENvcHlyaWdodCAoYykg MjAxMCBJRVRGIFRydXN0IGFuZCB0aGUgcGVyc29ucyBpZGVudGlmaWVkIGFzIHRoZQ0gICBkb2N1 bWVudCBhdXRob3JzLiAgQWxsIHJpZ2h0cyByZXNlcnZlZC4NDSAgIFRoaXMgZG9jdW1lbnQgaXMg c3ViamVjdCB0byBCQ1AgNzggYW5kIHRoZSBJRVRGIFRydXN0J3MgTGVnYWwNICAgUHJvdmlzaW9u cyBSZWxhdGluZyB0byBJRVRGIERvY3VtZW50cw0gICAoaHR0cDovL3RydXN0ZWUuaWV0Zi5vcmcv bGljZW5zZS1pbmZvKSBpbiBlZmZlY3Qgb24gdGhlIGRhdGUgb2YNDQ0NVmFsaW4sIGV0IGFsLiAg ICAgICAgICAgIEV4cGlyZXMgQXByaWwgMTUsIDIwMTEgICAgICAgICAgICAgICAgIFtQYWdlIDFd DQwNSW50ZXJuZXQtRHJhZnQgICAgICAgICAgICAgIENvZGVjIEd1aWRlbGluZXMgICAgICAgICAg ICAgICAgT2N0b2JlciAyMDEwDQ0NICAgcHVibGljYXRpb24gb2YgdGhpcyBkb2N1bWVudC4gIFBs ZWFzZSByZXZpZXcgdGhlc2UgZG9jdW1lbnRzDSAgIGNhcmVmdWxseSwgYXMgdGhleSBkZXNjcmli ZSB5b3VyIHJpZ2h0cyBhbmQgcmVzdHJpY3Rpb25zIHdpdGggcmVzcGVjdA0gICB0byB0aGlzIGRv Y3VtZW50LiAgQ29kZSBDb21wb25lbnRzIGV4dHJhY3RlZCBmcm9tIHRoaXMgZG9jdW1lbnQgbXVz dA0gICBpbmNsdWRlIFNpbXBsaWZpZWQgQlNEIExpY2Vuc2UgdGV4dCBhcyBkZXNjcmliZWQgaW4g U2VjdGlvbiA0LmUgb2YNICAgdGhlIFRydXN0IExlZ2FsIFByb3Zpc2lvbnMgYW5kIGFyZSBwcm92 aWRlZCB3aXRob3V0IHdhcnJhbnR5IGFzDSAgIGRlc2NyaWJlZCBpbiB0aGUgU2ltcGxpZmllZCBC U0QgTGljZW5zZS4NDQ1UYWJsZSBvZiBDb250ZW50cw0NICAgMS4gIEludHJvZHVjdGlvbiAuIC4g LiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuICAzDSAgIDIuICBE ZXZlbG9wbWVudCBQcm9jZXNzICAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAu IC4gLiAgNA0gICAzLiAgRXZhbHVhdGlvbiwgVGVzdGluZywgYW5kIENoYXJhY3Rlcml6YXRpb24g IC4gLiAuIC4gLiAuIC4gLiAuIC4gIDcNICAgNC4gIFJlcXVpcmVtZW50cyBDb25mb3JtYW5jZSAu IC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuICA4DSAgIDUuICBJbnRlbGxlY3R1 YWwgUHJvcGVydHkgIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAxMA0g ICA2LiAgUmVsYXRpb25zaGlwIHdpdGggT3RoZXIgU0RPcyAuIC4gLiAuIC4gLiAuIC4gLiAuIC4g LiAuIC4gLiAuIC4gMTINICAgNy4gIFNlY3VyaXR5IENvbnNpZGVyYXRpb25zICAuIC4gLiAuIC4g LiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIDE0DSAgIDguICBJQU5BIENvbnNpZGVyYXRpb25z ICAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAxNQ0gICA5LiAgQWNr bm93bGVkZ21lbnRzICAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAu IC4gMTYNICAgMTAuIFJlZmVyZW5jZXMgLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAu IC4gLiAuIC4gLiAuIC4gLiAuIDE3DSAgICAgMTAuMS4gIE5vcm1hdGl2ZSBSZWZlcmVuY2VzICAu IC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAxNw0gICAgIDEwLjIuICBJbmZvcm1h dGl2ZSBSZWZlcmVuY2VzICAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gMTcNICAg QXV0aG9ycycgQWRkcmVzc2VzIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4g LiAuIC4gLiAuIDE5DQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ1WYWxpbiwgZXQgYWwuICAg ICAgICAgICAgRXhwaXJlcyBBcHJpbCAxNSwgMjAxMSAgICAgICAgICAgICAgICAgW1BhZ2UgMl0N DA1JbnRlcm5ldC1EcmFmdCAgICAgICAgICAgICAgQ29kZWMgR3VpZGVsaW5lcyAgICAgICAgICAg ICAgICBPY3RvYmVyIDIwMTANDQ0xLiAgSW50cm9kdWN0aW9uDQ0gICBUaGlzIGRvY3VtZW50IGRl c2NyaWJlcyBhIHN1Z2dlc3RlZCBwcm9jZXNzIGZvciB3b3JrIGF0IHRoZSBJRVRGIG9uDSAgIHN0 YW5kYXJkaXphdGlvbiBvZiBhIGNvZGVjIHRoYXQgaXMgb3B0aW1pemVkIGZvciB1c2UgaW4gaW50 ZXJhY3RpdmUNICAgSW50ZXJuZXQgYXBwbGljYXRpb25zIGFuZCB0aGF0IGNhbiBiZSB3aWRlbHkg aW1wbGVtZW50ZWQgYW5kIGVhc2lseQ0gICBkaXN0cmlidXRlZCBhbW9uZyBhcHBsaWNhdGlvbiBk ZXZlbG9wZXJzLCBzZXJ2aWNlIG9wZXJhdG9ycywgYW5kIGVuZA0gICB1c2Vycy4NDQ0NDQ0NDQ0N DQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ1WYWxpbiwgZXQgYWwuICAgICAgICAg ICAgRXhwaXJlcyBBcHJpbCAxNSwgMjAxMSAgICAgICAgICAgICAgICAgW1BhZ2UgM10NDA1JbnRl cm5ldC1EcmFmdCAgICAgICAgICAgICAgQ29kZWMgR3VpZGVsaW5lcyAgICAgICAgICAgICAgICBP Y3RvYmVyIDIwMTANDQ0yLiAgRGV2ZWxvcG1lbnQgUHJvY2Vzcw0NICAgVGhlIHByb2Nlc3Mgb3V0 bGluZWQgaGVyZSBpcyBpbnRlbmRlZCB0byBtYXhpbWl6ZSBtYWtlIHRoZSB0cmFuc3BhcmVuY3kg b2YFDSAgIHdvcmsgb24gYW4gYXVkaW8gY29kZWMgd2l0aGluIHRoZSBJRVRGIHRyYW5zcGFyZW50 LCBwcmVkaWN0YWJsZSwgYW5kIHdlbGwgb3JnYW5pemVkLiAgU3VjaCB3b3JrIG1pZ2h0IGludm9s dmUNICAgZGV2ZWxvcG1lbnQgb2YgYSBjb21wbGV0ZWx5IG5ldyBjb2RlYywgYWRhcHRhdGlvbiBv ZiBhbiBleGlzdGluZw0gICBjb2RlYyB0byBtZWV0IHRoZSByZXF1aXJlbWVudHMsIG9yIGludGVn cmF0aW9uIGJldHdlZW4gdHdvIG9yIG1vcmUNICAgZXhpc3RpbmcgY29kZWNzIHRoYXQgcmVzdWx0 cyBpbiBhbiBpbXByb3ZlZCBjb2RlYyBjb21iaW5pbmcgdGhlIGJlc3QNICAgYXNwZWN0cyBvZiBl YWNoIGNvZGVjLiAgVG8gZW5hYmxlIGFzdWNoIHByb2NlZHVyYWwgdHJhbnNwYXJlbmN5LCB0aGUN ICAgY29udHJpYnV0b3Igb2YgYW4gZXhpc3RpbmcgY29kZWMgbXVzdCBiZSB3aWxsaW5nIHRvIGNl ZGUgY2hhbmdlDSAgIGNvbnRyb2wgdG8gdGhlIElFVEYgYW5kIHNob3VsZCBoYXZlIHN1ZmZpY2ll bnQga25vd2xlZGdlIG9mIHRoZSBjb2RlYw0gICB0byBhc3Npc3QgaW4gdGhlIHdvcmsgb2YgYWRh cHRpbmcgaXQgb3IgYXBwbHlpbmcgc29tZSBvZiBpdHMNICAgdGVjaG5vbG9neSB0byB0aGUgZGV2 ZWxvcG1lbnQgb3IgaW1wcm92ZW1uZXQgb2Ygb3RoZXIgY29kZWNzLg0gICBGdXJ0aGVybW9yZSwg Y29udHJpYnV0b3JzIG5lZWQgdG8gYmUgYXdhcmUgdGhhdCBhbnkgY29kZWMgdGhhdA0gICByZXN1 bHRzIGZyb20gd29yayB3aXRoaW4gdGhlIElFVEYgaXMgbGlrZWx5IHRvIGJlIGRpZmZlcmVudCBm cm9tIGFueQ0gICBleGlzdGluZyBjb2RlYyB0aGF0IHdhcyBjb250cmlidXRlZCB0byB0aGUgSW50 ZXJuZXQgU3RhbmRhcmRzDSAgIFByb2Nlc3MuDQ0gICBXb3JrIG9uIGNvZGVjIGRldmVsb3BtZW50 IGlzIGV4cGVjdGVkIHRvIHByb2NlZWQgYXMgZm9sbG93czoNDSAgIDEuICBJRVRGIHBhcnRpY2lw YW50cyB3aWxsIGlkZW50aWZ5IHRoZSByZXF1aXJlbWVudHMgdG8gYmUgbWV0IGJ5IGFuDSAgICAg ICBJbnRlcm5ldCBjb2RlYywgaW4gdGhlIGZvcm0gb2YgYW4gSW50ZXJuZXQtRHJhZnQuDQ0gICAy LiAgSW50ZXJlc3RlZCBwYXJ0aWVzIGFyZSBlbmNvdXJhZ2VkIHRvIG1ha2UgY29udHJpYnV0aW9u cyBwcm9wb3NpbmcNICAgICAgIGV4aXN0aW5nIG9yIG5ldyBjb2RlY3MsIG9yIGVsZW1lbnRzIHRo ZXJlb2YsIHRvIHRoZSBjb2RlYyBXRyBhcw0gICAgICAgbG9uZyBhcyB0aGVzZSBjb250cmlidXRp b25zIGFyZSB3aXRoaW4gdGhlIHNjb3BlIG9mIHRoZSBXRy4NICAgICAgIElkZWFsbHksIHRoZXNl IGNvbnRyaWJ1dGlvbnMgc2hvdWxkIGJlIGluIHRoZSBmb3JtIG9mIEludGVybmV0DSAgICAgICBE cmFmdHMsIGFsdGhvdWdoIG90aGVyIGZvcm1zIG9mIGNvbnRyaWJ1dGlvbnMgYXJlIGFsc28gcG9z c2libGUNICAgICAgIGFzIGRpc2N1c3NlZCBpbiBbUFJPQ0VTU10gYW5kIGluIHRoZSBJRVRGJ3Mg Tm90ZSBXZWxsLiAgQXMNICAgICAgIGFsd2F5cywgY29udHJpYnV0aW9ucyB0byB0aGUgSUVURiBh cmUgc3ViamVjdCwgYW1vbmcgb3RoZXINICAgICAgIHByb2Nlc3Mgb3JpZW50ZWQgUkZDcywgdG8g W1BST0NFU1NdLCBbVFJVU1RdLCBhbmQgW0lQUl0uDSAgICAgICBDb25zaWRlcmluZyB0aGUgZmll bGQgb2YgdGVjaG5vbG9neSwgSVBSIHRyYW5zcGFyZW5jeSBtYXkgYmUNICAgICAgIHBhcnRpY3Vs YXJseSBoaWdoIG9uIHRoZSBwcmlvcml0eSBsaXN0IG9mIG1hbnkgY29kZWMgV0cNICAgICAgIHBh cnRpY2lwYW50cy4gIEFjY29yZGluZ2x5LCBjb250cmlidXRvcnMgYXJlIHNwZWNpZmljYWxseQ0g ICAgICAgcmVtaW5kZWQgb2YgdGhlaXIgSVBSIGRpc2Nsb3N1cmUgcmVxdWlyZW1lbnQsIGFuZCBh bGwNICAgICAgIHBhcnRpY2lwYW50cyBhcmUgcmVtaW5kZWQgb2YgdGhlIHNvbGljaXRhdGlvbiBv ZiB0aGUgZGlzY2xvc3VyZQ0gICAgICAgb2YgdGhpcmQgcGFydHkgSVBSLCBib3RoIGFzIGNvZGlm aWVkIGluIFtJUFJdLg0NICAgMy4gIEFzIGNvbnRyaWJ1dGlvbnMgYXJlIHJlY2VpdmVkIGFuZCBk aXNjdXNzZWQgd2l0aGluIHRoZSB3b3JraW5nDSAgICAgICBncm91cCwgdGhlIGdyb3VwIHNob3Vs ZCBnYWluIGEgY2xlYXJlciB1bmRlcnN0YW5kaW5nIG9mIHdoYXQgaXMNICAgICAgIGFjaGlldmFi bGUgd2l0aGluIHRoZSBkZXNpZ24gc3BhY2UuICBBcyBhIHJlc3VsdCwgdGhlIGF1dGhvcnMgb2YN ICAgICAgIHRoZSByZXF1aXJlbWVudHMgZG9jdW1lbnQgc2hvdWxkIGl0ZXJhdGl2ZWx5IGNsYXJp ZnkgYW5kIGltcHJvdmUNICAgICAgIHRoZWlyIGRvY3VtZW50IHRvIHJlZmxlY3QgdGhlIGVtZXJn aW5nIHdvcmtpbmcgZ3JvdXAgY29uc2Vuc3VzLg0gICAgICAgVGhpcyBpcyBsaWtlbHkgdG8gaW52 b2x2ZSBjb2xsYWJvcmF0aW9uIHdpdGggSUVURiB3b3JraW5nIGdyb3Vwcw0gICAgICAgaW4gb3Ro ZXIgYXJlYXMsIHN1Y2ggYXMgY29sbGFib3JhdGlvbiB3aXRoIHdvcmtpbmcgZ3JvdXBzIGluIHRo ZQ0gICAgICAgVHJhbnNwb3J0IGFyZWEgdG8gaWRlbnRpZnkgaW1wb3J0YW50IGFzcGVjdHMgb2Yg cGFja2V0DSAgICAgICB0cmFuc21pc3Npb24gb3ZlciB0aGUgSW50ZXJuZXQgYW5kIHRvIHVuZGVy c3RhbmQgdGhlIGRlZ3JlZSBvZg0gICAgICAgcmF0ZSBhZGFwdGF0aW9uIGRlc2lyYWJsZSBhbmQs IHdpdGggd29ya2luZyBncm91cHMgaW4gdGhlIFJBSQ0gICAgICAgYXJlYSB0byBlbnN1cmUgdGhh dCBpbmZvcm1hdGlvbiBhYm91dCBhbmQgbmVnb3RpYXRpb24gb2YgdGhlDSAgICAgICBjb2RlYyBj YW4gYmUgZWFzaWx5IHJlcHJlc2VudGVkIGF0IHRoZSBzaWduYWxsaW5nIGxheWVyLCBhbmQgd2l0 aA0FDQ0NVmFsaW4sIGV0IGFsLiAgICAgICAgICAgIEV4cGlyZXMgQXByaWwgMTUsIDIwMTEgICAg ICAgICAgICAgICAgIFtQYWdlIDRdDQwNSW50ZXJuZXQtRHJhZnQgICAgICAgICAgICAgIENvZGVj IEd1aWRlbGluZXMgICAgICAgICAgICAgICAgT2N0b2JlciAyMDEwDQ0NICAgICAgIHdvcmtpbmcg Z3JvdXBzIGluIHRoZSBUcmFuc3BvcnQgYXJlYSB0byB1bmRlcnN0YW5kIHRoZSBkZWdyZWUgb2YN ICAgICAgIHJhdGUgYWRhcHRhdGlvbiBkZXNpcmFibGUuICBJbiBwYXJhbGxlbCB3aXRoIHRoaXMg d29yaywNICAgICAgIGludGVyZXN0ZWQgcGFydGllcyBzaG91bGQgZXZhbHVhdGUgdGhlIGNvbnRy aWJ1dGlvbnMgYXQgYSBoaWdoZXINICAgICAgIGxldmVsIHRvIHNlZSB3aGljaCByZXF1aXJlbWVu dHMgbWlnaHQgYmUgbWV0IGJ5IGVhY2ggY29kZWMuDQ0gICA0LiAgT25jZSBhIHN1ZmZpY2llbnQg bnVtYmVyIG9mIHByb3Bvc2FscyBoYXMgYmVlbiByZWNlaXZlZCwgdGhlDSAgICAgICBpbnRlcmVz dGVkIHBhcnRpZXMgd2lsbCBpZGVudGlmeSB0aGUgc3RyZW5ndGhzLCB3ZWFrbmVzc2VzLCBhbmQN ICAgICAgIGlubm92YXRpdmUgYXNwZWN0cyBvZiB0aGUgY29udHJpYnV0ZWQgY29kZWNzLiAgVGhp cyBzdGVwIHdpbGwNICAgICAgIGNvbnNpZGVyIG5vdCBvbmx5IHRoZSBjb2RlY3MgYXMgYSB3aG9s ZSwgYnV0IGFsc28ga2V5IGZlYXR1cmVzIG9mDSAgICAgICB0aGUgaW5kaXZpZHVhbCBhbGdvcml0 aG1zIChwcmVkaWN0b3JzLCBxdWFudGl6ZXJzLCB0cmFuc2Zvcm1zLA0gICAgICAgZXRjLikuDQ0g ICA1LiAgSXQgaXMgZXhwZWN0ZWQgdGhhdCBub25lIG9mIHRoZSBjb250cmlidXRlZCBjb2RlY3Mg d2lsbCBtZWV0IGFsbA0gICAgICAgb2YgdGhlIGRlZmluZWQgcmVxdWlyZW1lbnRzLiAgVGhlcmVm b3JlLCBpdCBpcyBleHBlY3RlZCB0aGF0IElFVEYNICAgICAgIHBhcnRpY2lwYW50cyB3aWxsIGFj Y2VwdCBhIF9iYXNlbGluZSBjb2RlY18gYXMgYSBXRyBpdGVtIHRvDSAgICAgICBmYWNpbGl0YXRl IHRoZSBkZXZlbG9wbWVudCBwcm9jZXNzLiAgVGhpcyBiYXNlbGluZSBjb2RlYyB3aWxsDSAgICAg ICBtZWV0IGFzIG1hbnkgb2YgdGhlIHJlcXVpcmVtZW50cyBhcyBwb3NzaWJsZSwgYnV0IHByb2Jh Ymx5IHdpbGwNICAgICAgIG5lZWQgdG8gYmUgYWRqdXN0ZWQgdGhyb3VnaCBhbiBpdGVyYXRpdmUg ZGV2ZWxvcG1lbnQgcHJvY2VzcyBpbg0gICAgICAgb3JkZXIgdG8gbWVldCBhbGwgb2YgdGhlIHJl cXVpcmVtZW50cyAob3IgYXMgbWFueSByZXF1aXJlbWVudHMgYXMNICAgICAgIHBvc3NpYmxlKS4g IFRoZSBiYXNlbGluZSBjb2RlYyBtaWdodCBiZSBvbmUgb2YgdGhlIGNvbnRyaWJ1dGVkDSAgICAg ICBjb2RlY3MgKGVzcGVjaWFsbHkgaWYgaXQgaXMgdGhlIG9ubHkgY29kZWMgdGhhdCBtZWV0cyBt b3N0IG9mIHRoZQ0gICAgICAgcmVxdWlyZW1lbnRzKSwgYSBjb21iaW5hdGlvbiBvZiB0d28gb3Ig bW9yZSBvZiB0aGUgY29udHJpYnV0ZWQNICAgICAgIGNvZGVjcywgb3IgYW4gZW50aXJlbHkgbmV3 IGNvZGVjLiAgTm9uZSBvZiB0aGUgZGVjaXNpb25zIHRha2VuIGF0DSAgICAgICB0aGlzIHN0ZXAg d2lsbCBiZSBkZWZpbml0aXZlLiAgSW4gcGFydGljdWxhciwgSUVURiBwYXJ0aWNpcGFudHMNICAg ICAgIHdpbGwgbm90IHByb3ZpZGUgYSAicnViYmVyIHN0YW1wIiBmb3IgYW55IGNvbnRyaWJ1dGVk IGNvZGVjLg0NICAgNi4gIElFVEYgcGFydGljaXBhbnRzIHNob3VsZCB0aGVuIGF0dGVtcHQgdG8g aXRlcmF0aXZlbHkgaW1wcm92ZSBlYWNoDSAgICAgICBjb21wb25lbnQgb2YgdGhlIGJhc2VsaW5l IGNvZGVjIHJlZmVyZW5jZSBpbXBsZW1lbnRhdGlvbiwgd2hlcmUNICAgICAgIGJ5ICJjb21wb25l bnQiIHdlIG1lYW4gaW5kaXZpZHVhbCBhbGdvcml0aG1zIHN1Y2ggYXMgcHJlZGljdG9ycywNICAg ICAgIHRyYW5zZm9ybXMsIHF1YW50aXplcnMsIGFuZCBlbnRyb3B5IGNvZGVycy4gIFRoZSBwYXJ0 aWNpcGFudHMNICAgICAgIHNob3VsZCBwcm9jZWVkIGJ5IHRyeWluZyBuZXcgZGVzaWducywgYXBw bHlpbmcgaWRlYXMgZnJvbSB0aGUNICAgICAgIGNvbnRyaWJ1dGVkIGNvZGVjcywgZXZhbHVhdGlu ZyAicHJvb2Ygb2YgY29uY2VwdCIgaWRlYXMsIGFuZA0gICAgICAgdXNpbmcgdGhlaXIgZXhwZXJ0 aXNlIGluIGNvZGVjIGRldmVsb3BtZW50IHRvIGltcHJvdmUgdGhlDSAgICAgICBiYXNlbGluZSBj b2RlYy4gIEFueSBhc3BlY3Qgb2YgdGhlIGJhc2VsaW5lIGNvZGVjIG1pZ2h0IGJlDSAgICAgICBj aGFuZ2VkIChldmVuIHRoZSBmdW5kYW1lbnRhbCBwcmluY2lwbGVzIG9mIHRoZSBjb2RlYykgb3Ig dGhlDSAgICAgICBwYXJ0aWNpcGFudHMgbWlnaHQgc3RhcnQgb3ZlciBlbnRpcmVseSBieSBzY3Jh cHBpbmcgdGhlIGJhc2VsaW5lDSAgICAgICBjb2RlYyBhbmQgZGVzaWduaW5nIGEgY29tcGxldGVs eSBuZXcgb25lLiAgVGhlIG92ZXJyaWRpbmcgZ29hbA0gICAgICAgc2hhbGwgYmUgdG8gZGVzaWdu IGEgY29kZWMgdGhhdCB3aWxsIG1lZXQgdGhlIHJlcXVpcmVtZW50cw0gICAgICAgZGVmaW5lZCBp biB0aGUgcmVxdWlyZW1lbnRzIGRvY3VtZW50LiAgR2l2ZW4gdGhlIElFVEYncyBvcGVuDSAgICAg ICBzdGFuZGFyZHMgcHJvY2VzcywgYW55IGludGVyZXN0ZWQgcGFydHkgd2lsbCBiZSBhYmxlIHRv DSAgICAgICBjb250cmlidXRlIHRvIHRoaXMgd29yaywgd2hldGhlciBvciBub3QgdGhleSBzdWJt aXR0ZWQgYW4NICAgICAgIEludGVybmV0LURyYWZ0IGZvciBvbmUgb2YgdGhlIGNvbnRyaWJ1dGVk IGNvZGVjcy4gIFRoZSBjb2RlYw0gICAgICAgaXRzZWxmIHNob3VsZCBiZSBub3JtYXRpdmVseSBz cGVjaWZpZWQgd2l0aCBjb2RlIGluIGFuIEludGVybmV0LQ0gICAgICAgRHJhZnQuDQ0gICA3LiAg SW4gcGFyYWxsZWwgd2l0aCB3b3JrIG9uIHRoZSBjb2RlYyByZWZlcmVuY2UgaW1wbGVtZW50YXRp b24sDSAgICAgICBkZXZlbG9wZXJzIGFuZCBvdGhlciBpbnRlcmVzdGVkIHBhcnRpZXMgc2hvdWxk IHBlcmZvcm0gZXZhbHVhdGlvbg0gICAgICAgb2YgdGhlIGNvZGVjIGFzIGRlc2NyaWJlZCB1bmRl ciBTZWN0aW9uIDMsIElFVEYgcGFydGljaXBhbnRzDQ0NDVZhbGluLCBldCBhbC4gICAgICAgICAg ICBFeHBpcmVzIEFwcmlsIDE1LCAyMDExICAgICAgICAgICAgICAgICBbUGFnZSA1XQ0MDUludGVy bmV0LURyYWZ0ICAgICAgICAgICAgICBDb2RlYyBHdWlkZWxpbmVzICAgICAgICAgICAgICAgIE9j dG9iZXIgMjAxMA0NDSAgICAgICBzaG91bGQgZGVmaW5lICh3aXRoaW4gdGhlIEFWVCBXb3JraW5n IEdyb3VwKSB0aGUgY29kZWMncyBwYXlsb2FkDSAgICAgICBmb3JtYXQgIGZvciB1c2Ugd2l0aCB0 aGUgUmVhbC10aW1lIFRyYW5zcG9ydCBQcm90b2NvbCBbUlRQXSAgIGluY2x1ZGluZyBhbiBhcHBy aW9yaWF0ZSByYXRlIGNvbnRyb2wFLCBhbmQNICAgICAgIGFwcGxpY2F0aW9uIGRldmVsb3BlcnMg c2hvdWxkIHN0YXJ0IHRlc3RpbmcgdGhlIGNvZGVjIGJ5DSAgICAgICBpbXBsZW1lbnRpbmcgaXQg aW4gY29kZSBhbmQgZGVwbG95aW5nIGl0IGluIGFjdHVhbCBJbnRlcm5ldA0gICAgICAgYXBwbGlj YXRpb25zIHRvIGlkZW50aWZ5IGFueSBwb3RlbnRpYWwgcHJvYmxlbXMuDQ0gICA4LiAgT25jZSBJ RVRGIHBhcnRpY2lwYW50cyBhZ3JlZSB0aGF0IHRoZSBjb2RlYyBiZWluZyBkZXZlbG9wZWQgbWVl dHMNICAgICAgIHRoZSByZXF1aXJlbWVudHMsIElFVEYgcGFydGljaXBhbnRzIGNhbiBiZWdpbiB0 aGUgdGFzayBvZg0gICAgICAgY2hhcmFjdGVyaXppbmcgdGhlIGNvZGVjLiAgVGhlIGNoYXJhY3Rl cml6YXRpb24gcHJvY2VzcyBpcw0gICAgICAgZGVzY3JpYmVkIHVuZGVyIFNlY3Rpb24gMy4NDQ0N DQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ1WYWxpbiwgZXQgYWwuICAgICAg ICAgICAgRXhwaXJlcyBBcHJpbCAxNSwgMjAxMSAgICAgICAgICAgICAgICAgW1BhZ2UgNl0NDA1J bnRlcm5ldC1EcmFmdCAgICAgICAgICAgICAgQ29kZWMgR3VpZGVsaW5lcyAgICAgICAgICAgICAg ICBPY3RvYmVyIDIwMTANDQ0zLiAgRXZhbHVhdGlvbiwgVGVzdGluZywgYW5kIENoYXJhY3Rlcml6 YXRpb24NDSAgIExhYiBldmFsdWF0aW9uIG9mIHRoZSBjb2RlYyBiZWluZyBkZXZlbG9wZWQgc2hv dWxkIGhhcHBlbiB0aHJvdWdob3V0DSAgIHRoZSBkZXZlbG9wbWVudCBwcm9jZXNzIGJlY2F1c2Ug aXQgd2lsbCBoZWxwIGVuc3VyZSB0aGF0IHByb2dyZXNzIGlzDSAgIGJlaW5nIG1hZGUgdG93YXJk IGZ1bGZpbGxtZW50IG9mIHRoZSByZXF1aXJlbWVudHMuICBUaGVyZSBhcmUgbWFueQ0gICB3YXlz IGluIHdoaWNoIGNvbnRpbnVvdXMgZXZhbHVhdGlvbiBjYW4gYmUgcGVyZm9ybWVkLiAgRm9yIG1p bm9yLA0gICB1bmNvbnRyb3ZlcnNpYWwgY2hhbmdlcyB0byB0aGUgY29kZWMgaXQgc2hvdWxkIHVz dWFsbHkgYmUgc3VmZmljaWVudA0gICB0byB1c2Ugb2JqZWN0aXZlIG1lYXN1cmVtZW50cyAoZS5n LiwgUEVTUSwgUEVBUSwgYW5kIFNlZ1NOUikNICAgdmFsaWRhdGVkIGJ5IGluZm9ybWFsIHN1Ympl Y3RpdmUgZXZhbHVhdGlvbi4gIEZvciBtb3JlIGNvbXBsZXgNICAgY2hhbmdlcyAoZS5nLiwgd2hl biBwc3ljaG9hY291c3RpYyBhc3BlY3RzIGFyZSBpbnZvbHZlZCkgb3IgZm9yDSAgIGNvbnRyb3Zl cnNpYWwgaXNzdWVzLCBpbnRlcm5hbCB0ZXN0aW5nIHNob3VsZCBiZSBwZXJmb3JtZWQuICBBbg0g ICBleGFtcGxlIG9mIGludGVybmFsIHRlc3Rpbmcgd291bGQgYmUgdG8gaGF2ZSBpbmRpdmlkdWFs IHBhcnRpY2lwYW50cw0gICByYXRlIHRoZSBkZWNvZGVkIHNhbXBsZXMgdXNpbmcgb25lIG9mIHRo ZSBlc3RhYmxpc2hlZCB0ZXN0aW5nDSAgIG1ldGhvZG9sb2dpZXMsIHN1Y2ggYXMgSVRVLVIgQlMu MTUzNCAoTVVTSFJBKS4NDSAgIFRocm91Z2hvdXQgdGhlIHByb2Nlc3MsIGl0IHdpbGwgYmUgaW1w b3J0YW50IHRvIG1ha2UgdXNlIG9mIHRoZQ0gICBJbnRlcm5ldCBjb21tdW5pdHkgYXQgbGFyZ2Ug Zm9yIHJlYWwtd29ybGQgZGlzdHJpYnV0ZWQgdGVzdGluZy4gIFRoaXMNICAgd2lsbCBlbmFibGUg bWFueSBkaWZmZXJlbnQgcGVvcGxlIHdpdGggZGlmZmVyZW50IGVxdWlwbWVudCBhbmQgdXNlDSAg IGNhc2VzIHRvIHRlc3QgdGhlIGNvZGVjIGFuZCByZXBvcnQgYW55IHByb2JsZW1zIHRoZXkgZXhw ZXJpZW5jZS4gIEluDSAgIHRoZSBzYW1lIHdheSwgdGhpcmQtcGFydHkgc29mdHdhcmUgZGV2ZWxv cGVycyB3aWxsIGJlIGVuY291cmFnZWQgdG8NICAgaW50ZWdyYXRlIHRoZSBjb2RlYyAod2l0aCBh IHdhcm5pbmcgYWJvdXQgdGhlIGJpdC1zdHJlYW0gbm90IGJlaW5nDSAgIGZpbmFsKSBhbmQgcHJv dmlkZSBmZWVkYmFjayBvbiBpdHMgcGVyZm9ybWFuY2UgaW4gcmVhbC13b3JsZCB1c2UNICAgY2Fz ZXMuDQ0gICBDaGFyYWN0ZXJpemF0aW9uIG9mIHRoZSBmaW5hbCBjb2RlYyBtdXN0IGJlIGJhc2Vk IG9uIHRoZSByZWZlcmVuY2UNICAgaW1wbGVtZW50YXRpb24gb25seSAoYW5kIG5vdCBvbiBhbnkg InByaXZhdGUgaW1wbGVtZW50YXRpb24iKS4gIFRoaXMNICAgY2FuIGJlIHBlcmZvcm1lZCBieSBp bmRlcGVuZGVudCB0ZXN0aW5nIGxhYnMgb3IsIGlmIHRoaXMgaXMgbm90DSAgIHBvc3NpYmxlLCB1 c2luZyB0aGUgdGVzdGluZyBsYWJzIG9mIHRoZSBvcmdhbml6YXRpb25zIHRoYXQgY29udHJpYnV0 ZQ0gICB0byB0aGUgSW50ZXJuZXQgU3RhbmRhcmRzIFByb2Nlc3MuICBQYWNrZXQgbG9zcyByb2J1 c3RuZXNzIHNob3VsZCBiZQ0gICBldmFsdWF0ZWQgdXNpbmcgYWN0dWFsIGxvc3MgcGF0dGVybnMg Y29sbGVjdGVkIGZyb20gdXNlIG92ZXIgdGhlDSAgIEludGVybmV0LCByYXRoZXIgdGhhbiB0aGVv cmV0aWNhbCBtb2RlbHMuICBUaGUgZ29hbHMgb2YgdGhlDSAgIGNoYXJhY3Rlcml6YXRpb24gcGhh c2UgYXJlIHRvOg0NICAgbyAgZW5zdXJlIHRoYXQgdGhlIHJlcXVpcmVtZW50cyBoYXZlIGJlZW4g ZnVsZmlsbGVkDQ0gICBvICBndWlkZSB0aGUgSUVTRyBpbiBpdHMgZXZhbHVhdGlvbiBvZiB0aGUg cmVzdWx0aW5nIHdvcmsNDSAgIG8gIGFzc2lzdCBhcHBsaWNhdGlvbiBkZXZlbG9wZXJzIGluIHVu ZGVyc3RhbmRpbmcgd2hldGhlciB0aGUgY29kZWMNICAgICAgaXMgc3VpdGFibGUgZm9yIGEgcGFy dGljdWxhciBhcHBsaWNhdGlvbg0NICAgVGhlIGV4YWN0IG1ldGhvZG9sb2d5IGZvciB0aGUgY2hh cmFjdGVyaXphdGlvbiBwaGFzZSBpcyBzdGlsbCBzdWJqZWN0DSAgIHRvIGRpc2N1c3Npb24gd2l0 aGluIHRoZSB3b3JraW5nIGdyb3VwLg0NDQ0NDQ0NDQ1WYWxpbiwgZXQgYWwuICAgICAgICAgICAg RXhwaXJlcyBBcHJpbCAxNSwgMjAxMSAgICAgICAgICAgICAgICAgW1BhZ2UgN10NDA1JbnRlcm5l dC1EcmFmdCAgICAgICAgICAgICAgQ29kZWMgR3VpZGVsaW5lcyAgICAgICAgICAgICAgICBPY3Rv YmVyIDIwMTANDQ00LiAgUHVibGlzaGluZyBhbmQgRGVzY3JpcHRpb24gb2YgdGhlIENvZGVjUmVx dWlyZW1lbnRzIENvbmZvcm1hbmNlDQ0gICBJdCBpcyBleHBlY3RlZCB0aGF0IHRoZSBjb2RlYyBh bGdvcml0aG1zIGFuZCB0aGUgY29kZWNzIHNvdXJjZSBjb2RlIHdpbGwgYmUgdmVyeSBjb21wbGV4 IGFuZCB3aWxsIGNvbnRhaW4gbWFueSBsaW5lcyBvZiBjb2RlLiBOZXZlcnRoZWxlc3MsIGl0IG11 c3QgYmUgYXZvaWRlZCB0aGF0IHRoZSBDb2RlYyByZW1haW5zIGEgYmxhY2sgYm94IHRoYXQgY2Fu IGhhcmRseSBiZSB1bmRlcnN0b29rIGJlY2F1c2Ugb3RoZXJ3aXNlIHRlc3RpbmcsIG1haW50YWlu aW5nLCBhbmQgZW5oYW5jaW5nIHdpbGwgZ2V0IGN1bWJlcnN0b25lLiBUaHVzIHRoZSBjb2RlY5Jz IHNwZWNpZmljYXRpb24gTVVTVCBoYXZlIHRoZSBmb2xsb3dpbmcgZmVhdHVyZXMuDSAgIEl0IGlz IHRoZSByZXNwb25zaWJpbGl0eSBvZiB0aGUgd29ya2luZyBncm91cCB0byBkZWZpbmUgY3JpdGVy aWEgZm9yDSAgIGV2YWx1YXRpbmcgY29uZm9ybWFuY2UsIGluY2x1ZGluZyBidXQgbm90IGxpbWl0 ZWQgdG8gY29tcGFyaXNvbiB0b29scw0gICBhbmQgdGVzdCB2ZWN0b3JzLiAgVGhlIGZvbGxvd2lu ZyB0ZXh0IHByb3ZpZGVzIHN1Z2dlc3Rpb25zIGZvcg0gICBjb25zaWRlcmF0aW9uIGJ5IHRoZSB3 b3JraW5nIGdyb3VwOg0JDQ0gICAxLiAgQW55IGNvZGVjIHNwZWNpZmllZCBieSB0aGUgSUVURiBN VVNUbXVzdCBiZSB3ZWxsIHNwZWNpZmllZC4gVGhpcyBNQVkgaW5jbHVkZSBpbmNsdWRlIHNvdXJj ZSBjb2RlIGZvciBhDSAgICAgICBub3JtYXRpdmUgQzg5IHJlZmVyZW5jZSBpbXBsZW1lbnRhdGlv biBvciBNQVkgY29udGFpbnMgYSBwb2ludGVyIHRvIGEgcmVmZXJlbmNlIGltcGxlbWVudGF0aW9u IG9mIGNvZGVjLg0NMi4gQmVjYXVzZSBvZiB0aGUgaHVnZSBleHRlbnQgb2YgdGhlIGNvZGVjknMg c291cmNlIGNvZGUsIGl0IGlzIGxpa2VseSB0aGF0IGV2ZW4gYWZ0ZXIgcHVibGlzaGluZyB0aGUg UkZDLCBidWdzIHdpbGwgYmUgZm91bmQgZnJvbSB0aW1lIHRvIHRpbWUuIEVpdGhlciBhbiBlcnJh dGEgb2YgdGhlIFJGQyBhbmQgaXRzIHNvZnR3YXJlIGRlc2NyaXB0aW9uIE1VU1QgYmUgbWFpbnRh aW5lZCBvciBhIHdlbGwgdmVyc2lvbml6ZWQgc29mdHdhcmUgcmVwb3NpdG9yeSB0aGF0IGluY2x1 ZGVzIHRoZSBjdXJyZW50IHJlZmVyZW5jZSBpbXBsZW1lbnRhdGlvbiBNVVNUIGJlIGtlcHQgcHVi bGljLiBDb21wcmVzc2VkIGZvcm1zIG9mIHRoZSBpbXBsZW1lbnRhdGlvbiBtYXkgbWFrZSBpdCBk aWZmaWN1bHQgdG8ga2VwdCB0cmFjayBvZiBtaW5vciBidWdzIGJlY2F1c2UgdGhlbiB0aGUgd2hv bGUgZmlsZSBtdXN0IGJlIHJlcGxhY2VkLg0NMy4gQWx0aG91Z2ggYSB0ZXh0IGRlc2NyaXB0aW9u IG9mIHRoZSBhbGdvcml0aG0gTVVTVCBiZSBwcm92aWRlZCwgaXRzIHVzZSBzaG91bGQgYmUgbGlt aXRlZCB0byBoZWxwaW5nIHRoZSByZWFkZXIgaW4gdW5kZXJzdGFuZGluZyB0aGUgc291cmNlDSAg ICAgICBjb2RlLiBTaG91bGQgdGhlIGRlc2NyaXB0aW9uIGNvbnRyYWRpY3QgdGhlIHNvdXJjZSBj b2RlLCB0aGUNICAgICAgIGxhdHRlciBzaGFsbCB0YWtlIHByZWNlZGVuY2UuDQ00LiBUaGUgY29k ZWMgc3BlY2lmaWNhdGlvbiBzaGFsbCBiZSAsIGRvY3VtZW50ZWQgaW4gYW4gSW50ZXJuZXQgRHJh ZnQNICAgICAgIGRlc3RpbmVkIGZvciBzdGFuZGFyZHMgdHJhY2sgUkZDLiAgBQ0NVGhpcyBpbXBs ZW1lbnRhdGlvbiB3aWxsIGJlDSAgICAgICB1c2VkIHRvIHZlcmlmeSBjb25mb3JtYW5jZSBvZiBh biBpbXBsZW1lbnRhdGlvbi4gIEFsdGhvdWdoIGEgdGV4dA0gICAgICAgZGVzY3JpcHRpb24gb2Yg dGhlIGFsZ29yaXRobSBzaG91bGQgYmUgcHJvdmlkZWQsIGl0cyB1c2Ugc2hvdWxkDSAgICAgICBi ZSBsaW1pdGVkIHRvIGhlbHBpbmcgdGhlIHJlYWRlciBpbiB1bmRlcnN0YW5kaW5nIHRoZSBzb3Vy Y2UNICAgICAgIGNvZGUuICBTaG91bGQgdGhlIGRlc2NyaXB0aW9uIGNvbnRyYWRpY3QgdGhlIHNv dXJjZSBjb2RlLCB0aGUNICAgICAgIGxhdHRlciBzaGFsbCB0YWtlIHByZWNlZGVuY2UuICBGb3Ig Y29udmVuaWVuY2UsIHRoZSBzb3VyY2UgY29kZQ0gICAgICAgbWF5IGJlIHByb3ZpZGVkIGluIGNv bXByZXNzZWQgZm9ybSwgd2l0aCBiYXNlNjQgZW5jb2RpbmcuDQ01LiBSZXF1aXJlbWVudHMgQ29u Zm9ybWFuY2UNDSAgIEl0IGlzIHRoZSByZXNwb25zaWJpbGl0eSBvZiB0aGUgd29ya2luZyBncm91 cCB0byBkZWZpbmUgY3JpdGVyaWEgZm9yDSAgIGV2YWx1YXRpbmcgY29uZm9ybWFuY2UsIGluY2x1 ZGluZyBidXQgbm90IGxpbWl0ZWQgdG8gY29tcGFyaXNvbiB0b29scw0gICBhbmQgdGVzdCB2ZWN0 b3JzLiBJbiBhbnkgY2FzZSwgdGhlIGNvZGVjIHNwZWNpZmljYXRpb24gd2lsbCBiZQ0gICAgICAg dXNlZCB0byB2ZXJpZnkgY29uZm9ybWFuY2Ugb2YgYW4gaW1wbGVtZW50YXRpb24uICBUaGUgZm9s bG93aW5nIHRleHQgcHJvdmlkZXMgc3VnZ2VzdGlvbnMgZm9yDSAgIGNvbnNpZGVyYXRpb24gYnkg dGhlIHdvcmtpbmcgZ3JvdXAgb24gdmVyaWZ5IHRoZSBjb25mb3JtYW5jZSBvZiB0aGUgY29kZWM6 Cw0FDSAgIDEyLiAgSXQgaXMgdGhlIGludGVudGlvbiBvZiB0aGUgZ3JvdXAgdG8gYWxsb3cgdGhl IGdyZWF0ZXN0IHBvc3NpYmxlDSAgICAgICBjaG9pY2Ugb2YgZnJlZWRvbSBpbiBpbXBsZW1lbnRp bmcgdGhlIHNwZWNpZmljYXRpb24uDSAgICAgICBBY2NvcmRpbmdseSwgdGhlIG51bWJlciBvZiBi aW5kaW5nIFJGQzIxMTkga2V5d29yZHMgaXMgZ29pbmcgdG8NICAgICAgIGJlIHRoZSBtaW5pbXVt IHN0aWxsIGFsbG93aW5nIGZvciBpbnRlcm9wYWJsZSBpbXBsZW1lbnRhdGlvbnMuDSAgICAgICBJ biBwcmFjdGljZSB0aGlzIGdlbmVyYWxseSBtYXkgbWVhbnMgdGhhdCBvbmx5IHRoZSBkZWNvZGVy IG5lZWRzIHRvDSAgICAgICBiZSBub3JtYXRpdmUFLCBzbyB0aGF0IHRoZSBlbmNvZGVyIGNhbiBp bXByb3ZlIG92ZXIgdGltZS4gIFRoaXMNICAgICAgIGFsc28gZW5hYmxlcyBkaWZmZXJlbnQgdHJh ZGVvZmZzIGJldHdlZW4gcXVhbGl0eSBhbmQgY29tcGxleGl0eS4NDSAgIDEzLiAgVG8gcmVkdWNl IHRoZSByaXNrIG9mIGJpYXMgdG93YXJkcyBjZXJ0YWluIENQVS9EU1AgYXJjaGl0ZWN0dXJlcywN ICAgICAgIGlkZWFsbHkgdGhlIGRlY29kZXIgc3BlY2lmaWNhdGlvbiBzaG91bGQgbm90IHJlcXVp cmUgImJpdC1leGFjdCINICAgICAgIGNvbmZvcm1hbmNlIHdpdGggdGhlIHJlZmVyZW5jZSBpbXBs ZW1lbnRhdGlvbi4gIFRoZSBvdXRwdXQgb2YgYQ0gICAgICAgZGVjb2RlciBpbXBsZW1lbnRhdGlv biBzaG91bGQgb25seSBiZSAiY2xvc2UgZW5vdWdoIiB0byB0aGUNICAgICAgIG91dHB1dCBvZiB0 aGUgcmVmZXJlbmNlIGRlY29kZXIuICBBIGNvbXBhcmlzb24gdG9vbCBtYXlzaG91bGQgYmUNICAg ICAgIHByb3ZpZGVkIGFsb25nIHdpdGggdGhlIGNvZGVjIHRvIHZlcmlmeSBvYmplY3RpdmVseSB0 aGF0IHRoZQ0gICAgICAgb3V0cHV0IG9mIGEgZGVjb2RlciBpcyBsaWtlbHkgdG8gYmUgcGVyY2Vw dHVhbGx5DSAgICAgICBpbmRpc3Rpbmd1aXNoYWJsZSBmcm9tIHRoYXQgb2YgdGhlIHJlZmVyZW5j ZSBkZWNvZGVyLiAgSG93ZXZlciwNICAgICAgIGFuIGltcGxlbWVudGF0aW9uIG1heSBzdGlsbCB3 aXNoIHRvIHByb2R1Y2UgYW4gb3V0cHV0IHRoYXQgaXMNICAgICAgIGJpdC1leGFjdCB3aXRoIHRo ZSByZWZlcmVuY2UgaW1wbGVtZW50YXRpb24gdG8gc2ltcGxpZnkgdGhlDSAgICAgICB0ZXN0aW5n IHByb2NlZHVyZS4NDSAgIDI0LiAgVG8gZW5zdXJlIGZyZWVkb20gb2YgaW1wbGVtZW50YXRpb24s IGRlY29kZXItc2lkZSBvbmx5IGVycm9yDSAgICAgICBjb25jZWFsbWVudCBkb2VzIG5vdCBuZWVk IHRvIGJlIHNwZWNpZmllZCwgYWx0aG91Z2ggdGhlIHJlZmVyZW5jZQ0gICAgICAgaW1wbGVtZW50 YXRpb24gbWF5c2hvdWxkIGluY2x1ZGUgdGhlIHNhbWUgUExDIGFsZ29yaXRobSBhcyB1c2VkIGlu DSAgICAgICB0aGUgdGVzdGluZyBwaGFzZS4gIElzIGl0IHVwIHRvIHRoZSB3b3JraW5nIGdyb3Vw IHRvIGRlY2lkZQ0gICAgICAgd2hldGhlciBtaW5pbXVtIHJlcXVpcmVtZW50cyBvbiBQTEMgcXVh bGl0eSB3aWxsIGJlIHJlcXVpcmVkIGZvcg0gICAgICAgY29tcGxpYW5jZSB3aXRoIHRoZSBzcGVj aWZpY2F0aW9uLiAgT2J2aW91c2x5LCBhbnkgaW5mb3JtYXRpb24NICAgICAgIHNpZ25hbGVkIGlu IHRoZSBiaXRzdHJlYW0gaW50ZW5kZWQgdG8gYWlkIFBMQyBuZWVkcyB0byBiZQ0gICAgICAgc3Bl Y2lmaWVkLg0NICAgMzUuICBBbiBlbmNvZGVyIGltcGxlbWVudGF0aW9uIG1heXNob3VsZCBub3Qg YmUgcmVxdWlyZWQgdG8gbWFrZSB1c2Ugb2YNICAgICAgIGFsbCB0aGUgImZlYXR1cmVzIiAodG9v bHMpIGluIHRoZSBiaXQtc3RyZWFtIGRlZmluaXRpb24uDQ0NDVZhbGluLCBldCBhbC4gICAgICAg ICAgICBFeHBpcmVzIEFwcmlsIDE1LCAyMDExICAgICAgICAgICAgICAgICBbUGFnZSA4XQ0MDUlu dGVybmV0LURyYWZ0ICAgICAgICAgICAgICBDb2RlYyBHdWlkZWxpbmVzICAgICAgICAgICAgICAg IE9jdG9iZXIgMjAxMA0NDSAgICAgICBIb3dldmVyLCB0aGUgY29kZWMgc3BlY2lmaWNhdGlvbiBt YXkgcmVxdWlyZSB0aGF0IGFuIGVuY29kZXINICAgICAgIGltcGxlbWVudGF0aW9uIGJlIGFibGUg dG8gZ2VuZXJhdGUgYW55IHBvc3NpYmxlIGJpdC1yYXRlLiAgVW5sZXNzDSAgICAgICBhIHBhcnRp Y3VsYXIgInByb2ZpbGUiIGlzIGRlZmluZWQgaW4gdGhlIHNwZWNpZmljYXRpb24sIHRoZQ0gICAg ICAgZGVjb2RlciBtdXN0IGJlIGFibGUgdG8gZGVjb2RlIGFsbCBmZWF0dXJlcyBvZiB0aGUgYml0 LXN0cmVhbS4NICAgICAgIFRoZSBkZWNvZGVyIG11c3QgYWxzbyBiZSBhYmxlIHRvIGhhbmRsZSBh bnkgY29tYmluYXRpb24gb2YgYml0cywNICAgICAgIGV2ZW4gY29tYmluYXRpb25zIHRoYXQgY2Fu bm90IGJlIGdlbmVyYXRlZCBieSB0aGUgcmVmZXJlbmNlDSAgICAgICBlbmNvZGVyLiAgSXQgbWF5 IGJlIGlzIHJlY29tbWVuZGVkIHRoYXQgdGhlIGRlY29kZXIgc3BlY2lmaWNhdGlvbiBzaGFsbA0g ICAgICAgZGVmaW5lIGV4YWN0bHkgaG93IHRoZSBkZWNvZGVyIHNob3VsZCByZWFjdCB0byAiaW1w b3NzaWJsZSINICAgICAgIHBhY2tldHMuICBIb3dldmVyLCBhbiBlbmNvZGVyIG11c3QgbmV2ZXIg Z2VuZXJhdGUgc3VjaCBwYWNrZXRzDSAgICAgICB0aGF0IGRvIG5vdCBjb25mb3JtIHRvIHRoZSBi aXQtc3RyZWFtIGRlZmluaXRpb24uDQ0gICA2LiAgQ29tcHJlc3NlZCB0ZXN0IHZlY3RvcnMgbWF5 c2hvdWxkIGJlIHByb3ZpZGVkIGFzIGEgbWVhbnMgdG8gdmVyaWZ5DSAgICAgICBjb25mb3JtYW5j ZSB3aXRoIHRoZSBkZWNvZGVyIHNwZWNpZmljYXRpb24uICBUaGVzZSB0ZXN0IHZlY3RvcnMNICAg ICAgIHNob3VsZCBleGVyY2lzZSBhbGwgcGF0aHMgaW4gdGhlIGRlY29kZXIgKDEwMCUgY29kZSBj b3ZlcmFnZSkgaWYgcG9zc2libGUuLg0NICAgNy4gIFdoaWxlIHRoZSBleGFjdCBlbmNvZGVyIG1h eSB3aWxsIG5vdCBiZSBzcGVjaWZpZWQsIGl0IGlzIG1heSBiZSByZWNvbW1lbmRlZA0gICAgICAg dG8gc3BlY2lmeSBvYmplY3RpdmUgbWVhc3VyZW1lbnQgdGFyZ2V0cyBmb3IgYW4gZW5jb2Rlciwg YmVsb3cNICAgICAgIHdoaWNoIHVzZSBvZiBhIHBhcnRpY3VsYXIgZW5jb2RlciBpbXBsZW1lbnRh dGlvbiBpcyBub3QNICAgICAgIHJlY29tbWVuZGVkLiAgRm9yIGV4YW1wbGUsIG9uZSBzdWNoIHNw ZWNpZmljYXRpb24gY291bGQgYmU6ICJ0aGUNICAgICAgIHVzZSBvZiBhbiBlbmNvZGVyIHdob3Nl IFBFU1EgTU9TIGlzIGxlc3MgdGhhbiAwLjEgYmVsb3cgdGhlDSAgICAgICByZWZlcmVuY2UgZW5j b2RlciBpbiB0aGUgZm9sbG93aW5nIGNvbmRpdGlvbnMgaXMgbm90DSAgICAgICByZWNvbW1lbmRl ZCIuDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NVmFsaW4sIGV0IGFsLiAgICAgICAgICAg IEV4cGlyZXMgQXByaWwgMTUsIDIwMTEgICAgICAgICAgICAgICAgIFtQYWdlIDldDQwNSW50ZXJu ZXQtRHJhZnQgICAgICAgICAgICAgIENvZGVjIEd1aWRlbGluZXMgICAgICAgICAgICAgICAgT2N0 b2JlciAyMDEwDQ0NNS4gIEludGVsbGVjdHVhbCBQcm9wZXJ0eQ0NICAgUHJvZHVjaW5nIGFuIHVu ZW5jdW1iZXJlZCBjb2RlYyBpcyBkZXNpcmFibGUgZm9yIHRoZSBmb2xsb3dpbmcNICAgcmVhc29u czoNDSAgIG8gIEl0IGlzIHRoZSBleHBlcmllbmNlIG9mIGEgd2lkZSB2YXJpZXR5IG9mIGFwcGxp Y2F0aW9uIGRldmVsb3BlcnMNICAgICAgYW5kIHNlcnZpY2UgcHJvdmlkZXJzIHRoYXQgZW5jdW1i cmFuY2VzIHN1Y2ggYXMgbGljZW5zaW5nIGFuZA0gICAgICByb3lhbHRpZXMgbWFrZSBpdCBkaWZm aWN1bHQgdG8gaW1wbGVtZW50LCBkZXBsb3ksIGFuZCBkaXN0cmlidXRlDSAgICAgIGF1ZGlvIGFw cGxpY2F0aW9ucyBmb3IgdXNlIGJ5IHRoZSBJbnRlcm5ldCBjb21tdW5pdHkuDQ0gICBvICBJdCBp cyBiZW5lZmljaWFsIHRvIGhhdmUgbG93LWNvc3Qgb3B0aW9ucyB3aGVuZXZlciBwb3NzaWJsZQ0g ICAgICBiZWNhdXNlIHN0YW5kYWxvbmUgdm9pY2Ugc2VydmljZXMgYXJlIGJlaW5nIGNvbW1vZGl0 aXplZCBhbmQNICAgICAgc21hbGwsIGlubm92YXRpdmUgZGV2ZWxvcG1lbnQgdGVhbXMgb2Z0ZW4g Y2Fubm90IGFmZm9yZCB0byBwYXkNICAgICAgcGVyLWNoYW5uZWwgbGljZW5zaW5nIGZlZXMgYW5k IHJveWFsdGllcy4NDSAgIG8gIE1hbnkgbWFya2V0IHNlZ21lbnRzIGFyZSBtb3ZpbmcgYXdheSBm cm9tIHNlbGxpbmcgaGFyZC1jb2RlZA0gICAgICBoYXJkd2FyZSBkZXZpY2VzIGFuZCB0b3dhcmQg ZnJlZWx5IGRpc3RyaWJ1dGluZyBlbmQtdXNlciBzb2Z0d2FyZTsNICAgICAgdGhpcyBpcyB0cnVl IG9mIG51bWVyb3VzIGxhcmdlIGFwcGxpY2F0aW9uIHByb3ZpZGVycyBhbmQgZXZlbg0gICAgICB0 ZWxjb3MgdGhlbXNlbHZlcy4NDSAgIG8gIENvbXBhdGliaWxpdHkgd2l0aCB0aGUgbGljZW5zaW5n IG9mIHR5cGljYWwgb3BlbiBzb3VyY2UNICAgICAgYXBwbGljYXRpb25zIGltcGxpZXMgdGhlIG5l ZWQgdG8gYXZvaWQgZW5jdW1icmFuY2VzLCBpbmNsdWRpbmcNICAgICAgZXZlbiB0aGUgcmVxdWly ZW1lbnQgdG8gb2J0YWluIGEgbGljZW5zZSBmb3IgaW1wbGVtZW50YXRpb24sDSAgICAgIGRlcGxv eW1lbnQsIG9yIHVzZSAoZXZlbiBpZiB0aGUgbGljZW5zZSBkb2VzIG5vdCByZXF1aXJlIHRoZQ0g ICAgICBwYXltZW50IG9mIGEgZmVlKS4NDSAgIFRoZXJlZm9yZSwgYSBjb2RlYyB0aGF0IGNhbiBi ZSB3aWRlbHkgaW1wbGVtZW50ZWQgYW5kIGVhc2lseQ0gICBkaXN0cmlidXRlZCBhbW9uZyBhcHBs aWNhdGlvbiBkZXZlbG9wZXJzLCBzZXJ2aWNlIG9wZXJhdG9ycywgYW5kIGVuZA0gICB1c2VycyBp cyBwcmVmZXJyZWQuICBNYW55IGV4aXN0aW5nIGNvZGVjcyB0aGF0IG1pZ2h0IGZ1bGZpbGwgc29t ZSBvcg0gICBtb3N0IG9mIHRoZSB0ZWNobmljYWwgYXR0cmlidXRlcyBsaXN0ZWQgYWJvdmUgYXJl IGVuY3VtYmVyZWQgaW4NICAgdmFyaW91cyB3YXlzLiAgRm9yIGV4YW1wbGUsIHBhdGVudCBob2xk ZXJzIG1pZ2h0IHJlcXVpcmUgdGhhdCB0aG9zZQ0gICB3aXNoaW5nIHRvIGltcGxlbWVudCB0aGUg Y29kZWMgaW4gc29mdHdhcmUsIGRlcGxveSB0aGUgY29kZWMgaW4gYQ0gICBzZXJ2aWNlLCBvciBk aXN0cmlidXRlIHRoZSBjb2RlYyBpbiBzb2Z0d2FyZSBvciBoYXJkd2FyZSBuZWVkIHRvDSAgIHJl cXVlc3QgYSBsaWNlbnNlLCBlbnRlciBpbnRvIGEgYnVzaW5lc3MgYWdyZWVtZW50LCBwYXkgbGlj ZW5zaW5nDSAgIGZlZXMgb3Igcm95YWx0aWVzLCBvciBhZGhlcmUgdG8gb3RoZXIgc3BlY2lhbCBj b25kaXRpb25zIG9yDSAgIHJlc3RyaWN0aW9ucy4gIEJlY2F1c2Ugc3VjaCBlbmN1bWJyYW5jZXMg aGF2ZSBtYWRlIGl0IGRpZmZpY3VsdCB0bw0gICB3aWRlbHkgaW1wbGVtZW50IGFuZCBlYXNpbHkg ZGlzdHJpYnV0ZSBoaWdoLXF1YWxpdHkgYXVkaW8gY29kZWNzDSAgIGFjcm9zcyB0aGUgZW50aXJl IEludGVybmV0IGNvbW11bml0eSwgdGhlIHdvcmtpbmcgZ3JvdXAgcHJlZmVycw0gICB1bmVuY3Vt YmVyZWQgdGVjaG5vbG9naWVzIGluIGEgd2F5IHRoYXQgaXMgY29uc2lzdGVudCB3aXRoIEJDUCA3 OCBhbmQNICAgQkNQIDc5LiAgSW4gcGFydGljdWxhciwgdGhlIHdvcmtpbmcgZ3JvdXAgc2hhbGwg aGVlZCB0aGUgcHJlZmVyZW5jZQ0gICBzdGF0ZWQgaW4gQkNQIDc5OiAiSW4gZ2VuZXJhbCwgSUVU RiB3b3JraW5nIGdyb3VwcyBwcmVmZXINICAgdGVjaG5vbG9naWVzIHdpdGggbm8ga25vd24gSVBS IGNsYWltcyBvciwgZm9yIHRlY2hub2xvZ2llcyB3aXRoDSAgIGNsYWltcyBhZ2FpbnN0IHRoZW0s IGFuIG9mZmVyIG9mIHJveWFsdHktZnJlZSBsaWNlbnNpbmcuIiAgQWx0aG91Z2gNICAgdGhpcyBw cmVmZXJlbmNlIGNhbm5vdCBndWFyYW50ZWUgdGhhdCB0aGUgd29ya2luZyBncm91cCB3aWxsIHBy b2R1Y2UNICAgYW4gdW5lbmN1bWJlcmVkIGNvZGVjLCB0aGUgd29ya2luZyBncm91cCBzaGFsbCBm b2xsb3cgQkNQIDc5LCBhbmQNICAgYWRoZXJlIHRvIHRoZSBzcGlyaXQgb2YgQkNQIDc5LiAgVGhl IHdvcmtpbmcgZ3JvdXAgY2Fubm90IGV4cGxpY2l0bHkNICAgcnVsZSBvdXQgdGhlIHBvc3NpYmls aXR5IG9mIGFkb3B0aW5nIGVuY3VtYmVyZWQgdGVjaG5vbG9naWVzOw0gICBob3dldmVyLCB0aGUg d29ya2luZyBncm91cCB3aWxsIHRyeSB0byBhdm9pZCBlbmN1bWJlcmVkIHRlY2hub2xvZ2llcw0N DQ1WYWxpbiwgZXQgYWwuICAgICAgICAgICAgRXhwaXJlcyBBcHJpbCAxNSwgMjAxMSAgICAgICAg ICAgICAgICBbUGFnZSAxMF0NDA1JbnRlcm5ldC1EcmFmdCAgICAgICAgICAgICAgQ29kZWMgR3Vp ZGVsaW5lcyAgICAgICAgICAgICAgICBPY3RvYmVyIDIwMTANDQ0gICB0aGF0IHJlcXVpcmUgcm95 YWx0aWVzIG9yIG90aGVyIGVuY3VtYnJhbmNlcyB0aGF0IHdvdWxkIHByZXZlbnQgc3VjaA0gICB0 ZWNobm9sb2dpZXMgZnJvbSBiZWluZyBlYXN5IHRvIHJlZGlzdHJpYnV0ZSBhbmQgdXNlLg0NICAg VGhlIGZvbGxvd2luZyBndWlkZWxpbmVzIHdpbGwgaGVscCB0byBtYXhpbWl6ZSB0aGUgb2RkcyB0 aGF0IHRoZQ0gICBjb2RlYyB3aWxsIGJlIHVuZW5jdW1iZXJlZDoNDSAgIDEuICBJbiBhY2NvcmRh bmNlIHdpdGggQkNQIDc5IFtJUFJdLCBjb250cmlidXRlZCBjb2RlY3Mgc2hvdWxkDSAgICAgICBw cmVmZXJhYmx5IHVzZSB0ZWNobm9sb2dpZXMgd2l0aCBubyBrbm93biBJUFIgY2xhaW1zIG9yDSAg ICAgICB0ZWNobm9sb2dpZXMgd2l0aCBhbiBvZmZlciBvZiByb3lhbHR5LWZyZWUgKFJGKSBsaWNl bnNpbmcuIFJGIGxpY2Vuc2luZyBzaGFsbCBhbHNvIGluY2x1ZGUgYW4gcHJpb3Itc3RhbmRhcmQg dGVzdGluZyBhbmQgZGV2ZWxvcG1lbnQgcGhhc2UgYW5kIGFuIHBvc3Qtc3RhbmRhcmQgYWR2YW5j ZW1lbnQgYW5kIGV4dGVuc2lvbiBwaGFzZS4NDSAgIDIuICBXaGVuZXZlciBwb3NzaWJsZSwgdGhl IHdvcmtpbmcgZ3JvdXAgc2hvdWxkIHVzZSB0ZWNobm9sb2dpZXMgdGhhdA0gICAgICAgYXJlIHBl cmNlaXZlZCBieSB0aGUgcGFydGljaXBhbnRzIHRvIGJlIHNhZmVyIHdpdGggcmVnYXJkIHRvIElQ Ug0gICAgICAgaXNzdWVzLg0NICAgMy4gIENvbnRyaWJ1dG9ycyBtdXN0IGRpc2Nsb3NlIElQUiBh cyBzcGVjaWZpZWQgaW4gQkNQIDc5Lg0NICAgNC4gIEluIGNhc2VzIHdoZXJlIG5vIFJGIGxpY2Vu c2UgY2FuIGJlIG9idGFpbmVkIHJlZ2FyZGluZyBhIHBhdGVudCwNICAgICAgIHRoZSBncm91cCBz aG91bGQgY29uc2lkZXIgYWx0ZXJuYXRpdmUgYWxnb3JpdGhtcyBvciBtZXRob2RzLCBldmVuDSAg ICAgICBpZiB0aGV5IHJlc3VsdCBpbiBsb3dlciBxdWFsaXR5LCBoaWdoZXIgY29tcGxleGl0eSwg b3Igb3RoZXJ3aXNlDSAgICAgICBsZXNzIGRlc2lyYWJsZSBjaGFyYWN0ZXJpc3RpY3MgKGluIG1v c3QgY2FzZXMsIHRoZSBkZWdyYWRhdGlvbg0gICAgICAgd2lsbCBsaWtlbHkgYmUgc21hbGwgb25j ZSB0aGUgYmVzdCBhbHRlcm5hdGl2ZSBoYXMgYmVlbg0gICAgICAgaWRlbnRpZmllZCkuDQ0gICA1 LiAgSW4gYWNjb3JkYW5jZSB3aXRoIEJDUCA3OCBbVFJVU1RdLCB0aGUgc291cmNlIGNvZGUgZm9y IHRoZQ0gICAgICAgcmVmZXJlbmNlIGltcGxlbWVudGF0aW9uIG11c3QgYmUgbWFkZSBhdmFpbGFi bGUgdW5kZXIgYSBCU0Qtc3R5bGUNICAgICAgIGxpY2Vuc2UgKG9yIHdoYXRldmVyIGxpY2Vuc2Ug aXMgZGVmaW5lZCBhcyBhY2NlcHRhYmxlIGJ5IHRoZSBJRVRGDSAgICAgICBUcnVzdCB3aGVuIHRo ZSBJbnRlcm5ldC1EcmFmdCBkZWZpbmluZyB0aGUgcmVmZXJlbmNlDSAgICAgICBpbXBsZW1lbnRh dGlvbiBpcyBwdWJsaXNoZWQpLg0NICAgSUVURiBwYXJ0aWNpcGFudHMgc2hvdWxkIGJlIGF3YXJl IHRoYXQsIGdpdmVuIHRoZSB3YXkgcGF0ZW50cyB3b3JrIGluDSAgIG1vc3QgY291bnRyaWVzLCB0 aGUgcmVzdWx0aW5nIGNvZGVjIGNhbiBuZXZlciBiZSBndWFyYW50ZWVkIHRvIGJlDSAgIGZyZWUg b2YgcGF0ZW50IGNsYWltcyBiZWNhdXNlIHNvbWUgcGF0ZW50cyBtYXkgbm90IGJlIGtub3duIHRv IHRoZQ0gICBjb250cmlidXRvcnMsIHNvbWUgcGF0ZW50IGFwcGxpY2F0aW9ucyBtYXkgbm90IGJl IGRpc2Nsb3NlZCBhdCB0aGUNICAgdGltZSB0aGUgY29kZWMgaXMgZGV2ZWxvcGVkLCBhbmQgb25s eSBjb3VydHMgb2YgbGF3IGNhbiBkZXRlcm1pbmUgdGhlDSAgIHZhbGlkaXR5IGFuZCBicmVhZHRo IG9mIHBhdGVudCBjbGFpbXMuICBIb3dldmVyLCB0aGVzZSBvYnNlcnZhdGlvbnMNICAgYXJlIG5v IGRpZmZlcmVudCB3aXRoaW4gdGhlIEludGVybmV0IFN0YW5kYXJkcyBQcm9jZXNzIHRoYW4gdGhl eSBhcmUNICAgZm9yIHN0YW5kYXJkaXphdGlvbiBvZiBjb2RlY3Mgd2l0aGluIG90aGVyIFNET3Mg KG9yIGRldmVsb3BtZW50IG9mDSAgIGNvZGVjcyBvdXRzaWRlIHRoZSBjb250ZXh0IG9mIGFueSBT RE8pLCBhbmQgZnVydGhlcm1vcmUgYXJlIG5vDSAgIGRpZmZlcmVudCBmb3IgY29kZWNzIHRoYW4g Zm9yIG90aGVyIHRlY2hub2xvZ2llcyB3b3JrZWQgb24gd2l0aGluIHRoZQ0gICBJRVRGLiAgSW4g YWxsIHRoZXNlIGNhc2VzLCB0aGUgYmVzdCBhcHByb2FjaCBpcyB0byBtaW5pbWl6ZSB0aGUgcmlz aw0gICBvZiB1bmtub3dpbmdseSBpbmN1cnJpbmcgZW5jcnVtYnJhbmNlIG9uIGV4aXN0aW5nIHBh dGVudHMuICBEZXNwaXRlDSAgIHRoZXNlIHByZWNhdXRpb25zLCBwYXJ0aWNpcGFudHMgbmVlZCB0 byB1bmRlcnN0YW5kIHRoYXQsIHByYWN0aWNhbGx5DSAgIHNwZWFraW5nLCBpdCBpcyBuZWFybHkg aW1wb3NzaWJsZSB0byBfZ3VhcmFudGVlXyB0aGF0IGltcGxlbWVudG9ycw0gICB3aWxsIG5vdCBp bmN1ciBlbmN1bWJyYW5jZSBvbiBleGlzdGluZyBwYXRlbnRzLg0NDQ0NDQ0NVmFsaW4sIGV0IGFs LiAgICAgICAgICAgIEV4cGlyZXMgQXByaWwgMTUsIDIwMTEgICAgICAgICAgICAgICAgW1BhZ2Ug MTFdDQwNSW50ZXJuZXQtRHJhZnQgICAgICAgICAgICAgIENvZGVjIEd1aWRlbGluZXMgICAgICAg ICAgICAgICAgT2N0b2JlciAyMDEwDQ0NNi4gIFJlbGF0aW9uc2hpcCB3aXRoIE90aGVyIFNET3MN DSAgIEl0IGlzIHVuZGVyc3Rvb2QgdGhhdCBvdGhlciBTRE9zIGFyZSBhbHNvIGludm9sdmVkIGlu IHRoZSBjb2RlYw0gICBkZXZlbG9wbWVudCBhbmQgc3RhbmRhcmRpemF0aW9uLCBpbmNsdWRpbmcg YnV0IG5vdCBuZWNlc3NhcmlseQ0gICBsaW1pdGVkIHRvOg0NICAgbyAgVGhlIFRlbGVjb21tdW5p Y2F0aW9uIFN0YW5kYXJkaXphdGlvbiBTZWN0b3IgKElUVS1UKSBvZiB0aGUNICAgICAgSW50ZXJu YXRpb25hbCBUZWxlY29tbXVuaWNhdGlvbiBVbmlvbiAoSVRVKSwgaW4gcGFydGljdWxhciBTdHVk eQ0gICAgICBHcm91cCAxNg0NICAgbyAgVGhlIE1vdmluZyBQaWN0dXJlIEV4cGVydHMgR3JvdXAg KE1QRUcpDQ0gICBvICBUaGUgRXVyb3BlYW4gVGVsZWNvbW11bmljYXRpb25zIFN0YW5kYXJkcyBJ bnN0aXR1dGUgKEVUU0kpDQ0gICBvICBUaGUgM3JkIEdlbmVyYXRpb24gUGFydG5lcnNoaXAgUHJv amVjdCAoM0dQUCkNDSAgIG8gIFRoZSAzcmQgR2VuZXJhdGlvbiBQYXJ0bmVyc2hpcCBQcm9qZWN0 IDIgKDNHUFAyKQ0NICAgSXQgaXMgaW1wb3J0YW50IHRvIGVuc3VyZSB0aGF0IHN1Y2ggd29yayBk b2VzIG5vdCBjb25zdGl0dXRlDSAgIHVuY29vcmRpbmF0ZWQgcHJvdG9jb2wgZGV2ZWxvcG1lbnQs IG9mIHRoZSBraW5kIGRlc2NyaWJlZCBpbg0gICBbVU5DT09SRF0gaW4gdGhlIGZvbGxvd2luZyBw cmluY2lwbGU6DQ0gICAgICBbVF1oZSBJQUIgY29uc2lkZXJzIGFuIGVzc2VudGlhbCBwcmluY2lw bGUgb2YgdGhlIHByb3RvY29sDSAgICAgIGRldmVsb3BtZW50IHByb2Nlc3MgdGhhdCBvbmx5IG9u ZSBTRE8gbWFpbnRhaW5zIGRlc2lnbiBhdXRob3JpdHkNICAgICAgZm9yIGEgZ2l2ZW4gcHJvdG9j b2wsIHdpdGggdGhhdCBTRE8gaGF2aW5nIHVsdGltYXRlIGF1dGhvcml0eSBvdmVyDSAgICAgIHRo ZSBhbGxvY2F0aW9uIG9mIHByb3RvY29sIHBhcmFtZXRlciBjb2RlLXBvaW50czsgZGVmaW5pbmcg dGhlDSAgICAgIGludGVuZGVkIHNlbWFudGljcywgaW50ZXJwcmV0YXRpb24sIGFuZCBhY3Rpb25z IGFzc29jaWF0ZWQgd2l0aA0gICAgICB0aG9zZSBjb2RlLXBvaW50cy4NDSAgIFRoZSB3b3JrIGVu dmlzaW9uZWQgYnkgdGhpcyBndWlkZWxpbmVzIGRvY3VtZW50IGlzIG5vdA0gICAidW5jb29yZGlu YXRlZCIgaW4gdGhlIHNlbnNlIGRlc2NyaWJlZCBpbiB0aGUgZm9yZWdvaW5nIHF1b3RlLCBmb3IN ICAgdGhlIGZvbGxvd2luZyByZWFzb25zOg0NICAgbyAgSW50ZXJuZXQgc2lnbmFsbGluZyB0ZWNo bm9sb2dpZXMgYXJlIGRlc2lnbmVkIHRvIGVuYWJsZSB0aGUNICAgICAgbmVnb3RpYXRpb24gb2Yg YW55IGNvZGVjcyB0aGF0IGFyZSBzdXBwb3J0ZWQgaW4gYSBwYXJ0aWN1bGFyDSAgICAgIGFwcGxp Y2F0aW9uIChzdWNoIHNpZ25hbGxpbmcgdGVjaG5vbG9naWVzIGluY2x1ZGUgdGhlIFNlc3Npb24N ICAgICAgSW5pdGlhdGlvbiBQcm90b2NvbCBbU0lQXSwgU2Vzc2lvbiBEZXNjcmlwdGlvbiBQcm90 b2NvbCBbU0RQXSwgYW5kDSAgICAgIHRoZSBFeHRlbnNpYmxlIE1lc3NhZ2luZyBhbmQgUHJlc2Vu Y2UgUHJvdG9jb2wgW1hNUFBdIGV4dGVuc2lvbnMNICAgICAgZm9yIG1lZGlhIG5lZ290aWF0aW9u IGFzIHNwZWNpZmllZCBpbiBbSmluZ2xlXSkuDQ0gICBvICBJbnRlcm5ldCB0cmFuc3BvcnQgdGVj aG5vbG9naWVzIHN1Y2ggYXMgdGhlIFJlYWwtdGltZSBUcmFuc3BvcnQNICAgICAgUHJvdG9jb2wg W1JUUF0gKGluY2x1ZGluZyBzZWN1cmUgdHJhbnNwb3J0IGFzIGRlc2NyaWJlZCBpbiBbU1JUUF0p DSAgICAgIGFyZSBkZXNpZ25lZCB0byBzdXBwb3J0IGFueSBjb2RlYyBmb3Igd2hpY2ggUlRQIHBh Y2tldGl6YXRpb24NICAgICAgcnVsZXMgaGF2ZSBiZWVuIGRlZmluZWQuDQ0gICBvICBUaGUgSUVU RiBjb2RlYyB3b3JraW5nIGdyb3VwIHdpbGwgZm9jdXMgb24gaXNzdWVzIHRoYXQgYXJlDSAgICAg IHNwZWNpZmljIHRvIHRoZSBJbnRlcm5ldCwgaW5jbHVkaW5nIHJvYnVzdG5lc3MgdG8gcGFja2V0 IGxvc3MgYW5kDSAgICAgIG90aGVyIGFzcGVjdHMgb2YgcGFja2V0IHRyYW5zbWlzc2lvbiBvdmVy IHRoZSBJbnRlcm5ldC4gIElzc3Vlcw0NDQ1WYWxpbiwgZXQgYWwuICAgICAgICAgICAgRXhwaXJl cyBBcHJpbCAxNSwgMjAxMSAgICAgICAgICAgICAgICBbUGFnZSAxMl0NDA1JbnRlcm5ldC1EcmFm dCAgICAgICAgICAgICAgQ29kZWMgR3VpZGVsaW5lcyAgICAgICAgICAgICAgICBPY3RvYmVyIDIw MTANDQ0gICAgICB0aGF0IGFyZSBzcGVjaWZpYyB0byBub24tSW50ZXJuZXQgdHJhbnNwb3J0cyAo ZS5nLiwgcmFkaW8NICAgICAgY29tbXVuaWNhdGlvbiBhbmQgY2lyY3VpdC1zd2l0Y2hlZCBuZXR3 b3JrcykgYXJlIHNwZWNpZmljYWxseSBvdXQNICAgICAgb2Ygc2NvcGUuDQ0gICBBbHRob3VnaCB0 aGVyZSBpcyBhbHJlYWR5IHN1ZmZpY2llbnQgY29kZWMgZXhwZXJ0aXNlIGF2YWlsYWJsZSBhbW9u Zw0gICBJRVRGIHBhcnRpY2lwYW50cyB0byBjb21wbGV0ZSB0aGUgZW52aXNpb25lZCB3b3JrLCBh ZGRpdGlvbmFsDSAgIGNvbnRyaWJ1dGlvbnMgYXJlIHdlbGNvbWUgd2l0aGluIHRoZSBmcmFtZXdv cmsgb2YgdGhlIEludGVybmV0DSAgIFN0YW5kYXJkcyBQcm9jZXNzLCBpbiB0aGUgZm9sbG93aW5n IHdheXM6DQ0gICBvICBJbmRpdmlkdWFscyB3aG8gYXJlIHRlY2huaWNhbCBjb250cmlidXRvcnMg dG8gY29kZWMgd29yayB3aXRoaW4NICAgICAgb3RoZXIgU0RPcyBjYW4gcGFydGljaXBhdGUgZGly ZWN0bHkgaW4gY29kZWMgd29yayB3aXRoaW4gdGhlIElFVEYuDQ0gICBvICBPdGhlciBTRE9zIGNh biBjb250cmlidXRlIHRoZWlyIGV4cGVydGlzZSAoZS5nLiwgY29kZWMNICAgICAgY2hhcmFjdGVy aXphdGlvbiBhbmQgZXZhbHVhdGlvbiB0ZWNobmlxdWVzKSBhbmQgdGh1cyBmYWNpbGl0YXRlDSAg ICAgIHRoZSB0ZXN0aW5nIG9mIGEgY29kZWMgcHJvZHVjZWQgYnkgdGhlIElFVEYuDQ0gICBvICBB bnkgU0RPIGNhbiBwcm92aWRlIGlucHV0IHRvIElFVEYgd29yayB0aHJvdWdoIGxpYWlzb24gc3Rh dGVtZW50cy4NDSAgIEhvd2V2ZXIsIGl0IGlzIGltcG9ydGFudCB0byBub3RlIHRoYXQgZmluYWwg cmVzcG9uc2liaWxpdHkgZm9yIHRoZQ0gICBkZXZlbG9wbWVudCBwcm9jZXNzIGFuZCB0aGUgcmVz dWx0aW5nIGNvZGVjIHdpbGwgcmVtYWluIHdpdGggdGhlIElFVEYNICAgYXMgZ292ZXJuZWQgYnkg QkNQIDkgW1BST0NFU1NdLg0NICAgRmluYWxseSwgdGhlcmUgaXMgcHJlY2VkZW50IGZvciB0aGUg Y29udHJpYnV0aW9uIG9mIGNvZGVjcyBkZXZlbG9wZWQNICAgZWxzZXdoZXJlIHRvIHRoZSBJVFUt VCAoZS5nLiwgQU1SIFdpZGViYW5kIHdhcyBzdGFuZGFyZGl6ZWQNICAgb3JpZ2luYWxseSB3aXRo aW4gM0dQUCkuICBUaGlzIGlzIGEgbW9kZWwgdG8gZXhwbG9yZSBhcyB0aGUgSUVURg0gICBjb29y ZGluYXRlcyBmdXJ0aGVyIHdpdGggdGhlIElUVS1UIGluIGFjY29yZGFuY2Ugd2l0aCB0aGUNICAg Y29sbGFib3JhdGlvbiBndWlkZWxpbmVzIGRlZmluZWQgaW4gW0NPTExBQl0uDQ03LiAgQ29sbGFi b3JhdGlvbiB3aXRoIE90aGVyIElFVEYgV29ya2luZyBHcm91cHMNDVRoZSBjb2RlYyBzaGFsbCBi ZSBvcHRpbWl6ZSB0byBzcGVjaWFsIHByb3BlcnRpZXMgb2YgdGhlIGludGVybmV0IGFuZCBzaG91 bGQgZml0IHdpbGwgaW50byB0aGUgSW50ZXJuZXQgcHJvdG9jb2wgYXJjaGl0ZWN0dXJlLiAgVGh1 cywgdGhlIGNvZGVjIHdvcmtpbmcgZ3JvdXAgc2hhbGwgY29vcGVyYXRlIHdpdGggdGhlIGZvbGxv d2luZyB3b3JraW5nIGdyb3Vwcy4NDVdpdGggdGhlIEFWVCBXRyB0byBkZWZpbmUgdGhlIGNvZGVj J3MgcGF5bG9hZCBmb3JtYXQgZm9yIHVzZSB3aXRoIHRoZQtSZWFsLXRpbWUgVHJhbnNwb3J0IFBy b3RvY29sIChSVFApIGFuZCBkaXNjdXNzIGlzc3VlcyByZWdhcmRpbmcgdGhlIHNlbGVjdGlvbiBv ZiBwcm9wZXIgY29kaW5nIG1vZGVzIGFuZCByYXRlcy4NQ29sbGFib3JhdGUgd2l0aCB3b3JraW5n IGdyb3VwcyBpbiB0aGUgVHJhbnNwb3J0IEFyZWEgdG8gaWRlbnRpZnkgaW1wb3J0YW50IGFzcGVj dHMgb2YgcGFja2V0IHRyYW5zbWlzc2lvbiBvdmVyIHRoZSBJbnRlcm5ldC4gVGhpcyBjb2xsYWJv cmF0aW9uIGVzcGVjaWFsbHkgYWRkcmVzc2VzIHRoZSBmcmVxdWVuY3kgYW5kIGNoYXJhY3Rlcmlz dGljcyBvZiBwYWNrZXQgbG9zc2VzLCBkZWxheSB2YXJpYW5jZXMgYW5kIHRoZSB0ZW1wb3JhbCBj b25uZWN0aW9uIG91dHRhZ2VzLiBBcyBSVFAgcnVucyBvbiB0b3Agb2YgdHJhbnNwb3J0IHByb3Rv Y29scywgdGhpcyBjb3ZlcnMgbm90IG9ubHkgcmF3IElQIHRyYW5zbWlzc2lvbiBvdmVyIHRoZSBJ bnRlcm5ldCBidXQgaW4gYWRkaXRpb24sIHRoZSBlZmZlY3Qgb2YgdHJhbnNwb3J0IHByb3RvY29s cyBzdWNoIGFzIFRDUCBhbmQgRENDUC4gVGhlc2UgaW5mb3JtYXRpb24gYXJlIGltcG9ydGFudCBh cyB0aGV5IGVmZmVjdCB0aGUgcGFja2V0IGxvc3MgY29uY2VhbG1lbnQgYWxnb3JpdGhtcyBhbmQg dGhlIGFsZ29yaXRobXMgdG8gY29uY2VhbCBjaGFuZ2VzIG9mIHRoZSBjb2RpbmcgcGFyYW1ldGVy cyBhbmQgcGxheW91dCB0aW1lLgsNQ29sbGFib3JhdGUgd2l0aCB3b3JraW5nIGdyb3VwcyBpbiB0 aGUgVHJhbnNwb3J0IEFyZWEgdG8gdW5kZXJzdGFuZAt0aGUgZGVncmVlIG9mIHJhdGUgYWRhcHRh dGlvbiBhbmQgY29uZ2VzdGlvbiBjb250cm9sIGRlc2lyYWJsZSwgYW5kIHRvIHJlZmxlY3QgdGhh dAt1bmRlcnN0YW5kaW5nIGluIHRoZSBkZXNpZ24gb2YgYSBjb2RlYyB0aGF0IGNhbiBhZGp1c3Qg aXN0IHRyYW5zbWlzc2lvbiBpbiBhIHdheSB0aGF0IG1pbmltaXplcyBkaXNydXB0aW9uIHRvIHRo ZSBhdWRpby4NDUNvbGxhYm9yYXRlIHdpdGggd29ya2luZyBncm91cHMgaW4gdGhlIFJBSSBBcmVh IHRvIGVuc3VyZSB0aGF0IGluZm9ybWF0aW9uIGFib3V0IGFuZCBuZWdvdGlhdGlvbiBvZiB0aGUg Y29kZWMgY2FuIGJlIGVhc2lseSByZXByZXNlbnRlZCBhdCB0aGUgc2lnbmFsaW5nIGxheWVyLiBU aGUgY29sbGFib3JhdGUgbWF5IGFsc28gYWRkcmVzcyB0aGUgaXNzdWUgd2hldGhlciB0aGUgbmVn b2F0aW9uIG9mIGNvZGVjIHBhcmFtZXRlcnMgc2hhbGwgb2NjdXIgd2l0aCBzaWduYWxsaW5nIHBy b3RvY29scyBsaWtlIFNJUCBvciBILjMyMyBvciBpbi1iYW5kIHVzaW5nIHRoZSBiaWRpcmVjdGlv bmFsIGF1ZGlvIHRyYW5zbWlzc2lvbiBwYXRoLg0NVGhlIGNvbGxhYm9yYXRpb24gd2l0aCBvdGhl ciBJRVRGIFdHIHNoYWxsIHRha2UgcGxhY2UgZHVyaW5nIHRoZSByZWd1bGFyIElFVEYgbWVldGlu Zy4gRm9yIGV4YW1wbGUsIGJyZWFrb3V0IG1lZXRpbmdzIG1heSBiZSBzY2hlZHVsZWQgdG8gYWRk cmVzcyB0aGUgYWJvdmUgbWVudGlvbmVkIHRvcGljcy4NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ1WYWxp biwgZXQgYWwuICAgICAgICAgICAgRXhwaXJlcyBBcHJpbCAxNSwgMjAxMSAgICAgICAgICAgICAg ICBbUGFnZSAxM10NDA1JbnRlcm5ldC1EcmFmdCAgICAgICAgICAgICAgQ29kZWMgR3VpZGVsaW5l cyAgICAgICAgICAgICAgICBPY3RvYmVyIDIwMTANDQ03LiAgU2VjdXJpdHkgQ29uc2lkZXJhdGlv bnMNDSAgIFRoZSBwcm9jZWR1cmFsIGd1aWRlbGluZXMgZm9yIGNvZGVjIGRldmVsb3BtZW50IGRv IG5vdCBoYXZlIHNlY3VyaXR5DSAgIGNvbnNpZGVyYXRpb25zLiAgSG93ZXZlciwgdGhlIHJlc3Vs dGluZyBjb2RlYyBuZWVkcyB0byB0YWtlDSAgIGFwcHJvcHJpYXRlIHNlY3VyaXR5IGNvbnNpZGVy YXRpb25zIGludG8gYWNjb3VudCwgZm9yIGV4YW1wbGUgYXMNICAgb3V0bGluZWQgaW4gW0RPU10g YW5kIFtTRUNHVUlERV0uDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0N DQ0NDVZhbGluLCBldCBhbC4gICAgICAgICAgICBFeHBpcmVzIEFwcmlsIDE1LCAyMDExICAgICAg ICAgICAgICAgIFtQYWdlIDE0XQ0MDUludGVybmV0LURyYWZ0ICAgICAgICAgICAgICBDb2RlYyBH dWlkZWxpbmVzICAgICAgICAgICAgICAgIE9jdG9iZXIgMjAxMA0NDTguICBJQU5BIENvbnNpZGVy YXRpb25zDQ0gICBUaGlzIGRvY3VtZW50IGhhcyBubyBhY3Rpb25zIGZvciBJQU5BLg0NDQ0NDQ0N DQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ1WYWxpbiwgZXQgYWwuICAg ICAgICAgICAgRXhwaXJlcyBBcHJpbCAxNSwgMjAxMSAgICAgICAgICAgICAgICBbUGFnZSAxNV0N DA1JbnRlcm5ldC1EcmFmdCAgICAgICAgICAgICAgQ29kZWMgR3VpZGVsaW5lcyAgICAgICAgICAg ICAgICBPY3RvYmVyIDIwMTANDQ05LiAgQWNrbm93bGVkZ21lbnRzDQ0gICBXZSB3b3VsZCBsaWtl IHRvIHRoYW5rIGFsbCB0aGUgb3RoZXIgcGVvcGxlIHdobyBjb250cmlidXRlZCBkaXJlY3RseQ0g ICBvciBpbmRpcmVjdGx5IHRvIHRoaXMgZG9jdW1lbnQsIGluY2x1ZGluZyBKYXNvbiBGaXNjaGws IEdyZWdvcnkNICAgTWF4d2VsbCwgQWxhbiBEdXJpYywgSm9uYXRoYW4gQ2hyaXN0ZW5zZW4sIEp1 bGlhbiBTcGl0dGthLCBNaWNoYWVsDSAgIEtuYXBwZSwgVGltb3RoeSBUZXJyaWJlcnJ5LCBDaHJp c3RpYW4gSG9lbmUsIFN0ZXBoYW4gV2VuZ2VyIGFuZCBIZW5yeQ0gICBTaW5ucmVpY2guICBXZSBh bHNvIGxpa2UgdG8gdGhhbmsgQ3VsbGVuIEplbm5pbmdzIGFuZCBHcmVnb3J5DSAgIExlYm92aXR6 IGZvciB0aGVpciBhZHZpY2UuICBTcGVjaWFsIHRoYW5rcyB0byBQZXRlciBTYWludC1BbmRyZSwg d2hvDSAgIG9yaWdpbmFsbHkgY28tYXV0aG9yZWQgdGhpcyBkb2N1bWVudC4NDQ0NDQ0NDQ0NDQ0N DQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NVmFsaW4sIGV0IGFsLiAgICAgICAgICAgIEV4 cGlyZXMgQXByaWwgMTUsIDIwMTEgICAgICAgICAgICAgICAgW1BhZ2UgMTZdDQwNSW50ZXJuZXQt RHJhZnQgICAgICAgICAgICAgIENvZGVjIEd1aWRlbGluZXMgICAgICAgICAgICAgICAgT2N0b2Jl ciAyMDEwDQ0NMTAuICBSZWZlcmVuY2VzDQ0xMC4xLiAgTm9ybWF0aXZlIFJlZmVyZW5jZXMNDSAg IFtJUFJdICAgICAgQnJhZG5lciwgUy4sICJJbnRlbGxlY3R1YWwgUHJvcGVydHkgUmlnaHRzIGlu IElFVEYNICAgICAgICAgICAgICBUZWNobm9sb2d5IiwgQkNQIDc5LCBSRkMgMzk3OSwgTWFyY2gg MjAwNS4NDSAgIFtQUk9DRVNTXSAgQnJhZG5lciwgUy4sICJUaGUgSW50ZXJuZXQgU3RhbmRhcmRz IFByb2Nlc3MgLS0gUmV2aXNpb24NICAgICAgICAgICAgICAzIiwgQkNQIDksIFJGQyAyMDI2LCBP Y3RvYmVyIDE5OTYuDQ0gICBbVFJVU1RdICAgIEJyYWRuZXIsIFMuIGFuZCBKLiBDb250cmVyYXMs ICJSaWdodHMgQ29udHJpYnV0b3JzIFByb3ZpZGUNICAgICAgICAgICAgICB0byB0aGUgSUVURiBU cnVzdCIsIEJDUCA3OCwgUkZDIDUzNzgsIE5vdmVtYmVyIDIwMDguDQ0xMC4yLiAgSW5mb3JtYXRp dmUgUmVmZXJlbmNlcw0NICAgW0NPTExBQl0gICBGaXNobWFuLCBHLiBhbmQgUy4gQnJhZG5lciwg IkludGVybmV0IEVuZ2luZWVyaW5nIFRhc2sNICAgICAgICAgICAgICBGb3JjZSBhbmQgSW50ZXJu YXRpb25hbCBUZWxlY29tbXVuaWNhdGlvbiBVbmlvbiAtDSAgICAgICAgICAgICAgVGVsZWNvbW11 bmljYXRpb25zIFN0YW5kYXJkaXphdGlvbiBTZWN0b3IgQ29sbGFib3JhdGlvbg0gICAgICAgICAg ICAgIEd1aWRlbGluZXMiLCBSRkMgMzM1NiwgQXVndXN0IDIwMDIuDQ0gICBbRE9TXSAgICAgIEhh bmRsZXksIE0uLCBSZXNjb3JsYSwgRS4sIGFuZCBJQUIsICJJbnRlcm5ldCBEZW5pYWwtb2YtDSAg ICAgICAgICAgICAgU2VydmljZSBDb25zaWRlcmF0aW9ucyIsIFJGQyA0NzMyLCBEZWNlbWJlciAy MDA2Lg0NICAgW0ppbmdsZV0gICBMdWR3aWcsIFMuLCBTYWludC1BbmRyZSwgUC4sIEVnYW4sIFMu LCBNY1F1ZWVuLCBSLiwgYW5kIEQuDSAgICAgICAgICAgICAgQ2lvbm9pdSwgIkppbmdsZSBSVFAg U2Vzc2lvbnMiLCBYU0YgWEVQIDAxNjcsIEp1bmUgMjAwOS4NDSAgIFtSVFBdICAgICAgU2NodWx6 cmlubmUsIEguLCBDYXNuZXIsIFMuLCBGcmVkZXJpY2ssIFIuLCBhbmQgVi4NICAgICAgICAgICAg ICBKYWNvYnNvbiwgIlJUUDogQSBUcmFuc3BvcnQgUHJvdG9jb2wgZm9yIFJlYWwtVGltZQ0gICAg ICAgICAgICAgIEFwcGxpY2F0aW9ucyIsIFNURCA2NCwgUkZDIDM1NTAsIEp1bHkgMjAwMy4NDSAg IFtTRFBdICAgICAgSGFuZGxleSwgTS4sIEphY29ic29uLCBWLiwgYW5kIEMuIFBlcmtpbnMsICJT RFA6IFNlc3Npb24NICAgICAgICAgICAgICBEZXNjcmlwdGlvbiBQcm90b2NvbCIsIFJGQyA0NTY2 LCBKdWx5IDIwMDYuDQ0gICBbU0VDR1VJREVdDSAgICAgICAgICAgICAgUmVzY29ybGEsIEUuIGFu ZCBCLiBLb3J2ZXIsICJHdWlkZWxpbmVzIGZvciBXcml0aW5nIFJGQw0gICAgICAgICAgICAgIFRl eHQgb24gU2VjdXJpdHkgQ29uc2lkZXJhdGlvbnMiLCBCQ1AgNzIsIFJGQyAzNTUyLA0gICAgICAg ICAgICAgIEp1bHkgMjAwMy4NDSAgIFtTSVBdICAgICAgUm9zZW5iZXJnLCBKLiwgU2NodWx6cmlu bmUsIEguLCBDYW1hcmlsbG8sIEcuLCBKb2huc3RvbiwNICAgICAgICAgICAgICBBLiwgUGV0ZXJz b24sIEouLCBTcGFya3MsIFIuLCBIYW5kbGV5LCBNLiwgYW5kIEUuDSAgICAgICAgICAgICAgU2No b29sZXIsICJTSVA6IFNlc3Npb24gSW5pdGlhdGlvbiBQcm90b2NvbCIsIFJGQyAzMjYxLA0gICAg ICAgICAgICAgIEp1bmUgMjAwMi4NDSAgIFtTUlRQXSAgICAgQmF1Z2hlciwgTS4sIE1jR3Jldywg RC4sIE5hc2x1bmQsIE0uLCBDYXJyYXJhLCBFLiwgYW5kIEsuDSAgICAgICAgICAgICAgTm9ycm1h biwgIlRoZSBTZWN1cmUgUmVhbC10aW1lIFRyYW5zcG9ydCBQcm90b2NvbCAoU1JUUCkiLA0gICAg ICAgICAgICAgIFJGQyAzNzExLCBNYXJjaCAyMDA0Lg0NICAgW1VOQ09PUkRdICBCcnlhbnQsIFMu IGFuZCBNLiBNb3Jyb3csICJVbmNvb3JkaW5hdGVkIFByb3RvY29sDQ0NDVZhbGluLCBldCBhbC4g ICAgICAgICAgICBFeHBpcmVzIEFwcmlsIDE1LCAyMDExICAgICAgICAgICAgICAgIFtQYWdlIDE3 XQ0MDUludGVybmV0LURyYWZ0ICAgICAgICAgICAgICBDb2RlYyBHdWlkZWxpbmVzICAgICAgICAg ICAgICAgIE9jdG9iZXIgMjAxMA0NDSAgICAgICAgICAgICAgRGV2ZWxvcG1lbnQgQ29uc2lkZXJl ZCBIYXJtZnVsIiwgUkZDIDU3MDQsIE5vdmVtYmVyIDIwMDkuDQ0gICBbWE1QUF0gICAgIFNhaW50 LUFuZHJlLCBQLiwgRWQuLCAiRXh0ZW5zaWJsZSBNZXNzYWdpbmcgYW5kIFByZXNlbmNlDSAgICAg ICAgICAgICAgUHJvdG9jb2wgKFhNUFApOiBDb3JlIiwgUkZDIDM5MjAsIE9jdG9iZXIgMjAwNC4N DQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ1WYWxpbiwgZXQg YWwuICAgICAgICAgICAgRXhwaXJlcyBBcHJpbCAxNSwgMjAxMSAgICAgICAgICAgICAgICBbUGFn ZSAxOF0NDA1JbnRlcm5ldC1EcmFmdCAgICAgICAgICAgICAgQ29kZWMgR3VpZGVsaW5lcyAgICAg ICAgICAgICAgICBPY3RvYmVyIDIwMTANDQ1BdXRob3JzJyBBZGRyZXNzZXMNDSAgIEplYW4tTWFy YyBWYWxpbg0gICBPY3Rhc2ljIEluYy4NICAgNDEwMSwgTW9sc29uIFN0cmVldA0gICBNb250cmVh bCwgUXVlYmVjDSAgIENhbmFkYQ0NICAgRW1haWw6IGplYW4tbWFyYy52YWxpbkBvY3Rhc2ljLmNv bQ0NDSAgIFNsYXZhIEJvcmlsaW4NICAgU1BJUklUIERTUA0NDSAgIEVtYWlsOiBib3JpbGluQHNw aXJpdGRzcC5uZXQNDQ0gICBLb2VuIFZvcw0gICBTa3lwZSBUZWNobm9sb2dpZXMgUy5BLg0gICBT dGFkc2dhcmRlbiA2DSAgIFN0b2NraG9sbSwgICAxMTY0NQ0gICBTd2VkZW4NDSAgIEVtYWlsOiBr b2VuLnZvc0Bza3lwZS5uZXQNDQ0gICBDaHJpc3RvcGhlciBNb250Z29tZXJ5DSAgIFhpcGguT3Jn IEZvdW5kYXRpb24NDQ0gICBFbWFpbDogeGlwaG1vbnRAeGlwaC5vcmcNDQ0gICBSYXltb25kIChK dWluLUh3ZXkpIENoZW4NICAgQnJvYWRjb20gQ29ycG9yYXRpb24NDQ0gICBFbWFpbDogcmNoZW5A YnJvYWRjb20uY29tDQ0NDQ0NDQ0NDQ0NDVZhbGluLCBldCBhbC4gICAgICAgICAgICBFeHBpcmVz IEFwcmlsIDE1LCAyMDExICAgICAgICAgICAgICAgIFtQYWdlIDE5XQ0MDQ0FhGGTIGNvbnRyYWRp Y3RzIHRoZSBzdGF0ZW1lbnQgYmVsb3cgdGhhdCBvbmUgb3IgbW9yZSBjb2RlYyB3aWxsIGJlIG1h ZGUuDQVNYXhpbWFsIHRyYW5zcGFyZW5jeT0gQmlnIGJyb3RoZXI/DQVUaGlzIGRvY3VtZW50IGRv ZXMgbm90IGRlc2NyaWJlIG9uIGhvdyB0aGUgY29vcGVyYXRpb24gd2l0aCBvdGhlciBXR3Mgc2hh bGwgbG9vayBsaWtlLiBUaGlzIGNvdWxkIGJlIHBsYW5uZWQgYW5kIGRlc2NyaWJlZCBoZXJlLg0F V2l0aG91dCByYXRlIGNvbnRyb2wsIHRoaXMgaGlnaCBzY2FsYWJsZSBjb2RlYyB3aWxsIHZhc3Qg aXRzIGNvZGluZyBlZmZpY2VuY3kgYmVjYXVzZSB0aGUgcHJvcGVyIGNvZGUgbWFkZSBpcyBub3Qg c2VsZWN0ZWQuIEFsc28sIGl0IGlzIHJlY29tbWVuZGVkIHRoYXQgUlRQIHBheWxvYWQgZm9ybWF0 cyBzaGFsbCBzdXBwb3J0IGNvbmdlc3Rpb24gY29udHJvbCAoZHJhZnQtaWV0Zi1hdnQtcnRwLWhv d3RvLTA2LCBTZWN0aW9uIDcuMikNBVN0YW5kYXJkIFJGQz8gRG9lc26SdCBpdCByZXF1aXJlIHR3 byBpbmRlcGVuZGVuZCBpbXBsZW1lbnRhdGlvbnMgOy0pDQVUaGlzIGlzIHRydWUgYnV0IHRoZSB3 b3JkaW5nIGluIHRoZSBmb2xsb3dpbmcgcGFyYWdyYXBoIHJlbWFpbnMgZGV0ZXJtaW5lIGZvciBz dWdnZXN0aW9ucw0FRG8gY29uY2VhbG1lbnQgYWxnb3JpdGhtcyBiZWxvbmcgdG8gdGhlIGRlY29k ZXI/DQ0NAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACAAA/AsAAP4LAAAKDAAACwwA AAwMAAB9GQAAhhkAAIoZAACLGQAAjhkAAJ4ZAACfGQAAoBkAAMkZAADoGQAA7RkAAPYZAAAIGwAA 89fHsMfzlIRx85RhlPOEUUHzAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHwEI gQRIAQAFaJVt6mYWaDgxSQBPSgMAUUoDAF5KAwAfAQiBBEgBAAVom23qZhZoxF9lAE9KAwBRSgMA XkoDAB8DagAAAAAWaDgxSQAwShEAT0oEAFFKBABVCAFeSgAAJQEIgQRIAQAFaJRt6mYVaDQ/BgAW aDgxSQBPSgMAUUoDAF5KAwAfAQiBBEgBAAVolG3qZhZoODFJAE9KAwBRSgMAXkoDADcACIEVaDQ/ BgAWaMU1yAAXaDgxSQBPSgMAUUoDAF5KAwBjSAEAZGgAAAAAZGgAAAAAZGiUbepmLAEIgQNqAAAA AARIAQAFaJNt6mYWaDgxSQAwShEAT0oEAFFKBABVCAFeSgAAAB8BCIEESAEABWiRbepmFmg4MUkA T0oDAFFKAwBeSgMANwAIgRVoND8GABZoxTXIABdoODFJAE9KAwBRSgMAXkoDAGNIAQBkaAAAAABk aAAAAABkaJFt6mYYFWg0PwYAFmjFNcgAT0oDAFFKAwBeSgMAEgAIAAABCAAAAggAAAMIAABMCAAA lQgAAN4IAAAnCQAAcAkAALkJAAACCgAASwoAAJQKAADdCgAAJgsAACcLAAAoCwAAZwsAAJsLAACc CwAApQsAAKYLAADuCwAAPwwAAIEMAACiDAAAowwAALcMAAC4DAAA+QwAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAAAAAAAAQPAGdkND8GAAAd+QwAAB0NAAAeDQAAYw0AAKUNAADt DQAAKg4AACsOAAB0DgAAvA4AAP4OAAA8DwAAPQ8AAHMPAAB0DwAAhQ8AAIYPAADJDwAA9A8AAPUP AAA2EAAAXxAAAKIQAACjEAAApBAAAKUQAADuEAAA8BAAADkRAAA6EQAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAAAAAAABA8AZ2Q0PwYAAB06EQAAOxEAAHsRAADEEQAADBIAAFIS AACVEgAAwRIAAMISAADDEgAA1RIAANYSAAAfEwAAaBMAALETAAD6EwAAQxQAAIwUAADVFAAAHhUA AGcVAACwFQAA+RUAAEIWAACLFgAAjBYAAI0WAACOFgAAjxYAAJAWAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAAAAAAAEDwBnZDQ/BgAAHZAWAACRFgAAkhYAAJMWAACUFgAAlRYA AJYWAACXFgAAmBYAAJkWAACaFgAAmxYAAJwWAACdFgAAnhYAAJ8WAACgFgAAoRYAAKIWAACjFgAA pBYAAKUWAACmFgAApxYAAPAWAADyFgAAOxcAADwXAAA9FwAAThcAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAAAAAAAAQPAGdkND8GAAAdThcAAE8XAACWFwAA3RcAACQYAABsGAAA dhgAAHcYAAB4GAAAeRgAAHoYAAB7GAAAfBgAAH0YAAB+GAAAfxgAAIAYAACBGAAAghgAAIMYAACE GAAAhRgAAIYYAACHGAAAiBgAAIkYAACKGAAAixgAAIwYAACNGAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAAAAAAABA8AZ2Q0PwYAAB2NGAAAjhgAAI8YAACQGAAAkRgAAJIYAACT GAAAlBgAAJUYAACWGAAAlxgAAJgYAACZGAAAmhgAAJsYAACcGAAAnRgAAJ4YAACfGAAAoBgAAKEY AACiGAAA6xgAAO0YAAA2GQAANxkAADgZAABQGQAAURkAAKAZAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAAAAAAAEDwBnZDQ/BgAAHaAZAAARGgAAVRoAAJsaAADjGgAAKxsAAG4b AAC3GwAA9xsAADgcAAB6HAAAwhwAAAMdAAAPHQAAEB0AAFAdAABRHQAAmR0AANIdAADTHQAAHB4A AGMeAACmHgAA7B4AADMfAAB1HwAAtx8AAPcfAAA7IAAAeiAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAAAAAAAAQPAGdkND8GAAAdCBsAAAkbAAANGwAAzSMAANIjAAAOJAAAEiQA ABMkAAC1JAAAvyQAAMAkAADBJAAAYSUAAIUlAADCJQAA0TEAANIxAADTMQAABjIAAO/TxrajtofG a8ZbxmuHxksvxjcACIEVaDQ/BgAWaMU1yAAXaAtgdQBPSgMAUUoDAF5KAwBjSAEAZGgAAAAAZGgA AAAAZGi1leomHwEIgQRIAQAFaLWV6iYWaAtgdQBPSgMAUUoDAF5KAwAfA2oAAAAAFmg4MUkAMEoR AE9KBABRSgQAVQgBXkoAADcACIEVaDQ/BgAWaMU1yAAXaDgxSQBPSgMAUUoDAF5KAwBjSAEAZGgA AAAAZGgAAAAAZGiYbepmNwAIgRVoND8GABZoxTXIABdoODFJAE9KAwBRSgMAXkoDAGNIAQBkaAAA AABkaAAAAABkaJdt6mYlAQiBBEgBAAVol23qZhVoND8GABZoODFJAE9KAwBRSgMAXkoDAB8BCIEE SAEABWiXbepmFmg4MUkAT0oDAFFKAwBeSgMAGBVoND8GABZoxTXIAE9KAwBRSgMAXkoDAAA3AAiB FWg0PwYAFmjFNcgAF2g4MUkAT0oDAFFKAwBeSgMAY0gBAGRoAAAAAGRoAAAAAGRolm3qZh8BCIEE SAEABWiWbepmFmg4MUkAT0oDAFFKAwBeSgMAABJ6IAAAuyAAAPggAAA/IQAAdSEAAHYhAAC8IQAA AyIAAEsiAACTIgAA2iIAACIjAABqIwAAqCMAAO4jAAAzJAAAdyQAAMAkAADCJAAAwyQAAMQkAAAN JQAADyUAAFglAABZJQAAWiUAAKIlAADhJQAAKSYAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAQPAGdkODFJAAAEDwBnZDQ/BgAAHCkmAABsJgAAbSYAALEmAAD4JgAAPScAAIYnAADM JwAA2icAANsnAAAjKAAAbCgAAK8oAAD0KAAAOykAAIIpAADLKQAAESoAAFoqAACgKgAA6SoAADAr AAB0KwAAdSsAAL4rAAAFLAAATSwAAJIsAADXLAAAGy0AAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAAAAAAAAQPAGdkND8GAAAdGy0AAFwtAACeLQAA4y0AACsuAABxLgAAsy4AAPcu AAA2LwAAdy8AALsvAAADMAAAETAAABIwAABWMAAAnzAAAOMwAADkMAAA5TAAAOYwAAAvMQAAMTEA AHoxAAB7MQAAfDEAAMQxAAA1MgAAdTIAALgyAADwMgAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAAAAAAABA8AZ2Q0PwYAAB0GMgAABzIAAAkyAAAuMgAALzIAAPY9AAABPgAAET4A ABM+AAAUPgAAGD4AAB0+AAA3PgAAVz4AAFg+AABnPgAAcj4AAIY+AACLPgAAkj4AAJM+AACXPgAA rj4AAMY+AAA1PwAA79/vyLurm6ubq5t/b19vX29Pb09vT18/AAAfAQiBBEgBAAVozpXqJhZoSj87 AE9KAwBRSgMAXkoDAB8BCIEESAEABWjMleomFmhKPzsAT0oDAFFKAwBeSgMAHwEIgQRIAQAFaM2V 6iYWaEo/OwBPSgMAUUoDAF5KAwAfAQiBBEgBAAVoy5XqJhZoSj87AE9KAwBRSgMAXkoDADcACIEV aDQ/BgAWaMU1yAAXaBAz6ABPSgMAUUoDAF5KAwBjSAEAZGgAAAAAZGgAAAAAZGjKleomHwEIgQRI AQAFaMqV6iYWaBAz6ABPSgMAUUoDAF5KAwAfAQiBBEgBAAVoyZXqJhZoEDPoAE9KAwBRSgMAXkoD ABgVaDQ/BgAWaMU1yABPSgMAUUoDAF5KAwAALAEIgQNqAAAAAARIAQAFaLaV6iYWaAtgdQAwShEA T0oEAFFKBABVCAFeSgAAAB8BCIEESAEABWi2leomFmgLYHUAT0oDAFFKAwBeSgMAHwEIgQRIAQAF aLWV6iYWaAtgdQBPSgMAUUoDAF5KAwAAGPAyAADxMgAAOjMAAHszAAC9MwAA3zMAAOAzAADhMwAA 4jMAAOMzAADkMwAA5TMAAOYzAADnMwAA6DMAAOkzAADqMwAA6zMAAOwzAADtMwAA7jMAAO8zAADw MwAA8TMAAPIzAADzMwAA9DMAAPUzAAD2MwAA9zMAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAAAAAAAAQPAGdkND8GAAAd9zMAAPgzAAD5MwAA+jMAAPszAAD8MwAA/TMAAP4zAAD/ MwAAADQAAAE0AAACNAAAAzQAAAQ0AAAFNAAABjQAAAc0AAAINAAAUTQAAFM0AACcNAAAnTQAAJ40 AADMNAAAzTQAABU1AABdNQAAozUAAOg1AAAwNgAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAAAAAAABA8AZ2Q0PwYAAB0wNgAAcDYAALI2AAD1NgAAODcAAIA3AADBNwAA8zcAAPQ3 AAA3OAAAgDgAAMY4AAAOOQAAVTkAAJs5AADfOQAA6TkAAOo5AAAwOgAAeDoAALs6AAAEOwAATDsA AJA7AADPOwAA8TsAAPI7AAApPAAAKjwAAGc8AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAAAAAAAEDwBnZDQ/BgAAHWc8AABoPAAArzwAAN48AADfPAAAKD0AAFM9AABUPQAAVT0A AFY9AABXPQAAWD0AAFk9AABaPQAAWz0AAFw9AAClPQAApz0AAPA9AADxPQAA8j0AADY+AAA3PgAA pz8AAO8/AAA4QAAAekAAAKFAAACjQAAApEAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAAAAAAAAQPAGdkND8GAAAdNT8AAD0/AABAPwAAQj8AAEY/AABQPwAAdD8AAKY/AACnPwAA oEAAAKFAAACiQAAAo0AAAMtAAADPQAAA79/vz9/Pv6+TempXSjoAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAfAQiBBEgBAAVowpXqJhZoEDPoAE9KAwBRSgMAXkoDABgV aDQ/BgAWaMU1yABPSgMAUUoDAF5KAwAAJQEIgQRIAQAFaMqV6iYVaDQ/BgAWaEo/OwBPSgMAUUoD AF5KAwAfAQiBBEgBAAVoypXqJhZoSj87AE9KAwBRSgMAXkoDADEACIEWaMU1yAAXaEo/OwBPSgMA UUoDAF5KAwBjSAEAZGgAAAAAZGgAAAAAZGjKleomNwAIgRVoND8GABZoxTXIABdoEDPoAE9KAwBR SgMAXkoDAGNIAQBkaAAAAABkaAAAAABkaMqV6iYfAQiBBEgBAAVoy5XqJhZoSj87AE9KAwBRSgMA XkoDAB8BCIEESAEABWjQleomFmhKPzsAT0oDAFFKAwBeSgMAHwEIgQRIAQAFaM+V6iYWaEo/OwBP SgMAUUoDAF5KAwAfAQiBBEgBAAVozpXqJhZoSj87AE9KAwBRSgMAXkoDAB8BCIEESAEABWjNleom FmhKPzsAT0oDAFFKAwBeSgMAAA7PQAAA00AAANRAAAD4QAAA/0AAACdBAAAxQQAAP0EAAFBBAABS QQAAaUEAAHtBAACAQQAAhkEAAJVBAACaQQAAvUEAAORBAADsQQAA8UEAAPJBAAATQgAAJ0IAAChC AAAsQgAANEIAADVCAAA2QgAA49bG49bG1ramtpa2hraWhnZmhmaGZlZmVkZWAB8BCIEESAEABWjI leomFmhKPzsAT0oDAFFKAwBeSgMAHwEIgQRIAQAFaMiV6iYWaBAz6ABPSgMAUUoDAF5KAwAfAQiB BEgBAAVox5XqJhZoEDPoAE9KAwBRSgMAXkoDAB8BCIEESAEABWjEleomFmgQM+gAT0oDAFFKAwBe SgMAHwEIgQRIAQAFaMaV6iYWaBAz6ABPSgMAUUoDAF5KAwAfAQiBBEgBAAVoxZXqJhZoEDPoAE9K AwBRSgMAXkoDAB8BCIEESAEABWjJleomFmgQM+gAT0oDAFFKAwBeSgMAHwEIgQRIAQAFaMOV6iYW aBAz6ABPSgMAUUoDAF5KAwAfAQiBBEgBAAVowpXqJhZoEDPoAE9KAwBRSgMAXkoDABgVaDQ/BgAW aMU1yABPSgMAUUoDAF5KAwAANwAIgRVoND8GABZoxTXIABdoEDPoAE9KAwBRSgMAXkoDAGNIAQBk aAAAAABkaAAAAABkaMKV6iYAG6RAAAASQQAAgkEAAINBAABZQwAAWkMAAOhDAAAsRAAAUUQAAFJE AACYRAAAxEQAAMVEAADhRAAAKkUAAHFFAAC1RQAA+kUAAEFGAACCRgAAg0YAAJ9GAACgRgAA6EYA ADFHAABzRwAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD1 AAAAAAAAAAAAAAAA8AAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPUAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAPUAAAAAAAAAAAAA AAD1AAAAAAAAAAAAAAAA5QAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAADgAAAAAAAAAAAAAAAA4AAA AAAAAAAAAAAAAOAAAAAAAAAAAAAAAADgAAAAAAAAAAAAAAAA2wAAAAAAAAAAAAAAAAAAAAAAAAAA AAAEDwBnZPB/BwAABA8AZ2QQM+gACw8AZ2RKPzsAb8YHAQEA0pXqJmQmAQAEDwBnZFwOeAAABA8A Z2RKPzsAAAQPAGdkND8GAAAZNkIAADdCAAA4QgAAQUIAAE1CAABRQgAAfUIAAIpCAAC8QgAA0UIA AOVCAADmQgAA+EIAAFhDAABZQwAAWkMAAF1DAABsQwAAbUMAAIpDAACOQwAAqkMAAKtDAABQRAAA UUQAAHZEAADv38/f78+/3++vn6+fj79/bH9sf2x/bFxMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAf AQiBBEgBAAVo1ZXqJhZoXA54AE9KAwBRSgMAXkoDAB8BCIEESAEABWjVleomFmgQM+gAT0oDAFFK AwBeSgMAJQEIgQRIAQAFaNSV6iYVaDQ/BgAWaEo/OwBPSgMAUUoDAF5KAwAfAQiBBEgBAAVo1JXq JhZoSj87AE9KAwBRSgMAXkoDAB8BCIEESAEABWjSleomFmgQM+gAT0oDAFFKAwBeSgMAHwEIgQRI AQAFaNOV6iYWaEo/OwBPSgMAUUoDAF5KAwAfAQiBBEgBAAVo0pXqJhZoSj87AE9KAwBRSgMAXkoD AB8BCIEESAEABWjHleomFmhKPzsAT0oDAFFKAwBeSgMAHwEIgQRIAQAFaMeV6iYWaBAz6ABPSgMA UUoDAF5KAwAfAQiBBEgBAAVoyJXqJhZoEDPoAE9KAwBRSgMAXkoDAB8BCIEESAEABWjRleomFmhK PzsAT0oDAFFKAwBeSgMAABl2RAAAd0QAAMJEAADDRAAAxEQAAMVEAAAaRQAAIEYAAIFGAACCRgAA g0YAAIZGAADj1sa2popuUopCMgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAHwEIgQRIAQAFaNeV6iYWaFwOeABPSgMAUUoDAF5KAwAfAQiBBEgBAAVoypXqJhZoxTXIAE9K AwBRSgMAXkoDADcACIEVaDQ/BgAWaMU1yAAXaEo/OwBPSgMAUUoDAF5KAwBjSAEAZGgAAAAAZGgA AAAAZGjSleomNwAIgRVoND8GABZoxTXIABdoSj87AE9KAwBRSgMAXkoDAGNIAQBkaAAAAABkaAAA AABkaNOV6iY3AAiBFWg0PwYAFmjFNcgAF2jwfwcAT0oDAFFKAwBeSgMAY0gBAGRoAAAAAGRoAAAA AGRo5pXqJh8BCIEESAEABWjYleomFmhcDngAT0oDAFFKAwBeSgMAHwEIgQRIAQAFaNWV6iYWaFwO eABPSgMAUUoDAF5KAwAfA2oAAAAAFmhcDngAMEoRAE9KBABRSgQAVQgBXkoAABgVaDQ/BgAWaMU1 yABPSgMAUUoDAF5KAwAANwAIgRVoND8GABZoxTXIABdoXA54AE9KAwBRSgMAXkoDAGNIAQBkaAAA AABkaAAAAABkaNWV6iYAC4ZGAACeRgAAn0YAAEZHAABURwAAVUcAAGpHAACsRwAA3kcAAP1HAAAB SAAAJkgAACdIAAAoSAAAKUgAACxIAAAtSAAALkgAAFxJAABgSQAAmEkAAOzc7My5zLnsqcyZ7IJ1 aFg8aFhoAAAAAAAAAAAAAAAAAAAAADcACIEVaDQ/BgAWaMU1yAAXaOdouABPSgMAUUoDAF5KAwBj SAEAZGgAAAAAZGgAAAAAZGgOmOpGHwEIgQRIAQAFaA6Y6kYWaOdouABPSgMAUUoDAF5KAwAYFWg0 PwYAFmjFNcgAT0oDAFFKAwBeSgMAABgVaDQ/BgAWaBAz6ABPSgMAUUoDAF5KAwAALAEIgQNqAAAA AARIAQAFaA+Y6kYWaOdouAAwShEAT0oEAFFKBABVCAFeSgAAAB8BCIEESAEABWjoleomFmjwfwcA T0oDAFFKAwBeSgMAHwEIgQRIAQAFaMqV6iYWaPB/BwBPSgMAUUoDAF5KAwAlAQiBBEgBAAVo55Xq JhVoND8GABZo8H8HAE9KAwBRSgMAXkoDAB8BCIEESAEABWjnleomFmjwfwcAT0oDAFFKAwBeSgMA HwEIgQRIAQAFaMqV6iYWaBAz6ABPSgMAUUoDAF5KAwAlAQiBBEgBAAVoypXqJhVoND8GABZoEDPo AE9KAwBRSgMAXkoDAAAUc0cAANhHAAAnSAAAKUgAAHFIAACtSAAA9EgAADpJAACFSQAAzEkAABRK AAAVSgAAX0oAAKdKAADuSgAAMUsAAHlLAAC9SwAA9UsAADxMAACBTAAAxEwAAN5MAADfTAAAJE0A AG1NAAC3TQAA+k0AAEJOAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD1 AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA9QAAAAAA AAAAAAAAAPUAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAPUAAAAAAAAAAAAA AAD1AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA9QAA AAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAPUAAAAAAAAA AAAAAAD1AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA 9QAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEDwBnZDQ/ BgAABA8AZ2QQM+gAAByYSQAAmUkAABhKAAAZSgAAGkoAAGxLAABvSwAAdUsAAOJMAADjTAAA5EwA AINNAACGTQAAjE0AAN9OAADgTgAA4U4AAP5OAAABTwAAB08AALVRAAC8UQAARFIAALFSAADSUgAA 1VIAAO/i0rbipori0rbiel7iporiel7ieuJe4k4AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAf AQiBBEgBAAVo65XqJhZo8H8HAE9KAwBRSgMAXkoDADcACIEVaDQ/BgAWaMU1yAAXaPB/BwBPSgMA UUoDAF5KAwBjSAEAZGgAAAAAZGgAAAAAZGjqleomHwEIgQRIAQAFaOqV6iYWaPB/BwBPSgMAUUoD AF5KAwA3AAiBFWg0PwYAFmjFNcgAF2jwfwcAT0oDAFFKAwBeSgMAY0gBAGRoAAAAAGRoAAAAAGRo 6ZXqJh8BCIEESAEABWjpleomFmjwfwcAT0oDAFFKAwBeSgMANwAIgRVoND8GABZoxTXIABdo8H8H AE9KAwBRSgMAXkoDAGNIAQBkaAAAAABkaAAAAABkaOiV6iYfAQiBBEgBAAVo6JXqJhZo8H8HAE9K AwBRSgMAXkoDABgVaDQ/BgAWaMU1yABPSgMAUUoDAF5KAwAAHwNqAAAAABZoXA54ADBKEQBPSgQA UUoEAFUIAV5KAAAAGUJOAACITgAAyU4AANtOAADcTgAAJ08AAGdPAABoTwAAaU8AAGpPAACzTwAA tU8AAP5PAAD/TwAAAFAAAERQAACNUAAA0FAAABZRAABeUQAAoVEAAPBRAAAzUgAAeVIAALJSAACz UgAA/VIAAERTAACXUwAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABA8AZ2TwfwcA AAQPAGdkND8GAAAc1VIAANtSAACIUwAAlVMAAJZTAAC3UwAAu1MAAMBTAADYUwAA31MAAKpiAACr YgAAy2IAAM1iAAD1YgAA+2IAAP9iAAABYwAAEGMAACpjAAAvYwAAMGMAAFB5AABReQAAVHkAAGJ5 AABneQAA49bG49bG49bG1ramlraGtpa2doZ21mZTQ1MfAQiBBEgBAAVo9ZXqJhZoNlGaAE9KAwBR SgMAXkoDACUBCIEESAEABWjwleomFWg0PwYAFmg2UZoAT0oDAFFKAwBeSgMAHwEIgQRIAQAFaPOV 6iYWaDZRmgBPSgMAUUoDAF5KAwAfAQiBBEgBAAVo75XqJhZo8H8HAE9KAwBRSgMAXkoDAB8BCIEE SAEABWgKmOpGFmiyFvoAT0oDAFFKAwBeSgMAHwEIgQRIAQAFaBCY6kYWaOdouABPSgMAUUoDAF5K AwAfAQiBBEgBAAVo7pXqJhZoshb6AE9KAwBRSgMAXkoDAB8BCIEESAEABWjuleomFmjwfwcAT0oD AFFKAwBeSgMAHwEIgQRIAQAFaOuV6iYWaPB/BwBPSgMAUUoDAF5KAwAYFWg0PwYAFmjFNcgAT0oD AFFKAwBeSgMAADcACIEVaDQ/BgAWaMU1yAAXaPB/BwBPSgMAUUoDAF5KAwBjSAEAZGgAAAAAZGgA AAAAZGjrleomABqXUwAAmFMAAOtTAAAxVAAAcFQAALhUAAD7VAAAN1UAAExVAABNVQAATlUAAE9V AABQVQAAUVUAAFJVAABTVQAAVFUAAFVVAABWVQAAV1UAAFhVAABZVQAAWlUAAFtVAABcVQAAXVUA AF5VAABfVQAAYFUAAGFVAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAAAAAAAE DwBnZDQ/BgAAHWFVAABiVQAAY1UAAGRVAABlVQAAZlUAAGdVAABoVQAAaVUAALJVAAC0VQAA/VUA AP5VAAD/VQAAGVYAABpWAABcVgAAaFYAAGlWAACwVgAA9FYAADtXAAB3VwAAeFcAALpXAAD9VwAA QlgAAHJYAABzWAAAtlgAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAAAAAAAAQP AGdkND8GAAAdtlgAAP9YAABDWQAAXFkAAF1ZAACbWQAA4FkAACNaAABmWgAAf1oAAIBaAADAWgAA CFsAAFBbAACTWwAA2lsAAB9cAABjXAAAqFwAAOdcAAAtXQAAcV0AALRdAAD9XQAARF4AAIFeAADE XgAAC18AAFNfAACYXwAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAAAAAAABA8A Z2Q0PwYAAB2YXwAA4F8AACFgAABpYAAAamAAAGtgAABsYAAAtWAAALdgAAAAYQAAAWEAAAJhAABK YQAAg2EAAIRhAADIYQAA52EAAOhhAAAqYgAAaWIAADFjAAAyYwAAe2MAAMNjAADSYwAA02MAABFk AAASZAAAWmQAAKNkAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAAAAAAAEDwBn ZDQ/BgAAHaNkAADrZAAAMWUAAHBlAACEZQAAhWUAAMdlAAAQZgAAWWYAAJVmAAC6ZgAAu2YAAARn AABJZwAAj2cAANVnAAAeaAAAZWgAAK1oAADzaAAANWkAAH5pAADGaQAADWoAAFVqAACbagAAzmoA AM9qAADQagAA0WoAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAAAAAAAAQPAGdk ND8GAAAd0WoAANJqAADTagAA1GoAANVqAAAeawAAIGsAAGlrAABqawAAa2sAAIxrAACNawAA0GsA ABJsAAAhbAAAImwAAGRsAACrbAAAumwAALtsAADpbAAA6mwAACttAAAsbQAAYG0AAGFtAACYbQAA mW0AANltAAAZbgAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAAAAAAABA8AZ2Q0 PwYAAB0ZbgAAQm4AAENuAACEbgAAy24AABRvAABZbwAAn28AALhvAAC5bwAA828AADlwAABTcAAA VHAAAJZwAADZcAAAHXEAAGZxAACtcQAA5HEAAOVxAAArcgAAdHIAALhyAADXcgAA2HIAABlzAABh cwAAp3MAAKhzAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAAAAAAAEDwBnZDQ/ BgAAHahzAACpcwAAqnMAAPNzAAD1cwAAPnQAAD90AABAdAAAgHQAAMh0AADYdAAA2XQAACF1AABi dQAApHUAANF1AADSdQAAGHYAAGF2AABidgAAn3YAAOV2AAAYdwAAGXcAAGJ3AABjdwAAqXcAAPJ3 AAAVeAAAFngAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAAAAAAAAQPAGdkND8G AAAdFngAAF54AACdeAAA4XgAAB55AABPeQAAUHkAAIF5AACCeQAAU3oAAFR6AAAKewAAcX0AAIJ+ AACDfgAA7H8AAO1/AACbgAAAnIAAAJ2AAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD1AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA4QAAAAAAAAAAAAAAAM0A AAAAAAAAAAAAAAC5AAAAAAAAAAAAAAAApQAAAAAAAAAAAAAAALkAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAAAAAAFBgAEmTwAAAA Z2RxdnIAb8YHAQEAA5jqRmQmAQBGDwAPhN8CXoTfAgATGAASZPAAAABnZHF2cgBvxgcBAQADmOpG ZCYBAEYPAAomAAtGAQAAExgAEmTwAAAAZ2RxdnIAb8YHAQEA+5XqJmQmAQBGDwAKJgALRgEAABMY ABJk8AAAAGdkNlGaAG/GBwEBAPmV6iZkJgEARg8ACiYAC0YBAAAEDwBnZDZRmgAABA8AZ2Q0PwYA ABNneQAAbXkAAHJ5AACAeQAAgXkAAIJ5AAABegAAUnoAAFN6AABUegAAXXoAAGN6AABnegAAvHoA AAh7AADv3+/MvKGGoW1dRCtEKwAAAAAAAAAAAAAAADABCIEESAEABWj4leomFmg2UZoAQioBQ0oU AE9KAgBRSgIAXkoCAGFKFABwaAAAAAAAMAEIgQRIAQAFaPWV6iYWaDZRmgBCKgFDShQAT0oCAFFK AgBeSgIAYUoUAHBoAAAAAAAfAQiBBEgBAAVo+JXqJhZoNlGaAE9KAwBRSgMAXkoDADEACIEWaDZR mgAXaDZRmgBPSgMAUUoDAF5KAwBjSAEAZGgAAAAAZGgAAAAAZGj3leomNAEIgQRIAQAFaPeV6iYW aDZRmgAwShoAQioBQ0oUAE9KAgBRSgIAXkoCAGFKFABwaAAAAAAANAEIgQRIAQAFaPaV6iYWaDZR mgAwShoAQioBQ0oUAE9KAgBRSgIAXkoCAGFKFABwaAAAAAAAHwEIgQRIAQAFaPWV6iYWaMU1yABP SgMAUUoDAF5KAwAlAQiBBEgBAAVo8JXqJhVoND8GABZoNlGaAE9KAwBRSgMAXkoDAB8BCIEESAEA BWjzleomFmg2UZoAT0oDAFFKAwBeSgMAHwEIgQRIAQAFaPCV6iYWaDZRmgBPSgMAUUoDAF5KAwAA Dgh7AAAKewAAS3sAAEx7AACHewAAjXsAAJt7AAC0ewAA3HsAAOB7AADiewAA5HsAAPh7AAD8ewAA GXwAALJ8AADJfAAAb30AAHB9AABxfQAA030AAOp9AABAfgAARH4AAIF+AADnzufO57Xntee157Xn tZyDaoOcSJxIg0gAAAAAAAAAAAAAAAAAAAAAAAAAAEMBCAEESAEABWj1leomFWhxdnIAFmg2UZoA icoHAQEA+5XqJoMqAUIqAUNKFABPSgIAUUoCAF5KAgBhShQAcGgAAAAAMAEIgQRIAQAFaAKY6kYW aHF2cgBCKgFDShQAT0oCAFFKAgBeSgIAYUoUAHBoAAAAAAAwAQiBBEgBAAVoAZjqRhZocXZyAEIq AUNKFABPSgIAUUoCAF5KAgBhShQAcGgAAAAAADABCIEESAEABWj7leomFmhxdnIAQioBQ0oUAE9K AgBRSgIAXkoCAGFKFABwaAAAAAAAMAEIgQRIAQAFaPqV6iYWaDZRmgBCKgFDShQAT0oCAFFKAgBe SgIAYUoUAHBoAAAAAAAwAQiBBEgBAAVo9ZXqJhZoNlGaAEIqAUNKFABPSgIAUUoCAF5KAgBhShQA cGgAAAAAADABCIEESAEABWj5leomFmg2UZoAQioBQ0oUAE9KAgBRSgIAXkoCAGFKFABwaAAAAAAY gX4AAIN+AADBfgAAwn4AAP5+AAD/fgAAIn8AACN/AADBfwAAxX8AANN/AADrfwAA7H8AAO1/AABR gAAAWoAAAGOAAACYgAAAmYAAAJqAAADokgAA6ZIAADKTAAAzkwAASJMAAFaTAABXkwAA58XnxefF 56yTrJPFhnZmdmZWZoZMSExESEwAAAYWaAtgdQAABhZoODFJAAATA2oAAAAAFmg4MUkAMEoRAFUI AR8BCIEESAEABWgRmOpGFmjnaLgAT0oDAFFKAwBeSgMAHwEIgQRIAQAFaAWY6kYWaHF2cgBPSgMA UUoDAF5KAwAfAQiBBEgBAAVoBJjqRhZocXZyAE9KAwBRSgMAXkoDABgVaDQ/BgAWaMU1yABPSgMA UUoDAF5KAwAAMAEIgQRIAQAFaAeY6kYWaHF2cgBCKgFDShQAT0oCAFFKAgBeSgIAYUoUAHBoAAAA AAAwAQiBBEgBAAVoBpjqRhZocXZyAEIqAUNKFABPSgIAUUoCAF5KAgBhShQAcGgAAAAAAEMBCAEE SAEABWj1leomFWhxdnIAFmg2UZoAicoHAQEAA5jqRoMqAUIqAUNKFABPSgIAUUoCAF5KAgBhShQA cGgAAAAAMAEIgQRIAQAFaAOY6kYWaHF2cgBCKgFDShQAT0oCAFFKAgBeSgIAYUoUAHBoAAAAABqd gAAAnoAAAJ+AAACggAAAoYAAAKKAAACjgAAApIAAAKWAAACmgAAAp4AAAKiAAACpgAAAqoAAAKuA AACsgAAArYAAAK6AAACvgAAA+IAAAPqAAABDgQAARIEAAEWBAABhgQAAYoEAAKqBAADpgQAALYIA AFKCAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAAAAAAAEDwBnZDQ/BgAAHVKC AABTggAAVIIAAFWCAABWggAAV4IAAFiCAABZggAAWoIAAFuCAABcggAAXYIAAF6CAABfggAAYIIA AGGCAABiggAAY4IAAGSCAABlggAAZoIAAGeCAABoggAAaYIAAGqCAABrggAAbIIAAG2CAABuggAA b4IAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAAAAAAAAQPAGdkND8GAAAdb4IA AHCCAABxggAAcoIAAHOCAAB0ggAAdYIAAHaCAAB3ggAAeIIAAHmCAAB6ggAAe4IAAHyCAAB9ggAA foIAAH+CAADIggAAyoIAABODAAAUgwAAFYMAAC2DAAAugwAAWIMAAFmDAABagwAAW4MAAFyDAABd gwAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAAAAAAABA8AZ2Q0PwYAAB1dgwAA XoMAAF+DAABggwAAYYMAAGKDAABjgwAAZIMAAGWDAABmgwAAZ4MAAGiDAABpgwAAaoMAAGuDAABs gwAAbYMAAG6DAABvgwAAcIMAAHGDAABygwAAc4MAAHSDAAB1gwAAdoMAAHeDAAB4gwAAeYMAAHqD AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAAAAAAAEDwBnZDQ/BgAAHXqDAAB7 gwAAfIMAAH2DAAB+gwAAf4MAAICDAACBgwAAgoMAAIODAACEgwAAhYMAAIaDAACHgwAAiIMAANGD AADTgwAAHIQAAB2EAAAehAAAMoQAADOEAAB7hAAAvoQAAASFAABNhQAAjoUAANaFAAD/hQAAAIYA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAAAAAAAAQPAGdkND8GAAAdAIYAAAGG AAAChgAAA4YAAASGAAAFhgAABoYAAAeGAAAIhgAACYYAAAqGAAALhgAADIYAAA2GAAAOhgAAD4YA ABCGAAARhgAAEoYAABOGAAAUhgAAFYYAABaGAAAXhgAAGIYAABmGAAAahgAAG4YAAByGAAAdhgAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAAAAAAABA8AZ2Q0PwYAAB0dhgAAHoYA AB+GAAAghgAAIYYAACKGAAAjhgAAJIYAACWGAAAmhgAAJ4YAACiGAAAphgAAcoYAAHSGAAC9hgAA voYAAL+GAADPhgAA0IYAAOyGAADthgAALocAAGeHAABohwAAr4cAAOCHAADhhwAAKogAAG2IAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAAAAAAAEDwBnZDQ/BgAAHW2IAABuiAAA jIgAAI2IAADSiAAAEokAAFiJAACKiQAAi4kAANKJAAASigAAE4oAAFyKAACjigAApIoAAOWKAAAm iwAAYIsAAGGLAACoiwAA4osAAOOLAADxiwAAN4wAAHmMAACSjAAAk4wAANqMAAAajQAAYI0AAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAAAAAAAAQPAGdkND8GAAAdYI0AAHmNAAB6 jQAAwo0AAAuOAAAvjgAAMI4AAHCOAABxjgAAco4AAHOOAAC8jgAAvo4AAAePAAAIjwAACY8AAFGP AABSjwAAmY8AANePAADYjwAA2Y8AANqPAADbjwAA3I8AAN2PAADejwAA348AAOCPAADhjwAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAAAAAAABA8AZ2Q0PwYAAB3hjwAA4o8AAOOP AADkjwAA5Y8AAOaPAADnjwAA6I8AAOmPAADqjwAA648AAOyPAADtjwAA7o8AAO+PAADwjwAA8Y8A APKPAADzjwAA9I8AAPWPAAD2jwAA948AAPiPAAD5jwAA+o8AAPuPAAD8jwAA/Y8AAP6PAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAAAAAAAEDwBnZDQ/BgAAHf6PAAD/jwAAAJAA AAGQAAACkAAAA5AAAASQAAAFkAAABpAAAE+QAABRkAAAmpAAAJuQAACckAAAr5AAALCQAADDkAAA 05AAAOqQAAD+kAAACJEAAAmRAAAvkQAAMJEAADGRAABCkQAAUJEAAFGRAABSkQAAcpEAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAAAAAAAAQPAGdkND8GAAAdcpEAAHORAAB0kQAA gJEAAJuRAACskQAAwpEAAMyRAADNkQAA6pEAAOuRAADskQAABpIAAB2SAAAekgAAH5IAADuSAAA8 kgAAPZIAAFmSAABxkgAAcpIAAHOSAACQkgAAkZIAAJKSAACTkgAAlJIAAJWSAACWkgAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAAAAAAABA8AZ2Q0PwYAAB2WkgAAl5IAAJiSAACZ kgAAmpIAAJuSAACckgAA5ZIAAOeSAADokgAAMpMAAFaTAADYkwAA0pQAABiVAABzlQAApZUAAKaV AACnlQAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+AAAAAAAAAAAAAAAAPgAAAAAAAAAAAAAAAD4AAAAAAAAAAAAAAAA+AAAAAAAAAAA AAAAAPgAAAAAAAAAAAAAAAD4AAAAAAAAAAAAAAAA+AAAAAAAAAAAAAAAAPYAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQAAAAESAAAEDwBnZDQ/BgAAEleTAABukwAAdpMAAI6T AACdkwAArpMAAMiTAADRkwAA2JMAANmTAAColAAAw5QAANKUAADTlAAA6ZQAAPyUAAASlQAAE5UA ABiVAAAZlQAAc5UAAHSVAACllQAAppUAAKeVAAD8+Pz4/PTw/Obw3/DV0c3RzdHDzdXRzbYAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAYFWg0PwYAFmjFNcgAT0oDAFFKAwBeSgMAABMDagAAAAAWaOdouAAwShEAVQgBBhZo 52i4AAAGFmhcDngAABMDagAAAAAWaFwOeAAwShEAVQgBDBVoC2B1ABZoC2B1AAATA2oAAAAAFmgL YHUAMEoRAFUIAQYWaAtgdQAABhZo0WehAAAGFmjEX2UAAAYWaDgxSQAYMgAxkGgBOnA0PwYAH7CC LiCwxkEhsDYFIrA3BSOQiQUkkG4EJbAAABewxAIYsMQCDJDEAgAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABqBBsAEgABAAsBDwAHAAQA BAAEAAAABAAIAAAAmAAAAJ4AAACeAAAAngAAAJ4AAACeAAAAngAAAJ4AAACeAAAANgYAADYGAAA2 BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAAdgIAAHYCAAB2AgAAdgIAAHYCAAB2AgAAdgIAAHYC AAB2AgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAAPgIAADYGAAA2BgAANgYAADYGAAA2BgAANgYA ADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAA NgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAAKgAAAA2BgAANgYAABYAAAA2BgAANgYAADYGAAA2 BgAANgYAADYGAAA2BgAANgYAALgAAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYG AAA2BgAANgYAADYGAABoAQAASAEAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYA ADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAA NgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2 BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYG AAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAAsAMA ADYGAAAyBgAAGAAAAMADAADQAwAA4AMAAPADAAAABAAAEAQAACAEAAAwBAAAQAQAAFAEAABgBAAA cAQAAIAEAACQBAAAwAMAANADAADgAwAA8AMAAAAEAAAQBAAAMgYAACgCAADYAQAA6AEAACAEAAAw BAAAQAQAAFAEAABgBAAAcAQAAIAEAACQBAAAwAMAANADAADgAwAA8AMAAAAEAAAQBAAAIAQAADAE AABABAAAUAQAAGAEAABwBAAAgAQAAJAEAADAAwAA0AMAAOADAADwAwAAAAQAABAEAAAgBAAAMAQA AEAEAABQBAAAYAQAAHAEAACABAAAkAQAAMADAADQAwAA4AMAAPADAAAABAAAEAQAACAEAAAwBAAA QAQAAFAEAABgBAAAcAQAAIAEAACQBAAAwAMAANADAADgAwAA8AMAAAAEAAAQBAAAIAQAADAEAABA BAAAUAQAAGAEAABwBAAAgAQAAJAEAADAAwAA0AMAAOADAADwAwAAAAQAABAEAAAgBAAAMAQAAEAE AABQBAAAYAQAAHAEAACABAAAkAQAADgBAABYAQAA+AEAAAgCAAAYAgAAVgIAAH4CAAAgAAAAT0oE AFBKBABRSgQAX0gBBG1IBwRuSAcEc0gHBHRIBwQAAAAATgAAYPH/AgBOAAwQAAByWKwAAAAIAFMA dABhAG4AZABhAHIAZAAAAAwAAAASZBQBAQAUpMgAGABDShYAX0gBBGFKFgBtSAcEc0gHBHRICQQA AAAAAAAAAAAAAAAAAAAAAABKAEFg8v+hAEoADA0AAAAAAAAQABkAQQBiAHMAYQB0AHoALQBTAHQA YQBuAGQAYQByAGQAcwBjAGgAcgBpAGYAdABhAHIAdAAAAAAAWABpAPP/swBYAAwdAAAAAAAAMAYP AE4AbwByAG0AYQBsAGUAIABUAGEAYgBlAGwAbABlAAAAHAAX9gMAADTWBgABCgNsADTWBgABBQMA AGH2AwAAAgALAAAAMABrIPT/wQAwAAANAAAAAAAAMAYLAEsAZQBpAG4AZQAgAEwAaQBzAHQAZQAA AAIADAAAAAAASgBaQAEA8gBKAAwMEAA0PwYAMAYIAE4AdQByACAAVABlAHgAdAAAAAwADwASZPAA AQAUpAAAFABDShUAT0oFAFFKBQBeSgUAYUoVAEYA/k+iAAEBRgAMAA8AND8GADAGDQBOAHUAcgAg AFQAZQB4AHQAIABaAGMAaABuAAAAFABDShUAT0oFAFFKBQBeSgUAYUoVAEAAJ0CiABEBQAAMDQAA ODFJADAGEABLAG8AbQBtAGUAbgB0AGEAcgB6AGUAaQBjAGgAZQBuAAAACABDShAAYUoQAD4AHkAB ACIBPgAMDRMAODFJADAGDQBLAG8AbQBtAGUAbgB0AGEAcgB0AGUAeAB0AAAAAgASAAgAQ0oUAGFK FABAAP4PogAxAUAADAESADgxSQAwBhIASwBvAG0AbQBlAG4AdABhAHIAdABlAHgAdAAgAFoAYwBo AG4AAAAEAHRICQQ+AGoAIQEiAT4ADA0VADgxSQAwBg4ASwBvAG0AbQBlAG4AdABhAHIAdABoAGUA bQBhAAAAAgAUAAYANQiBXAiBRAD+DzIBUQFEAAwBFAA4MUkAMAYTAEsAbwBtAG0AZQBuAHQAYQBy AHQAaABlAG0AYQAgAFoAYwBoAG4AAAAGADUIAVwIAVoAmQABAGIBWgAMDRcAODFJADAGEABTAHAA cgBlAGMAaABiAGwAYQBzAGUAbgB0AGUAeAB0AAAADAAWABJk8AABABSkAAAUAENKEABPSgYAUUoG AF5KBgBhShAAWgD+D6IAcQFaAAwBFgA4MUkAMAYVAFMAcAByAGUAYwBoAGIAbABhAHMAZQBuAHQA ZQB4AHQAIABaAGMAaABuAAAAGABDShAAT0oGAFFKBgBeSgYAYUoQAHRICQRkAF5gAQCCAWQADAkA ADZRmgAwBg4AUwB0AGEAbgBkAGEAcgBkACAAKABXAGUAYgApAAAAFgAYABJk8AABABOkZAAUpGQA WyQBXCQBGABDShgAT0oAAFBKAABRSgAAYUoYAHRIBwRCAP4vogCRAUIADAAAADZRmgAAABUAYQBw AHAAbABlAC0AYwBvAG4AdgBlAHIAdABlAGQALQBzAHAAYQBjAGUAAAAAADgA/m+iAKEBOAAMAAAA NlGaAAAAEABhAHAAcABsAGUALQBzAHQAeQBsAGUALQBzAHAAYQBuAAAAAABQSwMEFAAGAAgAAAAh AIKKvBP6AAAAHAIAABMAAABbQ29udGVudF9UeXBlc10ueG1srJHLasMwEEX3hf6D0LbYcroopdjO okl3fSzSDxjksS1qj4Q0Ccnfd+y4ULoILXQjEGLOmXtVro/joA4Yk/NU6VVeaIVkfeOoq/T77im7 1yoxUAODJ6z0CZNe19dX5e4UMCmZplTpnjk8GJNsjyOk3AckeWl9HIHlGjsTwH5Ah+a2KO6M9cRI nPHE0HX5KgtE16B6g8gvMIrHsKDw+/kMJICYC1irxzNhWqLSEMLgLLBEMAdqfugz37bOYuPtfhRp PoMX2M0EM79cYPU/6i/nBlvYD6y2R+niXH/EIf0t21JrLpNz/tS7kC4YLpe3tGHmv60/AQAA//8D AFBLAwQUAAYACAAAACEApdan58AAAAA2AQAACwAAAF9yZWxzLy5yZWxzhI/PasMwDIfvhb2D0X1R 0sMYJXYvpZBDL6N9AOEof2giG9sb69tPxwYKuwiEpO/3qT3+rov54ZTnIBaaqgbD4kM/y2jhdj2/ f4LJhaSnJQhbeHCGo3vbtV+8UNGjPM0xG6VItjCVEg+I2U+8Uq5CZNHJENJKRds0YiR/p5FxX9cf mJ4Z4DZM0/UWUtc3YK6PqMn/s8MwzJ5PwX+vLOVFBG43lExp5GKhqC/jU72QqGWq1B7Qtbj51v0B AAD//wMAUEsDBBQABgAIAAAAIQBreZYWgwAAAIoAAAAcAAAAdGhlbWUvdGhlbWUvdGhlbWVNYW5h Z2VyLnhtbAzMTQrDIBBA4X2hd5DZN2O7KEVissuuu/YAQ5waQceg0p/b1+XjgzfO3xTVm0sNWSyc Bw2KZc0uiLfwfCynG6jaSBzFLGzhxxXm6XgYybSNE99JyHNRfSPVkIWttd0g1rUr1SHvLN1euSRq PYtHV+jT9yniResrJgoCOP0BAAD//wMAUEsDBBQABgAIAAAAIQDjWkknlwYAAFUbAAAWAAAAdGhl bWUvdGhlbWUvdGhlbWUxLnhtbOxZTW8bRRi+I/EfRntvYyd2Gkd1qtixG2hSotgt6nG8O96dZnZn NTNO6htqj0hIiII4UIkbBwRUaiUu5dcEiqBI/Qu8M7O73onXJGkjWkF9SLyzz7zfXzO+eu1ezNAh EZLypO3VL9c8RBKfBzQJ296tYf/SmoekwkmAGU9I25sS6V3beP+9q3hdRSQmCPYnch23vUipdH1p SfqwjOVlnpIE3o25iLGCRxEuBQIfAd2YLS3XaqtLMaaJhxIcA9kdLKiU+NIWkTRMvI2cfI8Bj0RJ veAzMdDEibvHgIODuobIqewygQ4xa3vAKuBHQ3JPeYhhqeBF26uZj7e0cXUJr2ebmFqwt7Svbz7Z vmxDcLBseIpwVDCt9xutK1sFfQNgah7X6/W6vXpBzwCw74OqVpYyzUZ/rd7JaZZA9us87W6tWWu4 +BL9lTmZW51Op9nKZLFEDch+bczh12qrjc1lB29AFt+cwzc6m93uqoM3IItfncP3r7RWGy7egCJG k4M5tHZov59RLyBjzrYr4WsAX6tl8BkKoqEIL81izBO1MNhifJeLPiA0kmFFE6SmKRljHyK5i+OR oFhzwOsEl97YJV/OLWlmSPqCpqrtfZhiyIoZvZfPfnj57Ak6vv/0+P7Pxw8eHN//yRJydm3jJCzv evHd5389+gT9+eTbFw+/rMbLMv63Hz/99ZcvqoGQPzNxnn/1+Penj59//dkf3z+sgG8KPCrDhzQm Et0kR2ifx6CYsYorORmJ8+0YRpiWd2wmocQJ1lwq6PdU5KBvTjHLvOPI0SGuBW8LqB9VwOuTu47A g0hMFK3gfCOKHeAu56zDRaUVbmheJTMPJ0lYzVxMyrh9jA+reHdx4vi3N0mhcuZh6SjejYgj5h7D icIhSYhC+h0/IKRCuzuUOnbdpb7gko8VukNRB9NKkwzpyImm2aZtGoNfplU6g78d2+zeRh3OqrTe IocuErICswrhh4Q5ZryOJwrHVSSHOGZlg+9gFVUJOZgKv4zrSQWeDgnjqBcQaG0VUnwkQN+S029g KFmVbt9l09hFCkUPqmjuYM7LyC1+0I1wnFZhBzSJytgP5AGEKEZ7XFXBd7mbIfoZ/ICThe6+TYnj 7tOrwS0aOiLNAkS/mQhtRajVTgWOafJP5ZhRqMfW+hdXjqEAPv/mUYVP39ZCvAk9qSoTtk+U30W4 k0W3y0VA3/6au4UnyR6BMJ9vPO9K7ruS6/3nS+6ifD5roZ3VVii7em6wU7GZkePFI/KYMjZQU0Z2 pJmSJTSKoA+LeqM5IpLizJRG8DUr7A4uFNjsQYKrj6mKBhFOYcKue5pIKDPSoUQpl3C0M8uVtDUe pnRlD4ZNfWSwBUFitcsDu7yil/OTQUHGtJvQnD9zRiuawFmZrVzJiILar8KsroU6M7e6Ec3UOodb oTI4cV41WCysCRMIgrkFrLwKh3TNGk4mmJFA290239wtxgsX6SIZ4YBkPtJ6z/uobpyUx4q5DIDY qfCRPuadYrUSt5Ym+xrczuKkMrvGAna5917HS3kEz7ykE/dEOrKknJwsQUdtr9VcbnrIx2nbG8Oh Fr7GKXhd6qEPsxBuh3wlbNifmswmy2febOWKuUlQh4sKa/c5hZ06kAqptrCMbGiYV1kIsERzsvIv N8GsF6WAjfRXkGJlDYLhjUkBdnRdS8Zj4quys0sr2nb2MSulfKKIGETBERqxidjH4H4dqqBPQCXc TZiKoB/gJk1b27xyi3OWdOX7K4Oz65ilEc7KrU7RPJMt3ORxIYN5KokHulXKbpQ7vyom5S9IlXIY /89U0f0ErgpWAu0BH+5yBUY6X9seFyriUIXSiPp9AZODqR0QLXAbC68hqOBG2fwX5FD/tzlnaZi0 hhOf2qchEhT6kYoEIXtQlkz0nUKsnvUuS5JlhExElcSVqRV7RA4JG+oauKp7u4ciCHVTTbIyYHAn 4899zjJoFOohp5xvTg0peq/NgX978rHJDEq5ddgMNLn9CxEruqrdb7bnvbesiH4xG7MaeVYAs1Ir aGVp/4oinLPV2oo1p/FyMxcOvDivMSwWA1EKFz5I/4H+R4XPiAlj3VCHfB9qK4KfGjQxCBuI6kt2 8EC6QNrFEQxOdtEGkyZlTZuNTtpqebO+4Em34HvC2Fqys/j7nMYuhjOXnZOLF2nszMKOre3aQlOD Z0+mKCyN85OMcYz5Xav8wxMf3QVHb8EF/4QpaYIJflUSGEbPgckDSH7L0Wzd+BsAAP//AwBQSwME FAAGAAgAAAAhAA3RkJ+2AAAAGwEAACcAAAB0aGVtZS90aGVtZS9fcmVscy90aGVtZU1hbmFnZXIu eG1sLnJlbHOEj00KwjAUhPeCdwhvb9O6EJEm3YjQrdQDhOQ1DTY/JFHs7Q2uLAguh2G+mWm7l53J E2My3jFoqhoIOumVcZrBbbjsjkBSFk6J2TtksGCCjm837RVnkUsoTSYkUiguMZhyDidKk5zQilT5 gK44o49W5CKjpkHIu9BI93V9oPGbAXzFJL1iEHvVABmWUJr/s/04GolnLx8WXf5RQXPZhQUoosbM 4CObqkwEylu6usTfAAAA//8DAFBLAQItABQABgAIAAAAIQCCirwT+gAAABwCAAATAAAAAAAAAAAA AAAAAAAAAABbQ29udGVudF9UeXBlc10ueG1sUEsBAi0AFAAGAAgAAAAhAKXWp+fAAAAANgEAAAsA AAAAAAAAAAAAAAAAKwEAAF9yZWxzLy5yZWxzUEsBAi0AFAAGAAgAAAAhAGt5lhaDAAAAigAAABwA AAAAAAAAAAAAAAAAFAIAAHRoZW1lL3RoZW1lL3RoZW1lTWFuYWdlci54bWxQSwECLQAUAAYACAAA ACEA41pJJ5cGAABVGwAAFgAAAAAAAAAAAAAAAADRAgAAdGhlbWUvdGhlbWUvdGhlbWUxLnhtbFBL AQItABQABgAIAAAAIQAN0ZCftgAAABsBAAAnAAAAAAAAAAAAAAAAAJwJAAB0aGVtZS90aGVtZS9f cmVscy90aGVtZU1hbmFnZXIueG1sLnJlbHNQSwUGAAAAAAUABQBdAQAAlwoAAAAAPD94bWwgdmVy c2lvbj0iMS4wIiBlbmNvZGluZz0iVVRGLTgiIHN0YW5kYWxvbmU9InllcyI/Pg0KPGE6Y2xyTWFw IHhtbG5zOmE9Imh0dHA6Ly9zY2hlbWFzLm9wZW54bWxmb3JtYXRzLm9yZy9kcmF3aW5nbWwvMjAw Ni9tYWluIiBiZzE9Imx0MSIgdHgxPSJkazEiIGJnMj0ibHQyIiB0eDI9ImRrMiIgYWNjZW50MT0i YWNjZW50MSIgYWNjZW50Mj0iYWNjZW50MiIgYWNjZW50Mz0iYWNjZW50MyIgYWNjZW50ND0iYWNj ZW50NCIgYWNjZW50NT0iYWNjZW50NSIgYWNjZW50Nj0iYWNjZW50NiIgaGxpbms9ImhsaW5rIiBm b2xIbGluaz0iZm9sSGxpbmsiLz4FAGgAbwBlAG4AZQAKBAAAnhEAAMAcAAAuKgAAwjwAACdAAACY QQAAp40AAAEAaAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA2a5gEAEAaAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAca9gEAEAaAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAPLBgEAEAaAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAkU5nEAEAaAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAFlVnEAEAaAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAsGFnEAEAaAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAfVVnEBKY6kYAAAAAAAAA AAAAAAAAABKY6kYAAAAAAAAAAAAAAAAAABKY6kYAAAAAAAAAAAAAAAAAABKY6kYAAAAAAAAAAAAA AAAAABKY6kYAAAAAAAAAAAAAAAAAABKY6kYAAAAAAAAAAAAAAAAAABKY6kYAAAAAAAAAAAAAAAAA AAAAAABKAAAAbgAAAPAAAADqAQAAMAIAAIsCAAC9AgAAwAIAAAAAAACnjQAACAAAAAEAAAD///// AAgAAAgbAAAGMgAANT8AAM9AAAA2QgAAdkQAAIZGAACYSQAA1VIAAGd5AAAIewAAgX4AAFeTAACn lQAASwAAAFMAAABXAAAAXAAAAF0AAABfAAAAYAAAAGEAAABjAAAAZQAAAG8AAABwAAAAcQAAAH8A AAAACAAA+QwAADoRAACQFgAAThcAAI0YAACgGQAAeiAAACkmAAAbLQAA8DIAAPczAAAwNgAAZzwA AKRAAABzRwAAQk4AAJdTAABhVQAAtlgAAJhfAACjZAAA0WoAABluAACocwAAFngAAJ2AAABSggAA b4IAAF2DAAB6gwAAAIYAAB2GAABtiAAAYI0AAOGPAAD+jwAAcpEAAJaSAACnlQAATAAAAE0AAABO AAAATwAAAFAAAABRAAAAUgAAAFQAAABVAAAAVgAAAFgAAABZAAAAWgAAAFsAAABeAAAAYgAAAGQA AABmAAAAZwAAAGgAAABpAAAAagAAAGsAAABsAAAAbQAAAG4AAAByAAAAcwAAAHQAAAB1AAAAdgAA AHcAAAB4AAAAeQAAAHoAAAB7AAAAfAAAAH0AAAB+AAAADwAA8DgAAAAAAAbwGAAAAAIIAAACAAAA AQAAAAEAAAABAAAAAgAAAEAAHvEQAAAA//8AAAAA/wCAgIAA9wAAEAAPAALwkgAAABAACPAIAAAA AQAAAAEEAAAPAAPwMAAAAA8ABPAoAAAAAQAJ8BAAAAAAAAAAAAAAAAAAAAAAAAAAAgAK8AgAAAAA BAAABQAAAA8ABPBCAAAAEgAK8AgAAAABBAAAAA4AAFMAC/AeAAAAvwEAABAAywEAAAAA/wEAAAgA BAMJAAAAPwMBAAEAAAAR8AQAAAABAAAA//8HAAoAAAAAAdmuYBD/////AAAAAXGvYBD/////AAAA ATywYBD/////AAAAAZFOZxD/////AAAAARZVZxD/////AAAAAbBhZxD/////AAAAAX1VZxD///// /AMAAH0RAADaGgAACSoAAKw8AACsPwAAZkEAAOqKAAAAAAAAAQAAAAIAAAADAAAABAAAAAUAAAAG AAAACgQAAJ4RAADAHAAALioAAMI8AAAnQAAAmEEAAOqKAAAAAAAARgAAAEsAAABVAAAAWgAAAIgA AACPAAAAlQAAAJ0AAACeAAAApAAAANYAAADdAAAA3gAAAOUAAAChAQAApgEAAEACAABKAgAAFQMA ABwDAAAyAwAAPAMAAD0DAABAAwAAQQMAAEQDAABXAwAAXQMAAF4DAABhAwAAqQMAAK0DAACuAwAA tgMAALcDAAC/AwAAwAMAAMcDAADIAwAA0gMAANMDAADWAwAA1wMAANsDAADfAwAA6QMAAOoDAADt AwAA8QMAAPsDAAD+AwAABgQAAAcEAAAKBAAADAQAABEEAAASBAAAGAQAABkEAAAcBAAAKgQAADQE AAA7BAAAPgQAAEIEAABNBAAATgQAAFUEAABXBAAAYQQAAGMEAABvBAAAcAQAAHsEAAB9BAAAgAQA AIQEAACQBAAAkQQAAJkEAACaBAAAoAQAAKoEAACsBAAArQQAALEEAAC7BAAAvwQAAMkEAADOBAAA zwQAANEEAADfBAAA4wQAAOQEAADvBAAA8AQAAPQEAAD1BAAA+AQAAPwEAAAGBQAABwUAAAkFAAAR BQAAFAUAACoFAAAwBQAAMQUAADQFAAA9BQAARgUAAEcFAABJBQAASgUAAE0FAAB/BQAAgwUAAIQF AACJBQAAigUAAJAFAACRBQAAlAUAAJoFAACkBQAAsAUAALkFAAC6BQAAvAUAAMYFAADMBQAA0wUA ANcFAADYBQAA2gUAANsFAADiBQAA8AUAAPYFAAD3BQAA+QUAAPoFAAD8BQAANwYAAD0GAAA+BgAA QQYAAEIGAABHBgAASAYAAFEGAABYBgAAWwYAAF4GAABlBgAAZgYAAGgGAABpBgAAbAYAAG0GAABz BgAAdwYAAHoGAAB7BgAAfgYAAH8GAACBBgAAggYAAIkGAACLBgAAkwYAAJUGAACXBgAAmAYAAKEG AACiBgAApAYAAKUGAACqBgAAqwYAALQGAAC1BgAAtwYAALgGAAC7BgAAxgYAAMgGAADJBgAAywYA AMwGAADZBgAA2gYAANwGAADdBgAA4AYAAOoGAADwBgAA8QYAAPMGAAD0BgAA/QYAAAoHAAAMBwAA DQcAAA8HAAAQBwAAFAcAABUHAAAZBwAAGgcAAB8HAAAgBwAAJAcAACUHAAAnBwAAKQcAAC0HAAAx BwAAOQcAAEAHAABEBwAATgcAAFMHAABZBwAAXwcAAH4HAACEBwAApwcAAKoHAACrBwAArgcAAK8H AAC2BwAAtwcAAMEHAADCBwAAxAcAAMUHAADIBwAAzAcAANQHAADVBwAA3AcAAOMHAADpBwAA6gcA APIHAAD4BwAA/AcAAP0HAAAFCAAABggAAAgIAAAJCAAAEAgAABEIAAATCAAAGwgAAB4IAAAfCAAA IggAACgIAAAvCAAAOQgAAEMIAABECAAATAgAAE0IAABPCAAAVQgAAF4IAACMCAAAkggAAJYIAACZ CAAAmggAAJ4IAACfCAAAoQgAAKUIAACqCAAAvggAAMUIAAD5CAAA/ggAABIJAAAcCQAALAkAADMJ AAA+CQAASQkAAEoJAABMCQAATQkAAFEJAABSCQAAWgkAAF0JAABjCQAAZAkAAGoJAABrCQAAcAkA AHEJAAB6CQAAfgkAAIcJAACJCQAAiwkAAIwJAACQCQAAkQkAAJkJAACaCQAAngkAAJ8JAAClCQAA pgkAAKkJAACqCQAAtgkAALcJAAC7CQAAvAkAAMMJAADHCQAAyQkAAMoJAADOCQAAzwkAANcJAADq CQAA8wkAAPQJAAD4CQAA+QkAAP0JAAD+CQAABgoAAA8KAAAWCgAAFwoAACEKAAAmCgAALQoAAC4K AAAyCgAAMwoAADUKAAA2CgAAPwoAAEMKAABKCgAATwoAAFEKAABVCgAAWAoAAGUKAABvCgAAcAoA AHMKAAB0CgAAdwoAAHgKAACACgAAgQoAAIgKAACJCgAAkQoAAJIKAACUCgAAmAoAAKEKAAClCgAA qAoAAKkKAACzCgAAuAoAAL8KAADJCgAAywoAAN0KAADpCgAAMgsAADkLAAB7CwAAggsAAIQLAACH CwAAiAsAAJgLAAC4CwAAxAsAAMULAADQCwAAAQwAAA0MAABKDAAAVgwAAFcMAABbDAAAnAwAAKoM AADhDAAA7wwAACUNAAA0DQAARQ4AAEwOAABODgAAVw4AAKcOAACsDgAAwA4AAMcOAAD7DgAAAA8A ABQPAAAeDwAALg8AADUPAABBDwAATQ8AAFIPAABWDwAAVw8AAF8PAABgDwAAaQ8AAGwPAAB1DwAA dg8AAH0PAAB+DwAAgQ8AAIIPAACGDwAAhw8AAIkPAACKDwAAjQ8AAJkPAACoDwAAqQ8AAKsPAACu DwAAsw8AALQPAAC4DwAAuQ8AALsPAAC8DwAAxQ8AAMYPAADJDwAAyg8AAM0PAADRDwAA3A8AAOkP AAD1DwAA9g8AAPkPAAD6DwAA/g8AAP8PAAACEAAAAxAAAAUQAAAGEAAADBAAAA0QAAAYEAAAGRAA ABwQAAAdEAAAIxAAACcQAAAyEAAAMxAAADgQAAA5EAAARBAAAEUQAABPEAAAURAAAFgQAABZEAAA YhAAAGQQAABnEAAAbxAAAHQQAACiEAAApxAAALsQAADCEAAA9hAAAPsQAAAPEQAAGREAACkRAAAw EQAASBEAAE8RAABYEQAAXxEAAGARAABoEQAAaREAAG0RAABuEQAAcBEAAHERAAB5EQAAehEAAHwR AACGEQAAihEAAIsRAACOEQAAoxEAAKcRAACuEQAAsxEAALQRAAC5EQAAuhEAAMARAADBEQAAxBEA ANcRAAD2EQAA/hEAAAISAAADEgAACBIAAAkSAAAQEgAAFBIAAB8SAAAgEgAAIhIAACUSAAAvEgAA MBIAADMSAAA0EgAAORIAADsSAABFEgAARhIAAEgSAABMEgAAVBIAAFgSAABdEgAAXhIAAGASAABh EgAAZRIAAGYSAABpEgAAahIAAHYSAAB4EgAAehIAAHsSAACGEgAAhxIAAI4SAACPEgAAkhIAAJMS AACVEgAAlhIAAJoSAACeEgAAphIAAK4SAACyEgAAsxIAALoSAADBEgAAyRIAAMoSAADPEgAA0BIA ANkSAADaEgAA3RIAAN4SAADiEgAA5hIAAO0SAADuEgAA8BIAAPESAAD1EgAA9hIAAPsSAAD+EgAA ABMAAAETAAAHEwAADhMAABgTAAAZEwAAJRMAACcTAAAqEwAALhMAADkTAAA6EwAAPBMAAEATAABI EwAASRMAAE4TAABUEwAAVhMAAFcTAABeEwAAXxMAAGETAABiEwAAZhMAAGcTAABtEwAAcRMAAHgT AAB5EwAAexMAAHwTAAB/EwAAhRMAAIgTAACJEwAAjxMAAJATAACUEwAAlRMAAJ8TAACgEwAAqRMA AKoTAACsEwAArRMAALATAACxEwAAthMAALoTAAC8EwAAvRMAAMMTAADHEwAAyhMAAMsTAADPEwAA 0BMAANITAADTEwAA2xMAANwTAADeEwAA3xMAAOETAADiEwAA6hMAAOsTAADvEwAA8BMAAPITAADz EwAA9hMAAPoTAAAEFAAABRQAAAcUAAAIFAAACxQAAAwUAAAXFAAAGBQAABoUAAAbFAAAJhQAACcU AAApFAAAKhQAAC8UAAA7FAAARhQAAEgUAABUFAAAVRQAAFkUAABaFAAAXBQAAF0UAABfFAAAYBQA AGUUAABmFAAAahQAAGsUAABuFAAAbxQAAHQUAAB1FAAAeRQAAH0UAACEFAAAhRQAAIkUAACKFAAA jhQAAI8UAACVFAAAlhQAAJkUAACfFAAAoRQAAKIUAACoFAAAqRQAAKsUAACsFAAArhQAALkUAAC9 FAAAvhQAAMEUAADFFAAAzRQAAM4UAADTFAAA1BQAANgUAADdFAAA6BQAAOkUAADrFAAA7BQAAO8U AAAGFQAADRUAABsVAAAgFQAAIRUAACwVAAAtFQAALxUAADAVAAA4FQAAORUAADsVAAA8FQAAQxUA AEQVAABGFQAARxUAAE4VAABdFQAAaRUAAG8VAAB3FQAAeBUAAHsVAAB8FQAAiBUAAIkVAACLFQAA jBUAAI4VAACPFQAAkhUAAJMVAACVFQAAqRUAAK4VAACzFQAAthUAALwVAAC+FQAAyxUAANAVAADa FQAA5BUAAOUVAADsFQAA7RUAAPAVAADxFQAA+xUAAPwVAAD+FQAA/xUAAAMWAAAEFgAAERYAABIW AAAbFgAAIxYAACsWAAAsFgAALhYAAC8WAAAyFgAAOxYAAD0WAAA+FgAARhYAAEcWAABOFgAAUBYA AFIWAABTFgAAVhYAAFcWAABcFgAAYBYAAGIWAABqFgAAbhYAAG8WAABxFgAAchYAAHcWAAB4FgAA hRYAAIYWAACJFgAAihYAAJAWAACRFgAAlBYAAJUWAACaFgAAmxYAAJ0WAACeFgAAoRYAAK0WAAC0 FgAAthYAALsWAAC8FgAAyRYAAMoWAADQFgAA0RYAANMWAADXFgAA2hYAAOAWAADiFgAA8xYAAPkW AAD7FgAAAxcAAAQXAAAJFwAAChcAAA8XAAAQFwAAEhcAABMXAAAgFwAAIRcAACQXAAAqFwAAMhcA ADoXAAA8FwAAPRcAAEYXAABUFwAAVxcAAFsXAABeFwAAXxcAAGUXAAB8FwAAghcAAIQXAACRFwAA khcAAJQXAACVFwAAmBcAAJ4XAAChFwAAohcAAKkXAACrFwAAsBcAALEXAAC2FwAAvhcAAMUXAADG FwAAzhcAANUXAADXFwAA7BcAAO8XAAD+FwAACRgAAAoYAAANGAAADhgAABMYAAAUGAAAFhgAABcY AAAhGAAAJxgAADMYAAA0GAAANxgAADgYAAA6GAAAQhgAAE4YAABPGAAAUxgAAFcYAABaGAAAWxgA AGMYAABkGAAAaBgAAGkYAABrGAAAbBgAAHAYAABxGAAAdhgAAIEYAACNGAAAkBgAAJsYAACdGAAA qRgAAKoYAACtGAAArhgAALoYAADCGAAAyhgAAMsYAADNGAAAzhgAANMYAADYGAAA4hgAAOMYAADu GAAA8BgAAPMYAAD/GAAACxkAAAwZAAAPGQAAEBkAABgZAAAZGQAAGxkAABwZAAAfGQAAIBkAACwZ AAAtGQAALxkAADAZAAAzGQAANBkAAD4ZAABGGQAASBkAAEkZAABOGQAATxkAAFQZAABaGQAAXhkA AF8ZAABhGQAAYhkAAGoZAACAGQAAjRkAAI4ZAACRGQAAkhkAAJoZAACbGQAAnhkAAJ8ZAACoGQAA qRkAAK8ZAACwGQAAsxkAAMoZAADNGQAA1BkAANoZAADbGQAA3xkAAOIZAADpGQAA6hkAAPcZAAD4 GQAA+hkAAPsZAAD/GQAAABoAAAIaAAAKGgAAFBoAABUaAAAbGgAAHBoAAB8aAAAnGgAALBoAADQa AAA6GgAAPBoAAD8aAABAGgAARxoAAEgaAABKGgAAUhoAAFUaAABWGgAAYhoAAGMaAABrGgAAbBoA AHIaAABzGgAAfhoAAH8aAACGGgAAhxoAAIoaAACLGgAAkhoAAJoaAACfGgAAoBoAAKgaAACpGgAA qxoAAKwaAACzGgAAtBoAALcaAAC4GgAAwBoAAM8aAADYGgAA4RoAAOUaAADmGgAA6BoAAOkaAADv GgAA8BoAAPIaAADzGgAA+hoAAPsaAAAIGwAACRsAAA0bAAAbGwAAIRsAACwbAAAxGwAAMhsAADcb AAA+GwAAQBsAAEEbAABOGwAATxsAAFMbAABcGwAAYhsAAGYbAABpGwAAexsAAH8bAACAGwAAghsA AIMbAACLGwAAjBsAAJUbAACWGwAAnRsAAJ4bAACgGwAArxsAALsbAAC8GwAAwBsAAMEbAADEGwAA zhsAANEbAADSGwAA1BsAAOAbAADjGwAA5BsAAOobAADrGwAA7RsAAPobAAAEHAAABRwAAA4cAAAP HAAAEhwAABQcAAAYHAAAIRwAACccAAArHAAALhwAADocAAA+HAAAPxwAAEEcAABCHAAASBwAAEkc AABNHAAAThwAAFkcAABaHAAAXxwAAGAcAABjHAAAZBwAAG8cAABwHAAAchwAAHMcAAB2HAAAfhwA AIMcAACEHAAAhxwAAIgcAACKHAAAixwAAJEcAACSHAAAnRwAAJ4cAACgHAAAoRwAAKQcAAClHAAA rxwAALAcAAC1HAAAxBwAAMkcAADdHAAA5BwAABgdAAAdHQAAMR0AADsdAABLHQAAUh0AANEdAADV HQAA1h0AANodAADbHQAA3x0AAOgdAADyHQAA8x0AAPodAAD7HQAAAR4AAAIeAAAKHgAACx4AAA4e AAAPHgAAHB4AAB0eAAAfHgAAIh4AACgeAAAwHgAANR4AADYeAAA4HgAAOR4AADweAAA9HgAAQh4A AEMeAABPHgAAUB4AAFUeAABWHgAAWB4AAFkeAABcHgAAXR4AAF8eAABgHgAAZB4AAGUeAABqHgAA dB4AAHgeAAB7HgAAhR4AAIYeAACMHgAAjR4AAI8eAACQHgAAmR4AAJoeAACdHgAAnh4AAKIeAACj HgAAqx4AAK0eAACwHgAAuB4AAMIeAADDHgAAyh4AANAeAADYHgAA2R4AANweAADdHgAA5h4AAOge AADyHgAA9B4AAPceAAAKHwAAER8AABIfAAAUHwAAFR8AABgfAAAZHwAAJB8AAC4fAAAyHwAAMx8A ADcfAABEHwAATB8AAFEfAABVHwAAVh8AAFkfAABhHwAAYx8AAGYfAABrHwAAdh8AAHkfAAB6HwAA gh8AAIMfAACFHwAAjR8AAJAfAACcHwAAph8AAKgfAACyHwAAtB8AAL4fAADAHwAAyh8AAOIfAADk HwAA5R8AAOcfAADoHwAA8B8AAPEfAAD1HwAA9h8AAPofAAD7HwAA/R8AAP4fAAABIAAAAiAAAA0g AAAaIAAAHiAAACogAAAsIAAALSAAADAgAAAxIAAAOCAAADkgAABFIAAASCAAAFEgAABTIAAAVSAA AFYgAABYIAAAWSAAAGEgAABiIAAAZiAAAHMgAAB/IAAAhSAAAIsgAACPIAAAlyAAAJggAACdIAAA nyAAAKEgAACsIAAAriAAALYgAADAIAAAwSAAAMQgAADFIAAA0CAAANEgAADYIAAA2yAAAN8gAADg IAAA6CAAAOkgAADuIAAA+yAAAP8gAAAAIQAAAiEAAAMhAAAHIQAACCEAAAohAAALIQAADiEAAA8h AAAbIQAAHCEAAB4hAAAfIQAAJyEAAC0hAAA1IQAAQiEAAEYhAABHIQAASSEAAEohAABMIQAATSEA AFUhAABWIQAAXSEAAGshAAB2IQAAdyEAAH4hAACPIQAAkSEAAJIhAACWIQAAmyEAAJ0hAACeIQAA oSEAAKIhAACuIQAAsCEAALIhAACzIQAAtSEAALYhAAC6IQAAuyEAAMchAADIIQAAyiEAANIhAADa IQAA4iEAAOohAADrIQAA8CEAAPEhAAD2IQAA9yEAAPkhAAD6IQAA/SEAAP4hAAAAIgAAASIAAAQi AAAFIgAAECIAACAiAAAqIgAAKyIAAC0iAAAuIgAAMCIAADEiAAAzIgAANCIAADciAAA4IgAAPCIA AD0iAABCIgAAQyIAAEciAABIIgAATSIAAE4iAABSIgAAUyIAAFUiAABWIgAAWSIAAGEiAABtIgAA ciIAAH0iAAB+IgAAgCIAAIEiAACEIgAAhSIAAIciAACIIgAAjCIAAI0iAACPIgAAkCIAAJMiAACU IgAAnyIAAK8iAACxIgAAtSIAAL0iAAC+IgAAwSIAAMIiAADHIgAAzyIAANEiAADSIgAA1SIAANYi AADfIgAA4CIAAOUiAADmIgAA6CIAAPAiAAD0IgAA9SIAAPkiAAD/IgAAASMAABIjAAAcIwAAIyMA AC8jAABAIwAARyMAAEsjAABRIwAAUiMAAFcjAABZIwAAXCMAAF0jAABgIwAAYSMAAGwjAABtIwAA ciMAAIEjAACNIwAAjiMAAJQjAACVIwAAmSMAAJojAAChIwAAoiMAAKQjAAClIwAAsCMAALEjAAC4 IwAAuSMAAL0jAADFIwAAziMAAM8jAADRIwAA0iMAANUjAADWIwAA3iMAAN8jAADkIwAA5SMAAO4j AADvIwAA/SMAAP8jAAAEJAAADCQAAA4kAAAQJAAAGSQAABskAAAdJAAAHiQAACIkAAAuJAAAOCQA AD4kAABAJAAAQSQAAEskAABUJAAAXiQAAGAkAABqJAAAbCQAAG8kAABwJAAAdyQAAHgkAAB+JAAA hSQAAJEkAACZJAAAnyQAAKAkAACnJAAAqCQAAKokAACrJAAAsSQAALIkAAC1JAAAtiQAAL0kAAC/ JAAAxyQAAMgkAADNJAAAziQAANIkAADTJAAA1iQAAN4kAADpJAAA8iQAAPwkAAD+JAAAAyUAAAQl AAAGJQAAByUAAA4lAAAQJQAAFSUAABclAAAaJQAAIiUAACclAAAoJQAALSUAAC4lAAA3JQAAOyUA AEAlAABBJQAATCUAAE0lAABPJQAAUCUAAFclAABYJQAAWyUAAGMlAABrJQAAbCUAAHElAAB0JQAA dyUAAHglAAB+JQAAfyUAAIElAACCJQAAhSUAAIYlAACOJQAAjyUAAJQlAACVJQAAmiUAAJslAACd JQAApSUAAKwlAACuJQAAsiUAALMlAAC2JQAAwyUAAM0lAADOJQAA0CUAANElAADUJQAA1SUAANol AADcJQAA3iUAAN8lAADiJQAA6iUAAPYlAAD3JQAA/CUAAP0lAAACJgAAAyYAAAcmAAAIJgAAECYA ABEmAAATJgAAFCYAAB0mAAAeJgAAISYAACImAAAqJgAAMiYAADcmAAA4JgAAOyYAADwmAABFJgAA SCYAAFImAABTJgAAViYAAFcmAABaJgAAYSYAAGsmAABsJgAAcCYAAHgmAAB9JgAAfiYAAIAmAACB JgAAgyYAAI0mAACSJgAAkyYAAJcmAACdJgAAoSYAAKImAAClJgAApiYAALImAAC6JgAAwSYAAMUm AADIJgAAySYAANUmAADWJgAA3iYAAOEmAADmJgAA5yYAAOomAADrJgAA8SYAAP4mAAAHJwAACCcA AA8nAAARJwAAFCcAABUnAAAfJwAAICcAACUnAAArJwAALScAAC4nAAAyJwAAMycAADUnAAA9JwAA RycAAEgnAABKJwAASycAAE8nAABQJwAAVCcAAFYnAABdJwAAXicAAGAnAABlJwAAaScAAHcnAACR JwAAlCcAAJUnAACXJwAAmCcAAJsnAACcJwAApycAAK8nAACwJwAAsScAALQnAAC1JwAAuicAAMIn AADIJwAAyScAAM8nAADQJwAA0icAANMnAADeJwAA3ycAAOgnAADpJwAA7ScAAO4nAADyJwAACigA AA8oAAAlKAAAKSgAACooAAAuKAAAMigAADUoAAA2KAAAOygAADwoAABFKAAARigAAFQoAABdKAAA ZygAAGgoAABrKAAAbCgAAHEoAAByKAAAfCgAAH0oAACEKAAAhSgAAIsoAACMKAAAkygAAJQoAACe KAAApigAAKgoAACpKAAArCgAAK0oAACyKAAAsygAALUoAAC2KAAAvygAAMAoAADFKAAAxigAAM0o AADWKAAA4igAAOYoAADrKAAA/ygAAAYpAAA6KQAAPykAAFMpAABdKQAAbSkAAHQpAACDKQAAiSkA AIopAACQKQAAkikAAJgpAACZKQAAnCkAALApAACzKQAAtCkAALspAAC8KQAAwykAAMspAADRKQAA 0ykAANYpAADXKQAA2ikAANspAADfKQAA4ykAAO4pAAASKgAAEyoAAC8qAAAxKgAANCoAADwqAABH KgAASCoAAFIqAABTKgAAWSoAAFoqAABfKgAAYCoAAGcqAABoKgAAayoAAGwqAABxKgAAcioAAHQq AAB8KgAAiCoAAIkqAACLKgAAjyoAAJMqAACUKgAAlyoAAJgqAAChKgAAoioAAKQqAACoKgAArioA AL8qAADLKgAAzCoAAM4qAADPKgAA1yoAANgqAADbKgAA5ioAAO4qAAD4KgAA/CoAAAIrAAAOKwAA DysAABQrAAAVKwAAGSsAABorAAAdKwAAHisAACMrAAAkKwAAKSsAACorAAAzKwAANCsAADkrAABB KwAARCsAAEUrAABRKwAAWCsAAGQrAABlKwAAaCsAAGkrAABuKwAAbysAAHIrAABzKwAAdysAAHgr AAB6KwAAgisAAJArAACRKwAAlCsAAJUrAACaKwAAoSsAALErAACyKwAAuSsAALorAAC8KwAAxCsA AM0rAADOKwAA0ysAANQrAADbKwAACCwAAA0sAAAhLAAAKCwAAFwsAABhLAAAdSwAAH8sAACPLAAA liwAAK4sAAC1LAAAtywAALosAAC7LAAAyywAANQsAADeLAAA3ywAAOEsAADiLAAA5SwAAOYsAADr LAAA7CwAAPEsAADyLAAA+ywAAPwsAAACLQAACi0AABQtAAAYLQAAGy0AABwtAAAnLQAAKC0AAC8t AAAwLQAANy0AADgtAAA6LQAAQC0AAEQtAABFLQAASy0AAEwtAABQLQAAUS0AAFktAABaLQAAXC0A AGAtAABlLQAAZi0AAGotAABrLQAAcS0AAHItAAB9LQAAfi0AAIAtAACBLQAAhC0AAIUtAACRLQAA lC0AAJktAACaLQAAnS0AAJ4tAACiLQAApi0AAKotAACuLQAAsy0AALQtAAC+LQAAvy0AAMktAADK LQAAzS0AAM4tAADQLQAA0S0AANotAADdLQAA4C0AAOEtAADmLQAA6y0AAPotAAD7LQAAAi4AAAMu AAAFLgAABi4AAAkuAAAKLgAADy4AABAuAAASLgAAEy4AABkuAAAaLgAAIS4AACIuAAAkLgAAJS4A AC8uAAAzLgAANS4AADYuAAA5LgAAOi4AAEMuAABELgAAUC4AAGQuAABnLgAAaC4AAG4uAABzLgAA fC4AAH0uAAB/LgAAiS4AAJMuAACULgAAni4AAKEuAACkLgAApS4AAKkuAACqLgAAsS4AALUuAAC8 LgAAxC4AAMguAADJLgAA1y4AANguAADfLgAA4C4AAOMuAADkLgAA7C4AAO4uAADwLgAA8S4AAPQu AAD4LgAABS8AAAYvAAAMLwAADi8AABYvAAAXLwAAHi8AAB8vAAAlLwAAJi8AACgvAAApLwAAMi8A ADsvAABCLwAAQy8AAEUvAABGLwAATi8AAE8vAABWLwAAVy8AAFwvAABdLwAAXy8AAGAvAABiLwAA Yy8AAGcvAABzLwAAfy8AAIgvAACLLwAAjC8AAJMvAACULwAAmy8AAJwvAAChLwAAoi8AAKUvAACm LwAAqC8AAKkvAACsLwAArS8AALgvAAC5LwAAwC8AAMQvAADRLwAA2C8AANovAAD3LwAAATAAAAIw AAAFMAAABjAAAA0wAAAPMAAAETAAABcwAAAZMAAAGjAAACMwAAAkMAAAJjAAACcwAAArMAAALDAA AC8wAAAwMAAAMjAAADMwAAA2MAAAQzAAAEwwAABNMAAATzAAAFYwAABZMAAAXzAAAGQwAABlMAAA cDAAAHEwAAB4MAAAezAAAH8wAACIMAAAjjAAAI8wAACTMAAAnjAAAKQwAAClMAAAqTAAALQwAAC9 MAAAvjAAAMEwAADCMAAAxTAAAMkwAADOMAAAzzAAANEwAADSMAAA1jAAANcwAADaMAAA2zAAAOAw AADhMAAA5DAAAOUwAADrMAAA7DAAAO8wAADwMAAA+DAAAPkwAAD9MAAA/jAAAAgxAAARMQAAFDEA ABoxAAAdMQAAHzEAACoxAAArMQAAMzEAADQxAAA+MQAARDEAAEYxAABHMQAAUTEAAFIxAABUMQAA WDEAAGExAABiMQAAZTEAAGYxAABrMQAAbTEAAHExAAB0MQAAezEAAHwxAACBMQAAgjEAAIUxAACG MQAAkDEAAJUxAACaMQAApTEAAKgxAACpMQAAsDEAALExAAC5MQAAvTEAAMAxAADBMQAAzDEAANUx AADaMQAA2zEAAN4xAADiMQAA5zEAAO0xAAD9MQAA/jEAAAAyAAABMgAABDIAAAsyAAAQMgAAFjIA ABgyAAAZMgAAHjIAACIyAAAlMgAAJjIAAC8yAAAzMgAAQTIAAEIyAABGMgAASDIAAEsyAABTMgAA VjIAAGAyAABuMgAAczIAAHcyAAB7MgAAfjIAAH8yAACBMgAAgjIAAIsyAACMMgAAjjIAAI8yAACa MgAAmzIAAKIyAACjMgAApzIAAKgyAACqMgAArDIAAK4yAACvMgAAszIAALQyAAC2MgAAvjIAAMYy AADIMgAAzTIAAM4yAADRMgAA0jIAANkyAADaMgAA3jIAAN8yAADhMgAA4jIAAOUyAADmMgAA8zIA APQyAAD4MgAA+TIAAAMzAAAHMwAACTMAAAozAAANMwAAITMAACgzAAAyMwAANjMAADczAABBMwAA QjMAAEgzAABJMwAASzMAAE8zAABYMwAAWTMAAF4zAABfMwAAZTMAAGYzAABqMwAAazMAAHMzAAB0 MwAAfTMAAH4zAACCMwAAgzMAAIYzAACHMwAAizMAAIwzAACPMwAAnTMAAKMzAACkMwAAqDMAAKkz AAC0MwAAtTMAALszAADCMwAAxzMAAMgzAADKMwAAyzMAAM4zAADSMwAA4jMAAOMzAADoMwAA6TMA AOwzAADtMwAA7zMAAPgzAAD+MwAA/zMAAAM0AAAENAAABzQAAAg0AAAUNAAAFTQAABk0AAAaNAAA HjQAAB80AAAoNAAAMDQAADU0AAA2NAAAOTQAAEI0AABFNAAARjQAAFA0AABRNAAAUzQAAFQ0AABX NAAAWDQAAGE0AABiNAAAZjQAAG40AAB0NAAAdTQAAIA0AACBNAAAizQAAI80AACcNAAAnTQAAKQ0 AAClNAAAqDQAAKk0AACuNAAAtTQAALc0AAC4NAAAwDQAAME0AADENAAAxzQAANE0AADSNAAA3TQA AOY0AADrNAAA7DQAAPc0AAD4NAAA+zQAAPw0AAD/NAAAADUAABA1AAARNQAAFjUAABc1AAAZNQAA IDUAACc1AAArNQAALTUAAIc1AACwNQAAtTUAAMk1AADTNQAA4zUAAOo1AAD2NQAAeDYAAHk2AAC2 NgAAuDYAAGQ3AABmNwAApTcAAKs4AACuOAAArzgAALQ4AAC1OAAAvjgAAL84AADBOAAAwjgAAMU4 AADUOAAA1jgAANw4AADlOAAA5zgAAOs4AADwOAAA9zgAAAA5AAAGOQAABzkAAAs5AAAMOQAADzkA ACc5AAAwOQAAMTkAAD85AABAOQAAQjkAAEc5AABPOQAAUjkAAFk5AABaOQAAXDkAAF85AABoOQAA aTkAAHc5AAB4OQAAejkAAHs5AACAOQAAhjkAAI05AACOOQAAkDkAAJE5AACUOQAAlTkAAJk5AACa OQAAoDkAAKE5AACjOQAApDkAAKc5AACoOQAArzkAALA5AAC2OQAAtzkAALs5AAC9OQAAvzkAAMA5 AADCOQAAwzkAAMk5AADKOQAAzjkAAM85AADTOQAA2jkAAOQ5AADlOQAA6DkAAO45AADyOQAA+DkA APo5AAD7OQAAADoAAAE6AAAFOgAACzoAAA06AAAUOgAAGjoAAB46AAAkOgAAJToAACc6AAAoOgAA KzoAADA6AAAzOgAANDoAADc6AAA4OgAAQDoAAEE6AABMOgAAUjoAAFQ6AABVOgAAXzoAAGA6AABi OgAAajoAAHU6AAB2OgAAfjoAAH86AACJOgAAijoAAI46AACPOgAAlzoAAJg6AACbOgAAnDoAAKM6 AACkOgAArToAAK46AAC8OgAAwjoAAMQ6AADFOgAAyToAAMo6AADQOgAA0joAANw6AADdOgAA4joA AOM6AADlOgAA5joAAOk6AADqOgAA+DoAAPk6AAD8OgAAVzsAAF07AABlOwAAaDsAAGw7AABtOwAA eDsAAHk7AAB7OwAAfDsAAH87AACAOwAAiTsAAI87AACROwAAkjsAAJo7AACcOwAAnzsAAKA7AACj OwAApDsAAKo7AACrOwAArTsAALY7AAC4OwAAuTsAAMA7AADBOwAAxDsAAMU7AADLOwAAzzsAANw7 AADdOwAA4DsAAOE7AADnOwAA7zsAAPM7AAD1OwAA+zsAAPw7AAD/OwAAADwAAAs8AAAMPAAAFjwA ABc8AAAaPAAAGzwAACE8AAAiPAAAJjwAACg8AAArPAAAMzwAADk8AAA6PAAAPzwAAEA8AABEPAAA RTwAAE88AABZPAAAXjwAAF88AABsPAAAbTwAAHI8AABzPAAAdTwAAHg8AACCPAAAkjwAAJc8AACf PAAApzwAAKg8AACrPAAArDwAALU8AAC2PAAAuzwAAMU8AACCPgAAhj4AAJI+AACTPgAAnj4AAKM+ AAClPgAApj4AAKg+AACpPgAArD4AAK0+AAC7PgAAvD4AAL4+AAC/PgAAwj4AANE+AADTPgAA1D4A ANo+AADbPgAA4z4AAOQ+AADnPgAA6z4AAPU+AAD2PgAAAT8AAAM/AAAMPwAAHT8AAB8/AAAgPwAA Kj8AACs/AAAwPwAAND8AADc/AAA4PwAAPD8AAD0/AABEPwAARj8AAFY/AABXPwAAcj8AAHM/AACp PwAArD8AAK8/AACwPwAAxz8AAMg/AADTPwAA1D8AANc/AADYPwAAHkAAAB9AAAAmQAAAJ0AAAChA AAAsQAAALkAAADFAAABlQQAAZkEAABVCAAAYQgAAGkIAAB1CAAAfQgAAIEIAACZCAAAnQgAAKkIA ACtCAAAvQgAAMEIAADJCAAAzQgAAN0IAADhCAAA/QgAAQEIAAEdCAABQQgAAXUIAAGZCAABtQgAA bkIAAHFCAAByQgAAeUIAAHpCAACHQgAAiEIAAI5CAACTQgAAmkIAAJxCAAClQgAArkIAALlCAAC6 QgAAvkIAAL9CAADCQgAAw0IAAMxCAADNQgAA20IAAOJCAADoQgAA6UIAAOtCAAD1QgAA/EIAAP1C AAALQwAADEMAABJDAAATQwAAF0MAABhDAAAaQwAAHEMAACFDAAAiQwAAKEMAACpDAAAsQwAALUMA ADBDAAA4QwAAPkMAAD9DAABBQwAAQkMAAEVDAABGQwAAT0MAAFBDAABXQwAAWkMAAHVDAAB2QwAA eEMAAIBDAACIQwAAiUMAAI5DAACPQwAAk0MAAJRDAACXQwAAmEMAAJ1DAACeQwAAoEMAAKFDAACn QwAAqEMAALNDAAC0QwAAuEMAALlDAAC8QwAAxEMAAMpDAADLQwAAzUMAANBDAADXQwAA2EMAANpD AADbQwAA4UMAAOJDAADkQwAA5UMAAOdDAADoQwAA9EMAAPxDAAANRAAADkQAABJEAAATRAAAF0QA ABhEAAAaRAAAG0QAAB5EAAAfRAAAKEQAAClEAAAwRAAAM0QAADpEAABGRAAAVEQAAFVEAABYRAAA X0QAAGNEAABkRAAAZkQAAGdEAABuRAAAckQAAHhEAAB5RAAAfUQAAH5EAACARAAAiEQAAJFEAACS RAAAlkQAAJdEAACaRAAAm0QAAKREAAClRAAAs0QAALREAAC2RAAAt0QAAL9EAADARAAAw0QAAMtE AADSRAAA00QAANxEAADfRAAA5EQAAOdEAADpRAAA6kQAAPBEAADxRAAA+EQAAPlEAAD7RAAA/EQA AApFAAAMRQAAGEUAABlFAAAdRQAAHkUAACNFAAArRQAAO0UAADxFAABHRQAASEUAAEpFAABLRQAA VEUAAFZFAABeRQAAX0UAAGJFAABjRQAAbEUAAHRFAACCRQAAg0UAAIxFAACNRQAAlEUAAJVFAACY RQAAnUUAAKJFAACrRQAArEUAAK5FAACvRQAAs0UAAL5FAADBRQAAwkUAAMlFAADKRQAAz0UAANVF AADXRQAA2EUAANpFAADbRQAA3UUAAN5FAADhRQAA8EUAAPJFAADzRQAA+UUAAAFGAAAIRgAACUYA ABBGAAARRgAAHUYAACVGAAAsRgAAMkYAADRGAAA1RgAAPUYAAD5GAABBRgAASUYAAFNGAABURgAA WEYAAFlGAABcRgAAXUYAAGpGAABtRgAAdkYAAHhGAAB7RgAAfEYAAIdGAACPRgAAl0YAAJtGAACe RgAAn0YAAKhGAACpRgAAsUYAALJGAAC0RgAAtUYAALhGAAC9RgAAwkYAAMNGAADFRgAAxkYAAMhG AADQRgAA2UYAAORGAADnRgAA7kYAAO9GAAD9RgAA/kYAAAFHAAAMRwAADkcAAA9HAAAXRwAAGEcA ABpHAAAbRwAAH0cAACBHAAAjRwAAJEcAACZHAAAyRwAANUcAADdHAAA/RwAAQkcAAEdHAABMRwAA T0cAAFBHAABaRwAAW0cAAGVHAABqRwAAb0cAAINHAACKRwAAvkcAAMNHAADXRwAA4UcAAPFHAAD4 RwAAB0gAAA5IAAAQSAAAE0gAABRIAAAZSAAAGkgAACdIAAAoSAAAK0gAACxIAAAzSAAANEgAADhI AAA8SAAAQ0gAAEtIAABZSAAAWkgAAFxIAABdSAAAYUgAAGJIAABkSAAAZUgAAG1IAABuSAAAcUgA AHJIAAB6SAAAe0gAAH5IAACGSAAAjEgAAJZIAACgSAAAokgAAKlIAACrSAAArUgAAK5IAAC1SAAA uUgAALxIAAC9SAAAykgAAMxIAADPSAAA10gAAN5IAADkSAAA5kgAAOdIAADrSAAA7EgAAO5IAADv SAAA9UgAAPpIAAACSQAAA0kAAAVJAAAGSQAACUkAAApJAAAUSQAAIUkAAChJAAAzSQAANUkAADZJ AAA6SQAAO0kAAD1JAABFSQAASEkAAElJAABUSQAAVUkAAFdJAABYSQAAXEkAAGVJAABpSQAAakkA AHZJAAB3SQAAe0kAAHxJAACCSQAAg0kAAIVJAACQSQAAkkkAAJNJAACWSQAAl0kAAKBJAACoSQAA r0kAALJJAAC0SQAAtUkAALhJAAC5SQAAu0kAALxJAAC+SQAAv0kAAMpJAADLSQAAz0kAANBJAADT SQAA1EkAANtJAADcSQAA6UkAAOpJAADvSQAA90kAAP1JAAD+SQAABUoAAAZKAAAJSgAACkoAAA1K AAAOSgAAFUoAABZKAAAcSgAAHUoAACJKAAAjSgAAJUoAACdKAAAxSgAAOkoAAEFKAAC6SgAAxEoA AMVKAADJSgAAykoAANFKAADSSgAA1UoAANxKAADeSgAA30oAAOdKAADoSgAA6koAAO1KAADySgAA 80oAAPVKAAD2SgAA/EoAAARLAAAPSwAAEEsAABRLAAAVSwAAGEsAABlLAAAgSwAAIUsAAC5LAAA3 SwAAO0sAADxLAABDSwAAS0sAAFFLAABSSwAAWksAAF9LAABkSwAAaEsAAGtLAABsSwAAc0sAAHpL AAB+SwAAf0sAAIdLAACJSwAAi0sAAIxLAACUSwAAn0sAAKRLAAClSwAAqEsAAKlLAACuSwAAr0sA ALZLAAC3SwAAuksAAMRLAADGSwAAx0sAANBLAADSSwAA1EsAANVLAADXSwAA2EsAANtLAADcSwAA 3ksAAN9LAADqSwAA8ksAAPRLAAD1SwAA/EsAAP1LAAAGTAAAB0wAABJMAAATTAAAGkwAABtMAAAe TAAAIkwAAClMAAArTAAAMEwAADhMAAA9TAAAPkwAAEFMAABCTAAAREwAAEdMAABRTAAAUkwAAFlM AABaTAAAaEwAAGlMAABrTAAAd0wAAIJMAACFTAAAiEwAAIlMAACQTAAAkkwAAJVMAACbTAAAqEwA AKlMAACuTAAAr0wAALFMAAC0TAAAt0wAAL9MAADCTAAAw0wAAMVMAADJTAAA0EwAANFMAADWTAAA 4EwAAOJMAADjTAAA50wAAOhMAADsTAAA8UwAAPZMAAD3TAAA+kwAAAJNAAALTQAADE0AABNNAAAX TQAAGk0AABtNAAAkTQAAJU0AAC9NAAAwTQAAMk0AAD5NAABJTQAAaU0AAG5NAACCTQAAiU0AAL1N AADCTQAA1k0AAOBNAADwTQAA900AAANOAAAPTgAAHU4AACZOAAAqTgAANk4AADdOAAA8TgAAPU4A AD9OAABATgAASU4AAEpOAABNTgAATk4AAFFOAABSTgAAW04AAF9OAABmTgAAb04AAHFOAAByTgAA dE4AAHVOAAB4TgAAeU4AAINOAACETgAAhk4AAIlOAACNTgAAjk4AAJVOAACWTgAAmE4AAJlOAACk TgAApU4AAK9OAAC2TgAAuU4AALpOAADBTgAAwk4AAMtOAADMTgAA0E4AANFOAADdTgAA404AAOVO AADmTgAA704AAPBOAADzTgAA+k4AAANPAAAETwAACE8AAAlPAAALTwAADE8AABVPAAAWTwAAGE8A ABlPAAAiTwAAJE8AACpPAAAsTwAAL08AADBPAAA6TwAAQU8AAEZPAABHTwAAU08AAFRPAABXTwAA WE8AAFtPAABcTwAAXk8AAF9PAABiTwAAbE8AAHVPAAB+TwAAgE8AAIFPAACDTwAAhE8AAI5PAACP TwAAkU8AAJJPAACWTwAAl08AAJ9PAACgTwAAp08AAKhPAACwTwAAsU8AALlPAADATwAAx08AAMhP AADSTwAA008AANhPAADZTwAA4U8AAOJPAADlTwAA5k8AAOtPAADsTwAA+E8AAPlPAAD8TwAAA1AA AAhQAAAVUAAAIFAAACFQAAAmUAAAJ1AAACxQAAAtUAAAM1AAADRQAAA6UAAAO1AAAD1QAAA+UAAA QVAAAExQAABTUAAAVFAAAF1QAABeUAAAYlAAAGNQAABmUAAAZ1AAAHBQAAB5UAAAfVAAAH5QAACE UAAAhVAAAI1QAACOUAAAkVAAAJJQAACYUAAAmVAAAJ1QAACeUAAAolAAAKNQAACqUAAAq1AAALVQ AAC8UAAAxFAAAMVQAADMUAAAzVAAANBQAADRUAAA11AAANhQAADeUAAA31AAAOtQAADwUAAA9FAA APVQAAD9UAAABVEAAAlRAAAKUQAADFEAAA1RAAARUQAAElEAABRRAAAVUQAAHVEAACRRAAAvUQAA MFEAADlRAAA6UQAAPVEAAD5RAABCUQAASVEAAE9RAABQUQAAWlEAAGNRAABwUQAAcVEAAHVRAAB2 UQAAeVEAAHpRAACDUQAAhFEAAIZRAACHUQAAjlEAAJRRAACaUQAAoVEAAK1RAACuUQAAtVEAALZR AAC5UQAAulEAAL5RAAC/UQAAwVEAAMJRAADHUQAAyFEAANRRAADWUQAA31EAAOZRAADqUQAA61EA AO5RAADvUQAA+lEAAPtRAAD9UQAA/lEAAARSAAAHUgAADlIAAA9SAAASUgAAE1IAACFSAAApUgAA M1IAADVSAAA3UgAAOFIAADtSAAA9UgAAQVIAAEJSAABEUgAARVIAAEhSAABJUgAAUFIAAFFSAABV UgAAWlIAAGFSAABiUgAAZVIAAGxSAABzUgAAdFIAAHZSAAB5UgAAfFIAAINSAACMUgAAkFIAAJVS AACWUgAAmlIAAJtSAACeUgAAn1IAAKFSAACiUgAAqFIAADhXAABLVwAAUlcAAFlXAABlVwAAZlcA AGtXAABtVwAAcFcAAH9XAACEVwAAhVcAAItXAACUVwAAl1cAAJtXAAChVwAAolcAAKRXAAClVwAA qFcAAKlXAACvVwAAsFcAALJXAADOVwAA1FcAANVXAADfVwAA41cAAOdXAADsVwAA71cAAPBXAAD7 VwAA/FcAAP5XAAD/VwAAB1gAAAhYAAASWAAAE1gAAB9YAAAkWAAAK1gAAC1YAAAwWAAARFgAAEdY AABIWAAASlgAAEtYAABQWAAAUVgAAFtYAABcWAAAaFgAAGxYAABxWAAAhVgAAIxYAADAWAAAxVgA ANlYAADjWAAA81gAAPpYAAAFWQAACVkAAApZAAARWQAAElkAABtZAAAcWQAAHlkAAB9ZAAAkWQAA JVkAADFZAAAyWQAANlkAADdZAAA8WQAAPVkAAERZAABNWQAAWVkAAFpZAABeWQAAX1kAAGRZAABq WQAAbFkAAG1ZAAB5WQAAelkAAH1ZAAB+WQAAgVkAAItZAACUWQAAlVkAAJ9ZAAClWQAAqVkAAKpZ AACsWQAArVkAALVZAAC2WQAAuVkAALpZAAC+WQAAv1kAAMNZAADEWQAAx1kAAMtZAADQWQAA1lkA ANhZAADZWQAA5VkAAPJZAAD8WQAA/VkAAAFaAAAQWgAAG1oAACNaAAApWgAAMVoAADtaAAA8WgAA P1oAAEBaAABMWgAATVoAAFFaAABSWgAAVFoAAFVaAABaWgAAX1oAAGVaAABmWgAAaFoAAHBaAAB8 WgAAfVoAAIFaAACFWgAAiloAAItaAACNWgAAjloAAJpaAACpWgAAq1oAAM1aAADOWgAAAVsAAAJb AAAvWwAAOVsAAEFbAABCWwAASlsAAExbAABPWwAAXlsAAGRbAABlWwAAaFsAAGlbAAB1WwAAdlsA AHpbAACCWwAAhVsAAIZbAACPWwAAkFsAAJJbAACTWwAAllsAAJdbAACjWwAApFsAAKZbAACnWwAA qVsAAKpbAACvWwAAsFsAALRbAAC1WwAAu1sAALxbAAC+WwAAylsAANBbAADaWwAA5lsAAOxbAAD0 WwAA+VsAAPtbAAD8WwAABVwAABxcAAAhXAAAIlwAACdcAAAoXAAAKlwAAC5cAAA1XAAANlwAADlc AAA6XAAAPFwAAD1cAABFXAAARlwAAE9cAABhXAAAZFwAAGtcAABxXAAAclwAAHpcAACHXAAAkVwA AJJcAACUXAAAlVwAAJxcAACeXAAAolwAAKpcAACsXAAArVwAALFcAACyXAAAuFwAALxcAADBXAAA wlwAAMlcAADLXAAA0VwAANJcAADcXAAA3lwAAOBcAADhXAAA6lwAAPJcAAD2XAAA91wAAABdAAAB XQAAEF0AABVdAAAZXQAAGl0AAB9dAAAhXQAAJF0AACVdAAAwXQAAuF4AAL5eAADyYQAA+mEAAPth AAACYgAABWIAAAxiAAAQYgAAFWIAABZiAAAhYgAAI2IAAC9iAAAwYgAAUXEAAFRxAABhcQAAYnEA AHlxAAB6cQAAgHEAAIJxAAD/cQAAAnIAAFFyAABTcgAAWHIAAFlyAABtcgAAbnIAAJlyAACacgAA CHMAAApzAAAVcwAAFnMAAFVzAABWcwAAhnMAAIhzAADUcwAA1XMAABh0AAAadAAAx3QAAMl0AABu dQAAcXUAAHx1AAB9dQAA83UAAPV1AABQdgAAUXYAAIB2AACDdgAAjnYAAI92AADNdgAAznYAAOZ2 AADndgAACncAAAt3AAAhdwAAI3cAAOp3AADtdwAAQngAAER4AACZeAAAr3gAAOeKAADoigAA7YoA APiKAAD5igAA/IoAAP2KAAAGiwAAB4sAAAyLAAANiwAAEYsAABKLAAAViwAAFosAABiLAAAZiwAA HYsAAB6LAAAjiwAAKYsAACuLAAAsiwAAMIsAADuLAABHiwAATYsAAFSLAABXiwAAW4sAAFyLAABk iwAAZYsAAGmLAABuiwAAdosAAHqLAAB9iwAAfosAAIGLAACCiwAAjYsAAI6LAACSiwAAk4sAAJiL AACdiwAAoosAAKOLAACniwAAqIsAAKyLAACuiwAAsosAALOLAAC4iwAAuYsAALuLAAC8iwAAw4sA AMSLAADHiwAAyIsAANGLAADSiwAA1osAANmLAADgiwAA5osAAO2LAADviwAA84sAAPSLAAD4iwAA +YsAAAGMAAACjAAAB4wAAA2MAAARjAAAEowAABWMAAAWjAAAHIwAAB2MAAAmjAAAJ4wAAC6MAAAv jAAAMowAADqMAAA+jAAAP4wAAEOMAABEjAAARowAAEuMAABTjAAAW4wAAF2MAABejAAAYIwAAGGM AABsjAAAbYwAAHGMAAB2jAAAfYwAAH6MAACFjAAAhowAAIuMAACMjAAAk4wAAJSMAACejAAAn4wA AKaMAADFjAAAzIwAAOGMAADojAAA6YwAAOuMAADsjAAA84wAAPSMAAD3jAAA+IwAAAONAAAEjQAA E40AABmNAAAdjQAAHo0AACCNAAAhjQAAJY0AACqNAAAtjQAALo0AADWNAAA5jQAAPI0AAD2NAABG jQAAR40AAFCNAABRjQAAWI0AAFmNAABijQAAY40AAGaNAABnjQAAco0AAHeNAACCjQAAg40AAI2N AACOjQAAlI0AAJWNAACXjQAAmI0AAJuNAACcjQAAo40AAKiNAAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAAwAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAAwAcAAcAHAAHABwABwAcAAcAAwAHAAMABwADAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAADAAcAAwAHAAMABwADAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAMA BwAcAAcAHAAHABwABwADAAcAAwAHAAMABwADAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAADAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAMAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAAwAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHAAMABwADAAcAAwAHAAMABwADAAcAHAAHABwABwADAAcAAwAHAAMABwADAAcAAwAH AAMABwADAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHAAMABwADAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAD AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwADAAcAAwAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAADAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAAwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAMABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwAAwAHAAMABwADAAcAAwAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAADAAcAAwAcAAcAHAAHABwABwAc AAcAHAAHABwABwADAAcAAwAHAAMABwADAAcAAwAHAAMABwADAAcAAwAHAAMABwADAAcAAwAHAAMA BwADAAcAAwAHAAMABwADAAcAAwAHAAMABwADAAcAAwAHAAMABwADAAcAAwAHAAMABwADAAcAAwAH AAMABwADAAcAAwAHAAMABwAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcA HAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAc AAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwA BwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAHABwABwAcAAcAHAAH ABwABwAcAAcAAAAAANMFAADXBQAALgoAADIKAADbFgAA3xYAAGQYAABoGAAAN0sAADtLAAABVQAA BVUAAHhrAAB+awAA6IoAAKiNAAAHADMABwAzAAcAMwAHADMABwAzAAcAMwAHADMABwAHAAEAhGWd CuIXLOf/D/8P/w//D/8P/w//D/8P/w8QAAcAAAAXAAAAAAAAAAAAAAAAAAAAAAAAACAQAAAPhN8C EYSJ/l6E3wJghIn+QioAQ0oVAE9KAQBQSgQAUUoBAF5KAwBvKABwaAAAAP8BALfwAQAAABeAAAAA AAAAAAAAAAAAAAAAAAAAGRAAAA+EoAURhJj+XoSgBWCEmP5PSgMAUUoDAF5KAwBvKACHaAAAAACI SAAAAQBvAAEAAAAXgAAAAAAAAAAAAAAAAAAAAAAAABUQAAAPhHAIEYSY/l6EcAhghJj+T0oHAFFK BwBvKACHaAAAAACISAAAAQCn8AEAAAAXgAAAAAAAAAAAAAAAAAAAAAAAABUQAAAPhEALEYSY/l6E QAtghJj+T0oBAFFKAQBvKACHaAAAAACISAAAAQC38AEAAAAXgAAAAAAAAAAAAAAAAAAAAAAAABkQ AAAPhBAOEYSY/l6EEA5ghJj+T0oDAFFKAwBeSgMAbygAh2gAAAAAiEgAAAEAbwABAAAAF4AAAAAA AAAAAAAAAAAAAAAAAAAVEAAAD4TgEBGEmP5ehOAQYISY/k9KBwBRSgcAbygAh2gAAAAAiEgAAAEA p/ABAAAAF4AAAAAAAAAAAAAAAAAAAAAAAAAVEAAAD4SwExGEmP5ehLATYISY/k9KAQBRSgEAbygA h2gAAAAAiEgAAAEAt/ABAAAAF4AAAAAAAAAAAAAAAAAAAAAAAAAZEAAAD4SAFhGEmP5ehIAWYISY /k9KAwBRSgMAXkoDAG8oAIdoAAAAAIhIAAABAG8AAQAAABeAAAAAAAAAAAAAAAAAAAAAAAAAFRAA AA+EUBkRhJj+XoRQGWCEmP5PSgcAUUoHAG8oAIdoAAAAAIhIAAABAKfwAQAAAIRlnQoAAAAAAAAA AAAAAAD///////8BAAAAAAD//wEAAAASAFKYDEADAAcEBQAHBAEABwQDAAcEBQAHBAEABwQDAAcE BQAHBAEA9TuJHQAAAAAAAAAAAAECAAIAGAAAAAQAAAAIAAAA5QAAAAAAAAASAAAAND8GAPB/BwDU PRQADRssAEo/OwAvAUEAODFJAB1YTADEX2UAcXZyAAtgdQBcDngA8Fh6ADZRmgDRZ6EAvSOmAHJY rAAjTq8A52i4AMU1yAAQM+gA+XnrAC8r8QCyFvoAAAAAAOiKAADqigAAAAAAAAEAAAD/QAIQAAAA AAAAAKeNAABAAAAQAEAAAP//AgAAAAcAVQBuAGsAbgBvAHcAbgAFAGgAbwBlAG4AZQD//wIACAAA AAAAAAAAAAAAAAAAAAAAAAABAP//AgAAAAAAAAD//wAAAgD//wAAAAD//wAAAgD//wAAAAAJAAAA Rx6QAQAAAgIGAwUEBQIDBP8qAOBBeADACQAAAAAAAAD/AQAAAAAAAFQAaQBtAGUAcwAgAE4AZQB3 ACAAUgBvAG0AYQBuAAAANR6QAQIABQUBAgEHBgIFBwAAAAAAAAAQAAAAAAAAAAAAAACAAAAAAFMA eQBtAGIAbwBsAAAAMy6QAQAAAgsGBAICAgICBP8qAOBDeADACQAAAAAAAAD/AQAAAAAAAEEAcgBp AGEAbAAAAD89kAEAAAIHAwkCAgUCBAT/KgDgQ3gAwAkAAAAAAAAA/wEAAAAAAABDAG8AdQByAGkA ZQByACAATgBlAHcAAAA3LpABAAACDwUCAgIEAwIE/wIA4f+sAEAJAAAAAAAAAJ8BAAAAAAAAQwBh AGwAaQBiAHIAaQAAADk9kAEAAAILBgkCAgQDAgT/AgDh//wAQAkAAAAAAAAAnwEAAAAAAABDAG8A bgBzAG8AbABhAHMAAAA1LpABAAACCwYEAwUEBAIE/y4A4VtgAMApAAAAAAAAAP8BAQAAAAAAVABh AGgAbwBtAGEAAAA7DpABAgAFAAAAAAAAAAAAAAAAAAAAABAAAAAAAAAAAAAAAIAAAAAAVwBpAG4A ZwBkAGkAbgBnAHMAAABBHpABAAACBAUDBQQGAwIE7wIAoOsgAEIAAAAAAAAAAJ8BAAAAAAAAQwBh AG0AYgByAGkAYQAgAE0AYQB0AGgAAAAiAAQAMYmIGADwxAIAAKkBAAAAAJFt6mYSmOpGAAAAAAoA AAAAAAcTAADhdwAAFgBGAAAABAADkP8AAAAHEwAA4XcAABYARgAAAP8AAAAAAAAAIQMA8BAAAAAB AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAANgWJBbQAtACBgTIwAAAAAAAAAAAAAAAAAACiigAAoooA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAIAAAAAAAAAAAAAMoNRAPAQAAgA/P0BAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEhQAAAA AAnw/w8BACRQAAAAAAAA/AMAAP///3////9/AAAAAP///3////9/////f3JYrAAABAAAMgAAAAAA AAAAAAAAAAAAAAAAAAAAACEEAAAAAAAAAAAAAAAAAAAAAAAAEBwAAAgAAAAAAAAAAAB4AAAAeAAA AAAAAAAAAAAAoAUAAP//EgAAAAAAAAAAAAAAAAAAAAUAaABvAGUAbgBlAAUAaABvAGUAbgBlAAAA AAAAAAAAAAAAAAAAAAAAAAAAEAAAAAYAAAABAAAAAAAMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD+/wAABgECAAAAAAAAAAAAAAAAAAAAAAABAAAA 4IWf8vlPaBCrkQgAKyez2TAAAABAAQAADwAAAAEAAACAAAAAAgAAAIgAAAADAAAAlAAAAAQAAACg AAAABQAAALAAAAAHAAAAvAAAAAgAAADMAAAACQAAANwAAAASAAAA6AAAAAwAAAAIAQAADQAAABQB AAAOAAAAIAEAAA8AAAAoAQAAEAAAADABAAATAAAAOAEAAAIAAADkBAAAHgAAAAQAAAAAAAAAHgAA AAQAAAAAAAAAHgAAAAgAAABob2VuZQAAAB4AAAAEAAAAAAAAAB4AAAAIAAAATm9ybWFsAAAeAAAA CAAAAGhvZW5lAAAAHgAAAAQAAAAxMAAAHgAAABgAAABNaWNyb3NvZnQgT2ZmaWNlIFdvcmQAAABA AAAAAIZ4lxNrywFAAAAAAFzVUhJvywEDAAAAFgAAAAMAAAAHEwAAAwAAAOF3AAADAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAA/v8AAAYBAgAAAAAAAAAAAAAAAAAAAAAAAQAAAALVzdWcLhsQ k5cIACss+a4wAAAA/AAAAAwAAAABAAAAaAAAAA8AAABwAAAABQAAAJAAAAAGAAAAmAAAABEAAACg AAAAFwAAAKgAAAALAAAAsAAAABAAAAC4AAAAEwAAAMAAAAAWAAAAyAAAAA0AAADQAAAADAAAAN0A AAACAAAA5AQAAB4AAAAYAAAAVW5pdmVyc2l05HQgVPxiaW5nZW4AAAAAAwAAAP8AAAADAAAARgAA AAMAAACiigAAAwAAAAAADAALAAAAAAAAAAsAAAAAAAAACwAAAAAAAAALAAAAAAAAAB4QAAABAAAA AQAAAAAMEAAAAgAAAB4AAAAGAAAAVGl0ZWwAAwAAAAEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAEAAAACAAAAAwAAAAQAAAAFAAAABgAAAAcAAAAIAAAACQAAAAoAAAALAAAA DAAAAA0AAAAOAAAADwAAABAAAAARAAAAEgAAABMAAAAUAAAAFQAAABYAAAAXAAAAGAAAABkAAAAa AAAAGwAAABwAAAAdAAAAHgAAAB8AAAAgAAAAIQAAACIAAAAjAAAAJAAAACUAAAAmAAAAJwAAACgA AAApAAAAKgAAACsAAAAsAAAALQAAAC4AAAAvAAAAMAAAADEAAAAyAAAAMwAAADQAAAA1AAAANgAA ADcAAAA4AAAAOQAAADoAAAA7AAAAPAAAAD0AAAA+AAAAPwAAAEAAAABBAAAAQgAAAEMAAABEAAAA RQAAAEYAAABHAAAASAAAAEkAAABKAAAASwAAAEwAAABNAAAATgAAAE8AAABQAAAAUQAAAFIAAABT AAAAVAAAAFUAAABWAAAAVwAAAFgAAABZAAAAWgAAAFsAAABcAAAAXQAAAF4AAABfAAAAYAAAAGEA AABiAAAAYwAAAGQAAABlAAAAZgAAAGcAAABoAAAAaQAAAGoAAABrAAAAbAAAAG0AAABuAAAAbwAA AHAAAABxAAAAcgAAAHMAAAB0AAAAdQAAAHYAAAB3AAAAeAAAAHkAAAB6AAAAewAAAHwAAAB9AAAA fgAAAH8AAACAAAAA/v///4IAAACDAAAAhAAAAIUAAACGAAAAhwAAAIgAAACJAAAAigAAAIsAAACM AAAAjQAAAI4AAACPAAAAkAAAAJEAAACSAAAAkwAAAJQAAACVAAAAlgAAAJcAAACYAAAAmQAAAJoA AACbAAAAnAAAAJ0AAACeAAAAnwAAAKAAAAChAAAAogAAAKMAAACkAAAApQAAAKYAAACnAAAAqAAA AKkAAACqAAAAqwAAAKwAAACtAAAArgAAAK8AAACwAAAAsQAAALIAAACzAAAAtAAAALUAAAC2AAAA twAAALgAAAC5AAAAugAAALsAAAC8AAAAvQAAAL4AAAC/AAAAwAAAAMEAAADCAAAAwwAAAMQAAADF AAAA/v///8cAAADIAAAAyQAAAMoAAADLAAAAzAAAAM0AAAD+////zwAAANAAAADRAAAA0gAAANMA AADUAAAA1QAAAP7////9/////f///9kAAAD+/////v////7///////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////9SAG8AbwB0ACAARQBuAHQAcgB5AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAFgAFAf//////////AwAAAAYJAgAAAAAAwAAAAAAAAEYAAAAAAAAA AAAAAADAN4VcEm/LAdsAAACAAAAAAAAAADEAVABhAGIAbABlAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAOAAIB/////wUAAAD/////AAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgQAAAASIAAAAAAAAVwBvAHIAZABEAG8AYwB1AG0A ZQBuAHQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABoAAgEBAAAA//// //////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAANAABAAAAAAAFAFMA dQBtAG0AYQByAHkASQBuAGYAbwByAG0AYQB0AGkAbwBuAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAKAACAQIAAAAEAAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAMYA AAAAEAAAAAAAAAUARABvAGMAdQBtAGUAbgB0AFMAdQBtAG0AYQByAHkASQBuAGYAbwByAG0AYQB0 AGkAbwBuAAAAAAAAAAAAAAA4AAIB////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAzgAAAAAQAAAAAAAAAQBDAG8AbQBwAE8AYgBqAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgD///////////////8AAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAeQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP////////// /////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAA////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAQAAAP7///////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////8BAP7/AwoAAP////8GCQIAAAAAAMAAAAAAAABGJwAAAE1pY3Jvc29mdCBPZmZpY2Ug V29yZCA5Ny0yMDAzLURva3VtZW50AAoAAABNU1dvcmREb2MAEAAAAFdvcmQuRG9jdW1lbnQuOAD0 ObJxAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAA== ------=_NextPart_000_0013_01CB6F23.EDADCD00-- From Pochol@WebfootGames.com Mon Oct 18 19:00:29 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 1AC9E3A6AE9 for ; Mon, 18 Oct 2010 19:00:29 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: 0.001 X-Spam-Level: X-Spam-Status: No, score=0.001 tagged_above=-999 required=5 tests=[BAYES_50=0.001] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id uCHRoF+c6-n5 for ; Mon, 18 Oct 2010 19:00:25 -0700 (PDT) Received: from mail.webfootgames.com (mail.webfootgames.com [68.248.244.61]) by core3.amsl.com (Postfix) with ESMTP id 11B473A6A84 for ; Mon, 18 Oct 2010 19:00:22 -0700 (PDT) Received: from mail.WebfootGames.com (softdnserr [::ffff:127.0.0.1]) (AUTH: LOGIN pochol) by mail.webfootgames.com with esmtp; Mon, 18 Oct 2010 21:19:27 -0500 id 00000DF4.4CBD002F.00007EFD Message-ID: Date: Mon, 18 Oct 2010 21:19:27 -0500 From: "Pascal Pochol" To: codec@ietf.org X-Mailer: Webfoot's WebMail 1.6-CVS x-priority: 3 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mime-Autoconverted: from 8bit to 7bit by courier 0.64 Subject: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list Reply-To: Pochol@WebfootGames.com List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 02:18:35 -0000 Hello, we've been using speex and celt for voice and music but today I heard about Opus which sounds like a fantastic replacement for both these codecs. We're wondering if Opus will be released under the same type of license as speex and celt? SILK's license forbid its use in commercial software and we're worried that it might carry over to Opus. If not, as soon as Opus handles lower bitrates, stereo and fixed point decoding we'll be using it. -Pascal From jean-marc.valin@octasic.com Mon Oct 18 19:26:36 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 1716C3A6A84 for ; Mon, 18 Oct 2010 19:26:36 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.338 X-Spam-Level: X-Spam-Status: No, score=-2.338 tagged_above=-999 required=5 tests=[AWL=0.260, BAYES_00=-2.599, HTML_MESSAGE=0.001] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qhHN5gr4p3pD for ; Mon, 18 Oct 2010 19:26:34 -0700 (PDT) Received: from toroondcbmts07-srv.bellnexxia.net (toroondcbmts07-srv.bellnexxia.net [207.236.237.41]) by core3.amsl.com (Postfix) with ESMTP id 45C283A6A98 for ; Mon, 18 Oct 2010 19:26:34 -0700 (PDT) Received: from toip57-bus.srvr.bell.ca ([67.69.240.184]) by toroondcbmts07-srv.bellnexxia.net (InterMail vM.8.00.01.00 201-2244-105-20090324) with ESMTP id <20101019022803.WEXV26461.toroondcbmts07-srv.bellnexxia.net@toip57-bus.srvr.bell.ca> for ; Mon, 18 Oct 2010 22:28:03 -0400 Received: from toip36-bus.srvr.bell.ca ([67.69.240.37]) by toip57-bus.srvr.bell.ca with ESMTP; 18 Oct 2010 22:28:00 -0400 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: ApsEAMuXvExGNv5q/2dsb2JhbACiV8QBhUkE Received: from mail.octasic.com (HELO MAILEXCH.octasic.com) ([70.54.254.106]) by toip36-bus.srvr.bell.ca with ESMTP; 18 Oct 2010 22:27:59 -0400 X-MimeOLE: Produced By Microsoft Exchange V6.5 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01CB6F35.3E6BF313" Date: Mon, 18 Oct 2010 22:26:24 -0400 Message-ID: <390831ED3DF58E41A3D2FB82591E2C3605A34C4D@MAILEXCH.octasic.com> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: [codec] Opus codec licensing Thread-Index: ActvNCcg/6UdLTfYSPe62Egnu0SU9wAAN+7t References: From: "Jean-Marc Valin" To: , Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 02:26:36 -0000 This is a multi-part message in MIME format. ------_=_NextPart_001_01CB6F35.3E6BF313 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Opus is available under the BSD license. As for patents, Skype has = pledged to make them available royalty-free once the codec is accepted = as an IETF standard. Jean-Marc -----Original Message----- From: codec-bounces@ietf.org on behalf of Pascal Pochol Sent: Mon 10/18/2010 10:19 PM To: codec@ietf.org Subject: [codec] Opus codec licensing =20 Hello, we've been using speex and celt for voice and music but today I heard = about Opus which sounds like a fantastic replacement for both these codecs. We're wondering if Opus will be released under the same type of license = as speex and celt? SILK's license forbid its use in commercial software and we're worried that it might carry over to Opus. If not, as soon as Opus handles lower bitrates, stereo and fixed point decoding we'll be using = it. -Pascal _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec ------_=_NextPart_001_01CB6F35.3E6BF313 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable RE: [codec] Opus codec licensing

Opus is available under the BSD license. As for = patents, Skype has pledged to make them available royalty-free once the = codec is accepted as an IETF standard.

    Jean-Marc

-----Original Message-----
From: codec-bounces@ietf.org on behalf of Pascal Pochol
Sent: Mon 10/18/2010 10:19 PM
To: codec@ietf.org
Subject: [codec] Opus codec licensing

Hello,

we've been using speex and celt for voice and music but today I heard = about
Opus which sounds like a fantastic replacement for both these = codecs.

We're wondering if Opus will be released under the same type of license = as
speex and celt? SILK's license forbid its use in commercial software = and
we're worried that it might carry over to Opus. If not, as soon as = Opus
handles lower bitrates, stereo and fixed point decoding we'll be using = it.

-Pascal

_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org= /mailman/listinfo/codec

------_=_NextPart_001_01CB6F35.3E6BF313-- From Pochol@WebfootGames.com Tue Oct 19 06:32:58 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 57AB53A6826 for ; Tue, 19 Oct 2010 06:32:58 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.873 X-Spam-Level: X-Spam-Status: No, score=-1.873 tagged_above=-999 required=5 tests=[AWL=0.726, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id D45M815ZgDhQ for ; Tue, 19 Oct 2010 06:32:57 -0700 (PDT) Received: from mail.webfootgames.com (mail.webfootgames.com [68.248.244.61]) by core3.amsl.com (Postfix) with ESMTP id 4E7983A6814 for ; Tue, 19 Oct 2010 06:32:56 -0700 (PDT) Received: from mail.WebfootGames.com (softdnserr [::ffff:127.0.0.1]) (AUTH: LOGIN pochol) by mail.webfootgames.com with esmtp; Tue, 19 Oct 2010 08:52:11 -0500 id 00000DF4.4CBDA28B.0000217D Message-ID: <77af49d0332789f2b097aaaa2e453f5e@WebfootGames.com> Date: Tue, 19 Oct 2010 08:52:11 -0500 From: "Pascal Pochol" To: " Jean-Marc Valin " In-Reply-To: <390831ED3DF58E41A3D2FB82591E2C3605A34C4D@MAILEXCH.octasic.com> References: <390831ED3DF58E41A3D2FB82591E2C3605A34C4D@MAILEXCH.octasic.com> X-Mailer: Webfoot's WebMail 1.6-CVS x-priority: 3 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mime-Autoconverted: from 8bit to 7bit by courier 0.64 Cc: codec@ietf.org Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list Reply-To: Pochol@WebfootGames.com List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 13:32:58 -0000 On 9:26:24 pm 10/18/10 "Jean-Marc Valin" wrote: > Opus is available under the BSD license. As for patents, Skype has > pledged to make them available royalty-free once the codec is accepted > as an IETF standard. Great. I know it's still a draft now but how long do you roughly expect before it is accepted as a standard? Are we talking weeks? months? Or years? -Pascal From stpeter@stpeter.im Tue Oct 19 07:14:31 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id C85273A6809 for ; Tue, 19 Oct 2010 07:14:30 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.565 X-Spam-Level: X-Spam-Status: No, score=-102.565 tagged_above=-999 required=5 tests=[AWL=0.034, BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id p+I2K2dgI4Bx for ; Tue, 19 Oct 2010 07:14:29 -0700 (PDT) Received: from stpeter.im (stpeter.im [207.210.219.233]) by core3.amsl.com (Postfix) with ESMTP id 516EE3A6845 for ; Tue, 19 Oct 2010 07:14:28 -0700 (PDT) Received: from squire.local (dsl-179-124.dynamic-dsl.frii.net [216.17.179.124]) (Authenticated sender: stpeter) by stpeter.im (Postfix) with ESMTPSA id BC82040074; Tue, 19 Oct 2010 08:23:18 -0600 (MDT) Message-ID: <4CBDA81D.4090209@stpeter.im> Date: Tue, 19 Oct 2010 08:15:57 -0600 From: Peter Saint-Andre User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.2.9) Gecko/20100915 Thunderbird/3.1.4 MIME-Version: 1.0 To: Pochol@WebfootGames.com References: <390831ED3DF58E41A3D2FB82591E2C3605A34C4D@MAILEXCH.octasic.com> <77af49d0332789f2b097aaaa2e453f5e@WebfootGames.com> In-Reply-To: <77af49d0332789f2b097aaaa2e453f5e@WebfootGames.com> X-Enigmail-Version: 1.1.1 OpenPGP: url=http://www.saint-andre.com/me/stpeter.asc Content-Type: multipart/signed; protocol="application/pkcs7-signature"; micalg=sha1; boundary="------------ms020407060308040709080001" Cc: codec@ietf.org Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 14:14:31 -0000 This is a cryptographically signed message in MIME format. --------------ms020407060308040709080001 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable On 10/19/10 7:52 AM, Pascal Pochol wrote: > On 9:26:24 pm 10/18/10 "Jean-Marc Valin" = wrote: >> Opus is available under the BSD license. As for patents, Skype has >> pledged to make them available royalty-free once the codec is accepted= >> as an IETF standard. >=20 > Great. I know it's still a draft now but how long do you roughly expect= > before it is accepted as a standard? Are we talking weeks? months? Or y= ears? Definitely not weeks. Probably months. Hopefully not years. :) Peter --=20 Peter Saint-Andre https://stpeter.im/ --------------ms020407060308040709080001 Content-Type: application/pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIITzjCC BjQwggQcoAMCAQICASMwDQYJKoZIhvcNAQELBQAwfTELMAkGA1UEBhMCSUwxFjAUBgNVBAoT DVN0YXJ0Q29tIEx0ZC4xKzApBgNVBAsTIlNlY3VyZSBEaWdpdGFsIENlcnRpZmljYXRlIFNp Z25pbmcxKTAnBgNVBAMTIFN0YXJ0Q29tIENlcnRpZmljYXRpb24gQXV0aG9yaXR5MB4XDTA3 MTAyNDIxMDMzM1oXDTE3MTAyNDIxMDMzM1owgYwxCzAJBgNVBAYTAklMMRYwFAYDVQQKEw1T dGFydENvbSBMdGQuMSswKQYDVQQLEyJTZWN1cmUgRGlnaXRhbCBDZXJ0aWZpY2F0ZSBTaWdu aW5nMTgwNgYDVQQDEy9TdGFydENvbSBDbGFzcyAzIFByaW1hcnkgSW50ZXJtZWRpYXRlIENs aWVudCBDQTCCASIwDQYJKoZIhvcNAQEBBQADggEPADCCAQoCggEBALmjSW4SPiDKlAinvVeL ZOVfItiuP1aRHL530E7QUc9icCwL33+PH+Js1HAh8CgWFl34sOxx1FJyS/C4VLPRsqDfP72j tzCVUAL0DAxZ7wgzQvFz7x61jGxfhYhqYb1+PPOLkYBbkRIrPMg3dLEdKmXIYJYXDH+mB/V/ jLo73/Kb7h/rNoNg/oHHSv5Jolyvp5IY2btfcTBfW/telEFj5rDTX2juTvZ3Qhf3XQX5ca3Q 7A10zrUV/cWJOJ7F5RltbEIaboZmX5JBUb3FhUiAdBotehAX6DbDOuYoJtVxmGof6GuVGcPo 98K4TJf8FHo+UA9EOVDp/W7fCqKT4sXk/XkCAwEAAaOCAa0wggGpMA8GA1UdEwEB/wQFMAMB Af8wDgYDVR0PAQH/BAQDAgEGMB0GA1UdDgQWBBR7iZySlyShhEcCy3T8LvSs3DLl8zAfBgNV HSMEGDAWgBROC+8apEBbpRdphzDKNGhD0EGu8jBmBggrBgEFBQcBAQRaMFgwJwYIKwYBBQUH MAGGG2h0dHA6Ly9vY3NwLnN0YXJ0c3NsLmNvbS9jYTAtBggrBgEFBQcwAoYhaHR0cDovL3d3 dy5zdGFydHNzbC5jb20vc2ZzY2EuY3J0MFsGA1UdHwRUMFIwJ6AloCOGIWh0dHA6Ly93d3cu c3RhcnRzc2wuY29tL3Nmc2NhLmNybDAnoCWgI4YhaHR0cDovL2NybC5zdGFydHNzbC5jb20v c2ZzY2EuY3JsMIGABgNVHSAEeTB3MHUGCysGAQQBgbU3AQIBMGYwLgYIKwYBBQUHAgEWImh0 dHA6Ly93d3cuc3RhcnRzc2wuY29tL3BvbGljeS5wZGYwNAYIKwYBBQUHAgEWKGh0dHA6Ly93 d3cuc3RhcnRzc2wuY29tL2ludGVybWVkaWF0ZS5wZGYwDQYJKoZIhvcNAQELBQADggIBAGpd SbdLFMhirxK37V4gE00+uW74UdAXtDgQI3AsRZWtaRtKHgAxFBSteqz4kDkeAjH/1b+K8tQR 6cxSI2nho7qOaPW/UpzOfSS/MeKK/9vfM2lfs+uItXH7LWtvS9wD1erfH1a+BXHCrCp4LA1l fADDhRIiGTSS3i0Zu5xV3INNRHrCCCl6patltQ8RZTqzDMri7ombgIxjN51Zo7xV77EZcThV 0GA8iIN+7T53uHhUJpjfLIztHs/69OclRvHux9hCflfOm7GY5Sc4nqjfES+5XPArGGWiQSEk ez37QfXqsxO3oCHK4b3DFZysG4uyOuC/WL80ab3muQ3tgwjBhq0D3JZN5kvu5gSuNZPa1WrV hEgXkd6C7s5stqB6/htVpshG08jRz9DEutGM9oKQ1ncTivbfPNx7pILoHWvvT7N5i/puVoNu bPUmLXh/2wA6wzAzuuoONiIL14Xpw6jLSnqpaLWElo2yTIFZ/CU/nCvvpW1Dj1457P3Ci9bD 0RPkWSR+CuucpgxrEmaw4UOLxflzuYYaq1RJwygOO5K0s2bAWOcXpgteyUOnQ3d/EjJAWRri 2v0ubiq+4H3KUOMlbznlPAY/1T8YyyJPM88+Ueahe/AW1zoUwZayNcTnuM7cq6yBV8Wr3GOI LFXhtT0UVuJLChPMJKVKVsa7qNorlLkMMIIGxzCCBa+gAwIBAgICAIswDQYJKoZIhvcNAQEF BQAwgYwxCzAJBgNVBAYTAklMMRYwFAYDVQQKEw1TdGFydENvbSBMdGQuMSswKQYDVQQLEyJT ZWN1cmUgRGlnaXRhbCBDZXJ0aWZpY2F0ZSBTaWduaW5nMTgwNgYDVQQDEy9TdGFydENvbSBD bGFzcyAzIFByaW1hcnkgSW50ZXJtZWRpYXRlIENsaWVudCBDQTAeFw0xMDEwMTQwMTM2MzRa Fw0xMjEwMTQxMjAxMDdaMIHAMSAwHgYDVQQNExcyNzQ1ODEtOU5YMDRxeExEYjBvNDY5VDEL MAkGA1UEBhMCVVMxETAPBgNVBAgTCENvbG9yYWRvMQ8wDQYDVQQHEwZEZW52ZXIxLDAqBgNV BAsTI1N0YXJ0Q29tIFRydXN0ZWQgQ2VydGlmaWNhdGUgTWVtYmVyMRowGAYDVQQDExFQZXRl ciBTYWludC1BbmRyZTEhMB8GCSqGSIb3DQEJARYSc3RwZXRlckBzdHBldGVyLmltMIIBIjAN BgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEAuERvnrkpQTx9wbJfgxbNKEYvt0IilecZRUM6 wrbCzIUPCocuYhaAJcQoqIyHaKybPQ7f+DIGIAolAa3dHnNdlsXP2smTft/ZNpj10PIG5bil NAqLUYwmLJaEaqY7BMW8423U3blW43/luLJk/Pq4OsWcw7AK3LeVh1U/HOgqhin26N3h72X1 nbLEpZFrgcp8egmWtXLCbLBDMqUK3j6wjLldni79muzYEVqU0A5GqSeb8Wc4kIx8VI5yL24J KzinG2iVRP5ZDEbOZETzBXJabUsV56XSxqPG9DK6ke+ybCiL/wKV1HFqdtFB1y25lfvHgOP2 gyEApBKEDNjgLmKyyQIDAQABo4IC+zCCAvcwCQYDVR0TBAIwADALBgNVHQ8EBAMCBLAwHQYD VR0lBBYwFAYIKwYBBQUHAwIGCCsGAQUFBwMEMB0GA1UdDgQWBBS2EW2iNB+g0EibKJLBdv8I eLovVDAfBgNVHSMEGDAWgBR7iZySlyShhEcCy3T8LvSs3DLl8zAdBgNVHREEFjAUgRJzdHBl dGVyQHN0cGV0ZXIuaW0wggFCBgNVHSAEggE5MIIBNTCCATEGCysGAQQBgbU3AQICMIIBIDAu BggrBgEFBQcCARYiaHR0cDovL3d3dy5zdGFydHNzbC5jb20vcG9saWN5LnBkZjA0BggrBgEF BQcCARYoaHR0cDovL3d3dy5zdGFydHNzbC5jb20vaW50ZXJtZWRpYXRlLnBkZjCBtwYIKwYB BQUHAgIwgaowFBYNU3RhcnRDb20gTHRkLjADAgEBGoGRTGltaXRlZCBMaWFiaWxpdHksIHNl ZSBzZWN0aW9uICpMZWdhbCBMaW1pdGF0aW9ucyogb2YgdGhlIFN0YXJ0Q29tIENlcnRpZmlj YXRpb24gQXV0aG9yaXR5IFBvbGljeSBhdmFpbGFibGUgYXQgaHR0cDovL3d3dy5zdGFydHNz bC5jb20vcG9saWN5LnBkZjBjBgNVHR8EXDBaMCugKaAnhiVodHRwOi8vd3d3LnN0YXJ0c3Ns LmNvbS9jcnR1My1jcmwuY3JsMCugKaAnhiVodHRwOi8vY3JsLnN0YXJ0c3NsLmNvbS9jcnR1 My1jcmwuY3JsMIGOBggrBgEFBQcBAQSBgTB/MDkGCCsGAQUFBzABhi1odHRwOi8vb2NzcC5z dGFydHNzbC5jb20vc3ViL2NsYXNzMy9jbGllbnQvY2EwQgYIKwYBBQUHMAKGNmh0dHA6Ly93 d3cuc3RhcnRzc2wuY29tL2NlcnRzL3N1Yi5jbGFzczMuY2xpZW50LmNhLmNydDAjBgNVHRIE HDAahhhodHRwOi8vd3d3LnN0YXJ0c3NsLmNvbS8wDQYJKoZIhvcNAQEFBQADggEBADVtbXJG tKAr55xc/OUM546gXUybI72Bank0w739Mv+9BBNtq9rMEvCnLmSKhBi76c1mdXh6zXs8RQDo 6nR/aPabE3llF2T4z80smi9jfnl3y9dpu9TcgDoqDLZ7a2lBlW656XAAQzHjvLp2MC7/mxlg PYH2axa+q40mAYM20GbNsAEGbWQT1IqIh0BcLLsgbaMJHbyG/57zd9JLyMX3Vry1L1fJRQr3 GeLxMV5RtxN+mBgxrwFz/cOc09COiFExlsHgekpB5O43gqsAU16MXypyoSt4MrSfKTMHIGx6 2RF/M6vqUlvhi28gk2ZUvQ/+OX5+gjcZyooEzAAn4RuOKNswggbHMIIFr6ADAgECAgIAizAN BgkqhkiG9w0BAQUFADCBjDELMAkGA1UEBhMCSUwxFjAUBgNVBAoTDVN0YXJ0Q29tIEx0ZC4x KzApBgNVBAsTIlNlY3VyZSBEaWdpdGFsIENlcnRpZmljYXRlIFNpZ25pbmcxODA2BgNVBAMT L1N0YXJ0Q29tIENsYXNzIDMgUHJpbWFyeSBJbnRlcm1lZGlhdGUgQ2xpZW50IENBMB4XDTEw MTAxNDAxMzYzNFoXDTEyMTAxNDEyMDEwN1owgcAxIDAeBgNVBA0TFzI3NDU4MS05TlgwNHF4 TERiMG80NjlUMQswCQYDVQQGEwJVUzERMA8GA1UECBMIQ29sb3JhZG8xDzANBgNVBAcTBkRl bnZlcjEsMCoGA1UECxMjU3RhcnRDb20gVHJ1c3RlZCBDZXJ0aWZpY2F0ZSBNZW1iZXIxGjAY BgNVBAMTEVBldGVyIFNhaW50LUFuZHJlMSEwHwYJKoZIhvcNAQkBFhJzdHBldGVyQHN0cGV0 ZXIuaW0wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQC4RG+euSlBPH3Bsl+DFs0o Ri+3QiKV5xlFQzrCtsLMhQ8Khy5iFoAlxCiojIdorJs9Dt/4MgYgCiUBrd0ec12Wxc/ayZN+ 39k2mPXQ8gbluKU0CotRjCYsloRqpjsExbzjbdTduVbjf+W4smT8+rg6xZzDsArct5WHVT8c 6CqGKfbo3eHvZfWdssSlkWuBynx6CZa1csJssEMypQrePrCMuV2eLv2a7NgRWpTQDkapJ5vx ZziQjHxUjnIvbgkrOKcbaJVE/lkMRs5kRPMFclptSxXnpdLGo8b0MrqR77JsKIv/ApXUcWp2 0UHXLbmV+8eA4/aDIQCkEoQM2OAuYrLJAgMBAAGjggL7MIIC9zAJBgNVHRMEAjAAMAsGA1Ud DwQEAwIEsDAdBgNVHSUEFjAUBggrBgEFBQcDAgYIKwYBBQUHAwQwHQYDVR0OBBYEFLYRbaI0 H6DQSJsoksF2/wh4ui9UMB8GA1UdIwQYMBaAFHuJnJKXJKGERwLLdPwu9KzcMuXzMB0GA1Ud EQQWMBSBEnN0cGV0ZXJAc3RwZXRlci5pbTCCAUIGA1UdIASCATkwggE1MIIBMQYLKwYBBAGB tTcBAgIwggEgMC4GCCsGAQUFBwIBFiJodHRwOi8vd3d3LnN0YXJ0c3NsLmNvbS9wb2xpY3ku cGRmMDQGCCsGAQUFBwIBFihodHRwOi8vd3d3LnN0YXJ0c3NsLmNvbS9pbnRlcm1lZGlhdGUu cGRmMIG3BggrBgEFBQcCAjCBqjAUFg1TdGFydENvbSBMdGQuMAMCAQEagZFMaW1pdGVkIExp YWJpbGl0eSwgc2VlIHNlY3Rpb24gKkxlZ2FsIExpbWl0YXRpb25zKiBvZiB0aGUgU3RhcnRD b20gQ2VydGlmaWNhdGlvbiBBdXRob3JpdHkgUG9saWN5IGF2YWlsYWJsZSBhdCBodHRwOi8v d3d3LnN0YXJ0c3NsLmNvbS9wb2xpY3kucGRmMGMGA1UdHwRcMFowK6ApoCeGJWh0dHA6Ly93 d3cuc3RhcnRzc2wuY29tL2NydHUzLWNybC5jcmwwK6ApoCeGJWh0dHA6Ly9jcmwuc3RhcnRz c2wuY29tL2NydHUzLWNybC5jcmwwgY4GCCsGAQUFBwEBBIGBMH8wOQYIKwYBBQUHMAGGLWh0 dHA6Ly9vY3NwLnN0YXJ0c3NsLmNvbS9zdWIvY2xhc3MzL2NsaWVudC9jYTBCBggrBgEFBQcw AoY2aHR0cDovL3d3dy5zdGFydHNzbC5jb20vY2VydHMvc3ViLmNsYXNzMy5jbGllbnQuY2Eu Y3J0MCMGA1UdEgQcMBqGGGh0dHA6Ly93d3cuc3RhcnRzc2wuY29tLzANBgkqhkiG9w0BAQUF AAOCAQEANW1tcka0oCvnnFz85QznjqBdTJsjvYFqeTTDvf0y/70EE22r2swS8KcuZIqEGLvp zWZ1eHrNezxFAOjqdH9o9psTeWUXZPjPzSyaL2N+eXfL12m71NyAOioMtntraUGVbrnpcABD MeO8unYwLv+bGWA9gfZrFr6rjSYBgzbQZs2wAQZtZBPUioiHQFwsuyBtowkdvIb/nvN30kvI xfdWvLUvV8lFCvcZ4vExXlG3E36YGDGvAXP9w5zT0I6IUTGWweB6SkHk7jeCqwBTXoxfKnKh K3gytJ8pMwcgbHrZEX8zq+pSW+GLbyCTZlS9D/45fn6CNxnKigTMACfhG44o2zGCA80wggPJ AgEBMIGTMIGMMQswCQYDVQQGEwJJTDEWMBQGA1UEChMNU3RhcnRDb20gTHRkLjErMCkGA1UE CxMiU2VjdXJlIERpZ2l0YWwgQ2VydGlmaWNhdGUgU2lnbmluZzE4MDYGA1UEAxMvU3RhcnRD b20gQ2xhc3MgMyBQcmltYXJ5IEludGVybWVkaWF0ZSBDbGllbnQgQ0ECAgCLMAkGBSsOAwIa BQCgggIOMBgGCSqGSIb3DQEJAzELBgkqhkiG9w0BBwEwHAYJKoZIhvcNAQkFMQ8XDTEwMTAx OTE0MTU1N1owIwYJKoZIhvcNAQkEMRYEFAwDmEvgdWgDdk6lIzMOqjNxLRNXMF8GCSqGSIb3 DQEJDzFSMFAwCwYJYIZIAWUDBAECMAoGCCqGSIb3DQMHMA4GCCqGSIb3DQMCAgIAgDANBggq hkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBpAYJKwYBBAGCNxAEMYGWMIGT MIGMMQswCQYDVQQGEwJJTDEWMBQGA1UEChMNU3RhcnRDb20gTHRkLjErMCkGA1UECxMiU2Vj dXJlIERpZ2l0YWwgQ2VydGlmaWNhdGUgU2lnbmluZzE4MDYGA1UEAxMvU3RhcnRDb20gQ2xh c3MgMyBQcmltYXJ5IEludGVybWVkaWF0ZSBDbGllbnQgQ0ECAgCLMIGmBgsqhkiG9w0BCRAC CzGBlqCBkzCBjDELMAkGA1UEBhMCSUwxFjAUBgNVBAoTDVN0YXJ0Q29tIEx0ZC4xKzApBgNV BAsTIlNlY3VyZSBEaWdpdGFsIENlcnRpZmljYXRlIFNpZ25pbmcxODA2BgNVBAMTL1N0YXJ0 Q29tIENsYXNzIDMgUHJpbWFyeSBJbnRlcm1lZGlhdGUgQ2xpZW50IENBAgIAizANBgkqhkiG 9w0BAQEFAASCAQBiP2VJys4TVkXck4un3QEmDpdcm7b80KRmhDgYjsYPHrAgGdgokcMRwsmf 3RqlNA2payhdbseb1pENhumyFlPXGwrdD+KmjRrJldAeDq08zi8aV3T56krEojf8gtr8NUmR 3dv7ML6tsoXbk27kAxAsjDA69wuWSUL7AkWONU+JRP65eHIFDo6GYMsqEk65AgYaGqGthPyn WG4FvYM5govtar4DVGUWDJ910f9SA/oULX+bTefSbz7Hi11HcY/zIAt6f7fgSXfSBgm8iZ5o pRVsvPTJteUDLFf8Dh1Hd4KDhaJkmyifx6g3z0OHEx5PN2aT5A2Z0PY66Zg03CFgzRdtAAAA AAAA --------------ms020407060308040709080001-- From stewe@stewe.org Tue Oct 19 08:19:38 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id A9FDF3A68B1 for ; Tue, 19 Oct 2010 08:19:38 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.016 X-Spam-Level: X-Spam-Status: No, score=-1.016 tagged_above=-999 required=5 tests=[AWL=0.186, BAYES_00=-2.599, HTML_MESSAGE=0.001, MIME_QP_LONG_LINE=1.396] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id mrs8QaUDWEmV for ; Tue, 19 Oct 2010 08:19:34 -0700 (PDT) Received: from stewe.org (stewe.org [85.214.122.234]) by core3.amsl.com (Postfix) with ESMTP id D65273A6870 for ; Tue, 19 Oct 2010 08:19:31 -0700 (PDT) Received: from [192.168.1.108] (unverified [24.5.132.232]) by stewe.org (SurgeMail 3.9e) with ESMTP id 819265-1743317 for multiple; Tue, 19 Oct 2010 17:20:58 +0200 User-Agent: Microsoft-Entourage/12.26.0.100708 Date: Tue, 19 Oct 2010 08:20:47 -0700 From: Stephan Wenger To: Jean-Marc Valin , , Message-ID: Thread-Topic: [codec] Opus codec licensing Thread-Index: ActvNCcg/6UdLTfYSPe62Egnu0SU9wAAN+7tABsLZFo= In-Reply-To: <390831ED3DF58E41A3D2FB82591E2C3605A34C4D@MAILEXCH.octasic.com> Mime-version: 1.0 Content-type: multipart/alternative; boundary="B_3370321257_9559710" X-Originating-IP: 24.5.132.232 X-Authenticated-User: stewe@stewe.org X-ORBS-Stamp: Your IP (24.5.132.232) was found in the spamhaus database. http://www.spamhaus.net Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 15:19:38 -0000 > This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. --B_3370321257_9559710 Content-type: text/plain; charset="ISO-8859-1" Content-transfer-encoding: quoted-printable Hi, With =B3hat=B2: It is, in almost all cases, unwise for a third party to circumscribe=8Bor worse: abbreviate=8Blicensing terms provided by a rightholder. Skype=B9s licensing declaration can be found at https://datatracker.ietf.org/ipr/1297= . I note that the statement contains conditions beyond the acceptance of thei= r technology into an IETF standard. Stephan On 10.18.2010 19:26 , "Jean-Marc Valin" wrote= : > Opus is available under the BSD license. As for patents, Skype has pledge= d to > make them available royalty-free once the codec is accepted as an IETF > standard. >=20 > Jean-Marc >=20 > -----Original Message----- > From: codec-bounces@ietf.org on behalf of Pascal Pochol > Sent: Mon 10/18/2010 10:19 PM > To: codec@ietf.org > Subject: [codec] Opus codec licensing >=20 > Hello, >=20 > we've been using speex and celt for voice and music but today I heard abo= ut > Opus which sounds like a fantastic replacement for both these codecs. >=20 > We're wondering if Opus will be released under the same type of license a= s > speex and celt? SILK's license forbid its use in commercial software and > we're worried that it might carry over to Opus. If not, as soon as Opus > handles lower bitrates, stereo and fixed point decoding we'll be using it= . >=20 > -Pascal >=20 > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec >=20 >=20 >=20 > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec --B_3370321257_9559710 Content-type: text/html; charset="ISO-8859-1" Content-transfer-encoding: quoted-printable Re: [codec] Opus codec licensing Hi,
With “hat”:
It is, in almost all cases, unwise for a third party to circumscribe—= or worse: abbreviate—licensing terms provided by a rightholder.  = Skype’s licensing declaration can be found at https://datatracker.ietf.org/ipr/1297.  I n= ote that the statement contains conditions beyond the acceptance of their te= chnology into an IETF standard.
Stephan



On 10.18.2010 19:26 , "Jean-Marc Valin" <jean-marc.valin@octasic.com> wrote:

<= SPAN STYLE=3D'font-size:11pt'>Opus is available under the BSD license. As for = patents, Skype has pledged to make them available royalty-free once the code= c is accepted as an IETF standard.

    Jean-Marc

-----Original Message-----
From: codec-bounces@ietf.org on behalf= of Pascal Pochol
Sent: Mon 10/18/2010 10:19 PM
To: codec@ietf.org
Subject: [codec] Opus codec licensing

Hello,

we've been using speex and celt for voice and music but today I heard about=
Opus which sounds like a fantastic replacement for both these codecs.

We're wondering if Opus will be released under the same type of license as<= BR> speex and celt? SILK's license forbid its use in commercial software and we're worried that it might carry over to Opus. If not, as soon as Opus
handles lower bitrates, stereo and fixed point decoding we'll be using it.<= BR>
-Pascal

_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/= mailman/listinfo/codec


___________= ____________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/= mailman/listinfo/codec
--B_3370321257_9559710-- From stewe@stewe.org Tue Oct 19 08:24:49 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 4217C3A6875 for ; Tue, 19 Oct 2010 08:24:49 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.725 X-Spam-Level: X-Spam-Status: No, score=-1.725 tagged_above=-999 required=5 tests=[AWL=0.874, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Vv+cpH8O5lQK for ; Tue, 19 Oct 2010 08:24:48 -0700 (PDT) Received: from stewe.org (stewe.org [85.214.122.234]) by core3.amsl.com (Postfix) with ESMTP id 100933A683C for ; Tue, 19 Oct 2010 08:24:47 -0700 (PDT) Received: from [192.168.1.108] (unverified [24.5.132.232]) by stewe.org (SurgeMail 3.9e) with ESMTP id 819272-1743317 for ; Tue, 19 Oct 2010 17:26:18 +0200 User-Agent: Microsoft-Entourage/12.26.0.100708 Date: Tue, 19 Oct 2010 08:26:07 -0700 From: Stephan Wenger To: Message-ID: Thread-Topic: [codec] I-D Action:draft-ietf-codec-opus-00.txt Thread-Index: ActvofMm7xO3Qttmn0mBfZEEb4T8yQ== In-Reply-To: <20101015203002.C18F03A6B0D@core3.amsl.com> Mime-version: 1.0 Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit X-Originating-IP: 24.5.132.232 X-Authenticated-User: stewe@stewe.org X-ORBS-Stamp: Your IP (24.5.132.232) was found in the spamhaus database. http://www.spamhaus.net Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 15:24:49 -0000 Hi, I want to inquire the status of this draft. In many working groups, the filename "draft-ietf--xxx" indicates that the draft in question is a WG item of WG . Following this logic, it would appear that the "opus" draft is now a WG item of the codec WG. I don't recall a decision to than extent. If the draft were indeed accepted as a WG item, I would like to encourage those who made IPR statements related to it, to resubmit those statements with the new filename. This would help those of us who are searching through the IETF IPR tracker by WG name (which is a very common thing to do, at least for me). Thanks, Stephan On 10.15.2010 13:30 , "Internet-Drafts@ietf.org" wrote: > A New Internet-Draft is available from the on-line Internet-Drafts > directories. > This draft is a work item of the Internet Wideband Audio Codec Working Group > of the IETF. > > > Title : Definition of the Opus Audio Codec > Author(s) : J. Valin, K. Vos > Filename : draft-ietf-codec-opus-00.txt > Pages : 12 > Date : 2010-10-15 > > This document describes the Opus codec, designed for interactive > speech and audio transmission over the Internet. > > A URL for this Internet-Draft is: > http://www.ietf.org/internet-drafts/draft-ietf-codec-opus-00.txt > > Internet-Drafts are also available by anonymous FTP at: > ftp://ftp.ietf.org/internet-drafts/ > > Below is the data which will enable a MIME compliant mail reader > implementation to automatically retrieve the ASCII version of the > Internet-Draft. > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From jean-marc.valin@octasic.com Tue Oct 19 08:30:08 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 70C8E3A688A for ; Tue, 19 Oct 2010 08:30:08 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.382 X-Spam-Level: X-Spam-Status: No, score=-2.382 tagged_above=-999 required=5 tests=[AWL=0.217, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id A2AYG2j-LUQ1 for ; Tue, 19 Oct 2010 08:30:00 -0700 (PDT) Received: from toroondcbmts06-srv.bellnexxia.net (toroondcbmts06.bellnexxia.net [207.236.237.40]) by core3.amsl.com (Postfix) with ESMTP id 1B3443A6831 for ; Tue, 19 Oct 2010 08:29:59 -0700 (PDT) Received: from toip57-bus.srvr.bell.ca ([67.69.240.184]) by toroondcbmts06-srv.bellnexxia.net (InterMail vM.8.00.01.00 201-2244-105-20090324) with ESMTP id <20101019153129.PJQB18038.toroondcbmts06-srv.bellnexxia.net@toip57-bus.srvr.bell.ca>; Tue, 19 Oct 2010 11:31:29 -0400 Received: from toip36-bus.srvr.bell.ca ([67.69.240.37]) by toip57-bus.srvr.bell.ca with ESMTP; 19 Oct 2010 11:31:19 -0400 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: ApsEAJtOvUxGNv5q/2dsb2JhbACiWL9GhUoEjU8G Received: from mail.octasic.com (HELO MAILEXCH.octasic.com) ([70.54.254.106]) by toip36-bus.srvr.bell.ca with ESMTP; 19 Oct 2010 11:31:19 -0400 Received: from [10.100.60.27] ([10.100.60.27]) by MAILEXCH.octasic.com with Microsoft SMTPSVC(6.0.3790.4675); Tue, 19 Oct 2010 11:31:19 -0400 Message-ID: <4CBDB9C7.1020206@octasic.com> Date: Tue, 19 Oct 2010 11:31:19 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.12) Gecko/20100915 Lightning/1.0b1 Thunderbird/3.0.8 MIME-Version: 1.0 To: Stephan Wenger References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 19 Oct 2010 15:31:19.0211 (UTC) FILETIME=[AD3E2FB0:01CB6FA2] Cc: codec@ietf.org Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 15:30:08 -0000 Hi, draft-ietf-codec-opus-00.txt is indeed a WG item. It has had several names in the past, including draft-valin-codec-prototype and draft-valin-codec-definition, which may explain the confusion. This is the draft for which there was a hum during the last meeting. Jean-Marc On 10-10-19 11:26 AM, Stephan Wenger wrote: > Hi, > > I want to inquire the status of this draft. In many working groups, the > filename "draft-ietf--xxx" indicates that the draft in question is > a WG item of WG. Following this logic, it would appear that the > "opus" draft is now a WG item of the codec WG. I don't recall a decision to > than extent. > > If the draft were indeed accepted as a WG item, I would like to encourage > those who made IPR statements related to it, to resubmit those statements > with the new filename. This would help those of us who are searching > through the IETF IPR tracker by WG name (which is a very common thing to do, > at least for me). > > Thanks, > Stephan > > > On 10.15.2010 13:30 , "Internet-Drafts@ietf.org" > wrote: > >> A New Internet-Draft is available from the on-line Internet-Drafts >> directories. >> This draft is a work item of the Internet Wideband Audio Codec Working Group >> of the IETF. >> >> >> Title : Definition of the Opus Audio Codec >> Author(s) : J. Valin, K. Vos >> Filename : draft-ietf-codec-opus-00.txt >> Pages : 12 >> Date : 2010-10-15 >> >> This document describes the Opus codec, designed for interactive >> speech and audio transmission over the Internet. >> >> A URL for this Internet-Draft is: >> http://www.ietf.org/internet-drafts/draft-ietf-codec-opus-00.txt >> >> Internet-Drafts are also available by anonymous FTP at: >> ftp://ftp.ietf.org/internet-drafts/ >> >> Below is the data which will enable a MIME compliant mail reader >> implementation to automatically retrieve the ASCII version of the >> Internet-Draft. >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From stewe@stewe.org Tue Oct 19 09:03:34 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 2F47F3A68A0 for ; Tue, 19 Oct 2010 09:03:34 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.774 X-Spam-Level: X-Spam-Status: No, score=-1.774 tagged_above=-999 required=5 tests=[AWL=0.825, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id swQNhnN-KoGz for ; Tue, 19 Oct 2010 09:03:31 -0700 (PDT) Received: from stewe.org (stewe.org [85.214.122.234]) by core3.amsl.com (Postfix) with ESMTP id DBCB93A689C for ; Tue, 19 Oct 2010 09:03:30 -0700 (PDT) Received: from [192.168.1.108] (unverified [24.5.132.232]) by stewe.org (SurgeMail 3.9e) with ESMTP id 819299-1743317 for multiple; Tue, 19 Oct 2010 18:04:54 +0200 User-Agent: Microsoft-Entourage/12.27.0.100910 Date: Tue, 19 Oct 2010 09:04:45 -0700 From: Stephan Wenger To: Message-ID: Thread-Topic: [codec] I-D Action:draft-ietf-codec-opus-00.txt Thread-Index: Actvp1jJwYwK9yN760KBEit05he0lQ== In-Reply-To: <4CBDB9C7.1020206@octasic.com> Mime-version: 1.0 Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit X-Originating-IP: 24.5.132.232 X-Authenticated-User: stewe@stewe.org X-ORBS-Stamp: Your IP (24.5.132.232) was found in the spamhaus database. http://www.spamhaus.net Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 16:03:34 -0000 Hi Jean-Marc, Four things: First, indeed, there was a hum at the last meeting. Weak memory on my side. Apologies. Second, hums need to be reconfirmed on the mailing list, and that has not happened according to my read of the email archive. Third, I also missed the submission of draft-ietf-codec-description-00, which was really the time I should have complained. So I'm willing to assume (as apparently have the chairs, see point #2) that there has been an implied consensus of the WG to accept the codec-description draft. Which brings me to point #4: As this is now a WG item, any major change requires WG consensus. Selecting a marketing name, IMO, is such a major change. "Opus" is such a flashy name that certain participants and/or companies conceivably may not like it. For example, if I were working for a company that has in its portfolio an audio product named "Opus", I would object to the name change. So the thing you should have done, IMO, is to send an email to the list saying "The editors consider changing the name of our codec to Opus. Is that acceptable to the WG?". It appears to me that twice you guys (chairs included) have taken shortcuts with the IETF's procedures, as I understand them. That, IMO, fills up your quota for the next couple of years. Please be more conservative from now on. Regards, Stephan On 10.19.2010 08:31 , "Jean-Marc Valin" wrote: > Hi, > > draft-ietf-codec-opus-00.txt is indeed a WG item. It has had several names > in the past, including draft-valin-codec-prototype and > draft-valin-codec-definition, which may explain the confusion. This is the > draft for which there was a hum during the last meeting. > > Jean-Marc > > On 10-10-19 11:26 AM, Stephan Wenger wrote: >> Hi, >> >> I want to inquire the status of this draft. In many working groups, the >> filename "draft-ietf--xxx" indicates that the draft in question is >> a WG item of WG. Following this logic, it would appear that the >> "opus" draft is now a WG item of the codec WG. I don't recall a decision to >> than extent. >> >> If the draft were indeed accepted as a WG item, I would like to encourage >> those who made IPR statements related to it, to resubmit those statements >> with the new filename. This would help those of us who are searching >> through the IETF IPR tracker by WG name (which is a very common thing to do, >> at least for me). >> >> Thanks, >> Stephan >> >> >> On 10.15.2010 13:30 , "Internet-Drafts@ietf.org" >> wrote: >> >>> A New Internet-Draft is available from the on-line Internet-Drafts >>> directories. >>> This draft is a work item of the Internet Wideband Audio Codec Working Group >>> of the IETF. >>> >>> >>> Title : Definition of the Opus Audio Codec >>> Author(s) : J. Valin, K. Vos >>> Filename : draft-ietf-codec-opus-00.txt >>> Pages : 12 >>> Date : 2010-10-15 >>> >>> This document describes the Opus codec, designed for interactive >>> speech and audio transmission over the Internet. >>> >>> A URL for this Internet-Draft is: >>> http://www.ietf.org/internet-drafts/draft-ietf-codec-opus-00.txt >>> >>> Internet-Drafts are also available by anonymous FTP at: >>> ftp://ftp.ietf.org/internet-drafts/ >>> >>> Below is the data which will enable a MIME compliant mail reader >>> implementation to automatically retrieve the ASCII version of the >>> Internet-Draft. >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >> >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec > From hoene@uni-tuebingen.de Tue Oct 19 10:01:25 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 011DD3A68EF for ; Tue, 19 Oct 2010 10:01:25 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -5.343 X-Spam-Level: X-Spam-Status: No, score=-5.343 tagged_above=-999 required=5 tests=[AWL=0.906, BAYES_00=-2.599, HELO_EQ_DE=0.35, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fGt8VzhwfK43 for ; Tue, 19 Oct 2010 10:01:22 -0700 (PDT) Received: from mx05.uni-tuebingen.de (mx05.uni-tuebingen.de [134.2.3.4]) by core3.amsl.com (Postfix) with ESMTP id 641003A6908 for ; Tue, 19 Oct 2010 10:01:22 -0700 (PDT) Received: from hoeneT60 ([178.2.222.223]) (authenticated bits=0) by mx05.uni-tuebingen.de (8.13.6/8.13.6) with ESMTP id o9JH2ib1015622 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Tue, 19 Oct 2010 19:02:50 +0200 From: "Christian Hoene" To: "'Stephan Wenger'" , References: <4CBDB9C7.1020206@octasic.com> In-Reply-To: Date: Tue, 19 Oct 2010 19:02:47 +0200 Organization: =?UTF-8?Q?Universit=C3=A4t_T=C3=BCbingen?= Message-ID: <005b01cb6faf$7867a750$6936f5f0$@de> MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Mailer: Microsoft Office Outlook 12.0 Thread-Index: Actvp1jJwYwK9yN760KBEit05he0lQABo20A Content-language: de X-AntiVirus: NOT checked by Avira MailGate (version: 3.0.0-4; host: mx05) Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 17:01:25 -0000 Hi, the ITU had some problems with trademarks that were given to = standardized algorithms.=20 As far as I remember, PESQ or PEAQ are trademarked by some companies. = Since then, the ITU-T name their standards only by numbers. I do not have any problems with nicknames. However, I think the = trademark issue shall be addressed and it is important. I would prefer = that such as trademark is owned by the IETF if possible. Uspto.gov does not list an opus codec. But I do not know how fast they = (or the readers of this mailing list) are... With best regards, Christian --------------------------------------------------------------- Dr.-Ing. Christian Hoene Interactive Communication Systems (ICS), University of T=C3=BCbingen=20 Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532=20 http://www.net.uni-tuebingen.de/ -----Original Message----- From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf = Of Stephan Wenger Sent: Tuesday, October 19, 2010 6:05 PM To: codec@ietf.org Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt Hi Jean-Marc, Four things: First, indeed, there was a hum at the last meeting. Weak memory on my = side. Apologies. Second, hums need to be reconfirmed on the mailing list, and that has = not happened according to my read of the email archive. Third, I also missed the submission of draft-ietf-codec-description-00, which was really the time I should have complained. So I'm willing to assume (as apparently have the chairs, see point #2) = that there has been an implied consensus of the WG to accept the codec-description draft. Which brings me to point #4: As this is now a WG item, any major change requires WG consensus. = Selecting a marketing name, IMO, is such a major change. "Opus" is such a flashy = name that certain participants and/or companies conceivably may not like it. = For example, if I were working for a company that has in its portfolio an = audio product named "Opus", I would object to the name change. So the thing = you should have done, IMO, is to send an email to the list saying "The = editors consider changing the name of our codec to Opus. Is that acceptable to = the WG?". It appears to me that twice you guys (chairs included) have taken = shortcuts with the IETF's procedures, as I understand them. That, IMO, fills up = your quota for the next couple of years. Please be more conservative from = now on. Regards, Stephan On 10.19.2010 08:31 , "Jean-Marc Valin" = wrote: > Hi, >=20 > draft-ietf-codec-opus-00.txt is indeed a WG item. It has had several = names > in the past, including draft-valin-codec-prototype and > draft-valin-codec-definition, which may explain the confusion. This is = the > draft for which there was a hum during the last meeting. >=20 > Jean-Marc >=20 > On 10-10-19 11:26 AM, Stephan Wenger wrote: >> Hi, >>=20 >> I want to inquire the status of this draft. In many working groups, = the >> filename "draft-ietf--xxx" indicates that the draft in = question is >> a WG item of WG. Following this logic, it would appear that = the >> "opus" draft is now a WG item of the codec WG. I don't recall a = decision to >> than extent. >>=20 >> If the draft were indeed accepted as a WG item, I would like to = encourage >> those who made IPR statements related to it, to resubmit those = statements >> with the new filename. This would help those of us who are searching >> through the IETF IPR tracker by WG name (which is a very common thing = to do, >> at least for me). >>=20 >> Thanks, >> Stephan >>=20 >>=20 >> On 10.15.2010 13:30 , = "Internet-Drafts@ietf.org" >> wrote: >>=20 >>> A New Internet-Draft is available from the on-line Internet-Drafts >>> directories. >>> This draft is a work item of the Internet Wideband Audio Codec = Working Group >>> of the IETF. >>>=20 >>>=20 >>> Title : Definition of the Opus Audio Codec >>> Author(s) : J. Valin, K. Vos >>> Filename : draft-ietf-codec-opus-00.txt >>> Pages : 12 >>> Date : 2010-10-15 >>>=20 >>> This document describes the Opus codec, designed for interactive >>> speech and audio transmission over the Internet. >>>=20 >>> A URL for this Internet-Draft is: >>> http://www.ietf.org/internet-drafts/draft-ietf-codec-opus-00.txt >>>=20 >>> Internet-Drafts are also available by anonymous FTP at: >>> ftp://ftp.ietf.org/internet-drafts/ >>>=20 >>> Below is the data which will enable a MIME compliant mail reader >>> implementation to automatically retrieve the ASCII version of the >>> Internet-Draft. >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >>=20 >>=20 >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >=20 _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec From jean-marc.valin@octasic.com Tue Oct 19 10:32:24 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 789123A690C for ; Tue, 19 Oct 2010 10:32:24 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.413 X-Spam-Level: X-Spam-Status: No, score=-2.413 tagged_above=-999 required=5 tests=[AWL=0.186, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Q-hUODiSx-ds for ; Tue, 19 Oct 2010 10:32:22 -0700 (PDT) Received: from toroondcbmts05-srv.bellnexxia.net (toroondcbmts05-srv.bellnexxia.net [207.236.237.39]) by core3.amsl.com (Postfix) with ESMTP id 3F3543A690A for ; Tue, 19 Oct 2010 10:32:21 -0700 (PDT) Received: from toip56-bus.srvr.bell.ca ([67.69.240.142]) by toroondcbmts05-srv.bellnexxia.net (InterMail vM.8.00.01.00 201-2244-105-20090324) with ESMTP id <20101019173352.OZPQ21274.toroondcbmts05-srv.bellnexxia.net@toip56-bus.srvr.bell.ca>; Tue, 19 Oct 2010 13:33:52 -0400 Received: from toip38-bus.srvr.bell.ca ([67.69.240.39]) by toip56-bus.srvr.bell.ca with ESMTP; 19 Oct 2010 13:33:39 -0400 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: ApsEACVsvUxGNv5q/2dsb2JhbACiWcJoAoJ3glEEjU8G Received: from mail.octasic.com (HELO MAILEXCH.octasic.com) ([70.54.254.106]) by toip38-bus.srvr.bell.ca with ESMTP; 19 Oct 2010 13:33:39 -0400 Received: from [10.100.60.27] ([10.100.60.27]) by MAILEXCH.octasic.com with Microsoft SMTPSVC(6.0.3790.4675); Tue, 19 Oct 2010 13:33:38 -0400 Message-ID: <4CBDD672.2090202@octasic.com> Date: Tue, 19 Oct 2010 13:33:38 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.12) Gecko/20100915 Lightning/1.0b1 Thunderbird/3.0.8 MIME-Version: 1.0 To: Christian Hoene References: <4CBDB9C7.1020206@octasic.com> <005b01cb6faf$7867a750$6936f5f0$@de> In-Reply-To: <005b01cb6faf$7867a750$6936f5f0$@de> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit X-OriginalArrivalTime: 19 Oct 2010 17:33:38.0838 (UTC) FILETIME=[C4006360:01CB6FB3] Cc: codec@ietf.org Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 17:32:24 -0000 Hi Christian, I don't know the details of the ITU-T trademark issues, but in this case, the Opus name will not be controlled by any company (I'm not aware of anyone registering a trademark on the "Opus" name in this context). I'm not against the IETF owning the trademark, but I'm not sure any ownership is necessary to begin with. Cheers, Jean-Marc As a note, my company does use the term "Opus" in a different domain of application (digital signal processor architecture) and sees no issue with the IETF using it for an audio codec. Oh, and no, I'm not the one who originally suggested that name. On 10-10-19 01:02 PM, Christian Hoene wrote: > Hi, > > the ITU had some problems with trademarks that were given to standardized algorithms. > As far as I remember, PESQ or PEAQ are trademarked by some companies. Since then, the ITU-T name their standards only by numbers. > > I do not have any problems with nicknames. However, I think the trademark issue shall be addressed and it is important. I would prefer that such as trademark is owned by the IETF if possible. > > Uspto.gov does not list an opus codec. But I do not know how fast they (or the readers of this mailing list) are... > > With best regards, > > Christian > > --------------------------------------------------------------- > Dr.-Ing. Christian Hoene > Interactive Communication Systems (ICS), University of Tübingen > Sand 13, 72076 Tübingen, Germany, Phone +49 7071 2970532 > http://www.net.uni-tuebingen.de/ > > > -----Original Message----- > From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Stephan Wenger > Sent: Tuesday, October 19, 2010 6:05 PM > To: codec@ietf.org > Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt > > Hi Jean-Marc, > > Four things: > > First, indeed, there was a hum at the last meeting. Weak memory on my side. > Apologies. > > Second, hums need to be reconfirmed on the mailing list, and that has not > happened according to my read of the email archive. > > Third, I also missed the submission of draft-ietf-codec-description-00, > which was really the time I should have complained. > > So I'm willing to assume (as apparently have the chairs, see point #2) that > there has been an implied consensus of the WG to accept the > codec-description draft. Which brings me to point #4: > > As this is now a WG item, any major change requires WG consensus. Selecting > a marketing name, IMO, is such a major change. "Opus" is such a flashy name > that certain participants and/or companies conceivably may not like it. For > example, if I were working for a company that has in its portfolio an audio > product named "Opus", I would object to the name change. So the thing you > should have done, IMO, is to send an email to the list saying "The editors > consider changing the name of our codec to Opus. Is that acceptable to the > WG?". > > It appears to me that twice you guys (chairs included) have taken shortcuts > with the IETF's procedures, as I understand them. That, IMO, fills up your > quota for the next couple of years. Please be more conservative from now > on. > > Regards, > Stephan > > > > > On 10.19.2010 08:31 , "Jean-Marc Valin" wrote: > >> Hi, >> >> draft-ietf-codec-opus-00.txt is indeed a WG item. It has had several names >> in the past, including draft-valin-codec-prototype and >> draft-valin-codec-definition, which may explain the confusion. This is the >> draft for which there was a hum during the last meeting. >> >> Jean-Marc >> >> On 10-10-19 11:26 AM, Stephan Wenger wrote: >>> Hi, >>> >>> I want to inquire the status of this draft. In many working groups, the >>> filename "draft-ietf--xxx" indicates that the draft in question is >>> a WG item of WG. Following this logic, it would appear that the >>> "opus" draft is now a WG item of the codec WG. I don't recall a decision to >>> than extent. >>> >>> If the draft were indeed accepted as a WG item, I would like to encourage >>> those who made IPR statements related to it, to resubmit those statements >>> with the new filename. This would help those of us who are searching >>> through the IETF IPR tracker by WG name (which is a very common thing to do, >>> at least for me). >>> >>> Thanks, >>> Stephan >>> >>> >>> On 10.15.2010 13:30 , "Internet-Drafts@ietf.org" >>> wrote: >>> >>>> A New Internet-Draft is available from the on-line Internet-Drafts >>>> directories. >>>> This draft is a work item of the Internet Wideband Audio Codec Working Group >>>> of the IETF. >>>> >>>> >>>> Title : Definition of the Opus Audio Codec >>>> Author(s) : J. Valin, K. Vos >>>> Filename : draft-ietf-codec-opus-00.txt >>>> Pages : 12 >>>> Date : 2010-10-15 >>>> >>>> This document describes the Opus codec, designed for interactive >>>> speech and audio transmission over the Internet. >>>> >>>> A URL for this Internet-Draft is: >>>> http://www.ietf.org/internet-drafts/draft-ietf-codec-opus-00.txt >>>> >>>> Internet-Drafts are also available by anonymous FTP at: >>>> ftp://ftp.ietf.org/internet-drafts/ >>>> >>>> Below is the data which will enable a MIME compliant mail reader >>>> implementation to automatically retrieve the ASCII version of the >>>> Internet-Draft. >>>> _______________________________________________ >>>> codec mailing list >>>> codec@ietf.org >>>> https://www.ietf.org/mailman/listinfo/codec >>> >>> >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >> > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From stephen.botzko@gmail.com Tue Oct 19 10:43:10 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 828843A67C2 for ; Tue, 19 Oct 2010 10:43:10 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.711 X-Spam-Level: X-Spam-Status: No, score=-2.711 tagged_above=-999 required=5 tests=[AWL=-0.113, BAYES_00=-2.599, HTML_MESSAGE=0.001] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 344Ij8nvy9Ag for ; Tue, 19 Oct 2010 10:43:08 -0700 (PDT) Received: from mail-bw0-f44.google.com (mail-bw0-f44.google.com [209.85.214.44]) by core3.amsl.com (Postfix) with ESMTP id C2B0F3A67AB for ; Tue, 19 Oct 2010 10:43:07 -0700 (PDT) Received: by bwz14 with SMTP id 14so2076329bwz.31 for ; Tue, 19 Oct 2010 10:44:38 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type; bh=VEcfyqvl7Kxs1b2uuEP6blvht7U27c15UxGJoF48F/Q=; b=oNL128gbjqdhCuAd2GnES1aPgLnf5E/Jtanzrx6Bd6anm1MzfljGHrPLrI8aFHjmjw h0TysRqvW0IWZIRqnK62Dl0JPEH3oe6vFMcURCvTKXeR3k9nRjcbfz0u5SpAfaq0+6Ac W6p3WWuO+C5wMXQiwET2Hxn58aoZdkQjf+2g4= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=CZpswPLHbfVZRrYh90PjxqYZTzZxjCr7g315/K3SvezVALNlk7ccgGiArt2QC9zU5o hBVW4/q2WcDMIdYUaVAMZoTbmkoMv3OxN6LN3YNOZ2X9ocxuy7GzBWomnePa1+EHNmZ3 IlBoh0u1G0xgGFB39y8MwviDguu+dNC+PfOsg= MIME-Version: 1.0 Received: by 10.204.61.133 with SMTP id t5mr5937450bkh.4.1287510278311; Tue, 19 Oct 2010 10:44:38 -0700 (PDT) Received: by 10.204.80.213 with HTTP; Tue, 19 Oct 2010 10:44:38 -0700 (PDT) In-Reply-To: <4CBDD672.2090202@octasic.com> References: <4CBDB9C7.1020206@octasic.com> <005b01cb6faf$7867a750$6936f5f0$@de> <4CBDD672.2090202@octasic.com> Date: Tue, 19 Oct 2010 13:44:38 -0400 Message-ID: From: Stephen Botzko To: Jean-Marc Valin Content-Type: multipart/alternative; boundary=0016e6dee986c890640492fbd89c Cc: codec@ietf.org Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 17:43:10 -0000 --0016e6dee986c890640492fbd89c Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Registering trademarks can be expensive, esp. if you are doing it in multiple countries. I don't see any necessity for this - even if we had th= e money, I'd rather see it go for additional characterization/testing. Regards Stephen Botzko On Tue, Oct 19, 2010 at 1:33 PM, Jean-Marc Valin < jean-marc.valin@octasic.com> wrote: > Hi Christian, > > I don't know the details of the ITU-T trademark issues, but in this case, > the Opus name will not be controlled by any company (I'm not aware of any= one > registering a trademark on the "Opus" name in this context). I'm not agai= nst > the IETF owning the trademark, but I'm not sure any ownership is necessar= y > to begin with. > > Cheers, > > Jean-Marc > > As a note, my company does use the term "Opus" in a different domain of > application (digital signal processor architecture) and sees no issue wit= h > the IETF using it for an audio codec. Oh, and no, I'm not the one who > originally suggested that name. > > > On 10-10-19 01:02 PM, Christian Hoene wrote: > >> Hi, >> >> the ITU had some problems with trademarks that were given to standardize= d >> algorithms. >> As far as I remember, PESQ or PEAQ are trademarked by some companies. >> Since then, the ITU-T name their standards only by numbers. >> >> I do not have any problems with nicknames. However, I think the trademar= k >> issue shall be addressed and it is important. I would prefer that such a= s >> trademark is owned by the IETF if possible. >> >> Uspto.gov does not list an opus codec. But I do not know how fast they >> (or the readers of this mailing list) are... >> >> With best regards, >> >> Christian >> >> --------------------------------------------------------------- >> Dr.-Ing. Christian Hoene >> Interactive Communication Systems (ICS), University of T=C3=BCbingen >> Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532 >> >> http://www.net.uni-tuebingen.de/ >> >> >> -----Original Message----- >> From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf O= f >> Stephan Wenger >> Sent: Tuesday, October 19, 2010 6:05 PM >> To: codec@ietf.org >> Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt >> >> Hi Jean-Marc, >> >> Four things: >> >> First, indeed, there was a hum at the last meeting. Weak memory on my >> side. >> Apologies. >> >> Second, hums need to be reconfirmed on the mailing list, and that has no= t >> happened according to my read of the email archive. >> >> Third, I also missed the submission of draft-ietf-codec-description-00, >> which was really the time I should have complained. >> >> So I'm willing to assume (as apparently have the chairs, see point #2) >> that >> there has been an implied consensus of the WG to accept the >> codec-description draft. Which brings me to point #4: >> >> As this is now a WG item, any major change requires WG consensus. >> Selecting >> a marketing name, IMO, is such a major change. "Opus" is such a flashy >> name >> that certain participants and/or companies conceivably may not like it. >> For >> example, if I were working for a company that has in its portfolio an >> audio >> product named "Opus", I would object to the name change. So the thing y= ou >> should have done, IMO, is to send an email to the list saying "The edito= rs >> consider changing the name of our codec to Opus. Is that acceptable to >> the >> WG?". >> >> It appears to me that twice you guys (chairs included) have taken >> shortcuts >> with the IETF's procedures, as I understand them. That, IMO, fills up >> your >> quota for the next couple of years. Please be more conservative from no= w >> on. >> >> Regards, >> Stephan >> >> >> >> >> On 10.19.2010 08:31 , "Jean-Marc Valin" >> wrote: >> >> Hi, >>> >>> draft-ietf-codec-opus-00.txt is indeed a WG item. It has had several >>> names >>> in the past, including draft-valin-codec-prototype and >>> draft-valin-codec-definition, which may explain the confusion. This is >>> the >>> draft for which there was a hum during the last meeting. >>> >>> Jean-Marc >>> >>> On 10-10-19 11:26 AM, Stephan Wenger wrote: >>> >>>> Hi, >>>> >>>> I want to inquire the status of this draft. In many working groups, t= he >>>> filename "draft-ietf--xxx" indicates that the draft in questi= on >>>> is >>>> a WG item of WG. Following this logic, it would appear that >>>> the >>>> "opus" draft is now a WG item of the codec WG. I don't recall a >>>> decision to >>>> than extent. >>>> >>>> If the draft were indeed accepted as a WG item, I would like to >>>> encourage >>>> those who made IPR statements related to it, to resubmit those >>>> statements >>>> with the new filename. This would help those of us who are searching >>>> through the IETF IPR tracker by WG name (which is a very common thing = to >>>> do, >>>> at least for me). >>>> >>>> Thanks, >>>> Stephan >>>> >>>> >>>> On 10.15.2010 13:30 , "Internet-Drafts@ietf.org"< >>>> Internet-Drafts@ietf.org> >>>> wrote: >>>> >>>> A New Internet-Draft is available from the on-line Internet-Drafts >>>>> directories. >>>>> This draft is a work item of the Internet Wideband Audio Codec Workin= g >>>>> Group >>>>> of the IETF. >>>>> >>>>> >>>>> Title : Definition of the Opus Audio Codec >>>>> Author(s) : J. Valin, K. Vos >>>>> Filename : draft-ietf-codec-opus-00.txt >>>>> Pages : 12 >>>>> Date : 2010-10-15 >>>>> >>>>> This document describes the Opus codec, designed for interactive >>>>> speech and audio transmission over the Internet. >>>>> >>>>> A URL for this Internet-Draft is: >>>>> http://www.ietf.org/internet-drafts/draft-ietf-codec-opus-00.txt >>>>> >>>>> Internet-Drafts are also available by anonymous FTP at: >>>>> ftp://ftp.ietf.org/internet-drafts/ >>>>> >>>>> Below is the data which will enable a MIME compliant mail reader >>>>> implementation to automatically retrieve the ASCII version of the >>>>> Internet-Draft. >>>>> _______________________________________________ >>>>> codec mailing list >>>>> codec@ietf.org >>>>> https://www.ietf.org/mailman/listinfo/codec >>>>> >>>> >>>> >>>> _______________________________________________ >>>> codec mailing list >>>> codec@ietf.org >>>> https://www.ietf.org/mailman/listinfo/codec >>>> >>> >>> >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >> > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > --0016e6dee986c890640492fbd89c Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Registering trademarks can be expensive, esp. if you are doing it in multip= le countries.=A0 I don't see any necessity for this - even if we had th= e money, I'd rather see it go for additional characterization/testing.<= br>
Regards
Stephen Botzko

On Tue, Oct= 19, 2010 at 1:33 PM, Jean-Marc Valin <jean-marc.valin@octasic.com> = wrote:
Hi Christian,

I don't know the details of the ITU-T trademark issues, but in this cas= e, the Opus name will not be controlled by any company (I'm not aware o= f anyone registering a trademark on the "Opus" name in this conte= xt). I'm not against the IETF owning the trademark, but I'm not sur= e any ownership is necessary to begin with.

Cheers,

=A0 =A0 =A0 =A0Jean-Marc

As a note, my company does use the term "Opus" in a different dom= ain of application (digital signal processor architecture) and sees no issu= e with the IETF using it for an audio codec. Oh, and no, I'm not the on= e who originally suggested that name.


On 10-10-19 01:02 PM, Christian Hoene wrote:
Hi,

the ITU had some problems with trademarks that were given to standardized a= lgorithms.
As far as I remember, PESQ or PEAQ are trademarked by some companies. Since= then, the ITU-T name their standards only by numbers.

I do not have any problems with nicknames. However, I think the trademark i= ssue shall be addressed and it is important. I would prefer that such as tr= ademark is owned by the IETF if possible.

Uspto.gov does not list an opus codec. But I do not know how fast they =A0(= or the readers of this mailing list) are...

With best regards,

=A0Christian

---------------------------------------------------------------
Dr.-Ing. Christian Hoene
Interactive Communication Systems (ICS), University of T=C3=BCbingen
Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532

http://www.n= et.uni-tuebingen.de/


-----Original Message-----
From: codec-bou= nces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Stephan Wenger
Sent: Tuesday, October 19, 2010 6:05 PM
To: codec@ietf.org<= br> Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt

Hi Jean-Marc,

Four things:

First, indeed, there was a hum at the last meeting. =A0Weak memory on my si= de.
Apologies.

Second, hums need to be reconfirmed on the mailing list, and that has not happened according to my read of the email archive.

Third, I also missed the submission of draft-ietf-codec-description-00,
which was really the time I should have complained.

So I'm willing to assume (as apparently have the chairs, see point #2) = that
there has been an implied consensus of the WG to accept the
codec-description draft. =A0Which brings me to point #4:

As this is now a WG item, any major change requires WG consensus. =A0Select= ing
a marketing name, IMO, is such a major change. =A0"Opus" is such = a flashy name
that certain participants and/or companies conceivably may not like it. =A0= For
example, if I were working for a company that has in its portfolio an audio=
product named "Opus", I would object to the name change. =A0So th= e thing you
should have done, IMO, is to send an email to the list saying "The edi= tors
consider changing the name of our codec to Opus. =A0Is that acceptable to t= he
WG?".

It appears to me that twice you guys (chairs included) have taken shortcuts=
with the IETF's procedures, as I understand them. =A0That, IMO, fills u= p your
quota for the next couple of years. =A0Please be more conservative from now=
on.

Regards,
Stephan




On 10.19.2010 08:31 , "Jean-Marc Valin"<jean-marc.valin@octasic.com&= gt; =A0wrote:

Hi,

draft-ietf-codec-opus-00.txt is indeed a WG item. It has had several names<= br> in the past, including draft-valin-codec-prototype and
draft-valin-codec-definition, which may explain the confusion. This is the<= br> draft for which there was a hum during the last meeting.

Jean-Marc

On 10-10-19 11:26 AM, Stephan Wenger wrote:
Hi,

I want to inquire the status of this draft. =A0In many working groups, the<= br> filename "draft-ietf-<wg name>-xxx" indicates that the draf= t in question is
a WG item of WG<wg-name>. =A0Following this logic, it would appear th= at the
"opus" draft is now a WG item of the codec WG. =A0I don't rec= all a decision to
than extent.

If the draft were indeed accepted as a WG item, I would like to encourage those who made IPR statements related to it, to resubmit those statements with the new filename. =A0This would help those of us who are searching
through the IETF IPR tracker by WG name (which is a very common thing to do= ,
at least for me).

Thanks,
Stephan


On 10.15.2010 13:30 , "Internet-Drafts@ietf.org"<Internet-Drafts@ietf.org> wrote:

A New Internet-Draft is available from the on-line Internet-Drafts
directories.
This draft is a work item of the Internet Wideband Audio Codec Working Grou= p
of the IETF.


Title =A0 =A0 =A0 =A0 =A0 : Definition of the Opus Audio Codec
Author(s) =A0 =A0 =A0 : J. Valin, K. Vos
Filename =A0 =A0 =A0 =A0: draft-ietf-codec-opus-00.txt
Pages =A0 =A0 =A0 =A0 =A0 : 12
Date =A0 =A0 =A0 =A0 =A0 =A0: 2010-10-15

This document describes the Opus codec, designed for interactive
speech and audio transmission over the Internet.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-codec-op= us-00.txt

Internet-Drafts are also available by anonymous FTP at:
ftp://ftp= .ietf.org/internet-drafts/

Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.
_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec


_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec



_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec

_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec

_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec

--0016e6dee986c890640492fbd89c-- From stewe@stewe.org Tue Oct 19 10:54:44 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id B224E3A6838 for ; Tue, 19 Oct 2010 10:54:44 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.118 X-Spam-Level: X-Spam-Status: No, score=-1.118 tagged_above=-999 required=5 tests=[AWL=0.084, BAYES_00=-2.599, HTML_MESSAGE=0.001, MIME_QP_LONG_LINE=1.396] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id GIxdNTDM9wWX for ; Tue, 19 Oct 2010 10:54:35 -0700 (PDT) Received: from stewe.org (stewe.org [85.214.122.234]) by core3.amsl.com (Postfix) with ESMTP id 166BB3A67AB for ; Tue, 19 Oct 2010 10:54:33 -0700 (PDT) Received: from [192.168.1.108] (unverified [24.5.132.232]) by stewe.org (SurgeMail 3.9e) with ESMTP id 819368-1743317 for ; Tue, 19 Oct 2010 19:56:04 +0200 User-Agent: Microsoft-Entourage/12.27.0.100910 Date: Tue, 19 Oct 2010 10:55:44 -0700 From: Stephan Wenger To: "codec@ietf.org" Message-ID: Thread-Topic: [codec] I-D Action:draft-ietf-codec-opus-00.txt Thread-Index: ActvttncK1hlUPmMc0axXWkrqhXygA== In-Reply-To: Mime-version: 1.0 Content-type: multipart/alternative; boundary="B_3370330563_527642" X-Originating-IP: 24.5.132.232 X-Authenticated-User: stewe@stewe.org X-ORBS-Stamp: Your IP (24.5.132.232) was found in the spamhaus database. http://www.spamhaus.net Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 17:54:45 -0000 > This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. --B_3370330563_527642 Content-type: text/plain; charset="ISO-8859-1" Content-transfer-encoding: quoted-printable I agree with Stephen. In addition, there are kinds of marks that are not registered (and thus cannot be found in the USPTO database), but are still enforceable. (As an interesting (at least to me) side note, the trademark problematic ca= n also go the other way. There is a Chinese guy who tries to sell the world =B3his=B2 H.265 video codec=8Bwhich has nothing to do with the ITU=B9s effort towards a Recommendation that may bear this name a few years down the road. Would his activity =B3burn=B2 the designation of a future video codec as H.265 by the ITU? In the IETF, AFAIK, the combination of =B3RFC=B2 and a number is sufficiently well protected, and that=B9s what is really the main concern. I=B9m not sure the ITU has similarly strong mechanisms in place...) Stephan On 10.19.2010 10:44 , "Stephen Botzko" wrote: > Registering trademarks can be expensive, esp. if you are doing it in mult= iple > countries.=A0 I don't see any necessity for this - even if we had the money= , I'd > rather see it go for additional characterization/testing. >=20 > Regards > Stephen Botzko >=20 > On Tue, Oct 19, 2010 at 1:33 PM, Jean-Marc Valin > wrote: >> Hi Christian, >>=20 >> I don't know the details of the ITU-T trademark issues, but in this case= , the >> Opus name will not be controlled by any company (I'm not aware of anyone >> registering a trademark on the "Opus" name in this context). I'm not aga= inst >> the IETF owning the trademark, but I'm not sure any ownership is necessa= ry to >> begin with. >>=20 >> Cheers, >>=20 >> =A0 =A0 =A0 =A0Jean-Marc >>=20 >> As a note, my company does use the term "Opus" in a different domain of >> application (digital signal processor architecture) and sees no issue wi= th >> the IETF using it for an audio codec. Oh, and no, I'm not the one who >> originally suggested that name. >>=20 >>=20 >> On 10-10-19 01:02 PM, Christian Hoene wrote: >>> Hi, >>>=20 >>> the ITU had some problems with trademarks that were given to standardiz= ed >>> algorithms. >>> As far as I remember, PESQ or PEAQ are trademarked by some companies. S= ince >>> then, the ITU-T name their standards only by numbers. >>>=20 >>> I do not have any problems with nicknames. However, I think the tradema= rk >>> issue shall be addressed and it is important. I would prefer that such = as >>> trademark is owned by the IETF if possible. >>>=20 >>> Uspto.gov does not list an opus codec. But I do not know how fast they = =A0(or >>> the readers of this mailing list) are... >>>=20 >>> With best regards, >>>=20 >>> =A0Christian >>>=20 >>> --------------------------------------------------------------- >>> Dr.-Ing. Christian Hoene >>> Interactive Communication Systems (ICS), University of T=C3=BCbingen >>> Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532 >>>=20 >>> http://www.net.uni-tuebingen.de/ >>>=20 >>>=20 >>> -----Original Message----- >>> From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf = Of >>> Stephan Wenger >>> Sent: Tuesday, October 19, 2010 6:05 PM >>> To: codec@ietf.org >>> Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt >>>=20 >>> Hi Jean-Marc, >>>=20 >>> Four things: >>>=20 >>> First, indeed, there was a hum at the last meeting. =A0Weak memory on my = side. >>> Apologies. >>>=20 >>> Second, hums need to be reconfirmed on the mailing list, and that has n= ot >>> happened according to my read of the email archive. >>>=20 >>> Third, I also missed the submission of draft-ietf-codec-description-00, >>> which was really the time I should have complained. >>>=20 >>> So I'm willing to assume (as apparently have the chairs, see point #2) = that >>> there has been an implied consensus of the WG to accept the >>> codec-description draft. =A0Which brings me to point #4: >>>=20 >>> As this is now a WG item, any major change requires WG consensus. =A0Sele= cting >>> a marketing name, IMO, is such a major change. =A0"Opus" is such a flashy= name >>> that certain participants and/or companies conceivably may not like it.= =A0For >>> example, if I were working for a company that has in its portfolio an a= udio >>> product named "Opus", I would object to the name change. =A0So the thing = you >>> should have done, IMO, is to send an email to the list saying "The edit= ors >>> consider changing the name of our codec to Opus. =A0Is that acceptable to= the >>> WG?". >>>=20 >>> It appears to me that twice you guys (chairs included) have taken short= cuts >>> with the IETF's procedures, as I understand them. =A0That, IMO, fills up = your >>> quota for the next couple of years. =A0Please be more conservative from n= ow >>> on. >>>=20 >>> Regards, >>> Stephan >>>=20 >>>=20 >>>=20 >>>=20 >>> On 10.19.2010 08:31 , "Jean-Marc Valin" =A0w= rote: >>>=20 >>>> Hi, >>>>=20 >>>> draft-ietf-codec-opus-00.txt is indeed a WG item. It has had several n= ames >>>> in the past, including draft-valin-codec-prototype and >>>> draft-valin-codec-definition, which may explain the confusion. This is= the >>>> draft for which there was a hum during the last meeting. >>>>=20 >>>> Jean-Marc >>>>=20 >>>> On 10-10-19 11:26 AM, Stephan Wenger wrote: >>>>> Hi, >>>>>=20 >>>>> I want to inquire the status of this draft. =A0In many working groups, = the >>>>> filename "draft-ietf--xxx" indicates that the draft in quest= ion >>>>> is >>>>> a WG item of WG. =A0Following this logic, it would appear that= the >>>>> "opus" draft is now a WG item of the codec WG. =A0I don't recall a deci= sion >>>>> to >>>>> than extent. >>>>>=20 >>>>> If the draft were indeed accepted as a WG item, I would like to encou= rage >>>>> those who made IPR statements related to it, to resubmit those statem= ents >>>>> with the new filename. =A0This would help those of us who are searching >>>>> through the IETF IPR tracker by WG name (which is a very common thing= to >>>>> do, >>>>> at least for me). >>>>>=20 >>>>> Thanks, >>>>> Stephan >>>>>=20 >>>>>=20 >>>>> On 10.15.2010 13:30 , "Internet-Drafts@ietf.org" >>>>> wrote: >>>>>=20 >>>>>> A New Internet-Draft is available from the on-line Internet-Drafts >>>>>> directories. >>>>>> This draft is a work item of the Internet Wideband Audio Codec Worki= ng >>>>>> Group >>>>>> of the IETF. >>>>>>=20 >>>>>>=20 >>>>>> Title =A0 =A0 =A0 =A0 =A0 : Definition of the Opus Audio Codec >>>>>> Author(s) =A0 =A0 =A0 : J. Valin, K. Vos >>>>>> Filename =A0 =A0 =A0 =A0: draft-ietf-codec-opus-00.txt >>>>>> Pages =A0 =A0 =A0 =A0 =A0 : 12 >>>>>> Date =A0 =A0 =A0 =A0 =A0 =A0: 2010-10-15 >>>>>>=20 >>>>>> This document describes the Opus codec, designed for interactive >>>>>> speech and audio transmission over the Internet. >>>>>>=20 >>>>>> A URL for this Internet-Draft is: >>>>>> http://www.ietf.org/internet-drafts/draft-ietf-codec-opus-00.txt >>>>>>=20 >>>>>> Internet-Drafts are also available by anonymous FTP at: >>>>>> ftp://ftp.ietf.org/internet-drafts/ >>>>>>=20 >>>>>> Below is the data which will enable a MIME compliant mail reader >>>>>> implementation to automatically retrieve the ASCII version of the >>>>>> Internet-Draft. >>>>>> _______________________________________________ >>>>>> codec mailing list >>>>>> codec@ietf.org >>>>>> https://www.ietf.org/mailman/listinfo/codec >>>>>=20 >>>>>=20 >>>>> _______________________________________________ >>>>> codec mailing list >>>>> codec@ietf.org >>>>> https://www.ietf.org/mailman/listinfo/codec >>>>=20 >>>=20 >>>=20 >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >>>=20 >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >>=20 >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >=20 >=20 >=20 > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec --B_3370330563_527642 Content-type: text/html; charset="ISO-8859-1" Content-transfer-encoding: quoted-printable Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt I agree with Stephen.
In addition, there are kinds of marks that are not registered (and thus can= not be found in the USPTO database), but are still enforceable.

(As an interesting (at least to me) side note, the trademark problematic ca= n also go the other way.  There is a Chinese guy who tries to sell the = world “his” H.265 video codec—which has nothing to do with= the ITU’s effort towards a Recommendation that may bear this name a f= ew years down the road.  Would his activity “burn” the desi= gnation of a future video codec as H.265 by the ITU?  In the IETF, AFAI= K, the combination of  “RFC” and a number is sufficiently w= ell protected, and that’s what is really the main concern.  IR= 17;m not sure the ITU has similarly strong mechanisms in place...)

Stephan



On 10.19.2010 10:44 , "Stephen Botzko" <stephen.botzko@gmail.com> wrote:

<= SPAN STYLE=3D'font-size:11pt'>Registering trademarks can be expensive, esp. if= you are doing it in multiple countries.=A0 I don't see any necessity for this= - even if we had the money, I'd rather see it go for additional characteriz= ation/testing.

Regards
Stephen Botzko

On Tue, Oct 19, 2010 at 1:33 PM, Jean-Marc Valin <jean-marc.valin@octasic.com> wrote:
<= SPAN STYLE=3D'font-size:11pt'>Hi Christian,

I don't know the details of the ITU-T trademark issues, but in this case, t= he Opus name will not be controlled by any company (I'm not aware of anyone = registering a trademark on the "Opus" name in this context). I'm n= ot against the IETF owning the trademark, but I'm not sure any ownership is = necessary to begin with.

Cheers,

 =A0 =A0 =A0 =A0Jean-Marc

As a note, my company does use the term "Opus" in a different dom= ain of application (digital signal processor architecture) and sees no issue= with the IETF using it for an audio codec. Oh, and no, I'm not the one who = originally suggested that name.


On 10-10-19 01:02 PM, Christian Hoene wrote:
<= SPAN STYLE=3D'font-size:11pt'>Hi,

the ITU had some problems with trademarks that were given to standardized a= lgorithms.
As far as I remember, PESQ or PEAQ are trademarked by some companies. Since= then, the ITU-T name their standards only by numbers.

I do not have any problems with nicknames. However, I think the trademark i= ssue shall be addressed and it is important. I would prefer that such as tra= demark is owned by the IETF if possible.

Uspto.gov does not list an opus codec. But I do not know how fast they =A0(or= the readers of this mailing list) are...

With best regards,

 =A0Christian

---------------------------------------------------------------
Dr.-Ing. Christian Hoene
Interactive Communication Systems (ICS), University of TÃπbinge= n
Sand 13, 72076 TÃπbingen, Germany, Phone +49 7071 2970532

http://www.net.uni-tuebingen.de/=


-----Original Message-----
From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf= Of Stephan Wenger
Sent: Tuesday, October 19, 2010 6:05 PM
To: codec@ietf.org
Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt

Hi Jean-Marc,

Four things:

First, indeed, there was a hum at the last meeting. =A0Weak memory on my side= .
Apologies.

Second, hums need to be reconfirmed on the mailing list, and that has not happened according to my read of the email archive.

Third, I also missed the submission of draft-ietf-codec-description-00,
which was really the time I should have complained.

So I'm willing to assume (as apparently have the chairs, see point #2) that=
there has been an implied consensus of the WG to accept the
codec-description draft. =A0Which brings me to point #4:

As this is now a WG item, any major change requires WG consensus. =A0Selectin= g
a marketing name, IMO, is such a major change. =A0"Opus" is such a = flashy name
that certain participants and/or companies conceivably may not like it. =A0Fo= r
example, if I were working for a company that has in its portfolio an audio=
product named "Opus", I would object to the name change. =A0So the = thing you
should have done, IMO, is to send an email to the list saying "The edi= tors
consider changing the name of our codec to Opus. =A0Is that acceptable to the=
WG?".

It appears to me that twice you guys (chairs included) have taken shortcuts=
with the IETF's procedures, as I understand them. =A0That, IMO, fills up your=
quota for the next couple of years. =A0Please be more conservative from now on.

Regards,
Stephan




On 10.19.2010 08:31 , "Jean-Marc Valin"<jean-marc.valin@octasic.com> =A0wrote:

<= SPAN STYLE=3D'font-size:11pt'>Hi,

draft-ietf-codec-opus-00.txt is indeed a WG item. It has had several names<= BR> in the past, including draft-valin-codec-prototype and
draft-valin-codec-definition, which may explain the confusion. This is the<= BR> draft for which there was a hum during the last meeting.

Jean-Marc

On 10-10-19 11:26 AM, Stephan Wenger wrote:
<= SPAN STYLE=3D'font-size:11pt'>Hi,

I want to inquire the status of this draft. =A0In many working groups, the filename "draft-ietf-<wg name>-xxx" indicates that the draf= t in question is
a WG item of WG<wg-name>. =A0Following this logic, it would appear that= the
"opus" draft is now a WG item of the codec WG. =A0I don't recall a = decision to
than extent.

If the draft were indeed accepted as a WG item, I would like to encourage those who made IPR statements related to it, to resubmit those statements with the new filename. =A0This would help those of us who are searching
through the IETF IPR tracker by WG name (which is a very common thing to do= ,
at least for me).

Thanks,
Stephan


On 10.15.2010 13:30 , "Internet-Drafts@ietf.org"<Internet-Drafts@ietf.org&= gt;
wrote:

<= SPAN STYLE=3D'font-size:11pt'>A New Internet-Draft is available from the on-li= ne Internet-Drafts
directories.
This draft is a work item of the Internet Wideband Audio Codec Working Grou= p
of the IETF.


Title =A0 =A0 =A0 =A0 =A0 : Definition of the Opus Audio Codec
Author(s) =A0 =A0 =A0 : J. Valin, K. Vos
Filename =A0 =A0 =A0 =A0: draft-ietf-codec-opus-00.txt
Pages =A0 =A0 =A0 =A0 =A0 : 12
Date =A0 =A0 =A0 =A0 =A0 =A0: 2010-10-15

This document describes the Opus codec, designed for interactive
speech and audio transmission over the Internet.

A URL for this Internet-Draft is:
= http://www.ietf.org/internet-drafts/draft-ietf-codec-opus-00.txt

Internet-Drafts are also available by anonymous FTP at:
ftp://ftp.ietf.org/internet-d= rafts/

Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.
_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/= mailman/listinfo/codec
=

_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/= mailman/listinfo/codec
=
=

_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/= mailman/listinfo/codec

_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/= mailman/listinfo/codec
=
_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/= mailman/listinfo/codec
=

___________= ____________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/= mailman/listinfo/codec
--B_3370330563_527642-- From hoene@uni-tuebingen.de Tue Oct 19 11:11:30 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 9B1713A67AB for ; Tue, 19 Oct 2010 11:11:30 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -5.494 X-Spam-Level: X-Spam-Status: No, score=-5.494 tagged_above=-999 required=5 tests=[AWL=0.755, BAYES_00=-2.599, HELO_EQ_DE=0.35, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id df1CoqkSiZ9T for ; Tue, 19 Oct 2010 11:11:27 -0700 (PDT) Received: from mx05.uni-tuebingen.de (mx05.uni-tuebingen.de [134.2.3.4]) by core3.amsl.com (Postfix) with ESMTP id 99BF43A68B1 for ; Tue, 19 Oct 2010 11:11:25 -0700 (PDT) Received: from hoeneT60 ([178.2.222.223]) (authenticated bits=0) by mx05.uni-tuebingen.de (8.13.6/8.13.6) with ESMTP id o9JICkml025784 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Tue, 19 Oct 2010 20:12:52 +0200 From: "Christian Hoene" To: "'Stephan Wenger'" , References: <390831ED3DF58E41A3D2FB82591E2C3605A34C4D@MAILEXCH.octasic.com> In-Reply-To: Date: Tue, 19 Oct 2010 20:12:48 +0200 Organization: =?UTF-8?Q?Universit=C3=A4t_T=C3=BCbingen?= Message-ID: <000c01cb6fb9$41b95910$c52c0b30$@de> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_000D_01CB6FCA.05422910" X-Mailer: Microsoft Office Outlook 12.0 Thread-Index: ActvNCcg/6UdLTfYSPe62Egnu0SU9wAAN+7tABsLZFoABEUx8A== Content-language: de X-AntiVirus: NOT checked by Avira MailGate (version: 3.0.0-4; host: mx05) Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 18:11:30 -0000 This is a multi-part message in MIME format. ------=_NextPart_000_000D_01CB6FCA.05422910 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Hi =20 From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf = Of Stephan Wenger Sent: Tuesday, October 19, 2010 5:21 PM To: Jean-Marc Valin; Pochol@WebfootGames.com; codec@ietf.org Subject: Re: [codec] Opus codec licensing =20 Hi, With =E2=80=9Chat=E2=80=9D: It is, in almost all cases, unwise for a third party to = circumscribe=E2=80=94or worse: abbreviate=E2=80=94licensing terms = provided by a rightholder. Skype=E2=80=99s licensing declaration can be = found at https://datatracker.ietf.org/ipr/1297. I note that the = statement contains conditions beyond the acceptance of their technology = into an IETF standard. Stephan [Christian Hoene] +1 =20 It is planned to test the new codec by spreading it to users prior = standardization. However, under the terms of this licensing declaration, = this is not possible for anybody but Skype because it would require a = license from Skype. =20 Also, any enhancement of the codec, which does not conform to the = original standard anymore, will require a license. =20 Furthermore, any implementation of the codec that accidentally does not = confirm to the standard yet will require a license fee. =20 Clearly, these terms are unacceptable. Please, Koen, talk to your lawyer = again to weaken this license statements. This shall include at least a = draft versions of the standard. Also, updates of the standard shall not = require any new license anymore. =20 It is patent really essential for the opus codec?=20 =E2=80=9CA method of estimating noise in data containing voice = information and noise includes receiving the data as a sequence of input = values; transforming the data by applying a first non linear mapping to = the input values wherein the derivative function of the mapping = decreases in magnitude as the input values increase in magnitude = smoothing the transformed data; and transforming the smoothed = transformed data by applying a second non linear mapping that is = opposite to the first non linear mapping, to determine an estimate of = the noise in the inputted data.=E2=80=9D Anyhow, it should not be too difficult to circumvent the patent. Any = ideas?=20 =20 Christian =20 --------------------------------------------------------------- Dr.-Ing. Christian Hoene Interactive Communication Systems (ICS), University of T=C3=BCbingen=20 Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532=20 http://www.net.uni-tuebingen.de/ =20 On 10.18.2010 19:26 , "Jean-Marc Valin" = wrote: Opus is available under the BSD license. As for patents, Skype has = pledged to make them available royalty-free once the codec is accepted = as an IETF standard. Jean-Marc -----Original Message----- From: codec-bounces@ietf.org on behalf of Pascal Pochol Sent: Mon 10/18/2010 10:19 PM To: codec@ietf.org Subject: [codec] Opus codec licensing Hello, we've been using speex and celt for voice and music but today I heard = about Opus which sounds like a fantastic replacement for both these codecs. We're wondering if Opus will be released under the same type of license = as speex and celt? SILK's license forbid its use in commercial software and we're worried that it might carry over to Opus. If not, as soon as Opus handles lower bitrates, stereo and fixed point decoding we'll be using = it. -Pascal _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec _____ =20 _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec ------=_NextPart_000_000D_01CB6FCA.05422910 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Re: [codec] Opus codec licensing

Hi

 

From:= = codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Stephan Wenger
Sent: Tuesday, October 19, 2010 5:21 PM
To: Jean-Marc Valin; Pochol@WebfootGames.com; codec@ietf.org
Subject: Re: [codec] Opus codec licensing

 

Hi,
With =E2=80=9Chat=E2=80=9D:
It is, in almost all cases, unwise for a third party to = circumscribe=E2=80=94or worse: abbreviate=E2=80=94licensing terms provided by a rightholder. =  Skype=E2=80=99s licensing declaration can be found at https://datatracker.ietf.o= rg/ipr/1297.  I note that the statement contains conditions beyond the = acceptance of their technology into an IETF standard.
Stephan

[Christian Hoene] = +1

 

It is planned to test the new codec by spreading it to = users prior standardization. However, under the terms of this licensing = declaration, this is not possible for anybody but Skype because it would require a = license from Skype.

 

Also, any enhancement of the codec, which does not = conform to the original standard anymore, will require a = license.

 

Furthermore, any implementation of the codec that = accidentally does not confirm to the standard yet will require a license = fee.

 

Clearly, these terms are unacceptable. Please, Koen, talk = to your lawyer again to weaken this license statements. This shall include = at least a draft versions of the standard. Also, updates of the standard = shall not require any new license anymore.

 

It is patent really essential for the opus codec? =

=E2=80=9C= A method of estimating noise in data containing voice information and noise = includes receiving the data as a sequence of input values; transforming the data = by applying a first non linear mapping to the input values wherein the = derivative function of the mapping decreases in magnitude as the input values = increase in magnitude smoothing the transformed data; and transforming the smoothed transformed data by applying a second non linear mapping that is = opposite to the first non linear mapping, to determine an estimate of the noise in = the inputted data.=E2=80=9D

Anyhow, it should not be too difficult to circumvent the = patent. Any ideas?

 

Christian

 

----------------------------------------------------------= -----

Dr.-Ing. Christian Hoene

Interactive Communication Systems (ICS), University of = T=C3=BCbingen

Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 = 2970532

http://www.net.uni-tuebingen.de= /

 




On 10.18.2010 19:26 , "Jean-Marc Valin" <jean-marc.valin@octasic.com> wrote:

Opus is available under the BSD = license. As for patents, Skype has pledged to make them available royalty-free once = the codec is accepted as an IETF standard.

    Jean-Marc

-----Original Message-----
From: codec-bounces@ietf.org on = behalf of Pascal Pochol
Sent: Mon 10/18/2010 10:19 PM
To: codec@ietf.org
Subject: [codec] Opus codec licensing

Hello,

we've been using speex and celt for voice and music but today I heard = about
Opus which sounds like a fantastic replacement for both these = codecs.

We're wondering if Opus will be released under the same type of license = as
speex and celt? SILK's license forbid its use in commercial software = and
we're worried that it might carry over to Opus. If not, as soon as = Opus
handles lower bitrates, stereo and fixed point decoding we'll be using = it.

-Pascal

_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org= /mailman/listinfo/codec

_________________________= ______________________
codec mailing list
codec@ietf.org
https://www.ietf.org= /mailman/listinfo/codec

------=_NextPart_000_000D_01CB6FCA.05422910-- From roman@telurix.com Tue Oct 19 11:58:45 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id BB1C73A68C7 for ; Tue, 19 Oct 2010 11:58:45 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.373 X-Spam-Level: X-Spam-Status: No, score=-1.373 tagged_above=-999 required=5 tests=[AWL=0.603, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id n-Yo-gAwwELa for ; Tue, 19 Oct 2010 11:58:44 -0700 (PDT) Received: from mail-gw0-f44.google.com (mail-gw0-f44.google.com [74.125.83.44]) by core3.amsl.com (Postfix) with ESMTP id E51C93A67DA for ; Tue, 19 Oct 2010 11:58:43 -0700 (PDT) Received: by gwaa18 with SMTP id a18so463753gwa.31 for ; Tue, 19 Oct 2010 12:00:14 -0700 (PDT) Received: by 10.150.220.16 with SMTP id s16mr9866787ybg.149.1287514814718; Tue, 19 Oct 2010 12:00:14 -0700 (PDT) Received: from mail-yw0-f44.google.com (mail-yw0-f44.google.com [209.85.213.44]) by mx.google.com with ESMTPS id m12sm404836ybn.12.2010.10.19.12.00.13 (version=TLSv1/SSLv3 cipher=RC4-MD5); Tue, 19 Oct 2010 12:00:13 -0700 (PDT) Received: by ywa6 with SMTP id 6so1620611ywa.31 for ; Tue, 19 Oct 2010 12:00:12 -0700 (PDT) MIME-Version: 1.0 Received: by 10.42.214.138 with SMTP id ha10mr4280616icb.401.1287514812133; Tue, 19 Oct 2010 12:00:12 -0700 (PDT) Received: by 10.231.154.19 with HTTP; Tue, 19 Oct 2010 12:00:11 -0700 (PDT) In-Reply-To: <000c01cb6fb9$41b95910$c52c0b30$@de> References: <390831ED3DF58E41A3D2FB82591E2C3605A34C4D@MAILEXCH.octasic.com> <000c01cb6fb9$41b95910$c52c0b30$@de> Date: Tue, 19 Oct 2010 15:00:11 -0400 Message-ID: From: Roman Shpount To: Christian Hoene Content-Type: multipart/alternative; boundary=20cf301d41d40531ba0492fce710 Cc: codec@ietf.org Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 18:58:45 -0000 --20cf301d41d40531ba0492fce710 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: quoted-printable People get touchy here about discussing the actual patent claims, but as fa= r as I can see the only thing affected by this patent is the DTX portion of the codec. We can use some other VAD method for the Opus codec to work around the patent. P.S. I am pretty sure, I saw something very similar to what's described in this patent in OpenH323 about 10 years ago, but I am not the patent lawyer. _____________________________ Roman Shpount - www.telurix.com 2010/10/19 Christian Hoene > Hi > > > > *From:* codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] *On Behalf > Of *Stephan Wenger > *Sent:* Tuesday, October 19, 2010 5:21 PM > *To:* Jean-Marc Valin; Pochol@WebfootGames.com; codec@ietf.org > *Subject:* Re: [codec] Opus codec licensing > > > > Hi, > With =93hat=94: > It is, in almost all cases, unwise for a third party to circumscribe=97or > worse: abbreviate=97licensing terms provided by a rightholder. Skype=92s > licensing declaration can be found at > https://datatracker.ietf.org/ipr/1297. I note that the statement contain= s > conditions beyond the acceptance of their technology into an IETF standar= d. > Stephan > > *[Christian Hoene] *+1 > > > > It is planned to test the new codec by spreading it to users prior > standardization. However, under the terms of this licensing declaration, > this is not possible for anybody but Skype because it would require a > license from Skype. > > > > Also, any enhancement of the codec, which does not conform to the origina= l > standard anymore, will require a license. > > > > Furthermore, any implementation of the codec that accidentally does not > confirm to the standard yet will require a license fee. > > > > Clearly, these terms are unacceptable. Please, Koen, talk to your lawyer > again to weaken this license statements. This shall include at least a dr= aft > versions of the standard. Also, updates of the standard shall not require > any new license anymore. > > > > It is patent really essential for the opus codec? > > =93A method of estimating noise in data containing voice information and > noise includes receiving the data as a sequence of input values; > transforming the data by applying a first non linear mapping to the input > values wherein the derivative function of the mapping decreases in magnit= ude > as the input values increase in magnitude smoothing the transformed data; > and transforming the smoothed transformed data by applying a second non > linear mapping that is opposite to the first non linear mapping, to > determine an estimate of the noise in the inputted data.=94 > > Anyhow, it should not be too difficult to circumvent the patent. Any idea= s? > > > > > Christian > > > > --------------------------------------------------------------- > > Dr.-Ing. Christian Hoene > > Interactive Communication Systems (ICS), University of T=FCbingen > > Sand 13, 72076 T=FCbingen, Germany, Phone +49 7071 2970532 > > http://www.net.uni-tuebingen.de/ > > * * > > > > > On 10.18.2010 19:26 , "Jean-Marc Valin" > wrote: > > Opus is available under the BSD license. As for patents, Skype has pledge= d > to make them available royalty-free once the codec is accepted as an IETF > standard. > > Jean-Marc > > -----Original Message----- > From: codec-bounces@ietf.org on behalf of Pascal Pochol > Sent: Mon 10/18/2010 10:19 PM > To: codec@ietf.org > Subject: [codec] Opus codec licensing > > Hello, > > we've been using speex and celt for voice and music but today I heard abo= ut > Opus which sounds like a fantastic replacement for both these codecs. > > We're wondering if Opus will be released under the same type of license a= s > speex and celt? SILK's license forbid its use in commercial software and > we're worried that it might carry over to Opus. If not, as soon as Opus > handles lower bitrates, stereo and fixed point decoding we'll be using it= . > > -Pascal > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > ------------------------------ > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > --20cf301d41d40531ba0492fce710 Content-Type: text/html; charset=windows-1252 Content-Transfer-Encoding: quoted-printable People get touchy here about discussing the actual patent claims, but as fa= r as I can see the only thing affected by this patent is the DTX portion of= the codec. We can use some other VAD method for the Opus codec to work aro= und the patent.

P.S. I am pretty sure, I saw something very similar to what's descr= ibed in this patent in OpenH323 about 10 years ago, but I am not the patent= lawyer.
_____________________________
Roman Shpount - = www.telurix.com


2010/10/19 Christian Hoene <hoene@uni-tuebingen.de>

Hi

=A0

From:= codec-bounces@ietf.org [mailto:codec-b= ounces@ietf.org] On Behalf Of Stephan Wenger
Sent: Tuesday, October 19, 2010 5:21 PM
To: Jean-Marc Valin; Pochol@WebfootGames.com; codec@ietf.org
Subject: Re: [codec] Opus codec licensing

=A0

Hi,
With =93hat=94:
It is, in almost all cases, unwise for a third party to circumscribe=97or w= orse: abbreviate=97licensing terms provided by a rightholder. =A0Skype=92s licens= ing declaration can be found at https://datatracker.ietf.org/ipr/1297. =A0I note that the statement contains conditions beyond the acceptance of their technology into an IETF standard.
Stephan

[Christian Hoene] +1

=A0

It is planned to test the new codec by spreading it to= users prior standardization. However, under the terms of this licensing declarati= on, this is not possible for anybody but Skype because it would require a licen= se from Skype.

=A0

Also, any enhancement of the codec, which does not con= form to the original standard anymore, will require a license.

=A0

Furthermore, any implementation of the codec that acci= dentally does not confirm to the standard yet will require a license fee.

=A0

Clearly, these terms are unacceptable. Please, Koen, t= alk to your lawyer again to weaken this license statements. This shall include at least a draft versions of the standard. Also, updates of the standard shall= not require any new license anymore.

=A0

It is patent really essential for the opus codec?

=93A method of estimating noise in data containing voice information and noise includes receiving the data as a sequence of input values; transforming the data by applying a first non linear mapping to the input values wherein the derivat= ive function of the mapping decreases in magnitude as the input values increase= in magnitude smoothing the transformed data; and transforming the smoothed transformed data by applying a second non linear mapping that is opposite t= o the first non linear mapping, to determine an estimate of the noise in the inputted data.=94

Anyhow, it should not be too difficult to circumvent t= he patent. Any ideas?

=A0

Christian

=A0

-----------------------------= ----------------------------------

Dr.-Ing. Christian Hoene

Interactive Communication Sys= tems (ICS), University of T=FCbingen

Sand 13, 72076 T=FCbingen, Ge= rmany, Phone +49 7071 2970532

http://www.net.uni-tuebingen.de/

=A0




On 10.18.2010 19:26 , "Jean-Marc Valin" <jean-marc.valin@octasic.com= > wrote:

Opus is available under the BSD license. As for patents, Skype has pledged to make them available royalty-free once the codec is accepted as an IETF standard.

=A0=A0=A0=A0Jean-Marc

-----Original Message-----
From: codec-bou= nces@ietf.org on behalf of Pascal Pochol
Sent: Mon 10/18/2010 10:19 PM
To: codec@ietf.org<= br> Subject: [codec] Opus codec licensing

Hello,

we've been using speex and celt for voice and music but today I heard a= bout
Opus which sounds like a fantastic replacement for both these codecs.

We're wondering if Opus will be released under the same type of license= as
speex and celt? SILK's license forbid its use in commercial software an= d
we're worried that it might carry over to Opus. If not, as soon as Opus=
handles lower bitrates, stereo and fixed point decoding we'll be using = it.

-Pascal

_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec

_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec


_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec


--20cf301d41d40531ba0492fce710-- From koen.vos@skype.net Tue Oct 19 12:26:12 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id C83E53A6936 for ; Tue, 19 Oct 2010 12:26:12 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.258 X-Spam-Level: X-Spam-Status: No, score=-2.258 tagged_above=-999 required=5 tests=[AWL=0.340, BAYES_00=-2.599, HTML_MESSAGE=0.001] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9QWH3Ap997+F for ; Tue, 19 Oct 2010 12:26:11 -0700 (PDT) Received: from mx.skype.net (mx.skype.net [78.141.177.88]) by core3.amsl.com (Postfix) with ESMTP id 986D03A6931 for ; Tue, 19 Oct 2010 12:26:10 -0700 (PDT) Received: from mx.skype.net (localhost [127.0.0.1]) by mx.skype.net (Postfix) with ESMTP id DF4A81702; Tue, 19 Oct 2010 21:27:40 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=skype.net; h=date:from:to :cc:message-id:in-reply-to:subject:mime-version:content-type; s= mx; bh=r8rKmLtM6StsGC98zo0EblZgoLQ=; b=RHs5SftpxxBl3WJ8MNrhdF8fQ LIOUduNVKDhUwL4Syo1H7wkIKUqOrCY0SkQ41g1/XclHhlXL5+/pPDjOgAZHk/Zf eWvGRrxjnKEklXAJDh5Yjrzas53lNR0uLVSd32/a2r5RBMamPL0PZB9fK6op1M2p RZeDbscd96xKrqoFSQ= DomainKey-Signature: a=rsa-sha1; c=nofws; d=skype.net; h=date:from:to:cc :message-id:in-reply-to:subject:mime-version:content-type; q=dns ; s=mx; b=XYBHm9dOs9S/CW0P89Y04NLBupnkbf83/UxHCU1DqIniWvxZDp+eNT PWRmLRn8ugqTPUHY9T47JvTYBbm0u7yCZFJondcKKis2QxWloDr9R/08PREq0UWw dND3y9XL/uEi0kV3whjj+/I3KQd8DIOVPwYyc3BOxpc0nv112diNA= Received: from zimbra.skype.net (zimbra.skype.net [78.141.177.82]) by mx.skype.net (Postfix) with ESMTP id DD980CF; Tue, 19 Oct 2010 21:27:40 +0200 (CEST) Received: from localhost (localhost [127.0.0.1]) by zimbra.skype.net (Postfix) with ESMTP id C0BC61672687; Tue, 19 Oct 2010 21:27:40 +0200 (CEST) X-Virus-Scanned: amavisd-new at lu2-zimbra.skype.net Received: from zimbra.skype.net ([127.0.0.1]) by localhost (zimbra.skype.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id JiZwDUxZ+jsR; Tue, 19 Oct 2010 21:27:39 +0200 (CEST) Received: from zimbra.skype.net (lu2-zimbra.skype.net [78.141.177.82]) by zimbra.skype.net (Postfix) with ESMTP id A358C1672684; Tue, 19 Oct 2010 21:27:39 +0200 (CEST) Date: Tue, 19 Oct 2010 21:27:39 +0200 (CEST) From: Koen Vos To: Christian Hoene Message-ID: <1087600857.938363.1287516459556.JavaMail.root@lu2-zimbra> In-Reply-To: <000c01cb6fb9$41b95910$c52c0b30$@de> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_938362_2023225065.1287516459554" X-Originating-IP: [69.181.192.115] X-Mailer: Zimbra 6.0.6_GA_2330.UBUNTU8_64 (ZimbraWebClient - FF3.0 (Win)/6.0.6_GA_2330.UBUNTU8_64) Cc: codec@ietf.org Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 19:26:12 -0000 ------=_Part_938362_2023225065.1287516459554 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable If you modify the codec so much that it can't=20 From: "Christian Hoene" =20 To: "Stephan Wenger" , codec@ietf.org=20 Sent: Tuesday, October 19, 2010 11:12:48 AM=20 Subject: Re: [codec] Opus codec licensing=20 Re: [codec] Opus codec licensing=20 Hi=20 From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of S= tephan Wenger=20 Sent: Tuesday, October 19, 2010 5:21 PM=20 To: Jean-Marc Valin; Pochol@WebfootGames.com; codec@ietf.org=20 Subject: Re: [codec] Opus codec licensing=20 Hi,=20 With =E2=80=9Chat=E2=80=9D:=20 It is, in almost all cases, unwise for a third party to circumscribe=E2=80= =94or worse: abbreviate=E2=80=94licensing terms provided by a rightholder. = Skype=E2=80=99s licensing declaration can be found at https://datatracker.i= etf.org/ipr/1297 . I note that the statement contains conditions beyond the= acceptance of their technology into an IETF standard.=20 Stephan=20 [Christian Hoene] +1=20 It is planned to test the new codec by spreading it to users prior standard= ization. However, under the terms of this licensing declaration, this is no= t possible for anybody but Skype because it would require a license from Sk= ype.=20 Also, any enhancement of the codec, which does not conform to the original = standard anymore, will require a license.=20 Furthermore, any implementation of the codec that accidentally does not con= firm to the standard yet will require a license fee.=20 Clearly, these terms are unacceptable. Please, Koen, talk to your lawyer ag= ain to weaken this license statements. This shall include at least a draft = versions of the standard. Also, updates of the standard shall not require a= ny new license anymore.=20 It is patent really essential for the opus codec?=20 =E2=80=9C A method of estimating noise in data containing voice information= and noise includes receiving the data as a sequence of input values; trans= forming the data by applying a first non linear mapping to the input values= wherein the derivative function of the mapping decreases in magnitude as t= he input values increase in magnitude smoothing the transformed data; and t= ransforming the smoothed transformed data by applying a second non linear m= apping that is opposite to the first non linear mapping, to determine an es= timate of the noise in the inputted data.=E2=80=9D=20 Anyhow, it should not be too difficult to circumvent the patent. Any ideas?= =20 Christian=20 ---------------------------------------------------------------=20 Dr.-Ing. Christian Hoene=20 Interactive Communication Systems (ICS), University of T=C3=BCbingen=20 Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532=20 http://www.net.uni-tuebingen.de/=20 On 10.18.2010 19:26 , "Jean-Marc Valin" < jean-marc.valin@octasic.com > wro= te:=20 Opus is available under the BSD license. As for patents, Skype has pledged = to make them available royalty-free once the codec is accepted as an IETF s= tandard.=20 Jean-Marc=20 -----Original Message-----=20 From: codec-bounces@ietf.org on behalf of Pascal Pochol=20 Sent: Mon 10/18/2010 10:19 PM=20 To: codec@ietf.org=20 Subject: [codec] Opus codec licensing=20 Hello,=20 we've been using speex and celt for voice and music but today I heard about= =20 Opus which sounds like a fantastic replacement for both these codecs.=20 We're wondering if Opus will be released under the same type of license as= =20 speex and celt? SILK's license forbid its use in commercial software and=20 we're worried that it might carry over to Opus. If not, as soon as Opus=20 handles lower bitrates, stereo and fixed point decoding we'll be using it.= =20 -Pascal=20 _______________________________________________=20 codec mailing list=20 codec@ietf.org=20 https://www.ietf.org/mailman/listinfo/codec=20 _______________________________________________=20 codec mailing list=20 codec@ietf.org=20 https://www.ietf.org/mailman/listinfo/codec=20 _______________________________________________=20 codec mailing list=20 codec@ietf.org=20 https://www.ietf.org/mailman/listinfo/codec=20 ------=_Part_938362_2023225065.1287516459554 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable <= div style=3D'font-family: Times New Roman; font-size: 12pt; color: #000000'= >If you modify the codec so much that it can't


From: "Christian Hoene" <hoene@uni-tuebingen.de>
To: "Stephan We= nger" <stewe@stewe.org>, codec@ietf.org
Sent: Tuesday, Octo= ber 19, 2010 11:12:48 AM
Subject: Re: [codec] Opus codec licensin= g

Re: [codec] Opus codec licensing

Hi

 

From: code= c-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Stephan Wenger
Sent: Tuesday, October 19, 2010 5:21 PM
To: Jean-Marc Valin; Pochol@WebfootGames.com; codec@ietf.org
Subject: Re: [codec] Opus codec licensing

 

Hi, With =E2=80=9Chat=E2=80=9D:
It is, in almost all cases, unwise for a third party to circumscribe=E2=80= =94or worse: abbreviate=E2=80=94licensing terms provided by a rightholder.  Skype= =E2=80=99s licensing declaration can be found at https://datatracker.ietf.org/ipr/1297.  I note that the statement contains conditions beyond the acceptance o= f their technology into an IETF standard.
Stephan

[Christian Hoene] +1

 

It is planned to test the new codec by spreading it to users prior standardization. However, under the terms of this licensing declarati= on, this is not possible for anybody but Skype because it would require a licen= se from Skype.

 

Also, any enhancement of the codec, which does not conform to the original standard anymore, will require a license.

 

Furthermore, any implementation of the codec that accidentally does not confirm to the standard yet will require a license fee.

 

Clearly, these terms are unacceptable. Please, Koen, talk to your lawyer again to weaken this license statements. This shall include at least a draft versions of the standard. Also, updates of the standard shall= not require any new license anymore.

 

It is patent really essential for the opus codec?

=E2=80=9CA method of estimating noise in data containing voice information and noise includes receiving the data as a sequence of input values; transforming the data by applying a first non linear mapping to the input values wherein the derivat= ive function of the mapping decreases in magnitude as the input values increase= in magnitude smoothing the transformed data; and transforming the smoothed transformed data by applying a second non linear mapping that is opposite t= o the first non linear mapping, to determine an estimate of the noise in the inputted data.=E2=80=9D

Anyhow, it should not be too difficult to circumvent the patent. Any ideas?

 

Christian

 

-----------------------------= ----------------------------------

Dr.-Ing. Christian Hoene

Interactive Communication Sys= tems (ICS), University of T=C3=BCbingen

Sand 13, 72076 T=C3=BCbingen,= Germany, Phone +49 7071 2970532

http://www.net.uni-tuebingen.de/

 




On 10.18.2010 19:26 , "Jean-Marc Valin" <jea= n-marc.valin@octasic.com>= ; wrote:

Opus i= s available under the BSD license. As for patents, Skype has pledged to make them available royalty-free once the codec is accepted as an IETF standard.

    Jean-Marc

-----Original Message-----
From: codec-bounces@ie= tf.org on behalf of Pascal Pochol
Sent: Mon 10/18/2010 10:19 PM
To: codec@ietf.org
Subject: [codec] Opus codec licensing

Hello,

we've been using speex and celt for voice and music but today I heard about=
Opus which sounds like a fantastic replacement for both these codecs.

We're wondering if Opus will be released under the same type of license as<= br> speex and celt? SILK's license forbid its use in commercial software and we're worried that it might carry over to Opus. If not, as soon as Opus
handles lower bitrates, stereo and fixed point decoding we'll be using it.<= br>
-Pascal

_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec

_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec


_______________________________________________
codec mailing listcodec@ietf.org
https://www.ietf.org/mailman/listinfo/codec
------=_Part_938362_2023225065.1287516459554-- From koen.vos@skype.net Tue Oct 19 12:29:46 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 79A243A67AB for ; Tue, 19 Oct 2010 12:29:46 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.306 X-Spam-Level: X-Spam-Status: No, score=-2.306 tagged_above=-999 required=5 tests=[AWL=0.292, BAYES_00=-2.599, HTML_MESSAGE=0.001] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wZnEb1YjpgrH for ; Tue, 19 Oct 2010 12:29:44 -0700 (PDT) Received: from mx.skype.net (mx.skype.net [78.141.177.88]) by core3.amsl.com (Postfix) with ESMTP id 003833A6929 for ; Tue, 19 Oct 2010 12:29:43 -0700 (PDT) Received: from mx.skype.net (localhost [127.0.0.1]) by mx.skype.net (Postfix) with ESMTP id E67B21702; Tue, 19 Oct 2010 21:31:14 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=skype.net; h=date:from:to :cc:message-id:in-reply-to:subject:mime-version:content-type; s= mx; bh=if36IoaeqxtLc9XmJrCwaM8g3Kk=; b=vArzFpKhUeB6JW667Zzh5mUKN T8jwVeZROmxCQNw159F20xHKBS5r0L8ruJTFDDonnP8AFfDY9+k3wc0/fgr926Mh EoP1FWGHzL8kAEN0mY5l9R3XX9IiqVzGuODbNGdMOILP/tP8FptWivROokEIRPab OqM5n1SnFCIUr/3xgA= DomainKey-Signature: a=rsa-sha1; c=nofws; d=skype.net; h=date:from:to:cc :message-id:in-reply-to:subject:mime-version:content-type; q=dns ; s=mx; b=FZTcbT52c/wrCgqKdLjgoSpRvkqwSs4JkBffLPdso9cznMue0NL9TE +W1T747bYyjBzxMEol1SnDUrb6arwXt3aNQw8lbst9WokQccWHh/2IQv8T7+7siO EVfYjT2q6naQaAUg+zwdIa0u4Va8YQrKA2ZTAGBSU+dWDHoVqVAkA= Received: from zimbra.skype.net (zimbra.skype.net [78.141.177.82]) by mx.skype.net (Postfix) with ESMTP id E4C6ACF; Tue, 19 Oct 2010 21:31:14 +0200 (CEST) Received: from localhost (localhost [127.0.0.1]) by zimbra.skype.net (Postfix) with ESMTP id C481C1672687; Tue, 19 Oct 2010 21:31:14 +0200 (CEST) X-Virus-Scanned: amavisd-new at lu2-zimbra.skype.net Received: from zimbra.skype.net ([127.0.0.1]) by localhost (zimbra.skype.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id W-F-j+gFV9e5; Tue, 19 Oct 2010 21:31:13 +0200 (CEST) Received: from zimbra.skype.net (lu2-zimbra.skype.net [78.141.177.82]) by zimbra.skype.net (Postfix) with ESMTP id 74C121672684; Tue, 19 Oct 2010 21:31:13 +0200 (CEST) Date: Tue, 19 Oct 2010 21:31:13 +0200 (CEST) From: Koen Vos To: Christian Hoene Message-ID: <1559386036.938403.1287516673396.JavaMail.root@lu2-zimbra> In-Reply-To: <1389617873.938373.1287516517422.JavaMail.root@lu2-zimbra> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_938402_1267730389.1287516673393" X-Originating-IP: [69.181.192.115] X-Mailer: Zimbra 6.0.6_GA_2330.UBUNTU8_64 (ZimbraWebClient - FF3.0 (Win)/6.0.6_GA_2330.UBUNTU8_64) Cc: codec@ietf.org Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 19:29:46 -0000 ------=_Part_938402_1267730389.1287516673393 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Sorry, please ignore previous email - bad keyboard shortcut...=20 >> Furthermore, any implementation of the codec that accidentally does not = confirm to the standard yet will require a license fee.=20 If you modify the codec so much that it's no longer compatible to the stand= ard, then that modified codec is no longer "Opus". In that case you've crea= ted a new codec and I think it's only reasonable that the patent license do= esn't apply.=20 best,=20 koen.=20 From: "Koen Vos" =20 To: "Christian Hoene" =20 Cc: "Stephan Wenger" , codec@ietf.org=20 Sent: Tuesday, October 19, 2010 12:27:39 PM=20 Subject: Re: [codec] Opus codec licensing=20 If you modify the codec so much that it can't=20 From: "Christian Hoene" =20 To: "Stephan Wenger" , codec@ietf.org=20 Sent: Tuesday, October 19, 2010 11:12:48 AM=20 Subject: Re: [codec] Opus codec licensing=20 Re: [codec] Opus codec licensing=20 Hi=20 From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of S= tephan Wenger=20 Sent: Tuesday, October 19, 2010 5:21 PM=20 To: Jean-Marc Valin; Pochol@WebfootGames.com; codec@ietf.org=20 Subject: Re: [codec] Opus codec licensing=20 Hi,=20 With =E2=80=9Chat=E2=80=9D:=20 It is, in almost all cases, unwise for a third party to circumscribe=E2=80= =94or worse: abbreviate=E2=80=94licensing terms provided by a rightholder. = Skype=E2=80=99s licensing declaration can be found at https://datatracker.i= etf.org/ipr/1297 . I note that the statement contains conditions beyond the= acceptance of their technology into an IETF standard.=20 Stephan=20 [Christian Hoene] +1=20 It is planned to test the new codec by spreading it to users prior standard= ization. However, under the terms of this licensing declaration, this is no= t possible for anybody but Skype because it would require a license from Sk= ype.=20 Also, any enhancement of the codec, which does not conform to the original = standard anymore, will require a license.=20 Furthermore, any implementation of the codec that accidentally does not con= firm to the standard yet will require a license fee.=20 Clearly, these terms are unacceptable. Please, Koen, talk to your lawyer ag= ain to weaken this license statements. This shall include at least a draft = versions of the standard. Also, updates of the standard shall not require a= ny new license anymore.=20 It is patent really essential for the opus codec?=20 =E2=80=9C A method of estimating noise in data containing voice information= and noise includes receiving the data as a sequence of input values; trans= forming the data by applying a first non linear mapping to the input values= wherein the derivative function of the mapping decreases in magnitude as t= he input values increase in magnitude smoothing the transformed data; and t= ransforming the smoothed transformed data by applying a second non linear m= apping that is opposite to the first non linear mapping, to determine an es= timate of the noise in the inputted data.=E2=80=9D=20 Anyhow, it should not be too difficult to circumvent the patent. Any ideas?= =20 Christian=20 ---------------------------------------------------------------=20 Dr.-Ing. Christian Hoene=20 Interactive Communication Systems (ICS), University of T=C3=BCbingen=20 Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532=20 http://www.net.uni-tuebingen.de/=20 On 10.18.2010 19:26 , "Jean-Marc Valin" < jean-marc.valin@octasic.com > wro= te:=20 Opus is available under the BSD license. As for patents, Skype has pledged = to make them available royalty-free once the codec is accepted as an IETF s= tandard.=20 Jean-Marc=20 -----Original Message-----=20 From: codec-bounces@ietf.org on behalf of Pascal Pochol=20 Sent: Mon 10/18/2010 10:19 PM=20 To: codec@ietf.org=20 Subject: [codec] Opus codec licensing=20 Hello,=20 we've been using speex and celt for voice and music but today I heard about= =20 Opus which sounds like a fantastic replacement for both these codecs.=20 We're wondering if Opus will be released under the same type of license as= =20 speex and celt? SILK's license forbid its use in commercial software and=20 we're worried that it might carry over to Opus. If not, as soon as Opus=20 handles lower bitrates, stereo and fixed point decoding we'll be using it.= =20 -Pascal=20 _______________________________________________=20 codec mailing list=20 codec@ietf.org=20 https://www.ietf.org/mailman/listinfo/codec=20 _______________________________________________=20 codec mailing list=20 codec@ietf.org=20 https://www.ietf.org/mailman/listinfo/codec=20 _______________________________________________=20 codec mailing list=20 codec@ietf.org=20 https://www.ietf.org/mailman/listinfo/codec=20 ------=_Part_938402_1267730389.1287516673393 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable <= div style=3D'font-family: Times New Roman; font-size: 12pt; color: #000000'= >Sorry, please ignore previous email - bad keyboard shortcut...

>> Furthermore, an= y implementation of the codec that accidentally does not confirm to the standard yet will require a license fee.
=
If you modify the codec so much that it's no longer compatible to the s= tandard, then that modified codec is no longer "Opus".  In that case y= ou've created a new codec and I think it's only reasonable that the patent = license doesn't apply.

best,
koen.


From: "K= oen Vos" <koen.vos@skype.net>
To: "Christian Hoene" <hoe= ne@uni-tuebingen.de>
Cc: "Stephan Wenger" <stewe@stewe.org&= gt;, codec@ietf.org
Sent: Tuesday, October 19, 2010 12:27:39 PMSubject: Re: [codec] Opus codec licensing

If you modify the codec so much that it can't

From: "Christian Hoene" <hoene@uni-tuebingen.de>
<= b>To: "Stephan Wenger" <stewe@stewe.org>, codec@ietf.org
Se= nt: Tuesday, October 19, 2010 11:12:48 AM
Subject: Re: [codec= ] Opus codec licensing

Re: [codec] Opus codec licensing

Hi

 

From: code= c-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Stephan Wenger
Sent: Tuesday, October 19, 2010 5:21 PM
To: Jean-Marc Valin; Pochol@WebfootGames.com; codec@ietf.org
Subject: Re: [codec] Opus codec licensing

 

Hi, With =E2=80=9Chat=E2=80=9D:
It is, in almost all cases, unwise for a third party to circumscribe=E2=80= =94or worse: abbreviate=E2=80=94licensing terms provided by a rightholder.  Skype= =E2=80=99s licensing declaration can be found at https://datatracker.ietf.org/ipr/1297.  I note that the statement contains conditions beyond the acceptance o= f their technology into an IETF standard.
Stephan

[Christian Hoene] +1

 

It is planned to test the new codec by spreading it to users prior standardization. However, under the terms of this licensing declarati= on, this is not possible for anybody but Skype because it would require a licen= se from Skype.

 

Also, any enhancement of the codec, which does not conform to the original standard anymore, will require a license.

 

Furthermore, any implementation of the codec that accidentally does not confirm to the standard yet will require a license fee.

 

Clearly, these terms are unacceptable. Please, Koen, talk to your lawyer again to weaken this license statements. This shall include at least a draft versions of the standard. Also, updates of the standard shall= not require any new license anymore.

 

It is patent really essential for the opus codec?

=E2=80=9CA method of estimating noise in data containing voice information and noise includes receiving the data as a sequence of input values; transforming the data by applying a first non linear mapping to the input values wherein the derivat= ive function of the mapping decreases in magnitude as the input values increase= in magnitude smoothing the transformed data; and transforming the smoothed transformed data by applying a second non linear mapping that is opposite t= o the first non linear mapping, to determine an estimate of the noise in the inputted data.=E2=80=9D

Anyhow, it should not be too difficult to circumvent the patent. Any ideas?

 

Christian

 

-----------------------------= ----------------------------------

Dr.-Ing. Christian Hoene

Interactive Communication Sys= tems (ICS), University of T=C3=BCbingen

Sand 13, 72076 T=C3=BCbingen,= Germany, Phone +49 7071 2970532

http://www.net.uni-tuebingen.de/

 




On 10.18.2010 19:26 , "Jean-Marc Valin" <jea= n-marc.valin@octasic.com>= ; wrote:

Opus i= s available under the BSD license. As for patents, Skype has pledged to make them available royalty-free once the codec is accepted as an IETF standard.

    Jean-Marc

-----Original Message-----
From: codec-bounces@ie= tf.org on behalf of Pascal Pochol
Sent: Mon 10/18/2010 10:19 PM
To: codec@ietf.org
Subject: [codec] Opus codec licensing

Hello,

we've been using speex and celt for voice and music but today I heard about=
Opus which sounds like a fantastic replacement for both these codecs.

We're wondering if Opus will be released under the same type of license as<= br> speex and celt? SILK's license forbid its use in commercial software and we're worried that it might carry over to Opus. If not, as soon as Opus
handles lower bitrates, stereo and fixed point decoding we'll be using it.<= br>
-Pascal

_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec

_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec


_______________________________________________
codec mailing listcodec@ietf.org
https://www.ietf.org/mailman/listinfo/codec
------=_Part_938402_1267730389.1287516673393-- From stewe@stewe.org Tue Oct 19 13:09:30 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 2E14B3A692C for ; Tue, 19 Oct 2010 13:09:30 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.123 X-Spam-Level: X-Spam-Status: No, score=-1.123 tagged_above=-999 required=5 tests=[AWL=0.079, BAYES_00=-2.599, HTML_MESSAGE=0.001, MIME_QP_LONG_LINE=1.396] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id UlVmHRC7HgJ0 for ; Tue, 19 Oct 2010 13:09:19 -0700 (PDT) Received: from stewe.org (stewe.org [85.214.122.234]) by core3.amsl.com (Postfix) with ESMTP id 025583A6958 for ; Tue, 19 Oct 2010 13:09:18 -0700 (PDT) Received: from [192.168.1.108] (unverified [24.5.132.232]) by stewe.org (SurgeMail 3.9e) with ESMTP id 819448-1743317 for multiple; Tue, 19 Oct 2010 22:10:48 +0200 User-Agent: Microsoft-Entourage/12.27.0.100910 Date: Tue, 19 Oct 2010 13:10:38 -0700 From: Stephan Wenger To: Roman Shpount , Christian Hoene Message-ID: Thread-Topic: [codec] Opus codec licensing Thread-Index: ActvybJCgo2LcMQa10eod42K45G0lw== In-Reply-To: Mime-version: 1.0 Content-type: multipart/alternative; boundary="B_3370338647_909741" X-Originating-IP: 24.5.132.232 X-Authenticated-User: stewe@stewe.org X-ORBS-Stamp: Your IP (24.5.132.232) was found in the spamhaus database. http://www.spamhaus.net Cc: codec@ietf.org Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 20:09:30 -0000 > This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. --B_3370338647_909741 Content-type: text/plain; charset="ISO-8859-1" Content-transfer-encoding: quoted-printable Hi, On 10.19.2010 12:00 , "Roman Shpount" wrote: > People get touchy here about discussing the actual patent claims, [...] >=20 Indeed. I very strongly suggest to conduct patent related work elsewhere, and only among those interested and willing. You are exposing participants to claim language which can trigger all kinds of evil to those participatin= g and, arguably, even those only listening in. I can=B9t prevent you from doin= g so, but please keep the risks in mind when discussing claim language here: 1) possible willful infringement allegations, 2) potentially loosing participation because of input from legal departments due to the previous point, 3) coming to questionable, if not outright wrong, conclusions due to lack of legal experience, 4) changing a design based on those conclusions with or without removing the encumbrance. Stephan =20 >=20 > 2010/10/19 Christian Hoene >> Hi >> =A0 >>=20 >> From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf O= f >> Stephan Wenger >> Sent: Tuesday, October 19, 2010 5:21 PM >> To: Jean-Marc Valin; Pochol@WebfootGames.com; codec@ietf.org >> Subject: Re: [codec] Opus codec licensing >>=20 >> =A0 >> Hi, >> With =B3hat=B2: >> It is, in almost all cases, unwise for a third party to circumscribe=8Bor >> worse: abbreviate=8Blicensing terms provided by a rightholder. =A0Skype=B9s >> licensing declaration can be found at https://datatracker.ietf.org/ipr/1= 297. >> =A0I note that the statement contains conditions beyond the acceptance of = their >> technology into an IETF standard. >> Stephan >>=20 >>=20 >> [Christian Hoene] +1 >> =A0 >> It is planned to test the new codec by spreading it to users prior >> standardization. However, under the terms of this licensing declaration,= this >> is not possible for anybody but Skype because it would require a license= from >> Skype. >> =A0 >> Also, any enhancement of the codec, which does not conform to the origin= al >> standard anymore, will require a license. >> =A0 >> Furthermore, any implementation of the codec that accidentally does not >> confirm to the standard yet will require a license fee. >> =A0 >> Clearly, these terms are unacceptable. Please, Koen, talk to your lawyer >> again to weaken this license statements. This shall include at least a d= raft >> versions of the standard. Also, updates of the standard shall not requir= e any >> new license anymore. >> =A0 >> It is patent really essential for the opus codec? >> =B3A method of estimating noise in data containing voice information and n= oise >> includes receiving the data as a sequence of input values; transforming = the >> data by applying a first non linear mapping to the input values wherein = the >> derivative function of the mapping decreases in magnitude as the input v= alues >> increase in magnitude smoothing the transformed data; and transforming t= he >> smoothed transformed data by applying a second non linear mapping that i= s >> opposite to the first non linear mapping, to determine an estimate of th= e >> noise in the inputted data.=B2 >> Anyhow, it should not be too difficult to circumvent the patent. Any ide= as? >> =A0 >> Christian >> =A0 >> --------------------------------------------------------------- >> Dr.-Ing. Christian Hoene >> Interactive Communication Systems (ICS), University of T=FCbingen >> Sand 13, 72076 T=FCbingen, Germany, Phone +49 7071 2970532 >> http://www.net.uni-tuebingen.de/ >>=20 >> =A0 >>=20 >>=20 >>=20 >> On 10.18.2010 19:26 , "Jean-Marc Valin" > > wrote: >> Opus is available under the BSD license. As for patents, Skype has pledg= ed to >> make them available royalty-free once the codec is accepted as an IETF >> standard. >>=20 >> =A0=A0=A0=A0Jean-Marc >>=20 >> -----Original Message----- >> From: codec-bounces@ietf.org on behalf = of >> Pascal Pochol >> Sent: Mon 10/18/2010 10:19 PM >> To: codec@ietf.org >> Subject: [codec] Opus codec licensing >>=20 >> Hello, >>=20 >> we've been using speex and celt for voice and music but today I heard ab= out >> Opus which sounds like a fantastic replacement for both these codecs. >>=20 >> We're wondering if Opus will be released under the same type of license = as >> speex and celt? SILK's license forbid its use in commercial software and >> we're worried that it might carry over to Opus. If not, as soon as Opus >> handles lower bitrates, stereo and fixed point decoding we'll be using i= t. >>=20 >> -Pascal >>=20 >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >>=20 >>=20 >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >>=20 >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >>=20 >>=20 >>=20 --B_3370338647_909741 Content-type: text/html; charset="ISO-8859-1" Content-transfer-encoding: quoted-printable Re: [codec] Opus codec licensing Hi,

On 10.19.2010 12:00 , "Roman Shpount" <roman@telurix.com> wrote:

<= SPAN STYLE=3D'font-size:11pt'>People get touchy here about discussing the actu= al patent claims, [...]

= Indeed.  I very strongly suggest to conduc= t patent related work elsewhere, and only among those interested and willing= .  You are exposing participants to claim language which can trigger al= l kinds of evil to those participating and, arguably, even those only listen= ing in.  I can’t prevent you from doing so, but please keep the r= isks in mind when discussing claim language here: 1) possible willful infrin= gement allegations, 2) potentially loosing participation because of input fr= om legal departments due to the previous point, 3) coming to questionable, i= f not outright wrong, conclusions due to lack of legal experience, 4) changi= ng a design based on those conclusions with or without removing the encumbra= nce.

Stephan

 
<= SPAN STYLE=3D'font-size:11pt'>
2010/10/19 Christian Hoene <hoene@uni-t= uebingen.de>
<= SPAN STYLE=3D'font-size:11pt'>Hi
=A0

From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Stephan= Wenger
Sent: Tuesday, October 19, 2010 5:21 PM
To: Jean-Marc Valin; Pochol@Webfoo= tGames.com; codec@ietf.org
Subject: Re: [codec] Opus codec licensing

=A0
Hi,
With “hat”:
It is, in almost all cases, unwise for a third party to circumscribe—= or worse: abbreviate—licensing terms provided by a rightholder. =A0Skype= ’s licensing declaration can be found at https://datatracker.ietf.org/ipr/1297. =A0I note that t= he statement contains conditions beyond the acceptance of their technology i= nto an IETF standard.
Stephan


[Christian Hoene] +1
=A0
It is planned to test the new codec by spreading it to users prior standard= ization. However, under the terms of this licensing declaration, this is not= possible for anybody but Skype because it would require a license from Skyp= e.
=A0
Also, any enhancement of the codec, which does not conform to the original = standard anymore, will require a license.
=A0
Furthermore, any implementation of the codec that accidentally does not con= firm to the standard yet will require a license fee.
=A0
Clearly, these terms are unacceptable. Please, Koen, talk to your lawyer ag= ain to weaken this license statements. This shall include at least a draft v= ersions of the standard. Also, updates of the standard shall not require any= new license anymore.
=A0
It is patent really essential for the opus codec?
“A method of estima= ting noise in data containing voice information and noise includes receiving= the data as a sequence of input values; transforming the data by applying a= first non linear mapping to the input values wherein the derivative functio= n of the mapping decreases in magnitude as the input values increase in magn= itude smoothing the transformed data; and transforming the smoothed transfor= med data by applying a second non linear mapping that is opposite to the fir= st non linear mapping, to determine an estimate of the noise in the inputted= data.”
Anyhow, it should not be too dif= ficult to circumvent the patent. Any ideas?
=A0
Christian
=A0
-------------------------------------------------= --------------
Dr.-Ing. Christian Hoene
Interactive Communication Systems (ICS), University of Tübingen
Sand 13, 72076 Tübingen, Germany, Phone +49 7071 2970532
http://www.net.uni-tuebingen.de/=

=A0



On 10.18.2010 19:26 , "Jean-Marc Valin" <jean-marc.valin@octasic.com <http://jean-marc.valin@octasic.com> > wrote:=
Opus is available under the BSD license. As for patents, Skype has pledged = to make them available royalty-free once the codec is accepted as an IETF st= andard.

=A0=A0=A0=A0Jean-Marc

-----Original Message-----
From: codec-bounces@ietf.org <http://codec-bounces@ietf.org> &nb= sp;on behalf of Pascal Pochol
Sent: Mon 10/18/2010 10:19 PM
To: codec@ietf.org <http://codec@ietf.org>
Subject: [codec] Opus codec licensing

Hello,

we've been using speex and celt for voice and music but today I heard about=
Opus which sounds like a fantastic replacement for both these codecs.

We're wondering if Opus will be released under the same type of license as<= BR> speex and celt? SILK's license forbid its use in commercial software and we're worried that it might carry over to Opus. If not, as soon as Opus
handles lower bitrates, stereo and fixed point decoding we'll be using it.<= BR>
-Pascal

_______________________________________________
codec mailing list
codec@ietf.org <http://codec@ietf.org>
https://www.ietf.org/= mailman/listinfo/codec

_______________________________________________
codec mailing list
codec@ietf.org <http://codec@ietf.org>
https://www.ietf.org/= mailman/listinfo/codec

_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/= mailman/listinfo/codec

=

--B_3370338647_909741-- From stewe@stewe.org Tue Oct 19 13:12:53 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 3F2D73A6956 for ; Tue, 19 Oct 2010 13:12:53 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.126 X-Spam-Level: X-Spam-Status: No, score=-1.126 tagged_above=-999 required=5 tests=[AWL=0.076, BAYES_00=-2.599, HTML_MESSAGE=0.001, MIME_QP_LONG_LINE=1.396] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3HkphB3a1L-z for ; Tue, 19 Oct 2010 13:12:47 -0700 (PDT) Received: from stewe.org (stewe.org [85.214.122.234]) by core3.amsl.com (Postfix) with ESMTP id 4D7D13A6958 for ; Tue, 19 Oct 2010 13:12:47 -0700 (PDT) Received: from [192.168.1.108] (unverified [24.5.132.232]) by stewe.org (SurgeMail 3.9e) with ESMTP id 819450-1743317 for multiple; Tue, 19 Oct 2010 22:14:07 +0200 User-Agent: Microsoft-Entourage/12.27.0.100910 Date: Tue, 19 Oct 2010 13:14:00 -0700 From: Stephan Wenger To: Christian Hoene , Message-ID: Thread-Topic: [codec] Opus codec licensing Thread-Index: ActvNCcg/6UdLTfYSPe62Egnu0SU9wAAN+7tABsLZFoABEUx8AAF+F/5 In-Reply-To: <000c01cb6fb9$41b95910$c52c0b30$@de> Mime-version: 1.0 Content-type: multipart/alternative; boundary="B_3370338846_968376" X-Originating-IP: 24.5.132.232 X-Authenticated-User: stewe@stewe.org X-ORBS-Stamp: Your IP (24.5.132.232) was found in the spamhaus database. http://www.spamhaus.net Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Oct 2010 20:12:53 -0000 > This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. --B_3370338846_968376 Content-type: text/plain; charset="ISO-8859-1" Content-transfer-encoding: quoted-printable Hi Christian, On 10.19.2010 11:12 , "Christian Hoene" wrote: > [...]=20 > Clearly, these terms are unacceptable. > [...] > =20 Let me suggest that you (and everyone else) mark their opinion on licensing terms as your personal opinion. What=B9s acceptable or inacceptable to you may very well be acceptable or inacceptable to me. Same goes with compliance to the charter. Stephan >=20 >=20 > On 10.18.2010 19:26 , "Jean-Marc Valin" > wrote: > Opus is available under the BSD license. As for patents, Skype has pledge= d to > make them available royalty-free once the codec is accepted as an IETF > standard. >=20 > Jean-Marc >=20 > -----Original Message----- > From: codec-bounces@ietf.org on behalf of Pascal Pochol > Sent: Mon 10/18/2010 10:19 PM > To: codec@ietf.org > Subject: [codec] Opus codec licensing >=20 > Hello, >=20 > we've been using speex and celt for voice and music but today I heard abo= ut > Opus which sounds like a fantastic replacement for both these codecs. >=20 > We're wondering if Opus will be released under the same type of license a= s > speex and celt? SILK's license forbid its use in commercial software and > we're worried that it might carry over to Opus. If not, as soon as Opus > handles lower bitrates, stereo and fixed point decoding we'll be using it= . >=20 > -Pascal >=20 > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec >=20 >=20 > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec >=20 --B_3370338846_968376 Content-type: text/html; charset="ISO-8859-1" Content-transfer-encoding: quoted-printable Re: [codec] Opus codec licensing Hi Christian,

On 10.19.2010 11:12 , "Christian Hoene" <hoene@uni-tuebingen.de> wrote:

<= SPAN STYLE=3D'font-size:11pt'>[...]
Clearly, these terms are unacceptable.
[...]

Let me suggest that you (and everyone else) mark th= eir opinion on licensing terms as your personal opinion.  What’s = acceptable or inacceptable to you may very well be acceptable or inacceptabl= e to me.

Same goes with compliance to the charter.

Stephan

<= SPAN STYLE=3D'font-size:11pt'>

On 10.18.2010 19:26 , "Jean-Marc Valin" <jean-marc.valin@octasic.com <jean-marc.valin@octasic.com> > wrote:
Opus is available under the BSD license. As for patents, Skype has pledged = to make them available royalty-free once the codec is accepted as an IETF st= andard.

    Jean-Marc

-----Original Message-----
From: codec-bounces@ietf.org on behalf= of Pascal Pochol
Sent: Mon 10/18/2010 10:19 PM
To: codec@ietf.org
Subject: [codec] Opus codec licensing

Hello,

we've been using speex and celt for voice and music but today I heard about=
Opus which sounds like a fantastic replacement for both these codecs.

We're wondering if Opus will be released under the same type of license as<= BR> speex and celt? SILK's license forbid its use in commercial software and we're worried that it might carry over to Opus. If not, as soon as Opus
handles lower bitrates, stereo and fixed point decoding we'll be using it.<= BR>
-Pascal

_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/= mailman/listinfo/codec

_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/= mailman/listinfo/codec

--B_3370338846_968376-- From jdrosen@jdrosen.net Tue Oct 19 17:41:40 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id C616B3A697D for ; Tue, 19 Oct 2010 17:41:40 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.279 X-Spam-Level: X-Spam-Status: No, score=-102.279 tagged_above=-999 required=5 tests=[AWL=0.320, BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9ttGVchrJzkg for ; Tue, 19 Oct 2010 17:41:38 -0700 (PDT) Received: from ecbiz71.inmotionhosting.com (ecbiz71.inmotionhosting.com [173.205.126.243]) by core3.amsl.com (Postfix) with ESMTP id A62333A68E3 for ; Tue, 19 Oct 2010 17:41:38 -0700 (PDT) Received: from [98.109.139.97] (helo=[192.168.1.8]) by ecbiz71.inmotionhosting.com with esmtpsa (TLSv1:AES256-SHA:256) (Exim 4.69) (envelope-from ) id 1P8Mln-0004v9-2N for codec@ietf.org; Tue, 19 Oct 2010 20:43:07 -0400 Message-ID: <4CBE3B1D.8080105@jdrosen.net> Date: Tue, 19 Oct 2010 20:43:09 -0400 From: Jonathan Rosenberg User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.2.9) Gecko/20100915 Thunderbird/3.1.4 MIME-Version: 1.0 To: codec@ietf.org References: <4CBDB9C7.1020206@octasic.com> <005b01cb6faf$7867a750$6936f5f0$@de> <4CBDD672.2090202@octasic.com> In-Reply-To: <4CBDD672.2090202@octasic.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - ecbiz71.inmotionhosting.com X-AntiAbuse: Original Domain - ietf.org X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - jdrosen.net Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 00:41:40 -0000 IETF has a long history of standards with names - ICE, SIP, POP, IMAP, etc. These are also acronyms, but increasingly the acronyms are highly contrived and are basically names (DECADE, ENUM, SALUD) and in some cases just names (Sieve). However, IETF has not ever filed for trademarks on any of these, and I see absolutely no reason to start now. -Jonathan R. On 10/19/2010 1:33 PM, Jean-Marc Valin wrote: > Hi Christian, > > I don't know the details of the ITU-T trademark issues, but in this > case, the Opus name will not be controlled by any company (I'm not aware > of anyone registering a trademark on the "Opus" name in this context). > I'm not against the IETF owning the trademark, but I'm not sure any > ownership is necessary to begin with. > > Cheers, > > Jean-Marc > > As a note, my company does use the term "Opus" in a different domain of > application (digital signal processor architecture) and sees no issue > with the IETF using it for an audio codec. Oh, and no, I'm not the one > who originally suggested that name. > > On 10-10-19 01:02 PM, Christian Hoene wrote: >> Hi, >> >> the ITU had some problems with trademarks that were given to >> standardized algorithms. >> As far as I remember, PESQ or PEAQ are trademarked by some companies. >> Since then, the ITU-T name their standards only by numbers. >> >> I do not have any problems with nicknames. However, I think the >> trademark issue shall be addressed and it is important. I would prefer >> that such as trademark is owned by the IETF if possible. >> >> Uspto.gov does not list an opus codec. But I do not know how fast they >> (or the readers of this mailing list) are... >> >> With best regards, >> >> Christian >> >> --------------------------------------------------------------- >> Dr.-Ing. Christian Hoene >> Interactive Communication Systems (ICS), University of Tübingen >> Sand 13, 72076 Tübingen, Germany, Phone +49 7071 2970532 >> http://www.net.uni-tuebingen.de/ >> >> >> -----Original Message----- >> From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf >> Of Stephan Wenger >> Sent: Tuesday, October 19, 2010 6:05 PM >> To: codec@ietf.org >> Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt >> >> Hi Jean-Marc, >> >> Four things: >> >> First, indeed, there was a hum at the last meeting. Weak memory on my >> side. >> Apologies. >> >> Second, hums need to be reconfirmed on the mailing list, and that has not >> happened according to my read of the email archive. >> >> Third, I also missed the submission of draft-ietf-codec-description-00, >> which was really the time I should have complained. >> >> So I'm willing to assume (as apparently have the chairs, see point #2) >> that >> there has been an implied consensus of the WG to accept the >> codec-description draft. Which brings me to point #4: >> >> As this is now a WG item, any major change requires WG consensus. >> Selecting >> a marketing name, IMO, is such a major change. "Opus" is such a flashy >> name >> that certain participants and/or companies conceivably may not like >> it. For >> example, if I were working for a company that has in its portfolio an >> audio >> product named "Opus", I would object to the name change. So the thing you >> should have done, IMO, is to send an email to the list saying "The >> editors >> consider changing the name of our codec to Opus. Is that acceptable to >> the >> WG?". >> >> It appears to me that twice you guys (chairs included) have taken >> shortcuts >> with the IETF's procedures, as I understand them. That, IMO, fills up >> your >> quota for the next couple of years. Please be more conservative from now >> on. >> >> Regards, >> Stephan >> >> >> >> >> On 10.19.2010 08:31 , "Jean-Marc Valin" >> wrote: >> >>> Hi, >>> >>> draft-ietf-codec-opus-00.txt is indeed a WG item. It has had several >>> names >>> in the past, including draft-valin-codec-prototype and >>> draft-valin-codec-definition, which may explain the confusion. This >>> is the >>> draft for which there was a hum during the last meeting. >>> >>> Jean-Marc >>> >>> On 10-10-19 11:26 AM, Stephan Wenger wrote: >>>> Hi, >>>> >>>> I want to inquire the status of this draft. In many working groups, the >>>> filename "draft-ietf--xxx" indicates that the draft in >>>> question is >>>> a WG item of WG. Following this logic, it would appear that >>>> the >>>> "opus" draft is now a WG item of the codec WG. I don't recall a >>>> decision to >>>> than extent. >>>> >>>> If the draft were indeed accepted as a WG item, I would like to >>>> encourage >>>> those who made IPR statements related to it, to resubmit those >>>> statements >>>> with the new filename. This would help those of us who are searching >>>> through the IETF IPR tracker by WG name (which is a very common >>>> thing to do, >>>> at least for me). >>>> >>>> Thanks, >>>> Stephan >>>> >>>> >>>> On 10.15.2010 13:30 , >>>> "Internet-Drafts@ietf.org" >>>> wrote: >>>> >>>>> A New Internet-Draft is available from the on-line Internet-Drafts >>>>> directories. >>>>> This draft is a work item of the Internet Wideband Audio Codec >>>>> Working Group >>>>> of the IETF. >>>>> >>>>> >>>>> Title : Definition of the Opus Audio Codec >>>>> Author(s) : J. Valin, K. Vos >>>>> Filename : draft-ietf-codec-opus-00.txt >>>>> Pages : 12 >>>>> Date : 2010-10-15 >>>>> >>>>> This document describes the Opus codec, designed for interactive >>>>> speech and audio transmission over the Internet. >>>>> >>>>> A URL for this Internet-Draft is: >>>>> http://www.ietf.org/internet-drafts/draft-ietf-codec-opus-00.txt >>>>> >>>>> Internet-Drafts are also available by anonymous FTP at: >>>>> ftp://ftp.ietf.org/internet-drafts/ >>>>> >>>>> Below is the data which will enable a MIME compliant mail reader >>>>> implementation to automatically retrieve the ASCII version of the >>>>> Internet-Draft. >>>>> _______________________________________________ >>>>> codec mailing list >>>>> codec@ietf.org >>>>> https://www.ietf.org/mailman/listinfo/codec >>>> >>>> >>>> _______________________________________________ >>>> codec mailing list >>>> codec@ietf.org >>>> https://www.ietf.org/mailman/listinfo/codec >>> >> >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec -- Jonathan D. Rosenberg, Ph.D. SkypeID: jdrosen Chief Technology Strategist Mobile: +1 (732) 766-2496 Skype SkypeIn: +1 (408) 465-0361 jdrosen@skype.net http://www.skype.com jdrosen@jdrosen.net http://www.jdrosen.net From jean-marc.valin@usherbrooke.ca Tue Oct 19 18:23:48 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id E5CEB3A697B for ; Tue, 19 Oct 2010 18:23:48 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.223 X-Spam-Level: X-Spam-Status: No, score=-2.223 tagged_above=-999 required=5 tests=[AWL=0.376, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id cjymjrvl3C2U for ; Tue, 19 Oct 2010 18:23:47 -0700 (PDT) Received: from relais.videotron.ca (relais.videotron.ca [24.201.245.36]) by core3.amsl.com (Postfix) with ESMTP id 5E02E3A6976 for ; Tue, 19 Oct 2010 18:23:47 -0700 (PDT) MIME-version: 1.0 Content-transfer-encoding: 8BIT Content-type: text/plain; charset=ISO-8859-1 Received: from [192.168.1.14] ([70.81.109.112]) by VL-MR-MRZ20.ip.videotron.ca (Sun Java(tm) System Messaging Server 6.3-8.01 (built Dec 16 2008; 32bit)) with ESMTP id <0LAK00167EM1DQ00@VL-MR-MRZ20.ip.videotron.ca> for codec@ietf.org; Tue, 19 Oct 2010 21:25:14 -0400 (EDT) Message-id: <4CBE4508.8040000@usherbrooke.ca> Date: Tue, 19 Oct 2010 21:25:28 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.12) Gecko/20100915 Thunderbird/3.0.8 To: Jonathan Rosenberg References: <4CBDB9C7.1020206@octasic.com> <005b01cb6faf$7867a750$6936f5f0$@de> <4CBDD672.2090202@octasic.com> <4CBE3B1D.8080105@jdrosen.net> In-reply-to: <4CBE3B1D.8080105@jdrosen.net> X-Enigmail-Version: 1.0.1 Cc: codec@ietf.org Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 01:23:49 -0000 OK great! I didn't see any reason either, but was leaving the door open in case someone insisted. Glad that's not the case :-) Jean-Marc On 10-10-19 08:43 PM, Jonathan Rosenberg wrote: > IETF has a long history of standards with names - ICE, SIP, POP, IMAP, > etc. These are also acronyms, but increasingly the acronyms are highly > contrived and are basically names (DECADE, ENUM, SALUD) and in some > cases just names (Sieve). However, IETF has not ever filed for > trademarks on any of these, and I see absolutely no reason to start now. > > -Jonathan R. > > > > On 10/19/2010 1:33 PM, Jean-Marc Valin wrote: >> Hi Christian, >> >> I don't know the details of the ITU-T trademark issues, but in this >> case, the Opus name will not be controlled by any company (I'm not aware >> of anyone registering a trademark on the "Opus" name in this context). >> I'm not against the IETF owning the trademark, but I'm not sure any >> ownership is necessary to begin with. >> >> Cheers, >> >> Jean-Marc >> >> As a note, my company does use the term "Opus" in a different domain of >> application (digital signal processor architecture) and sees no issue >> with the IETF using it for an audio codec. Oh, and no, I'm not the one >> who originally suggested that name. >> >> On 10-10-19 01:02 PM, Christian Hoene wrote: >>> Hi, >>> >>> the ITU had some problems with trademarks that were given to >>> standardized algorithms. >>> As far as I remember, PESQ or PEAQ are trademarked by some companies. >>> Since then, the ITU-T name their standards only by numbers. >>> >>> I do not have any problems with nicknames. However, I think the >>> trademark issue shall be addressed and it is important. I would prefer >>> that such as trademark is owned by the IETF if possible. >>> >>> Uspto.gov does not list an opus codec. But I do not know how fast they >>> (or the readers of this mailing list) are... >>> >>> With best regards, >>> >>> Christian >>> >>> --------------------------------------------------------------- >>> Dr.-Ing. Christian Hoene >>> Interactive Communication Systems (ICS), University of Tübingen >>> Sand 13, 72076 Tübingen, Germany, Phone +49 7071 2970532 >>> http://www.net.uni-tuebingen.de/ >>> >>> >>> -----Original Message----- >>> From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf >>> Of Stephan Wenger >>> Sent: Tuesday, October 19, 2010 6:05 PM >>> To: codec@ietf.org >>> Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt >>> >>> Hi Jean-Marc, >>> >>> Four things: >>> >>> First, indeed, there was a hum at the last meeting. Weak memory on my >>> side. >>> Apologies. >>> >>> Second, hums need to be reconfirmed on the mailing list, and that has >>> not >>> happened according to my read of the email archive. >>> >>> Third, I also missed the submission of draft-ietf-codec-description-00, >>> which was really the time I should have complained. >>> >>> So I'm willing to assume (as apparently have the chairs, see point #2) >>> that >>> there has been an implied consensus of the WG to accept the >>> codec-description draft. Which brings me to point #4: >>> >>> As this is now a WG item, any major change requires WG consensus. >>> Selecting >>> a marketing name, IMO, is such a major change. "Opus" is such a flashy >>> name >>> that certain participants and/or companies conceivably may not like >>> it. For >>> example, if I were working for a company that has in its portfolio an >>> audio >>> product named "Opus", I would object to the name change. So the thing >>> you >>> should have done, IMO, is to send an email to the list saying "The >>> editors >>> consider changing the name of our codec to Opus. Is that acceptable to >>> the >>> WG?". >>> >>> It appears to me that twice you guys (chairs included) have taken >>> shortcuts >>> with the IETF's procedures, as I understand them. That, IMO, fills up >>> your >>> quota for the next couple of years. Please be more conservative from now >>> on. >>> >>> Regards, >>> Stephan >>> >>> >>> >>> >>> On 10.19.2010 08:31 , "Jean-Marc Valin" >>> wrote: >>> >>>> Hi, >>>> >>>> draft-ietf-codec-opus-00.txt is indeed a WG item. It has had several >>>> names >>>> in the past, including draft-valin-codec-prototype and >>>> draft-valin-codec-definition, which may explain the confusion. This >>>> is the >>>> draft for which there was a hum during the last meeting. >>>> >>>> Jean-Marc >>>> >>>> On 10-10-19 11:26 AM, Stephan Wenger wrote: >>>>> Hi, >>>>> >>>>> I want to inquire the status of this draft. In many working groups, >>>>> the >>>>> filename "draft-ietf--xxx" indicates that the draft in >>>>> question is >>>>> a WG item of WG. Following this logic, it would appear that >>>>> the >>>>> "opus" draft is now a WG item of the codec WG. I don't recall a >>>>> decision to >>>>> than extent. >>>>> >>>>> If the draft were indeed accepted as a WG item, I would like to >>>>> encourage >>>>> those who made IPR statements related to it, to resubmit those >>>>> statements >>>>> with the new filename. This would help those of us who are searching >>>>> through the IETF IPR tracker by WG name (which is a very common >>>>> thing to do, >>>>> at least for me). >>>>> >>>>> Thanks, >>>>> Stephan >>>>> >>>>> >>>>> On 10.15.2010 13:30 , >>>>> "Internet-Drafts@ietf.org" >>>>> wrote: >>>>> >>>>>> A New Internet-Draft is available from the on-line Internet-Drafts >>>>>> directories. >>>>>> This draft is a work item of the Internet Wideband Audio Codec >>>>>> Working Group >>>>>> of the IETF. >>>>>> >>>>>> >>>>>> Title : Definition of the Opus Audio Codec >>>>>> Author(s) : J. Valin, K. Vos >>>>>> Filename : draft-ietf-codec-opus-00.txt >>>>>> Pages : 12 >>>>>> Date : 2010-10-15 >>>>>> >>>>>> This document describes the Opus codec, designed for interactive >>>>>> speech and audio transmission over the Internet. >>>>>> >>>>>> A URL for this Internet-Draft is: >>>>>> http://www.ietf.org/internet-drafts/draft-ietf-codec-opus-00.txt >>>>>> >>>>>> Internet-Drafts are also available by anonymous FTP at: >>>>>> ftp://ftp.ietf.org/internet-drafts/ >>>>>> >>>>>> Below is the data which will enable a MIME compliant mail reader >>>>>> implementation to automatically retrieve the ASCII version of the >>>>>> Internet-Draft. >>>>>> _______________________________________________ >>>>>> codec mailing list >>>>>> codec@ietf.org >>>>>> https://www.ietf.org/mailman/listinfo/codec >>>>> >>>>> >>>>> _______________________________________________ >>>>> codec mailing list >>>>> codec@ietf.org >>>>> https://www.ietf.org/mailman/listinfo/codec >>>> >>> >>> >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >>> >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec > From hoene@uni-tuebingen.de Tue Oct 19 23:01:22 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 1F8C73A69CA for ; Tue, 19 Oct 2010 23:01:22 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -5.602 X-Spam-Level: X-Spam-Status: No, score=-5.602 tagged_above=-999 required=5 tests=[AWL=0.647, BAYES_00=-2.599, HELO_EQ_DE=0.35, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id dO6t60HaleoR for ; Tue, 19 Oct 2010 23:01:20 -0700 (PDT) Received: from mx05.uni-tuebingen.de (mx05.uni-tuebingen.de [134.2.3.4]) by core3.amsl.com (Postfix) with ESMTP id D436F3A69CB for ; Tue, 19 Oct 2010 23:01:19 -0700 (PDT) Received: from hoeneT60 ([178.2.222.223]) (authenticated bits=0) by mx05.uni-tuebingen.de (8.13.6/8.13.6) with ESMTP id o9K62hau010835 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Wed, 20 Oct 2010 08:02:49 +0200 From: "Christian Hoene" To: "'Jonathan Rosenberg'" , References: <4CBDB9C7.1020206@octasic.com> <005b01cb6faf$7867a750$6936f5f0$@de> <4CBDD672.2090202@octasic.com> <4CBE3B1D.8080105@jdrosen.net> In-Reply-To: <4CBE3B1D.8080105@jdrosen.net> Date: Wed, 20 Oct 2010 08:02:45 +0200 Organization: =?UTF-8?Q?Universit=C3=A4t_T=C3=BCbingen?= Message-ID: <004d01cb701c$6ef1b2c0$4cd51840$@de> MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Mailer: Microsoft Office Outlook 12.0 Thread-Index: Actv78+K7yOmRUthTp2l/XPmDw3j/QALCguA Content-language: de X-AntiVirus: NOT checked by Avira MailGate (version: 3.0.0-4; host: mx05) Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 06:01:22 -0000 Dear Jonathan, is Skype registering Opus as trademark? With best regards, Christian --------------------------------------------------------------- Dr.-Ing. Christian Hoene Interactive Communication Systems (ICS), University of T=C3=BCbingen=20 Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532=20 http://www.net.uni-tuebingen.de/ -----Original Message----- From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf = Of Jonathan Rosenberg Sent: Wednesday, October 20, 2010 2:43 AM To: codec@ietf.org Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt IETF has a long history of standards with names - ICE, SIP, POP, IMAP,=20 etc. These are also acronyms, but increasingly the acronyms are highly=20 contrived and are basically names (DECADE, ENUM, SALUD) and in some=20 cases just names (Sieve). However, IETF has not ever filed for=20 trademarks on any of these, and I see absolutely no reason to start now. -Jonathan R. On 10/19/2010 1:33 PM, Jean-Marc Valin wrote: > Hi Christian, > > I don't know the details of the ITU-T trademark issues, but in this > case, the Opus name will not be controlled by any company (I'm not = aware > of anyone registering a trademark on the "Opus" name in this context). > I'm not against the IETF owning the trademark, but I'm not sure any > ownership is necessary to begin with. > > Cheers, > > Jean-Marc > > As a note, my company does use the term "Opus" in a different domain = of > application (digital signal processor architecture) and sees no issue > with the IETF using it for an audio codec. Oh, and no, I'm not the one > who originally suggested that name. > > On 10-10-19 01:02 PM, Christian Hoene wrote: >> Hi, >> >> the ITU had some problems with trademarks that were given to >> standardized algorithms. >> As far as I remember, PESQ or PEAQ are trademarked by some companies. >> Since then, the ITU-T name their standards only by numbers. >> >> I do not have any problems with nicknames. However, I think the >> trademark issue shall be addressed and it is important. I would = prefer >> that such as trademark is owned by the IETF if possible. >> >> Uspto.gov does not list an opus codec. But I do not know how fast = they >> (or the readers of this mailing list) are... >> >> With best regards, >> >> Christian >> >> --------------------------------------------------------------- >> Dr.-Ing. Christian Hoene >> Interactive Communication Systems (ICS), University of = T=C3=83=C2=BCbingen >> Sand 13, 72076 T=C3=83=C2=BCbingen, Germany, Phone +49 7071 2970532 >> http://www.net.uni-tuebingen.de/ >> >> >> -----Original Message----- >> From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On = Behalf >> Of Stephan Wenger >> Sent: Tuesday, October 19, 2010 6:05 PM >> To: codec@ietf.org >> Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt >> >> Hi Jean-Marc, >> >> Four things: >> >> First, indeed, there was a hum at the last meeting. Weak memory on my >> side. >> Apologies. >> >> Second, hums need to be reconfirmed on the mailing list, and that has = not >> happened according to my read of the email archive. >> >> Third, I also missed the submission of = draft-ietf-codec-description-00, >> which was really the time I should have complained. >> >> So I'm willing to assume (as apparently have the chairs, see point = #2) >> that >> there has been an implied consensus of the WG to accept the >> codec-description draft. Which brings me to point #4: >> >> As this is now a WG item, any major change requires WG consensus. >> Selecting >> a marketing name, IMO, is such a major change. "Opus" is such a = flashy >> name >> that certain participants and/or companies conceivably may not like >> it. For >> example, if I were working for a company that has in its portfolio an >> audio >> product named "Opus", I would object to the name change. So the thing = you >> should have done, IMO, is to send an email to the list saying "The >> editors >> consider changing the name of our codec to Opus. Is that acceptable = to >> the >> WG?". >> >> It appears to me that twice you guys (chairs included) have taken >> shortcuts >> with the IETF's procedures, as I understand them. That, IMO, fills up >> your >> quota for the next couple of years. Please be more conservative from = now >> on. >> >> Regards, >> Stephan >> >> >> >> >> On 10.19.2010 08:31 , "Jean-Marc Valin" >> wrote: >> >>> Hi, >>> >>> draft-ietf-codec-opus-00.txt is indeed a WG item. It has had several >>> names >>> in the past, including draft-valin-codec-prototype and >>> draft-valin-codec-definition, which may explain the confusion. This >>> is the >>> draft for which there was a hum during the last meeting. >>> >>> Jean-Marc >>> >>> On 10-10-19 11:26 AM, Stephan Wenger wrote: >>>> Hi, >>>> >>>> I want to inquire the status of this draft. In many working groups, = the >>>> filename "draft-ietf--xxx" indicates that the draft in >>>> question is >>>> a WG item of WG. Following this logic, it would appear = that >>>> the >>>> "opus" draft is now a WG item of the codec WG. I don't recall a >>>> decision to >>>> than extent. >>>> >>>> If the draft were indeed accepted as a WG item, I would like to >>>> encourage >>>> those who made IPR statements related to it, to resubmit those >>>> statements >>>> with the new filename. This would help those of us who are = searching >>>> through the IETF IPR tracker by WG name (which is a very common >>>> thing to do, >>>> at least for me). >>>> >>>> Thanks, >>>> Stephan >>>> >>>> >>>> On 10.15.2010 13:30 , >>>> "Internet-Drafts@ietf.org" >>>> wrote: >>>> >>>>> A New Internet-Draft is available from the on-line Internet-Drafts >>>>> directories. >>>>> This draft is a work item of the Internet Wideband Audio Codec >>>>> Working Group >>>>> of the IETF. >>>>> >>>>> >>>>> Title : Definition of the Opus Audio Codec >>>>> Author(s) : J. Valin, K. Vos >>>>> Filename : draft-ietf-codec-opus-00.txt >>>>> Pages : 12 >>>>> Date : 2010-10-15 >>>>> >>>>> This document describes the Opus codec, designed for interactive >>>>> speech and audio transmission over the Internet. >>>>> >>>>> A URL for this Internet-Draft is: >>>>> http://www.ietf.org/internet-drafts/draft-ietf-codec-opus-00.txt >>>>> >>>>> Internet-Drafts are also available by anonymous FTP at: >>>>> ftp://ftp.ietf.org/internet-drafts/ >>>>> >>>>> Below is the data which will enable a MIME compliant mail reader >>>>> implementation to automatically retrieve the ASCII version of the >>>>> Internet-Draft. >>>>> _______________________________________________ >>>>> codec mailing list >>>>> codec@ietf.org >>>>> https://www.ietf.org/mailman/listinfo/codec >>>> >>>> >>>> _______________________________________________ >>>> codec mailing list >>>> codec@ietf.org >>>> https://www.ietf.org/mailman/listinfo/codec >>> >> >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec --=20 Jonathan D. Rosenberg, Ph.D. SkypeID: jdrosen Chief Technology Strategist Mobile: +1 (732) 766-2496 Skype SkypeIn: +1 (408) = 465-0361 jdrosen@skype.net http://www.skype.com jdrosen@jdrosen.net http://www.jdrosen.net _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec From hoene@uni-tuebingen.de Wed Oct 20 00:51:08 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id A935F3A690E for ; Wed, 20 Oct 2010 00:51:08 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.919 X-Spam-Level: X-Spam-Status: No, score=-4.919 tagged_above=-999 required=5 tests=[AWL=1.329, BAYES_00=-2.599, HELO_EQ_DE=0.35, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fp9NBJm0RS9c for ; Wed, 20 Oct 2010 00:51:06 -0700 (PDT) Received: from mx05.uni-tuebingen.de (mx05.uni-tuebingen.de [134.2.3.4]) by core3.amsl.com (Postfix) with ESMTP id A9AAE3A69D0 for ; Wed, 20 Oct 2010 00:51:05 -0700 (PDT) Received: from hoeneT60 (u-173-c044.cs.uni-tuebingen.de [134.2.173.44]) (authenticated bits=0) by mx05.uni-tuebingen.de (8.13.6/8.13.6) with ESMTP id o9K7qbJh027922 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Wed, 20 Oct 2010 09:52:37 +0200 From: "Christian Hoene" To: "'Stephan Wenger'" , References: <000c01cb6fb9$41b95910$c52c0b30$@de> In-Reply-To: Date: Wed, 20 Oct 2010 09:52:39 +0200 Organization: =?UTF-8?Q?Universit=C3=A4t_T=C3=BCbingen?= Message-ID: <005701cb702b$c5038d00$4f0aa700$@de> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0058_01CB703C.888C5D00" X-Mailer: Microsoft Office Outlook 12.0 Thread-Index: ActvNCcg/6UdLTfYSPe62Egnu0SU9wAAN+7tABsLZFoABEUx8AAF+F/5ABfbS/A= Content-language: de X-AntiVirus: NOT checked by Avira MailGate (version: 3.0.0-4; host: mx05) Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 07:51:08 -0000 This is a multi-part message in MIME format. ------=_NextPart_000_0058_01CB703C.888C5D00 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Hello, =20 as long as the IPR license conditions remain unclear to everybody = without a legal background =E2=80=93 and even those with legal = background describe that they could be unacceptable =E2=80=93 may I = remind that the guidelines draft contains a sentence stating: =20 =E2=80=9CIn cases where no RF license can be obtained regarding a = patent, the group should consider alternative algorithms or methods, = even if they result in lower quality, higher complexity, or otherwise = less desirable characteristics (in most cases, the degradation will = likely be small once the best alternative has been identified).=E2=80=9C =20 Thus, I kindly want to remind that there is work to do and that any help = from any side is highly appreciated. I will try to do my very best to = overcome this current drawback.=20 What are your personal opinions, does the patent really cover only the = DTX part of the Opus codec? =20 With best regards, =20 Christian Hoene =20 PS: This, as, my previous emails, are just my personal opinions. =20 PPS: Does the IPR statement #1297 only covers the draft-vos-silk-01? Is it = require to file a similar statement for the opus draft? =20 =20 =20 =20 =20 =20 --------------------------------------------------------------- Dr.-Ing. Christian Hoene Interactive Communication Systems (ICS), University of T=C3=BCbingen=20 Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532=20 http://www.net.uni-tuebingen.de/ =20 From: Stephan Wenger [mailto:stewe@stewe.org]=20 Sent: Tuesday, October 19, 2010 10:14 PM To: Christian Hoene; codec@ietf.org Cc: 'Alfons Martin' Subject: Re: [codec] Opus codec licensing =20 Hi Christian, On 10.19.2010 11:12 , "Christian Hoene" wrote: [...]=20 Clearly, these terms are unacceptable.=20 [...] Let me suggest that you (and everyone else) mark their opinion on = licensing terms as your personal opinion. What=E2=80=99s acceptable or = inacceptable to you may very well be acceptable or inacceptable to me. Same goes with compliance to the charter. Stephan On 10.18.2010 19:26 , "Jean-Marc Valin" > wrote: Opus is available under the BSD license. As for patents, Skype has = pledged to make them available royalty-free once the codec is accepted = as an IETF standard. Jean-Marc -----Original Message----- From: codec-bounces@ietf.org on behalf of Pascal Pochol Sent: Mon 10/18/2010 10:19 PM To: codec@ietf.org Subject: [codec] Opus codec licensing Hello, we've been using speex and celt for voice and music but today I heard = about Opus which sounds like a fantastic replacement for both these codecs. We're wondering if Opus will be released under the same type of license = as speex and celt? SILK's license forbid its use in commercial software and we're worried that it might carry over to Opus. If not, as soon as Opus handles lower bitrates, stereo and fixed point decoding we'll be using = it. -Pascal _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec _____ =20 _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec ------=_NextPart_000_0058_01CB703C.888C5D00 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Re: [codec] Opus codec licensing

Hello,

 

as long as the IPR license conditions remain unclear to everybody without a legal background =E2=80=93 and even those with legal = background describe that they could be unacceptable =E2=80=93 may I remind that the = guidelines draft contains a sentence stating:

 

=E2=80=9CIn cases where no RF license can be obtained regarding a patent, the group = should consider alternative algorithms or methods, even if they result in lower quality, higher complexity, or otherwise less desirable characteristics = (in most cases, the degradation will likely be small once the best = alternative has been identified).=E2=80=9C

 

Thus, I kindly want to remind that there is work to do = and that any help from any side is highly appreciated. I will try to do my very best = to overcome this current drawback.

What are your personal opinions, does the patent really = cover only the DTX part of the Opus codec?

 

With best regards,

 

=C2=A0Christian Hoene

 

PS:

This, as, my previous emails, are just my personal = opinions.

 

PPS:

Does the IPR statement #1297 only covers the = draft-vos-silk-01? Is it require to file a similar statement for the opus = draft?

=C2=A0

 

 

=C2=A0

 

 

----------------------------------------------------------= -----

Dr.-Ing. Christian Hoene

Interactive Communication Systems (ICS), University of = T=C3=BCbingen

Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 = 2970532

http://www.net.uni-tuebingen.de= /

 

From:= Stephan = Wenger [mailto:stewe@stewe.org]
Sent: Tuesday, October 19, 2010 10:14 PM
To: Christian Hoene; codec@ietf.org
Cc: 'Alfons Martin'
Subject: Re: [codec] Opus codec licensing

 

Hi Christian,

On 10.19.2010 11:12 , "Christian Hoene" <hoene@uni-tuebingen.de> = wrote:

[...] =
Clearly, these terms are unacceptable.
[...]

Let me suggest that = you (and everyone else) mark their opinion on licensing terms as your personal = opinion.  What=E2=80=99s acceptable or inacceptable to you may very well be = acceptable or inacceptable to me.

Same goes with compliance to the charter.

Stephan



On 10.18.2010 19:26 , "Jean-Marc Valin" <jean-marc.valin@octasic.com = <jean-marc.valin@octasic.com> = > wrote:
Opus is available under the BSD license. As for patents, Skype has = pledged to make them available royalty-free once the codec is accepted as an IETF standard.

    Jean-Marc

-----Original Message-----
From: codec-bounces@ietf.org on = behalf of Pascal Pochol
Sent: Mon 10/18/2010 10:19 PM
To: codec@ietf.org
Subject: [codec] Opus codec licensing

Hello,

we've been using speex and celt for voice and music but today I heard = about
Opus which sounds like a fantastic replacement for both these = codecs.

We're wondering if Opus will be released under the same type of license = as
speex and celt? SILK's license forbid its use in commercial software = and
we're worried that it might carry over to Opus. If not, as soon as = Opus
handles lower bitrates, stereo and fixed point decoding we'll be using = it.

-Pascal

_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org= /mailman/listinfo/codec

_________________________= ______________________
codec mailing list
codec@ietf.org
https://www.ietf.org= /mailman/listinfo/codec

------=_NextPart_000_0058_01CB703C.888C5D00-- From stephen.botzko@gmail.com Wed Oct 20 03:59:45 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 889733A63CB for ; Wed, 20 Oct 2010 03:59:45 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.703 X-Spam-Level: X-Spam-Status: No, score=-2.703 tagged_above=-999 required=5 tests=[AWL=-0.105, BAYES_00=-2.599, HTML_MESSAGE=0.001] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5a1V8pJKiFzc for ; Wed, 20 Oct 2010 03:59:44 -0700 (PDT) Received: from mail-bw0-f44.google.com (mail-bw0-f44.google.com [209.85.214.44]) by core3.amsl.com (Postfix) with ESMTP id 561E93A63C9 for ; Wed, 20 Oct 2010 03:59:42 -0700 (PDT) Received: by bwz14 with SMTP id 14so2722293bwz.31 for ; Wed, 20 Oct 2010 04:01:14 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type; bh=vNbecCJlqIAkhslf/pUwAtRX/hdnjVplp5OI0cXBF3c=; b=nbeJ/jwWez7BxBYgyUT7Pfo3L+xbRzahHzPZS/rThylV7L06nzT/wgcKYoVQtmDxxo MstXrLU5EHVjGCMPqvQfR9v8rP+lsiPqOLG6dQKJ9fcWmzib+Ct4DIqdpGl1UTMFYEmG /JrfvCkkhe0MWuRfMrucov2vtoV8WQion9Wx8= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=RA1bUiuyNGSswg3z6H1gzTjzXii6RBtguW6K04kJGCYr8wRsLM4DM+sJBWwP6tA6z1 zEV1VGKkqKRflqFwOQvtmUDFI2TcuTgyg7eUVcFvP+7m6KvXNDJfBTS2LS011aqZM3jx cDDxwOJW91lOBA2NTBxRbUGHX2EY+y3IgD+So= MIME-Version: 1.0 Received: by 10.204.76.80 with SMTP id b16mr3073139bkk.160.1287572474392; Wed, 20 Oct 2010 04:01:14 -0700 (PDT) Received: by 10.204.80.213 with HTTP; Wed, 20 Oct 2010 04:01:14 -0700 (PDT) In-Reply-To: <005701cb702b$c5038d00$4f0aa700$@de> References: <000c01cb6fb9$41b95910$c52c0b30$@de> <005701cb702b$c5038d00$4f0aa700$@de> Date: Wed, 20 Oct 2010 07:01:14 -0400 Message-ID: From: Stephen Botzko To: Christian Hoene Content-Type: multipart/alternative; boundary=00163649908bf5707204930a53dc Cc: codec@ietf.org Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 10:59:45 -0000 --00163649908bf5707204930a53dc Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: quoted-printable Is it possible to create a CODEC-IPR mailing list (or some such) to allow folks to use the alternative list when chatting about patents and work-arounds? That would address some of Stephan's concerns, but still allow the discussion to archived to an IETF list. I'd prefer to keep out of the patent discussions, I suspect there are others who would rather not get those posts either. Stephen Botzko 2010/10/20 Christian Hoene > Hello, > > > > as long as the IPR license conditions remain unclear to everybody without= a > legal background =96 and even those with legal background describe that t= hey > could be unacceptable =96 may I remind that the guidelines draft contains= a > sentence stating: > > > > =93In cases where no RF license can be obtained regarding a patent, the g= roup > should consider alternative algorithms or methods, even if they result in > lower quality, higher complexity, or otherwise less desirable > characteristics (in most cases, the degradation will likely be small once > the best alternative has been identified).=93 > > > > Thus, I kindly want to remind that there is work to do and that any help > from any side is highly appreciated. I will try to do my very best to > overcome this current drawback. > > What are your personal opinions, does the patent really cover only the DT= X > part of the Opus codec? > > > > With best regards, > > > > Christian Hoene > > > > PS: > > This, as, my previous emails, are just my personal opinions. > > > > PPS: > > Does the IPR statement #1297 only covers the draft-vos-silk-01? Is it > require to file a similar statement for the opus draft? > > > > > > > > > > > > > > --------------------------------------------------------------- > > Dr.-Ing. Christian Hoene > > Interactive Communication Systems (ICS), University of T=FCbingen > > Sand 13, 72076 T=FCbingen, Germany, Phone +49 7071 2970532 > > http://www.net.uni-tuebingen.de/ > > > > *From:* Stephan Wenger [mailto:stewe@stewe.org] > *Sent:* Tuesday, October 19, 2010 10:14 PM > *To:* Christian Hoene; codec@ietf.org > *Cc:* 'Alfons Martin' > > *Subject:* Re: [codec] Opus codec licensing > > > > Hi Christian, > > > On 10.19.2010 11:12 , "Christian Hoene" wrote: > > [...] > Clearly, these terms are unacceptable. > [...] > > Let me suggest that you (and everyone else) mark their opinion on licensi= ng > terms as your personal opinion. What=92s acceptable or inacceptable to y= ou > may very well be acceptable or inacceptable to me. > > Same goes with compliance to the charter. > > Stephan > > > > On 10.18.2010 19:26 , "Jean-Marc Valin" jean-marc.valin@octasic.com> > wrote: > Opus is available under the BSD license. As for patents, Skype has pledge= d > to make them available royalty-free once the codec is accepted as an IETF > standard. > > Jean-Marc > > -----Original Message----- > From: codec-bounces@ietf.org on behalf of Pascal Pochol > Sent: Mon 10/18/2010 10:19 PM > To: codec@ietf.org > Subject: [codec] Opus codec licensing > > Hello, > > we've been using speex and celt for voice and music but today I heard abo= ut > Opus which sounds like a fantastic replacement for both these codecs. > > We're wondering if Opus will be released under the same type of license a= s > speex and celt? SILK's license forbid its use in commercial software and > we're worried that it might carry over to Opus. If not, as soon as Opus > handles lower bitrates, stereo and fixed point decoding we'll be using it= . > > -Pascal > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > ------------------------------ > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > --00163649908bf5707204930a53dc Content-Type: text/html; charset=windows-1252 Content-Transfer-Encoding: quoted-printable Is it possible to create a CODEC-IPR mailing list (or some such) to allow f= olks to use the alternative list when chatting about patents and work-aroun= ds?

That would address some of Stephan's concerns, but still all= ow the discussion to archived to an IETF list.=A0 I'd prefer to keep ou= t of the patent discussions, I suspect there are others who would rather no= t get those posts either.

Stephen Botzko

2010/10/20 Christian H= oene <hoene@= uni-tuebingen.de>

Hello,

=A0

as long as the IPR license conditions remain unclear t= o everybody without a legal background =96 and even those with legal backgrou= nd describe that they could be unacceptable =96 may I remind that the guidelin= es draft contains a sentence stating:

=A0

=93= In cases where no RF license can be obtained regarding a patent, the group sho= uld consider alternative algorithms or methods, even if they result in lower quality, higher complexity, or otherwise less desirable characteristics (in most cases, the degradation will likely be small once the best alternative = has been identified).=93

=A0=

Thus, I kindly want to remind that there is work to do= and that any help from any side is highly appreciated. I will try to do my very best to overcome this current drawback.

What are your personal opinions, does the patent reall= y cover only the DTX part of the Opus codec?

=A0

With best regards,

=A0

=A0Christian Hoene

=A0

PS:

This, as, my previous emails, are just my personal opi= nions.

=A0

PPS:

Does the IPR statement #1297 only covers the draft-vos= -silk-01? Is it require to file a similar statement for the opus draft?

=A0

=A0

=A0

=A0=

=A0

=A0

-----------------------------= ----------------------------------

Dr.-Ing. Christian Hoene

Interactive Communication Sys= tems (ICS), University of T=FCbingen

Sand 13, 72076 T=FCbingen, Ge= rmany, Phone +49 7071 2970532

http://www.net.uni-tuebingen.de/

=A0

From:= Stephan Wenger [mailto:stewe@stewe.or= g]
Sent: Tuesday, October 19, 2010 10:14 PM
To: Christian Hoene; codec@ietf.org
Cc: 'Alfons Martin'


Subject: Re: [codec] Opus codec licensing

=A0

Hi Christian,



On 10.19.2010 11:12 , "Christian Hoene" <hoene@uni-tuebingen.de> wrote= :

[...]
Clearly, these terms are unacceptable.
[...]

Let me suggest that y= ou (and everyone else) mark their opinion on licensing terms as your personal opini= on. =A0What=92s acceptable or inacceptable to you may very well be acceptable o= r inacceptable to me.

Same goes with compliance to the charter.

Stephan



On 10.18.2010 19:26 , "Jean-Marc Valin" <jean-marc.valin@octasic.com= <jean-= marc.valin@octasic.com> > wrote:
Opus is available under the BSD license. As for patents, Skype has pledged = to make them available royalty-free once the codec is accepted as an IETF standard.

=A0=A0=A0=A0Jean-Marc

-----Original Message-----
From: codec-bou= nces@ietf.org on behalf of Pascal Pochol
Sent: Mon 10/18/2010 10:19 PM
To: codec@ietf.org<= br> Subject: [codec] Opus codec licensing

Hello,

we've been using speex and celt for voice and music but today I heard a= bout
Opus which sounds like a fantastic replacement for both these codecs.

We're wondering if Opus will be released under the same type of license= as
speex and celt? SILK's license forbid its use in commercial software an= d
we're worried that it might carry over to Opus. If not, as soon as Opus=
handles lower bitrates, stereo and fixed point decoding we'll be using = it.

-Pascal

_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec

_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec


_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec


--00163649908bf5707204930a53dc-- From fluffy@cisco.com Wed Oct 20 06:40:47 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 7EA963A67F9 for ; Wed, 20 Oct 2010 06:40:47 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -110.52 X-Spam-Level: X-Spam-Status: No, score=-110.52 tagged_above=-999 required=5 tests=[AWL=0.079, BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 7sr9hRminN-u for ; Wed, 20 Oct 2010 06:40:42 -0700 (PDT) Received: from sj-iport-1.cisco.com (sj-iport-1.cisco.com [171.71.176.70]) by core3.amsl.com (Postfix) with ESMTP id BF7ED3A67D3 for ; Wed, 20 Oct 2010 06:40:38 -0700 (PDT) Authentication-Results: sj-iport-1.cisco.com; dkim=neutral (message not signed) header.i=none X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AvsEAEqOvkyrR7Ht/2dsb2JhbAChWnGhf5xRhUoEhFWFdoME X-IronPort-AV: E=Sophos;i="4.57,355,1283731200"; d="scan'208";a="372735108" Received: from sj-core-1.cisco.com ([171.71.177.237]) by sj-iport-1.cisco.com with ESMTP; 20 Oct 2010 13:42:12 +0000 Received: from [192.168.4.2] (rcdn-fluffy-8711.cisco.com [10.99.9.18]) by sj-core-1.cisco.com (8.13.8/8.14.3) with ESMTP id o9KDgBFO017443; Wed, 20 Oct 2010 13:42:11 GMT From: Cullen Jennings Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Date: Wed, 20 Oct 2010 07:42:10 -0600 Message-Id: To: Stephan Wenger , codec@ietf.org Mime-Version: 1.0 (Apple Message framework v1081) X-Mailer: Apple Mail (2.1081) Subject: [codec] Seperate list for IPR discutions? X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 13:40:47 -0000 =09 Stephan, as our WG technical advisor on topics like this, do you have = advice on if we should set up a separate list for people that want to = talk about patent issues related to the standards from this WG? =46rom a = practical point of view, someone could easily set up a separate list = either on the ietf.org servers or not on ietf servers.=20= From fluffy@cisco.com Wed Oct 20 06:42:46 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id BE5FA3A6819 for ; Wed, 20 Oct 2010 06:42:46 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -108.547 X-Spam-Level: X-Spam-Status: No, score=-108.547 tagged_above=-999 required=5 tests=[AWL=-1.847, BAYES_00=-2.599, FB_CIALIS_LEO3=3.899, RCVD_IN_DNSWL_HI=-8, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id u2xzViSxBdoA for ; Wed, 20 Oct 2010 06:42:45 -0700 (PDT) Received: from rtp-iport-2.cisco.com (rtp-iport-2.cisco.com [64.102.122.149]) by core3.amsl.com (Postfix) with ESMTP id 97C4D3A680D for ; Wed, 20 Oct 2010 06:42:42 -0700 (PDT) Authentication-Results: rtp-iport-2.cisco.com; dkim=neutral (message not signed) header.i=none X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AvsEAMOOvkxAZnwM/2dsb2JhbAChWnGhd5xOgnmCUQSEVYV2gwQ X-IronPort-AV: E=Sophos;i="4.57,355,1283731200"; d="scan'208";a="172923540" Received: from rtp-core-1.cisco.com ([64.102.124.12]) by rtp-iport-2.cisco.com with ESMTP; 20 Oct 2010 13:44:15 +0000 Received: from [192.168.4.2] (rcdn-fluffy-8711.cisco.com [10.99.9.18]) by rtp-core-1.cisco.com (8.13.8/8.14.3) with ESMTP id o9KDiEWX015816; Wed, 20 Oct 2010 13:44:14 GMT Mime-Version: 1.0 (Apple Message framework v1081) Content-Type: text/plain; charset=us-ascii From: Cullen Jennings In-Reply-To: Date: Wed, 20 Oct 2010 07:44:13 -0600 Content-Transfer-Encoding: quoted-printable Message-Id: References: To: Stephan Wenger X-Mailer: Apple Mail (2.1081) Cc: codec@ietf.org Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 13:42:46 -0000 So just as a point of IETF process, my understanding is the selecting of = a WG draft is just a chair decision and not an WG consensus topic - of = course the draft needs to be changed such that it reflects the WG = consensus so really all the chairs are doing here is picking which = virtual piece of paper will be used to write on. Often chairs takes hums = to get input on this but given the lack of alternatives, and the = consensus developing around this draft, it seemed like a pretty easy = decision. If we made a mistake or too much of a short cut here, we = should go fix that.=20 On the name topic, typically the WG just lets the authors of the draft = pick a name as its often viewed as a more editorial than actual = technical topic. If someone has some serious problem with Opus and wants = something different, lets talk about it but I was hoping not to spend = too much time on such a bike shed topic. Clearly if Skype has = trademarked Opus then we are going to be having a long conversation = about it but I'm working on the assumption that they have not done that. = We will see what they say to the question asked.=20 Your point about updating IPR declaration is excellent - once we are = stable on the name, we should all bug folks update the IPR disclosure to = point at the new draft. I have been trying to keep the datatracker = "replaced by" up to date so there is some hope of seeing what previous = drafts are closely related to the newer draft.=20 On Oct 19, 2010, at 10:04 AM, Stephan Wenger wrote: > Hi Jean-Marc, >=20 > Four things: >=20 > First, indeed, there was a hum at the last meeting. Weak memory on my = side. > Apologies. >=20 > Second, hums need to be reconfirmed on the mailing list, and that has = not > happened according to my read of the email archive. >=20 > Third, I also missed the submission of = draft-ietf-codec-description-00, > which was really the time I should have complained. >=20 > So I'm willing to assume (as apparently have the chairs, see point #2) = that > there has been an implied consensus of the WG to accept the > codec-description draft. Which brings me to point #4: >=20 > As this is now a WG item, any major change requires WG consensus. = Selecting > a marketing name, IMO, is such a major change. "Opus" is such a = flashy name > that certain participants and/or companies conceivably may not like = it. For > example, if I were working for a company that has in its portfolio an = audio > product named "Opus", I would object to the name change. So the thing = you > should have done, IMO, is to send an email to the list saying "The = editors > consider changing the name of our codec to Opus. Is that acceptable = to the > WG?". >=20 > It appears to me that twice you guys (chairs included) have taken = shortcuts > with the IETF's procedures, as I understand them. That, IMO, fills up = your > quota for the next couple of years. Please be more conservative from = now > on. >=20 > Regards, > Stephan >=20 >=20 >=20 >=20 > On 10.19.2010 08:31 , "Jean-Marc Valin" = wrote: >=20 >> Hi, >>=20 >> draft-ietf-codec-opus-00.txt is indeed a WG item. It has had several = names >> in the past, including draft-valin-codec-prototype and >> draft-valin-codec-definition, which may explain the confusion. This = is the >> draft for which there was a hum during the last meeting. >>=20 >> Jean-Marc >>=20 >> On 10-10-19 11:26 AM, Stephan Wenger wrote: >>> Hi, >>>=20 >>> I want to inquire the status of this draft. In many working groups, = the >>> filename "draft-ietf--xxx" indicates that the draft in = question is >>> a WG item of WG. Following this logic, it would appear = that the >>> "opus" draft is now a WG item of the codec WG. I don't recall a = decision to >>> than extent. >>>=20 >>> If the draft were indeed accepted as a WG item, I would like to = encourage >>> those who made IPR statements related to it, to resubmit those = statements >>> with the new filename. This would help those of us who are = searching >>> through the IETF IPR tracker by WG name (which is a very common = thing to do, >>> at least for me). >>>=20 >>> Thanks, >>> Stephan >>>=20 >>>=20 >>> On 10.15.2010 13:30 , = "Internet-Drafts@ietf.org" >>> wrote: >>>=20 >>>> A New Internet-Draft is available from the on-line Internet-Drafts >>>> directories. >>>> This draft is a work item of the Internet Wideband Audio Codec = Working Group >>>> of the IETF. >>>>=20 >>>>=20 >>>> Title : Definition of the Opus Audio Codec >>>> Author(s) : J. Valin, K. Vos >>>> Filename : draft-ietf-codec-opus-00.txt >>>> Pages : 12 >>>> Date : 2010-10-15 >>>>=20 >>>> This document describes the Opus codec, designed for interactive >>>> speech and audio transmission over the Internet. >>>>=20 >>>> A URL for this Internet-Draft is: >>>> http://www.ietf.org/internet-drafts/draft-ietf-codec-opus-00.txt >>>>=20 >>>> Internet-Drafts are also available by anonymous FTP at: >>>> ftp://ftp.ietf.org/internet-drafts/ >>>>=20 >>>> Below is the data which will enable a MIME compliant mail reader >>>> implementation to automatically retrieve the ASCII version of the >>>> Internet-Draft. >>>> _______________________________________________ >>>> codec mailing list >>>> codec@ietf.org >>>> https://www.ietf.org/mailman/listinfo/codec >>>=20 >>>=20 >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >>=20 >=20 >=20 > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From jdrosen@jdrosen.net Wed Oct 20 09:16:51 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 563A43A69E6 for ; Wed, 20 Oct 2010 09:16:46 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.333 X-Spam-Level: X-Spam-Status: No, score=-102.333 tagged_above=-999 required=5 tests=[AWL=0.266, BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ndYC9Y7aEKPg for ; Wed, 20 Oct 2010 09:16:29 -0700 (PDT) Received: from ecbiz71.inmotionhosting.com (ecbiz71.inmotionhosting.com [173.205.126.243]) by core3.amsl.com (Postfix) with ESMTP id 655A03A68E8 for ; Wed, 20 Oct 2010 09:15:48 -0700 (PDT) Received: from [98.109.139.97] (helo=[192.168.1.8]) by ecbiz71.inmotionhosting.com with esmtpsa (TLSv1:AES256-SHA:256) (Exim 4.69) (envelope-from ) id 1P8bLr-0002wl-DK; Wed, 20 Oct 2010 12:17:19 -0400 Message-ID: <4CBF1611.20607@jdrosen.net> Date: Wed, 20 Oct 2010 12:17:21 -0400 From: Jonathan Rosenberg User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.2.9) Gecko/20100915 Thunderbird/3.1.4 MIME-Version: 1.0 To: Christian Hoene References: <4CBDB9C7.1020206@octasic.com> <005b01cb6faf$7867a750$6936f5f0$@de> <4CBDD672.2090202@octasic.com> <4CBE3B1D.8080105@jdrosen.net> <004d01cb701c$6ef1b2c0$4cd51840$@de> In-Reply-To: <004d01cb701c$6ef1b2c0$4cd51840$@de> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - ecbiz71.inmotionhosting.com X-AntiAbuse: Original Domain - ietf.org X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - jdrosen.net Cc: codec@ietf.org Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 16:16:52 -0000 No. Since the Opus codec is a product of the IETF it would make no sense for an individual company to trademark the name of an IETF output. -Jonathan R. On 10/20/2010 2:02 AM, Christian Hoene wrote: > Dear Jonathan, > > is Skype registering Opus as trademark? > > With best regards, > > Christian > > > --------------------------------------------------------------- > Dr.-Ing. Christian Hoene > Interactive Communication Systems (ICS), University of Tübingen > Sand 13, 72076 Tübingen, Germany, Phone +49 7071 2970532 > http://www.net.uni-tuebingen.de/ > > > -----Original Message----- > From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Jonathan Rosenberg > Sent: Wednesday, October 20, 2010 2:43 AM > To: codec@ietf.org > Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt > > IETF has a long history of standards with names - ICE, SIP, POP, IMAP, > etc. These are also acronyms, but increasingly the acronyms are highly > contrived and are basically names (DECADE, ENUM, SALUD) and in some > cases just names (Sieve). However, IETF has not ever filed for > trademarks on any of these, and I see absolutely no reason to start now. > > -Jonathan R. > > > > On 10/19/2010 1:33 PM, Jean-Marc Valin wrote: >> Hi Christian, >> >> I don't know the details of the ITU-T trademark issues, but in this >> case, the Opus name will not be controlled by any company (I'm not aware >> of anyone registering a trademark on the "Opus" name in this context). >> I'm not against the IETF owning the trademark, but I'm not sure any >> ownership is necessary to begin with. >> >> Cheers, >> >> Jean-Marc >> >> As a note, my company does use the term "Opus" in a different domain of >> application (digital signal processor architecture) and sees no issue >> with the IETF using it for an audio codec. Oh, and no, I'm not the one >> who originally suggested that name. >> >> On 10-10-19 01:02 PM, Christian Hoene wrote: >>> Hi, >>> >>> the ITU had some problems with trademarks that were given to >>> standardized algorithms. >>> As far as I remember, PESQ or PEAQ are trademarked by some companies. >>> Since then, the ITU-T name their standards only by numbers. >>> >>> I do not have any problems with nicknames. However, I think the >>> trademark issue shall be addressed and it is important. I would prefer >>> that such as trademark is owned by the IETF if possible. >>> >>> Uspto.gov does not list an opus codec. But I do not know how fast they >>> (or the readers of this mailing list) are... >>> >>> With best regards, >>> >>> Christian >>> >>> --------------------------------------------------------------- >>> Dr.-Ing. Christian Hoene >>> Interactive Communication Systems (ICS), University of Tübingen >>> Sand 13, 72076 Tübingen, Germany, Phone +49 7071 2970532 >>> http://www.net.uni-tuebingen.de/ >>> >>> >>> -----Original Message----- >>> From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf >>> Of Stephan Wenger >>> Sent: Tuesday, October 19, 2010 6:05 PM >>> To: codec@ietf.org >>> Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt >>> >>> Hi Jean-Marc, >>> >>> Four things: >>> >>> First, indeed, there was a hum at the last meeting. Weak memory on my >>> side. >>> Apologies. >>> >>> Second, hums need to be reconfirmed on the mailing list, and that has not >>> happened according to my read of the email archive. >>> >>> Third, I also missed the submission of draft-ietf-codec-description-00, >>> which was really the time I should have complained. >>> >>> So I'm willing to assume (as apparently have the chairs, see point #2) >>> that >>> there has been an implied consensus of the WG to accept the >>> codec-description draft. Which brings me to point #4: >>> >>> As this is now a WG item, any major change requires WG consensus. >>> Selecting >>> a marketing name, IMO, is such a major change. "Opus" is such a flashy >>> name >>> that certain participants and/or companies conceivably may not like >>> it. For >>> example, if I were working for a company that has in its portfolio an >>> audio >>> product named "Opus", I would object to the name change. So the thing you >>> should have done, IMO, is to send an email to the list saying "The >>> editors >>> consider changing the name of our codec to Opus. Is that acceptable to >>> the >>> WG?". >>> >>> It appears to me that twice you guys (chairs included) have taken >>> shortcuts >>> with the IETF's procedures, as I understand them. That, IMO, fills up >>> your >>> quota for the next couple of years. Please be more conservative from now >>> on. >>> >>> Regards, >>> Stephan >>> >>> >>> >>> >>> On 10.19.2010 08:31 , "Jean-Marc Valin" >>> wrote: >>> >>>> Hi, >>>> >>>> draft-ietf-codec-opus-00.txt is indeed a WG item. It has had several >>>> names >>>> in the past, including draft-valin-codec-prototype and >>>> draft-valin-codec-definition, which may explain the confusion. This >>>> is the >>>> draft for which there was a hum during the last meeting. >>>> >>>> Jean-Marc >>>> >>>> On 10-10-19 11:26 AM, Stephan Wenger wrote: >>>>> Hi, >>>>> >>>>> I want to inquire the status of this draft. In many working groups, the >>>>> filename "draft-ietf--xxx" indicates that the draft in >>>>> question is >>>>> a WG item of WG. Following this logic, it would appear that >>>>> the >>>>> "opus" draft is now a WG item of the codec WG. I don't recall a >>>>> decision to >>>>> than extent. >>>>> >>>>> If the draft were indeed accepted as a WG item, I would like to >>>>> encourage >>>>> those who made IPR statements related to it, to resubmit those >>>>> statements >>>>> with the new filename. This would help those of us who are searching >>>>> through the IETF IPR tracker by WG name (which is a very common >>>>> thing to do, >>>>> at least for me). >>>>> >>>>> Thanks, >>>>> Stephan >>>>> >>>>> >>>>> On 10.15.2010 13:30 , >>>>> "Internet-Drafts@ietf.org" >>>>> wrote: >>>>> >>>>>> A New Internet-Draft is available from the on-line Internet-Drafts >>>>>> directories. >>>>>> This draft is a work item of the Internet Wideband Audio Codec >>>>>> Working Group >>>>>> of the IETF. >>>>>> >>>>>> >>>>>> Title : Definition of the Opus Audio Codec >>>>>> Author(s) : J. Valin, K. Vos >>>>>> Filename : draft-ietf-codec-opus-00.txt >>>>>> Pages : 12 >>>>>> Date : 2010-10-15 >>>>>> >>>>>> This document describes the Opus codec, designed for interactive >>>>>> speech and audio transmission over the Internet. >>>>>> >>>>>> A URL for this Internet-Draft is: >>>>>> http://www.ietf.org/internet-drafts/draft-ietf-codec-opus-00.txt >>>>>> >>>>>> Internet-Drafts are also available by anonymous FTP at: >>>>>> ftp://ftp.ietf.org/internet-drafts/ >>>>>> >>>>>> Below is the data which will enable a MIME compliant mail reader >>>>>> implementation to automatically retrieve the ASCII version of the >>>>>> Internet-Draft. >>>>>> _______________________________________________ >>>>>> codec mailing list >>>>>> codec@ietf.org >>>>>> https://www.ietf.org/mailman/listinfo/codec >>>>> >>>>> >>>>> _______________________________________________ >>>>> codec mailing list >>>>> codec@ietf.org >>>>> https://www.ietf.org/mailman/listinfo/codec >>>> >>> >>> >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >>> >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec > -- Jonathan D. Rosenberg, Ph.D. SkypeID: jdrosen Chief Technology Strategist Mobile: +1 (732) 766-2496 Skype SkypeIn: +1 (408) 465-0361 jdrosen@skype.net http://www.skype.com jdrosen@jdrosen.net http://www.jdrosen.net From stewe@stewe.org Wed Oct 20 09:42:47 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 7C9613A69F1 for ; Wed, 20 Oct 2010 09:42:47 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.828 X-Spam-Level: X-Spam-Status: No, score=-1.828 tagged_above=-999 required=5 tests=[AWL=0.771, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id j0OdFxxtAbiH for ; Wed, 20 Oct 2010 09:42:46 -0700 (PDT) Received: from stewe.org (stewe.org [85.214.122.234]) by core3.amsl.com (Postfix) with ESMTP id D133A3A69F9 for ; Wed, 20 Oct 2010 09:39:12 -0700 (PDT) Received: from [192.168.1.108] (unverified [24.5.132.232]) by stewe.org (SurgeMail 3.9e) with ESMTP id 820455-1743317 for multiple; Wed, 20 Oct 2010 18:40:45 +0200 User-Agent: Microsoft-Entourage/12.27.0.100910 Date: Wed, 20 Oct 2010 09:40:33 -0700 From: Stephan Wenger To: "fluffy@cisco.com" , Message-ID: Thread-Topic: Seperate list for IPR discutions? Thread-Index: ActwdYOBYXx4y0JWLU+z0ln69pD9Ow== In-Reply-To: Mime-version: 1.0 Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit X-Originating-IP: 24.5.132.232 X-Authenticated-User: stewe@stewe.org X-ORBS-Stamp: Your IP (24.5.132.232) was found in the spamhaus database. http://www.spamhaus.net Subject: Re: [codec] Seperate list for IPR discutions? X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 16:42:47 -0000 Hi, Yes, I think a separate mailing list would be an excellent idea. There is IMO no danger to the IETF when using an IETF-hosted list. But, as you mentioned, there are also other options. Whatever is the easiest for those interested in this work. My preference would be that this list has a subscription process independent from the subscription to codec@ietf.org. As the minimum, there must be an opt-out process. I would recommend that the maintainers of the list, periodically, make people on codec@ietf.org aware of the existence of the list and the ongoing discussions, in a summarized form. I also don't mind if suggested design choices spill over to codec@ietf; after all, this is the list where technical discussions are being held. The borderline would be citations of patent language, including patent numbers and claim language. Personally, I surely do not want to be involved on this hypothetical new list. Stephan On 10.20.2010 06:42 , "fluffy@cisco.com" wrote: > > Stephan, as our WG technical advisor on topics like this, do you have advice > on if we should set up a separate list for people that want to talk about > patent issues related to the standards from this WG? From a practical point of > view, someone could easily set up a separate list either on the ietf.org > servers or not on ietf servers. From hoene@uni-tuebingen.de Wed Oct 20 10:01:09 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 3CAAD3A6967 for ; Wed, 20 Oct 2010 10:01:09 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -5.683 X-Spam-Level: X-Spam-Status: No, score=-5.683 tagged_above=-999 required=5 tests=[AWL=0.566, BAYES_00=-2.599, HELO_EQ_DE=0.35, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 2KrpJDmqQvXm for ; Wed, 20 Oct 2010 10:01:07 -0700 (PDT) Received: from mx05.uni-tuebingen.de (mx05.uni-tuebingen.de [134.2.3.4]) by core3.amsl.com (Postfix) with ESMTP id 4FFAC3A68FB for ; Wed, 20 Oct 2010 10:01:07 -0700 (PDT) Received: from hoeneT60 ([178.2.222.223]) (authenticated bits=0) by mx05.uni-tuebingen.de (8.13.6/8.13.6) with ESMTP id o9KH2WW9014332 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Wed, 20 Oct 2010 19:02:38 +0200 From: "Christian Hoene" To: "'Jonathan Rosenberg'" References: <4CBDB9C7.1020206@octasic.com> <005b01cb6faf$7867a750$6936f5f0$@de> <4CBDD672.2090202@octasic.com> <4CBE3B1D.8080105@jdrosen.net> <004d01cb701c$6ef1b2c0$4cd51840$@de> <4CBF1611.20607@jdrosen.net> In-Reply-To: <4CBF1611.20607@jdrosen.net> Date: Wed, 20 Oct 2010 19:02:34 +0200 Organization: =?utf-8?Q?Universit=C3=A4t_T=C3=BCbingen?= Message-ID: <002401cb7078$9b6bc9b0$d2435d10$@de> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable X-Mailer: Microsoft Office Outlook 12.0 Thread-Index: ActwclOFJcInlZfoSqS69qEDtgVYXgAAukHA Content-Language: de X-AntiVirus: NOT checked by Avira MailGate (version: 3.0.0-4; host: mx05) Cc: codec@ietf.org Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 17:01:09 -0000 Dear Jonathan Rosenberg and Jean-Marc Valin, Thank you very much for answering fast and clearly.=20 As the decision on the name of the codec was made behind closed doors, I = had to ask the question on the trademark. Now, after publishing the = codec's name, it should be more difficult for a third party to clam a = trademark on it. Sincerely Christian PS: Yes, Opus is an excellent name that has a nice allusion to music. --------------------------------------------------------------- Dr.-Ing. Christian Hoene Interactive Communication Systems (ICS), University of T=C3=BCbingen=20 Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532=20 http://www.net.uni-tuebingen.de/ -----Original Message----- From: Jonathan Rosenberg [mailto:jdrosen@jdrosen.net]=20 Sent: Wednesday, October 20, 2010 6:17 PM To: Christian Hoene Cc: codec@ietf.org Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt No. Since the Opus codec is a product of the IETF it would make no sense for = an individual company to trademark the name of an IETF output. -Jonathan R. On 10/20/2010 2:02 AM, Christian Hoene wrote: > Dear Jonathan, > > is Skype registering Opus as trademark? > > With best regards, > > Christian > > > --------------------------------------------------------------- > Dr.-Ing. Christian Hoene > Interactive Communication Systems (ICS), University of T=C3=BCbingen > Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532 > http://www.net.uni-tuebingen.de/ > > > -----Original Message----- > From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf = Of Jonathan Rosenberg > Sent: Wednesday, October 20, 2010 2:43 AM > To: codec@ietf.org > Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt > > IETF has a long history of standards with names - ICE, SIP, POP, IMAP, > etc. These are also acronyms, but increasingly the acronyms are highly > contrived and are basically names (DECADE, ENUM, SALUD) and in some > cases just names (Sieve). However, IETF has not ever filed for > trademarks on any of these, and I see absolutely no reason to start = now. > > -Jonathan R. > > > > On 10/19/2010 1:33 PM, Jean-Marc Valin wrote: >> Hi Christian, >> >> I don't know the details of the ITU-T trademark issues, but in this >> case, the Opus name will not be controlled by any company (I'm not = aware >> of anyone registering a trademark on the "Opus" name in this = context). >> I'm not against the IETF owning the trademark, but I'm not sure any >> ownership is necessary to begin with. >> >> Cheers, >> >> Jean-Marc >> >> As a note, my company does use the term "Opus" in a different domain = of >> application (digital signal processor architecture) and sees no issue >> with the IETF using it for an audio codec. Oh, and no, I'm not the = one >> who originally suggested that name. >> >> On 10-10-19 01:02 PM, Christian Hoene wrote: >>> Hi, >>> >>> the ITU had some problems with trademarks that were given to >>> standardized algorithms. >>> As far as I remember, PESQ or PEAQ are trademarked by some = companies. >>> Since then, the ITU-T name their standards only by numbers. >>> >>> I do not have any problems with nicknames. However, I think the >>> trademark issue shall be addressed and it is important. I would = prefer >>> that such as trademark is owned by the IETF if possible. >>> >>> Uspto.gov does not list an opus codec. But I do not know how fast = they >>> (or the readers of this mailing list) are... >>> >>> With best regards, >>> >>> Christian >>> >>> --------------------------------------------------------------- >>> Dr.-Ing. Christian Hoene >>> Interactive Communication Systems (ICS), University of = T=C3=83=C2=BCbingen >>> Sand 13, 72076 T=C3=83=C2=BCbingen, Germany, Phone +49 7071 2970532 >>> http://www.net.uni-tuebingen.de/ >>> >>> >>> -----Original Message----- >>> From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On = Behalf >>> Of Stephan Wenger >>> Sent: Tuesday, October 19, 2010 6:05 PM >>> To: codec@ietf.org >>> Subject: Re: [codec] I-D Action:draft-ietf-codec-opus-00.txt >>> >>> Hi Jean-Marc, >>> >>> Four things: >>> >>> First, indeed, there was a hum at the last meeting. Weak memory on = my >>> side. >>> Apologies. >>> >>> Second, hums need to be reconfirmed on the mailing list, and that = has not >>> happened according to my read of the email archive. >>> >>> Third, I also missed the submission of = draft-ietf-codec-description-00, >>> which was really the time I should have complained. >>> >>> So I'm willing to assume (as apparently have the chairs, see point = #2) >>> that >>> there has been an implied consensus of the WG to accept the >>> codec-description draft. Which brings me to point #4: >>> >>> As this is now a WG item, any major change requires WG consensus. >>> Selecting >>> a marketing name, IMO, is such a major change. "Opus" is such a = flashy >>> name >>> that certain participants and/or companies conceivably may not like >>> it. For >>> example, if I were working for a company that has in its portfolio = an >>> audio >>> product named "Opus", I would object to the name change. So the = thing you >>> should have done, IMO, is to send an email to the list saying "The >>> editors >>> consider changing the name of our codec to Opus. Is that acceptable = to >>> the >>> WG?". >>> >>> It appears to me that twice you guys (chairs included) have taken >>> shortcuts >>> with the IETF's procedures, as I understand them. That, IMO, fills = up >>> your >>> quota for the next couple of years. Please be more conservative from = now >>> on. >>> >>> Regards, >>> Stephan >>> >>> >>> >>> >>> On 10.19.2010 08:31 , "Jean-Marc Valin" >>> wrote: >>> >>>> Hi, >>>> >>>> draft-ietf-codec-opus-00.txt is indeed a WG item. It has had = several >>>> names >>>> in the past, including draft-valin-codec-prototype and >>>> draft-valin-codec-definition, which may explain the confusion. This >>>> is the >>>> draft for which there was a hum during the last meeting. >>>> >>>> Jean-Marc >>>> >>>> On 10-10-19 11:26 AM, Stephan Wenger wrote: >>>>> Hi, >>>>> >>>>> I want to inquire the status of this draft. In many working = groups, the >>>>> filename "draft-ietf--xxx" indicates that the draft in >>>>> question is >>>>> a WG item of WG. Following this logic, it would appear = that >>>>> the >>>>> "opus" draft is now a WG item of the codec WG. I don't recall a >>>>> decision to >>>>> than extent. >>>>> >>>>> If the draft were indeed accepted as a WG item, I would like to >>>>> encourage >>>>> those who made IPR statements related to it, to resubmit those >>>>> statements >>>>> with the new filename. This would help those of us who are = searching >>>>> through the IETF IPR tracker by WG name (which is a very common >>>>> thing to do, >>>>> at least for me). >>>>> >>>>> Thanks, >>>>> Stephan >>>>> >>>>> >>>>> On 10.15.2010 13:30 , >>>>> "Internet-Drafts@ietf.org" >>>>> wrote: >>>>> >>>>>> A New Internet-Draft is available from the on-line = Internet-Drafts >>>>>> directories. >>>>>> This draft is a work item of the Internet Wideband Audio Codec >>>>>> Working Group >>>>>> of the IETF. >>>>>> >>>>>> >>>>>> Title : Definition of the Opus Audio Codec >>>>>> Author(s) : J. Valin, K. Vos >>>>>> Filename : draft-ietf-codec-opus-00.txt >>>>>> Pages : 12 >>>>>> Date : 2010-10-15 >>>>>> >>>>>> This document describes the Opus codec, designed for interactive >>>>>> speech and audio transmission over the Internet. >>>>>> >>>>>> A URL for this Internet-Draft is: >>>>>> http://www.ietf.org/internet-drafts/draft-ietf-codec-opus-00.txt >>>>>> >>>>>> Internet-Drafts are also available by anonymous FTP at: >>>>>> ftp://ftp.ietf.org/internet-drafts/ >>>>>> >>>>>> Below is the data which will enable a MIME compliant mail reader >>>>>> implementation to automatically retrieve the ASCII version of the >>>>>> Internet-Draft. >>>>>> _______________________________________________ >>>>>> codec mailing list >>>>>> codec@ietf.org >>>>>> https://www.ietf.org/mailman/listinfo/codec >>>>> >>>>> >>>>> _______________________________________________ >>>>> codec mailing list >>>>> codec@ietf.org >>>>> https://www.ietf.org/mailman/listinfo/codec >>>> >>> >>> >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >>> >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec > --=20 Jonathan D. Rosenberg, Ph.D. SkypeID: jdrosen Chief Technology Strategist Mobile: +1 (732) 766-2496 Skype SkypeIn: +1 (408) = 465-0361 jdrosen@skype.net http://www.skype.com jdrosen@jdrosen.net http://www.jdrosen.net From stephen.botzko@gmail.com Wed Oct 20 10:02:12 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id DE8E83A6935 for ; Wed, 20 Oct 2010 10:02:12 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.697 X-Spam-Level: X-Spam-Status: No, score=-2.697 tagged_above=-999 required=5 tests=[AWL=-0.099, BAYES_00=-2.599, HTML_MESSAGE=0.001] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id gees3FNJw-Ld for ; Wed, 20 Oct 2010 10:02:11 -0700 (PDT) Received: from mail-bw0-f44.google.com (mail-bw0-f44.google.com [209.85.214.44]) by core3.amsl.com (Postfix) with ESMTP id 383583A68F9 for ; Wed, 20 Oct 2010 10:02:11 -0700 (PDT) Received: by bwz14 with SMTP id 14so3050750bwz.31 for ; Wed, 20 Oct 2010 10:03:43 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type; bh=9C/CIwKmX9eUWO30WL7LCUnmtRPhtwS+IGmFXk5lYS0=; b=gSfhxzIZkLAclUkcLeX2BmKiD0EpqnWA/g3BfJAprC2PcTtG7PvX8PWwgoJquslzux k6Z/xJisB7jvn9NRCJQCM0Y/Q9c/DbewfiYMyRBoI7R9mwA739SNoFueShhcjp0vdEG7 2JgbHYEv5YiHkRfsRBRn2opxTI7Xsksgf5sp8= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=TIPm6Lp5TEjQ15oPrs0lX0lJxcbvJU3rYRZlO8f1j+m5+EYrvb8bqNJZzI3AfJTwtT +BrXaV0fRMXdgil5UGBt3un/t03loE5xY1o67yG1lHSFycoMUjQzrnhEep27dFWbkeAY Ty+lGfd9CQIcsZt0U70i7aVyDlQZnrbWMwXTI= MIME-Version: 1.0 Received: by 10.204.47.226 with SMTP id o34mr6980625bkf.191.1287594221985; Wed, 20 Oct 2010 10:03:41 -0700 (PDT) Received: by 10.204.80.213 with HTTP; Wed, 20 Oct 2010 10:03:41 -0700 (PDT) In-Reply-To: References: Date: Wed, 20 Oct 2010 13:03:41 -0400 Message-ID: From: Stephen Botzko To: Stephan Wenger Content-Type: multipart/alternative; boundary=0016e6db5c26375eeb04930f6460 Cc: codec@ietf.org Subject: Re: [codec] Seperate list for IPR discutions? X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 17:02:13 -0000 --0016e6db5c26375eeb04930f6460 Content-Type: text/plain; charset=ISO-8859-1 +1, I'd like to see it handled precisely this way. I think it provides a good way forward, and allows those of us who do not want exposure to patent details to continue to participate on the main list. Stephen Botzko On Wed, Oct 20, 2010 at 12:40 PM, Stephan Wenger wrote: > Hi, > > Yes, I think a separate mailing list would be an excellent idea. There is > IMO no danger to the IETF when using an IETF-hosted list. But, as you > mentioned, there are also other options. Whatever is the easiest for those > interested in this work. > > My preference would be that this list has a subscription process > independent > from the subscription to codec@ietf.org. As the minimum, there must be an > opt-out process. > > I would recommend that the maintainers of the list, periodically, make > people on codec@ietf.org aware of the existence of the list and the > ongoing > discussions, in a summarized form. I also don't mind if suggested design > choices spill over to codec@ietf; after all, this is the list where > technical discussions are being held. The borderline would be citations of > patent language, including patent numbers and claim language. > > Personally, I surely do not want to be involved on this hypothetical new > list. > > Stephan > > > On 10.20.2010 06:42 , "fluffy@cisco.com" wrote: > > > > > Stephan, as our WG technical advisor on topics like this, do you have > advice > > on if we should set up a separate list for people that want to talk about > > patent issues related to the standards from this WG? From a practical > point of > > view, someone could easily set up a separate list either on the ietf.org > > servers or not on ietf servers. > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > --0016e6db5c26375eeb04930f6460 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable +1, I'd like to see it handled precisely this way.=A0

I think i= t provides a good way forward, and allows those of us who do not want expos= ure to patent details to continue to participate on the main list.

Stephen Botzko


On Wed, Oct 20, 2010 a= t 12:40 PM, Stephan Wenger <stewe@stewe.org> wrote:
Hi,

Yes, I think a separate mailing list would be an excellent idea. =A0There i= s
IMO no danger to the IETF when using an IETF-hosted list. =A0But, as you mentioned, there are also other options. =A0Whatever is the easiest for tho= se
interested in this work.

My preference would be that this list has a subscription process independen= t
from the subscription to codec@ietf.org. =A0As the minimum, there must be an
opt-out process.

I would recommend that the maintainers of the list, periodically, make
people on codec@ietf.org aware of the= existence of the list and the ongoing
discussions, in a summarized form. =A0I also don't mind if suggested de= sign
choices spill over to codec@ietf; after all, this is the list where
technical discussions are being held. =A0The borderline would be citations = of
patent language, including patent numbers and claim language.

Personally, I surely do not want to be involved on this hypothetical new list.

Stephan

--0016e6db5c26375eeb04930f6460-- From jean-marc.valin@octasic.com Wed Oct 20 10:26:35 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id C33683A69A4 for ; Wed, 20 Oct 2010 10:26:35 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.436 X-Spam-Level: X-Spam-Status: No, score=-2.436 tagged_above=-999 required=5 tests=[AWL=0.163, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Ny3B0z8NpKEd for ; Wed, 20 Oct 2010 10:26:34 -0700 (PDT) Received: from toroondcbmts07-srv.bellnexxia.net (toroondcbmts07-srv.bellnexxia.net [207.236.237.41]) by core3.amsl.com (Postfix) with ESMTP id 519783A68F6 for ; Wed, 20 Oct 2010 10:26:34 -0700 (PDT) Received: from toip56-bus.srvr.bell.ca ([67.69.240.142]) by toroondcbmts07-srv.bellnexxia.net (InterMail vM.8.00.01.00 201-2244-105-20090324) with ESMTP id <20101020172806.PFWF26461.toroondcbmts07-srv.bellnexxia.net@toip56-bus.srvr.bell.ca>; Wed, 20 Oct 2010 13:28:06 -0400 Received: from toip53-bus.srvr.bell.ca ([67.69.240.54]) by toip56-bus.srvr.bell.ca with ESMTP; 20 Oct 2010 13:27:56 -0400 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: ApsEAN69vkxGNv5q/2dsb2JhbACiN79thUoEjU8G Received: from mail.octasic.com (HELO MAILEXCH.octasic.com) ([70.54.254.106]) by toip53-bus.srvr.bell.ca with ESMTP; 20 Oct 2010 13:27:55 -0400 Received: from [10.100.60.27] ([10.100.60.27]) by MAILEXCH.octasic.com with Microsoft SMTPSVC(6.0.3790.4675); Wed, 20 Oct 2010 13:27:55 -0400 Message-ID: <4CBF269B.8090609@octasic.com> Date: Wed, 20 Oct 2010 13:27:55 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.12) Gecko/20100915 Lightning/1.0b1 Thunderbird/3.0.8 MIME-Version: 1.0 To: Stephen Botzko References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 20 Oct 2010 17:27:55.0318 (UTC) FILETIME=[21A92560:01CB707C] Cc: codec@ietf.org Subject: Re: [codec] Seperate list for IPR discutions? X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 17:26:35 -0000 Hi, Just to make things clear, I think there's two sets of issues regarding IPR. One is discussing the exact terms of IPR disclosure/licensing and the other is discussion about patent themselves. I think the former should remain on this list. OTOH, I think the latter is (unfortunately) best dealt with in private. I don't think there is anything to be gained in discussing patent claims in a publicly archived list. I don't think even having that on a separate list is a good idea and I'm not interested in participating in such a list should it be created. Jean-Marc On 10-10-20 01:03 PM, Stephen Botzko wrote: > +1, I'd like to see it handled precisely this way. > > I think it provides a good way forward, and allows those of us who do > not want exposure to patent details to continue to participate on the > main list. > > Stephen Botzko > > > On Wed, Oct 20, 2010 at 12:40 PM, Stephan Wenger > wrote: > > Hi, > > Yes, I think a separate mailing list would be an excellent idea. > There is > IMO no danger to the IETF when using an IETF-hosted list. But, as you > mentioned, there are also other options. Whatever is the easiest > for those > interested in this work. > > My preference would be that this list has a subscription process > independent > from the subscription to codec@ietf.org . As > the minimum, there must be an > opt-out process. > > I would recommend that the maintainers of the list, periodically, make > people on codec@ietf.org aware of the > existence of the list and the ongoing > discussions, in a summarized form. I also don't mind if suggested > design > choices spill over to codec@ietf; after all, this is the list where > technical discussions are being held. The borderline would be > citations of > patent language, including patent numbers and claim language. > > Personally, I surely do not want to be involved on this hypothetical new > list. > > Stephan > > > On 10.20.2010 06:42 , "fluffy@cisco.com " > > wrote: > > > > > Stephan, as our WG technical advisor on topics like this, do you > have advice > > on if we should set up a separate list for people that want to > talk about > > patent issues related to the standards from this WG? From a > practical point of > > view, someone could easily set up a separate list either on the > ietf.org > > servers or not on ietf servers. > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From stephen.botzko@gmail.com Wed Oct 20 10:40:56 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id C319D3A68C8 for ; Wed, 20 Oct 2010 10:40:56 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.691 X-Spam-Level: X-Spam-Status: No, score=-2.691 tagged_above=-999 required=5 tests=[AWL=-0.093, BAYES_00=-2.599, HTML_MESSAGE=0.001] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id kiB2MM-GBBgi for ; Wed, 20 Oct 2010 10:40:55 -0700 (PDT) Received: from mail-bw0-f44.google.com (mail-bw0-f44.google.com [209.85.214.44]) by core3.amsl.com (Postfix) with ESMTP id 0A7F03A689C for ; Wed, 20 Oct 2010 10:40:54 -0700 (PDT) Received: by bwz14 with SMTP id 14so3087286bwz.31 for ; Wed, 20 Oct 2010 10:42:28 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type; bh=EBB5+FFCXOc6pPmd+e33EQzoS283EbzpNMeoMqkUbgM=; b=LM8+65j0J3oVpQO0bMXIwdTeClALelR+MjngiL4KskIMazA/SHVM7o7n9XYqClufXN ifPH9T54mAmAWc6/uuTSlmfZIyijpvNg1mE0RhJYO1gCHOqSiTdQ03FDNxmcSbd24UiB YVTfNjC46NVOiCjy+uyPhbpbdW0JV7FnH3fB4= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=c+kzIko0aJaHWvp3DnVwzcZ3ZUENW4NHKfD+e/XiS2vJWyZCzvOYvue2eyO7gESwWl c1hrd/Xg72v3L0ZhuVnKP6fM/AXMPIpdhQT1vX+VsM1TjD8Y2XIHvXUhYEZrNsO1RVo4 0MUxAw/ACHlocBhoHIZI//53pZCINNRCSzEgs= MIME-Version: 1.0 Received: by 10.204.68.10 with SMTP id t10mr6816412bki.77.1287596547756; Wed, 20 Oct 2010 10:42:27 -0700 (PDT) Received: by 10.204.80.213 with HTTP; Wed, 20 Oct 2010 10:42:27 -0700 (PDT) In-Reply-To: <4CBF269B.8090609@octasic.com> References: <4CBF269B.8090609@octasic.com> Date: Wed, 20 Oct 2010 13:42:27 -0400 Message-ID: From: Stephen Botzko To: Jean-Marc Valin Content-Type: multipart/alternative; boundary=001636b2bbebd7d45304930feede Cc: codec@ietf.org Subject: Re: [codec] Seperate list for IPR discutions? X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 17:40:56 -0000 --001636b2bbebd7d45304930feede Content-Type: text/plain; charset=ISO-8859-1 Stephan's suggestion on the boundary "citations of patent language, including patent numbers and claim language" is consistent with your boundary. Detailed analysis of proposed work-arounds, would presumably include patent language. General discussion of license terms would not. Since I wouldn't subscribe or otherwise access the new list, it doesn't matter to me if it is publicly archived or not. Based on past IETF experience, it is quite difficult to convince posters to go off-line for a private discussion. One practical issue is that the folks who do want to discuss this stuff have no easy way to identify themselves. I'm thinking that providing the second list might make that easier. Though perhaps no one but Christian is willing to talk about patent claims? Stephen Botzko On Wed, Oct 20, 2010 at 1:27 PM, Jean-Marc Valin < jean-marc.valin@octasic.com> wrote: > Hi, > > Just to make things clear, I think there's two sets of issues regarding > IPR. One is discussing the exact terms of IPR disclosure/licensing and the > other is discussion about patent themselves. I think the former should > remain on this list. OTOH, I think the latter is (unfortunately) best dealt > with in private. I don't think there is anything to be gained in discussing > patent claims in a publicly archived list. I don't think even having that on > a separate list is a good idea and I'm not interested in participating in > such a list should it be created. > > Jean-Marc > > > On 10-10-20 01:03 PM, Stephen Botzko wrote: > >> +1, I'd like to see it handled precisely this way. >> >> I think it provides a good way forward, and allows those of us who do >> not want exposure to patent details to continue to participate on the >> main list. >> >> Stephen Botzko >> >> >> On Wed, Oct 20, 2010 at 12:40 PM, Stephan Wenger > > wrote: >> >> Hi, >> >> Yes, I think a separate mailing list would be an excellent idea. >> There is >> IMO no danger to the IETF when using an IETF-hosted list. But, as you >> mentioned, there are also other options. Whatever is the easiest >> for those >> interested in this work. >> >> My preference would be that this list has a subscription process >> independent >> from the subscription to codec@ietf.org . As >> >> the minimum, there must be an >> opt-out process. >> >> I would recommend that the maintainers of the list, periodically, make >> people on codec@ietf.org aware of the >> >> existence of the list and the ongoing >> discussions, in a summarized form. I also don't mind if suggested >> design >> choices spill over to codec@ietf; after all, this is the list where >> technical discussions are being held. The borderline would be >> citations of >> patent language, including patent numbers and claim language. >> >> Personally, I surely do not want to be involved on this hypothetical >> new >> list. >> >> Stephan >> >> >> On 10.20.2010 06:42 , "fluffy@cisco.com " >> >> > wrote: >> >> > >> > Stephan, as our WG technical advisor on topics like this, do you >> have advice >> > on if we should set up a separate list for people that want to >> talk about >> > patent issues related to the standards from this WG? From a >> practical point of >> > view, someone could easily set up a separate list either on the >> ietf.org >> >> > servers or not on ietf servers. >> >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> >> https://www.ietf.org/mailman/listinfo/codec >> >> >> >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >> > > --001636b2bbebd7d45304930feede Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Stephan's suggestion on the boundary=A0 "citations of patent langu= age, including patent numbers and claim language" is consistent with y= our boundary.=A0 Detailed analysis of proposed work-arounds, would presumab= ly include patent language.=A0 General discussion of license terms would no= t.

Since I wouldn't subscribe or otherwise access the new list, it doe= sn't matter to me if it is publicly archived or not.=A0

Based o= n past IETF experience, it is quite difficult to convince posters to go off= -line for a private discussion.=A0 One practical issue is that the folks wh= o do want to discuss this stuff have no easy way to identify themselves.=A0= I'm thinking that providing the second list might make that easier.=A0= Though perhaps no one but Christian is willing to talk about patent claims= ?

Stephen Botzko

On Wed, Oct 20, 2010 a= t 1:27 PM, Jean-Marc Valin <jean-marc.valin@octasic.com> wrote:
<= blockquote class=3D"gmail_quote" style=3D"margin: 0pt 0pt 0pt 0.8ex; border= -left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"> Hi,

Just to make things clear, I think there's two sets of issues regarding= IPR. One is discussing the exact terms of IPR disclosure/licensing and the= other is discussion about patent themselves. I think the former should rem= ain on this list. OTOH, I think the latter is (unfortunately) best dealt wi= th in private. I don't think there is anything to be gained in discussi= ng patent claims in a publicly archived list. I don't think even having= that on a separate list is a good idea and I'm not interested in parti= cipating in such a list should it be created.

=A0 =A0 Jean-Marc


On 10-10-20 01:03 PM, Stephen Botzko wrote:
+1, I'd like to see it handled precisely this way.

I think it provides a good way forward, and allows those of us who do
not want exposure to patent details to continue to participate on the
main list.

Stephen Botzko


On Wed, Oct 20, 2010 at 12:40 PM, Stephan Wenger <stewe@stewe.org
<mailto:stewe@stewe= .org>> wrote:

=A0 =A0Hi,

=A0 =A0Yes, I think a separate mailing list would be an excellent idea. =A0 =A0 =A0There is
=A0 =A0IMO no danger to the IETF when using an IETF-hosted list. =A0But, a= s you
=A0 =A0mentioned, there are also other options. =A0Whatever is the easiest=
=A0 =A0for those
=A0 =A0interested in this work.

=A0 =A0My preference would be that this list has a subscription process =A0 =A0independent
=A0 =A0from the subscription to codec@ietf.org <mailto:codec@ietf.org>. =A0As

=A0 =A0the minimum, there must be an
=A0 =A0opt-out process.

=A0 =A0I would recommend that the maintainers of the list, periodically, m= ake
=A0 =A0people on codec= @ietf.org <mailto:codec@ietf.org> aware of the

=A0 =A0existence of the list and the ongoing
=A0 =A0discussions, in a summarized form. =A0I also don't mind if sugg= ested
=A0 =A0design
=A0 =A0choices spill over to codec@ietf; after all, this is the list where=
=A0 =A0technical discussions are being held. =A0The borderline would be =A0 =A0citations of
=A0 =A0patent language, including patent numbers and claim language.

=A0 =A0Personally, I surely do not want to be involved on this hypothetica= l new
=A0 =A0list.

=A0 =A0Stephan


=A0 =A0On 10.20.2010 06:42 , "fluffy@cisco.com <mailto:fluffy@cisco.com>"
<= br> =A0 =A0<fluffy@ci= sco.com <mailto:fluffy@cisco.com>> wrote:

=A0 =A0 >
=A0 =A0 > Stephan, as our WG technical advisor on topics like this, do = you
=A0 =A0have advice
=A0 =A0 > on if we should set up a separate list for people that want t= o
=A0 =A0talk about
=A0 =A0 > patent issues related to the standards from this WG? From a =A0 =A0practical point of
=A0 =A0 > view, someone could easily set up a separate list either on t= he
=A0 =A0ietf.org <http://ietf.org>

=A0 =A0 > servers or not on ietf servers.


=A0 =A0_______________________________________________
=A0 =A0codec mailing list
=A0 =A0codec@ietf.org<= /a> <mailto:codec@ie= tf.org>

=A0 =A0https://www.ietf.org/mailman/listinfo/codec




_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec


--001636b2bbebd7d45304930feede-- From gmaxwell@juniper.net Wed Oct 20 10:49:14 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 4460C3A68FA for ; Wed, 20 Oct 2010 10:49:14 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -5.884 X-Spam-Level: X-Spam-Status: No, score=-5.884 tagged_above=-999 required=5 tests=[AWL=0.214, BAYES_00=-2.599, HTML_MESSAGE=0.001, MIME_BAD_LINEBREAK=0.5, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id YMavQFeXdItl for ; Wed, 20 Oct 2010 10:49:12 -0700 (PDT) Received: from exprod7og119.obsmtp.com (exprod7og119.obsmtp.com [64.18.2.16]) by core3.amsl.com (Postfix) with ESMTP id 4879C3A693C for ; Wed, 20 Oct 2010 10:49:09 -0700 (PDT) Received: from source ([66.129.224.36]) (using TLSv1) by exprod7ob119.postini.com ([64.18.6.12]) with SMTP ID DSNKTL8r7EEDkiGscUiuFSFLWttv6nVyD9Ia@postini.com; Wed, 20 Oct 2010 10:50:43 PDT Received: from EMBX01-HQ.jnpr.net ([fe80::c821:7c81:f21f:8bc7]) by P-EMHUB01-HQ.jnpr.net ([fe80::fc92:eb1:759:2c72%11]) with mapi; Wed, 20 Oct 2010 10:50:00 -0700 From: Gregory Maxwell To: "'stephen.botzko@gmail.com'" , "'jean-marc.valin@octasic.com'" Date: Wed, 20 Oct 2010 10:50:00 -0700 Thread-Topic: [codec] Seperate list for IPR discutions? Thread-Index: Actwfi08ukuKx/BjSx+Q0cCrHKI+EgAAQdGq Message-ID: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US Content-Type: multipart/alternative; boundary="_000_BCB3F026FAC4C145A4A3330806FEFDA93B74D97C49EMBX01HQjnprn_" MIME-Version: 1.0 Cc: "'codec@ietf.org'" Subject: Re: [codec] Seperate list for IPR discutions? X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 17:49:14 -0000 --_000_BCB3F026FAC4C145A4A3330806FEFDA93B74D97C49EMBX01HQjnprn_ Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 DQpUaGVyZSBpcywgaGFzIGJlZW4sIGFuZCB3aWxsIGJlIHBsZW50eSBvZiBkaXNjdXNzaW9uIGlu IF9wcml2YXRlXy4gQmVpbmcgX3dpbGxmdWxseV8gaWdub3JhbnQgb2YgeW91ciBvYmxpZ2F0aW9u cyB3aGljaCB5b3Ugd291bGQgb3RoZXJ3aXNlIGtub3cgYWJvdXQgaXMganVzdCB3aWxsZnVsbHkg aW5mcmluZ2luZyBpbiBhIHByZXR0eSBkcmVzcy4NCg0KVGhlIGdvYWwgb2YgdGhlIGRldmVsb3Bl cnMgaGVyZSBpcyB0byBub3QgaW5mcmluZ2Ugb24gdGhpcmQgcGFydCBwYXRlbnRzLiBGdWxsIHN0 b3AuDQoNClRoZSBvYmplY3Rpb24gaXMgdG8gYSBwdWJsaWMgZGlzY3Vzc2lvbi0gYmVjYXVzZSBw YXRlbnQgZGlzY3Vzc2lvbiBtdXN0IGJlIGluZm9ybWVkIGJ5IHRoZSByZWxldmFudCBleHBlcnRz LCBiZWNhdXNlIGl0IG5vIG9uZSB3YW50cyB0byBiZSBkcmFnZ2VkIGluIHRvIHRlc3RpZnkgb24g cmFuZG9tIHBhdGVudHMsIGFuZCBiZWNhdXNlIGV4cG9zaW5nIHlvdXIgc3RyYXRlZ3kgdG8gYSBw b3NzaWJsZSBhdHRhY2tlciBtYWtlcyBpdCB3ZWFrZXIsIGV0Yy4NCg0KDQoNCg0KDQpfX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fXw0KRnJvbTogY29kZWMtYm91bmNlc0BpZXRmLm9yZyA8 Y29kZWMtYm91bmNlc0BpZXRmLm9yZz4NClRvOiBKZWFuLU1hcmMgVmFsaW4gPGplYW4tbWFyYy52 YWxpbkBvY3Rhc2ljLmNvbT4NCkNjOiBjb2RlY0BpZXRmLm9yZyA8Y29kZWNAaWV0Zi5vcmc+DQpT ZW50OiBXZWQgT2N0IDIwIDEwOjQyOjI3IDIwMTANClN1YmplY3Q6IFJlOiBbY29kZWNdIFNlcGVy YXRlIGxpc3QgZm9yIElQUiBkaXNjdXRpb25zPw0KDQpTdGVwaGFuJ3Mgc3VnZ2VzdGlvbiBvbiB0 aGUgYm91bmRhcnkgICJjaXRhdGlvbnMgb2YgcGF0ZW50IGxhbmd1YWdlLCBpbmNsdWRpbmcgcGF0 ZW50IG51bWJlcnMgYW5kIGNsYWltIGxhbmd1YWdlIiBpcyBjb25zaXN0ZW50IHdpdGggeW91ciBi b3VuZGFyeS4gIERldGFpbGVkIGFuYWx5c2lzIG9mIHByb3Bvc2VkIHdvcmstYXJvdW5kcywgd291 bGQgcHJlc3VtYWJseSBpbmNsdWRlIHBhdGVudCBsYW5ndWFnZS4gIEdlbmVyYWwgZGlzY3Vzc2lv biBvZiBsaWNlbnNlIHRlcm1zIHdvdWxkIG5vdC4NCg0KU2luY2UgSSB3b3VsZG4ndCBzdWJzY3Jp YmUgb3Igb3RoZXJ3aXNlIGFjY2VzcyB0aGUgbmV3IGxpc3QsIGl0IGRvZXNuJ3QgbWF0dGVyIHRv IG1lIGlmIGl0IGlzIHB1YmxpY2x5IGFyY2hpdmVkIG9yIG5vdC4NCg0KQmFzZWQgb24gcGFzdCBJ RVRGIGV4cGVyaWVuY2UsIGl0IGlzIHF1aXRlIGRpZmZpY3VsdCB0byBjb252aW5jZSBwb3N0ZXJz IHRvIGdvIG9mZi1saW5lIGZvciBhIHByaXZhdGUgZGlzY3Vzc2lvbi4gIE9uZSBwcmFjdGljYWwg aXNzdWUgaXMgdGhhdCB0aGUgZm9sa3Mgd2hvIGRvIHdhbnQgdG8gZGlzY3VzcyB0aGlzIHN0dWZm IGhhdmUgbm8gZWFzeSB3YXkgdG8gaWRlbnRpZnkgdGhlbXNlbHZlcy4gIEknbSB0aGlua2luZyB0 aGF0IHByb3ZpZGluZyB0aGUgc2Vjb25kIGxpc3QgbWlnaHQgbWFrZSB0aGF0IGVhc2llci4gIFRo b3VnaCBwZXJoYXBzIG5vIG9uZSBidXQgQ2hyaXN0aWFuIGlzIHdpbGxpbmcgdG8gdGFsayBhYm91 dCBwYXRlbnQgY2xhaW1zPw0KDQpTdGVwaGVuIEJvdHprbw0KDQpPbiBXZWQsIE9jdCAyMCwgMjAx MCBhdCAxOjI3IFBNLCBKZWFuLU1hcmMgVmFsaW4gPGplYW4tbWFyYy52YWxpbkBvY3Rhc2ljLmNv bTxtYWlsdG86amVhbi1tYXJjLnZhbGluQG9jdGFzaWMuY29tPj4gd3JvdGU6DQpIaSwNCg0KSnVz dCB0byBtYWtlIHRoaW5ncyBjbGVhciwgSSB0aGluayB0aGVyZSdzIHR3byBzZXRzIG9mIGlzc3Vl cyByZWdhcmRpbmcgSVBSLiBPbmUgaXMgZGlzY3Vzc2luZyB0aGUgZXhhY3QgdGVybXMgb2YgSVBS IGRpc2Nsb3N1cmUvbGljZW5zaW5nIGFuZCB0aGUgb3RoZXIgaXMgZGlzY3Vzc2lvbiBhYm91dCBw YXRlbnQgdGhlbXNlbHZlcy4gSSB0aGluayB0aGUgZm9ybWVyIHNob3VsZCByZW1haW4gb24gdGhp cyBsaXN0LiBPVE9ILCBJIHRoaW5rIHRoZSBsYXR0ZXIgaXMgKHVuZm9ydHVuYXRlbHkpIGJlc3Qg ZGVhbHQgd2l0aCBpbiBwcml2YXRlLiBJIGRvbid0IHRoaW5rIHRoZXJlIGlzIGFueXRoaW5nIHRv IGJlIGdhaW5lZCBpbiBkaXNjdXNzaW5nIHBhdGVudCBjbGFpbXMgaW4gYSBwdWJsaWNseSBhcmNo aXZlZCBsaXN0LiBJIGRvbid0IHRoaW5rIGV2ZW4gaGF2aW5nIHRoYXQgb24gYSBzZXBhcmF0ZSBs aXN0IGlzIGEgZ29vZCBpZGVhIGFuZCBJJ20gbm90IGludGVyZXN0ZWQgaW4gcGFydGljaXBhdGlu ZyBpbiBzdWNoIGEgbGlzdCBzaG91bGQgaXQgYmUgY3JlYXRlZC4NCg0KICAgIEplYW4tTWFyYw0K DQoNCk9uIDEwLTEwLTIwIDAxOjAzIFBNLCBTdGVwaGVuIEJvdHprbyB3cm90ZToNCisxLCBJJ2Qg bGlrZSB0byBzZWUgaXQgaGFuZGxlZCBwcmVjaXNlbHkgdGhpcyB3YXkuDQoNCkkgdGhpbmsgaXQg cHJvdmlkZXMgYSBnb29kIHdheSBmb3J3YXJkLCBhbmQgYWxsb3dzIHRob3NlIG9mIHVzIHdobyBk bw0Kbm90IHdhbnQgZXhwb3N1cmUgdG8gcGF0ZW50IGRldGFpbHMgdG8gY29udGludWUgdG8gcGFy dGljaXBhdGUgb24gdGhlDQptYWluIGxpc3QuDQoNClN0ZXBoZW4gQm90emtvDQoNCg0KT24gV2Vk LCBPY3QgMjAsIDIwMTAgYXQgMTI6NDAgUE0sIFN0ZXBoYW4gV2VuZ2VyIDxzdGV3ZUBzdGV3ZS5v cmc8bWFpbHRvOnN0ZXdlQHN0ZXdlLm9yZz4NCjxtYWlsdG86c3Rld2VAc3Rld2Uub3JnPG1haWx0 bzpzdGV3ZUBzdGV3ZS5vcmc+Pj4gd3JvdGU6DQoNCiAgIEhpLA0KDQogICBZZXMsIEkgdGhpbmsg YSBzZXBhcmF0ZSBtYWlsaW5nIGxpc3Qgd291bGQgYmUgYW4gZXhjZWxsZW50IGlkZWEuDQogICAg IFRoZXJlIGlzDQogICBJTU8gbm8gZGFuZ2VyIHRvIHRoZSBJRVRGIHdoZW4gdXNpbmcgYW4gSUVU Ri1ob3N0ZWQgbGlzdC4gIEJ1dCwgYXMgeW91DQogICBtZW50aW9uZWQsIHRoZXJlIGFyZSBhbHNv IG90aGVyIG9wdGlvbnMuICBXaGF0ZXZlciBpcyB0aGUgZWFzaWVzdA0KICAgZm9yIHRob3NlDQog ICBpbnRlcmVzdGVkIGluIHRoaXMgd29yay4NCg0KICAgTXkgcHJlZmVyZW5jZSB3b3VsZCBiZSB0 aGF0IHRoaXMgbGlzdCBoYXMgYSBzdWJzY3JpcHRpb24gcHJvY2Vzcw0KICAgaW5kZXBlbmRlbnQN CiAgIGZyb20gdGhlIHN1YnNjcmlwdGlvbiB0byBjb2RlY0BpZXRmLm9yZzxtYWlsdG86Y29kZWNA aWV0Zi5vcmc+IDxtYWlsdG86Y29kZWNAaWV0Zi5vcmc8bWFpbHRvOmNvZGVjQGlldGYub3JnPj4u ICBBcw0KDQogICB0aGUgbWluaW11bSwgdGhlcmUgbXVzdCBiZSBhbg0KICAgb3B0LW91dCBwcm9j ZXNzLg0KDQogICBJIHdvdWxkIHJlY29tbWVuZCB0aGF0IHRoZSBtYWludGFpbmVycyBvZiB0aGUg bGlzdCwgcGVyaW9kaWNhbGx5LCBtYWtlDQogICBwZW9wbGUgb24gY29kZWNAaWV0Zi5vcmc8bWFp bHRvOmNvZGVjQGlldGYub3JnPiA8bWFpbHRvOmNvZGVjQGlldGYub3JnPG1haWx0bzpjb2RlY0Bp ZXRmLm9yZz4+IGF3YXJlIG9mIHRoZQ0KDQogICBleGlzdGVuY2Ugb2YgdGhlIGxpc3QgYW5kIHRo ZSBvbmdvaW5nDQogICBkaXNjdXNzaW9ucywgaW4gYSBzdW1tYXJpemVkIGZvcm0uICBJIGFsc28g ZG9uJ3QgbWluZCBpZiBzdWdnZXN0ZWQNCiAgIGRlc2lnbg0KICAgY2hvaWNlcyBzcGlsbCBvdmVy IHRvIGNvZGVjQGlldGY7IGFmdGVyIGFsbCwgdGhpcyBpcyB0aGUgbGlzdCB3aGVyZQ0KICAgdGVj aG5pY2FsIGRpc2N1c3Npb25zIGFyZSBiZWluZyBoZWxkLiAgVGhlIGJvcmRlcmxpbmUgd291bGQg YmUNCiAgIGNpdGF0aW9ucyBvZg0KICAgcGF0ZW50IGxhbmd1YWdlLCBpbmNsdWRpbmcgcGF0ZW50 IG51bWJlcnMgYW5kIGNsYWltIGxhbmd1YWdlLg0KDQogICBQZXJzb25hbGx5LCBJIHN1cmVseSBk byBub3Qgd2FudCB0byBiZSBpbnZvbHZlZCBvbiB0aGlzIGh5cG90aGV0aWNhbCBuZXcNCiAgIGxp c3QuDQoNCiAgIFN0ZXBoYW4NCg0KDQogICBPbiAxMC4yMC4yMDEwIDA2OjQyICwgImZsdWZmeUBj aXNjby5jb208bWFpbHRvOmZsdWZmeUBjaXNjby5jb20+IDxtYWlsdG86Zmx1ZmZ5QGNpc2NvLmNv bTxtYWlsdG86Zmx1ZmZ5QGNpc2NvLmNvbT4+Ig0KDQogICA8Zmx1ZmZ5QGNpc2NvLmNvbTxtYWls dG86Zmx1ZmZ5QGNpc2NvLmNvbT4gPG1haWx0bzpmbHVmZnlAY2lzY28uY29tPG1haWx0bzpmbHVm ZnlAY2lzY28uY29tPj4+IHdyb3RlOg0KDQogICAgPg0KICAgID4gU3RlcGhhbiwgYXMgb3VyIFdH IHRlY2huaWNhbCBhZHZpc29yIG9uIHRvcGljcyBsaWtlIHRoaXMsIGRvIHlvdQ0KICAgaGF2ZSBh ZHZpY2UNCiAgICA+IG9uIGlmIHdlIHNob3VsZCBzZXQgdXAgYSBzZXBhcmF0ZSBsaXN0IGZvciBw ZW9wbGUgdGhhdCB3YW50IHRvDQogICB0YWxrIGFib3V0DQogICAgPiBwYXRlbnQgaXNzdWVzIHJl bGF0ZWQgdG8gdGhlIHN0YW5kYXJkcyBmcm9tIHRoaXMgV0c/IEZyb20gYQ0KICAgcHJhY3RpY2Fs IHBvaW50IG9mDQogICAgPiB2aWV3LCBzb21lb25lIGNvdWxkIGVhc2lseSBzZXQgdXAgYSBzZXBh cmF0ZSBsaXN0IGVpdGhlciBvbiB0aGUNCiAgIGlldGYub3JnPGh0dHA6Ly9pZXRmLm9yZz4gPGh0 dHA6Ly9pZXRmLm9yZz4NCg0KICAgID4gc2VydmVycyBvciBub3Qgb24gaWV0ZiBzZXJ2ZXJzLg0K DQoNCiAgIF9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fDQog ICBjb2RlYyBtYWlsaW5nIGxpc3QNCiAgIGNvZGVjQGlldGYub3JnPG1haWx0bzpjb2RlY0BpZXRm Lm9yZz4gPG1haWx0bzpjb2RlY0BpZXRmLm9yZzxtYWlsdG86Y29kZWNAaWV0Zi5vcmc+Pg0KDQog ICBodHRwczovL3d3dy5pZXRmLm9yZy9tYWlsbWFuL2xpc3RpbmZvL2NvZGVjDQoNCg0KDQoNCl9f X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fDQpjb2RlYyBtYWls aW5nIGxpc3QNCmNvZGVjQGlldGYub3JnPG1haWx0bzpjb2RlY0BpZXRmLm9yZz4NCmh0dHBzOi8v d3d3LmlldGYub3JnL21haWxtYW4vbGlzdGluZm8vY29kZWMNCg0KDQo= --_000_BCB3F026FAC4C145A4A3330806FEFDA93B74D97C49EMBX01HQjnprn_ Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: base64 PHA+PGZvbnQgc2l6ZT0yIGNvbG9yPW5hdnkgZmFjZT1BcmlhbD4NCjxicj5UaGVyZSBpcywgaGFz IGJlZW4sIGFuZCB3aWxsIGJlIHBsZW50eSBvZiBkaXNjdXNzaW9uIGluIF9wcml2YXRlXy4gICBC ZWluZyBfd2lsbGZ1bGx5XyBpZ25vcmFudCBvZiB5b3VyIG9ibGlnYXRpb25zIHdoaWNoIHlvdSB3 b3VsZCBvdGhlcndpc2Uga25vdyBhYm91dCBpcyBqdXN0IHdpbGxmdWxseSBpbmZyaW5naW5nIGlu IGEgcHJldHR5IGRyZXNzLjxicj48YnI+VGhlIGdvYWwgb2YgdGhlIGRldmVsb3BlcnMgaGVyZSBp cyB0byBub3QgaW5mcmluZ2Ugb24gdGhpcmQgcGFydCBwYXRlbnRzLiBGdWxsIHN0b3AuPGJyPjxi cj5UaGUgb2JqZWN0aW9uIGlzIHRvIGEgcHVibGljIGRpc2N1c3Npb24tIGJlY2F1c2UgcGF0ZW50 IGRpc2N1c3Npb24gbXVzdCBiZSBpbmZvcm1lZCBieSB0aGUgcmVsZXZhbnQgZXhwZXJ0cywgYmVj YXVzZSBpdCBubyBvbmUgd2FudHMgdG8gYmUgZHJhZ2dlZCBpbiB0byB0ZXN0aWZ5IG9uIHJhbmRv bSBwYXRlbnRzLCBhbmQgYmVjYXVzZSBleHBvc2luZyB5b3VyIHN0cmF0ZWd5IHRvIGEgcG9zc2li bGUgYXR0YWNrZXIgbWFrZXMgaXQgd2Vha2VyLCBldGMuPGJyPjxicj48YnI+DTxicj4NPGJyPjwv Zm9udD48L3A+DQo8cD48aHIgc2l6ZT0yIHdpZHRoPSIxMDAlIiBhbGlnbj1jZW50ZXIgdGFiaW5k ZXg9LTE+DQo8Zm9udCBmYWNlPVRhaG9tYSBzaXplPTI+DQo8Yj5Gcm9tPC9iPjogY29kZWMtYm91 bmNlc0BpZXRmLm9yZyAmbHQ7Y29kZWMtYm91bmNlc0BpZXRmLm9yZyZndDsNPGJyPjxiPlRvPC9i PjogSmVhbi1NYXJjIFZhbGluICZsdDtqZWFuLW1hcmMudmFsaW5Ab2N0YXNpYy5jb20mZ3Q7DTxi cj48Yj5DYzwvYj46IGNvZGVjQGlldGYub3JnICZsdDtjb2RlY0BpZXRmLm9yZyZndDsNPGJyPjxi PlNlbnQ8L2I+OiBXZWQgT2N0IDIwIDEwOjQyOjI3IDIwMTA8YnI+PGI+U3ViamVjdDwvYj46IFJl OiBbY29kZWNdIFNlcGVyYXRlIGxpc3QgZm9yIElQUiBkaXNjdXRpb25zPw08YnI+PC9mb250Pjwv cD4NClN0ZXBoYW4mIzM5O3Mgc3VnZ2VzdGlvbiBvbiB0aGUgYm91bmRhcnnCoCAmcXVvdDtjaXRh dGlvbnMgb2YgcGF0ZW50IGxhbmd1YWdlLCBpbmNsdWRpbmcgcGF0ZW50IG51bWJlcnMgYW5kIGNs YWltIGxhbmd1YWdlJnF1b3Q7IGlzIGNvbnNpc3RlbnQgd2l0aCB5b3VyIGJvdW5kYXJ5LsKgIERl dGFpbGVkIGFuYWx5c2lzIG9mIHByb3Bvc2VkIHdvcmstYXJvdW5kcywgd291bGQgcHJlc3VtYWJs eSBpbmNsdWRlIHBhdGVudCBsYW5ndWFnZS7CoCBHZW5lcmFsIGRpc2N1c3Npb24gb2YgbGljZW5z ZSB0ZXJtcyB3b3VsZCBub3QuPGJyPg0KPGJyPlNpbmNlIEkgd291bGRuJiMzOTt0IHN1YnNjcmli ZSBvciBvdGhlcndpc2UgYWNjZXNzIHRoZSBuZXcgbGlzdCwgaXQgZG9lc24mIzM5O3QgbWF0dGVy IHRvIG1lIGlmIGl0IGlzIHB1YmxpY2x5IGFyY2hpdmVkIG9yIG5vdC7CoCA8YnI+PGJyPkJhc2Vk IG9uIHBhc3QgSUVURiBleHBlcmllbmNlLCBpdCBpcyBxdWl0ZSBkaWZmaWN1bHQgdG8gY29udmlu Y2UgcG9zdGVycyB0byBnbyBvZmYtbGluZSBmb3IgYSBwcml2YXRlIGRpc2N1c3Npb24uwqAgT25l IHByYWN0aWNhbCBpc3N1ZSBpcyB0aGF0IHRoZSBmb2xrcyB3aG8gZG8gd2FudCB0byBkaXNjdXNz IHRoaXMgc3R1ZmYgaGF2ZSBubyBlYXN5IHdheSB0byBpZGVudGlmeSB0aGVtc2VsdmVzLsKgIEkm IzM5O20gdGhpbmtpbmcgdGhhdCBwcm92aWRpbmcgdGhlIHNlY29uZCBsaXN0IG1pZ2h0IG1ha2Ug dGhhdCBlYXNpZXIuwqAgVGhvdWdoIHBlcmhhcHMgbm8gb25lIGJ1dCBDaHJpc3RpYW4gaXMgd2ls bGluZyB0byB0YWxrIGFib3V0IHBhdGVudCBjbGFpbXM/PGJyPg0KPGJyPlN0ZXBoZW4gQm90emtv PGJyPjxicj48ZGl2IGNsYXNzPSJnbWFpbF9xdW90ZSI+T24gV2VkLCBPY3QgMjAsIDIwMTAgYXQg MToyNyBQTSwgSmVhbi1NYXJjIFZhbGluIDxzcGFuIGRpcj0ibHRyIj4mbHQ7PGEgaHJlZj0ibWFp bHRvOmplYW4tbWFyYy52YWxpbkBvY3Rhc2ljLmNvbSI+amVhbi1tYXJjLnZhbGluQG9jdGFzaWMu Y29tPC9hPiZndDs8L3NwYW4+IHdyb3RlOjxicj48YmxvY2txdW90ZSBjbGFzcz0iZ21haWxfcXVv dGUiIHN0eWxlPSJtYXJnaW46IDBwdCAwcHQgMHB0IDAuOGV4OyBib3JkZXItbGVmdDogMXB4IHNv bGlkIHJnYigyMDQsIDIwNCwgMjA0KTsgcGFkZGluZy1sZWZ0OiAxZXg7Ij4NCkhpLDxicj4NCjxi cj4NCkp1c3QgdG8gbWFrZSB0aGluZ3MgY2xlYXIsIEkgdGhpbmsgdGhlcmUmIzM5O3MgdHdvIHNl dHMgb2YgaXNzdWVzIHJlZ2FyZGluZyBJUFIuIE9uZSBpcyBkaXNjdXNzaW5nIHRoZSBleGFjdCB0 ZXJtcyBvZiBJUFIgZGlzY2xvc3VyZS9saWNlbnNpbmcgYW5kIHRoZSBvdGhlciBpcyBkaXNjdXNz aW9uIGFib3V0IHBhdGVudCB0aGVtc2VsdmVzLiBJIHRoaW5rIHRoZSBmb3JtZXIgc2hvdWxkIHJl bWFpbiBvbiB0aGlzIGxpc3QuIE9UT0gsIEkgdGhpbmsgdGhlIGxhdHRlciBpcyAodW5mb3J0dW5h dGVseSkgYmVzdCBkZWFsdCB3aXRoIGluIHByaXZhdGUuIEkgZG9uJiMzOTt0IHRoaW5rIHRoZXJl IGlzIGFueXRoaW5nIHRvIGJlIGdhaW5lZCBpbiBkaXNjdXNzaW5nIHBhdGVudCBjbGFpbXMgaW4g YSBwdWJsaWNseSBhcmNoaXZlZCBsaXN0LiBJIGRvbiYjMzk7dCB0aGluayBldmVuIGhhdmluZyB0 aGF0IG9uIGEgc2VwYXJhdGUgbGlzdCBpcyBhIGdvb2QgaWRlYSBhbmQgSSYjMzk7bSBub3QgaW50 ZXJlc3RlZCBpbiBwYXJ0aWNpcGF0aW5nIGluIHN1Y2ggYSBsaXN0IHNob3VsZCBpdCBiZSBjcmVh dGVkLjxicj4NCg0KPGJyPg0KIMKgIMKgIEplYW4tTWFyYzxkaXYgY2xhc3M9ImltIj48YnI+DQo8 YnI+DQpPbiAxMC0xMC0yMCAwMTowMyBQTSwgU3RlcGhlbiBCb3R6a28gd3JvdGU6PGJyPg0KPC9k aXY+PGJsb2NrcXVvdGUgY2xhc3M9ImdtYWlsX3F1b3RlIiBzdHlsZT0ibWFyZ2luOiAwcHQgMHB0 IDBwdCAwLjhleDsgYm9yZGVyLWxlZnQ6IDFweCBzb2xpZCByZ2IoMjA0LCAyMDQsIDIwNCk7IHBh ZGRpbmctbGVmdDogMWV4OyI+PGRpdiBjbGFzcz0iaW0iPg0KKzEsIEkmIzM5O2QgbGlrZSB0byBz ZWUgaXQgaGFuZGxlZCBwcmVjaXNlbHkgdGhpcyB3YXkuPGJyPg0KPGJyPg0KSSB0aGluayBpdCBw cm92aWRlcyBhIGdvb2Qgd2F5IGZvcndhcmQsIGFuZCBhbGxvd3MgdGhvc2Ugb2YgdXMgd2hvIGRv PGJyPg0Kbm90IHdhbnQgZXhwb3N1cmUgdG8gcGF0ZW50IGRldGFpbHMgdG8gY29udGludWUgdG8g cGFydGljaXBhdGUgb24gdGhlPGJyPg0KbWFpbiBsaXN0Ljxicj4NCjxicj4NClN0ZXBoZW4gQm90 emtvPGJyPg0KPGJyPg0KPGJyPg0KT24gV2VkLCBPY3QgMjAsIDIwMTAgYXQgMTI6NDAgUE0sIFN0 ZXBoYW4gV2VuZ2VyICZsdDs8YSBocmVmPSJtYWlsdG86c3Rld2VAc3Rld2Uub3JnIiB0YXJnZXQ9 Il9ibGFuayI+c3Rld2VAc3Rld2Uub3JnPC9hPjxicj48L2Rpdj48ZGl2IGNsYXNzPSJpbSI+DQom bHQ7bWFpbHRvOjxhIGhyZWY9Im1haWx0bzpzdGV3ZUBzdGV3ZS5vcmciIHRhcmdldD0iX2JsYW5r Ij5zdGV3ZUBzdGV3ZS5vcmc8L2E+Jmd0OyZndDsgd3JvdGU6PGJyPg0KPGJyPg0KIMKgIMKgSGks PGJyPg0KPGJyPg0KIMKgIMKgWWVzLCBJIHRoaW5rIGEgc2VwYXJhdGUgbWFpbGluZyBsaXN0IHdv dWxkIGJlIGFuIGV4Y2VsbGVudCBpZGVhLjxicj4NCiDCoCDCoCDCoFRoZXJlIGlzPGJyPg0KIMKg IMKgSU1PIG5vIGRhbmdlciB0byB0aGUgSUVURiB3aGVuIHVzaW5nIGFuIElFVEYtaG9zdGVkIGxp c3QuIMKgQnV0LCBhcyB5b3U8YnI+DQogwqAgwqBtZW50aW9uZWQsIHRoZXJlIGFyZSBhbHNvIG90 aGVyIG9wdGlvbnMuIMKgV2hhdGV2ZXIgaXMgdGhlIGVhc2llc3Q8YnI+DQogwqAgwqBmb3IgdGhv c2U8YnI+DQogwqAgwqBpbnRlcmVzdGVkIGluIHRoaXMgd29yay48YnI+DQo8YnI+DQogwqAgwqBN eSBwcmVmZXJlbmNlIHdvdWxkIGJlIHRoYXQgdGhpcyBsaXN0IGhhcyBhIHN1YnNjcmlwdGlvbiBw cm9jZXNzPGJyPg0KIMKgIMKgaW5kZXBlbmRlbnQ8YnI+PC9kaXY+DQogwqAgwqBmcm9tIHRoZSBz dWJzY3JpcHRpb24gdG8gPGEgaHJlZj0ibWFpbHRvOmNvZGVjQGlldGYub3JnIiB0YXJnZXQ9Il9i bGFuayI+Y29kZWNAaWV0Zi5vcmc8L2E+ICZsdDttYWlsdG86PGEgaHJlZj0ibWFpbHRvOmNvZGVj QGlldGYub3JnIiB0YXJnZXQ9Il9ibGFuayI+Y29kZWNAaWV0Zi5vcmc8L2E+Jmd0Oy4gwqBBczxk aXYgY2xhc3M9ImltIj48YnI+DQogwqAgwqB0aGUgbWluaW11bSwgdGhlcmUgbXVzdCBiZSBhbjxi cj4NCiDCoCDCoG9wdC1vdXQgcHJvY2Vzcy48YnI+DQo8YnI+DQogwqAgwqBJIHdvdWxkIHJlY29t bWVuZCB0aGF0IHRoZSBtYWludGFpbmVycyBvZiB0aGUgbGlzdCwgcGVyaW9kaWNhbGx5LCBtYWtl PGJyPjwvZGl2Pg0KIMKgIMKgcGVvcGxlIG9uIDxhIGhyZWY9Im1haWx0bzpjb2RlY0BpZXRmLm9y ZyIgdGFyZ2V0PSJfYmxhbmsiPmNvZGVjQGlldGYub3JnPC9hPiAmbHQ7bWFpbHRvOjxhIGhyZWY9 Im1haWx0bzpjb2RlY0BpZXRmLm9yZyIgdGFyZ2V0PSJfYmxhbmsiPmNvZGVjQGlldGYub3JnPC9h PiZndDsgYXdhcmUgb2YgdGhlPGRpdiBjbGFzcz0iaW0iPjxicj4NCiDCoCDCoGV4aXN0ZW5jZSBv ZiB0aGUgbGlzdCBhbmQgdGhlIG9uZ29pbmc8YnI+DQogwqAgwqBkaXNjdXNzaW9ucywgaW4gYSBz dW1tYXJpemVkIGZvcm0uIMKgSSBhbHNvIGRvbiYjMzk7dCBtaW5kIGlmIHN1Z2dlc3RlZDxicj4N CiDCoCDCoGRlc2lnbjxicj4NCiDCoCDCoGNob2ljZXMgc3BpbGwgb3ZlciB0byBjb2RlY0BpZXRm OyBhZnRlciBhbGwsIHRoaXMgaXMgdGhlIGxpc3Qgd2hlcmU8YnI+DQogwqAgwqB0ZWNobmljYWwg ZGlzY3Vzc2lvbnMgYXJlIGJlaW5nIGhlbGQuIMKgVGhlIGJvcmRlcmxpbmUgd291bGQgYmU8YnI+ DQogwqAgwqBjaXRhdGlvbnMgb2Y8YnI+DQogwqAgwqBwYXRlbnQgbGFuZ3VhZ2UsIGluY2x1ZGlu ZyBwYXRlbnQgbnVtYmVycyBhbmQgY2xhaW0gbGFuZ3VhZ2UuPGJyPg0KPGJyPg0KIMKgIMKgUGVy c29uYWxseSwgSSBzdXJlbHkgZG8gbm90IHdhbnQgdG8gYmUgaW52b2x2ZWQgb24gdGhpcyBoeXBv dGhldGljYWwgbmV3PGJyPg0KIMKgIMKgbGlzdC48YnI+DQo8YnI+DQogwqAgwqBTdGVwaGFuPGJy Pg0KPGJyPg0KPGJyPjwvZGl2Pg0KIMKgIMKgT24gMTAuMjAuMjAxMCAwNjo0MiAsICZxdW90Ozxh IGhyZWY9Im1haWx0bzpmbHVmZnlAY2lzY28uY29tIiB0YXJnZXQ9Il9ibGFuayI+Zmx1ZmZ5QGNp c2NvLmNvbTwvYT4gJmx0O21haWx0bzo8YSBocmVmPSJtYWlsdG86Zmx1ZmZ5QGNpc2NvLmNvbSIg dGFyZ2V0PSJfYmxhbmsiPmZsdWZmeUBjaXNjby5jb208L2E+Jmd0OyZxdW90OzxkaXYgY2xhc3M9 ImltIj48YnI+DQogwqAgwqAmbHQ7PGEgaHJlZj0ibWFpbHRvOmZsdWZmeUBjaXNjby5jb20iIHRh cmdldD0iX2JsYW5rIj5mbHVmZnlAY2lzY28uY29tPC9hPiAmbHQ7bWFpbHRvOjxhIGhyZWY9Im1h aWx0bzpmbHVmZnlAY2lzY28uY29tIiB0YXJnZXQ9Il9ibGFuayI+Zmx1ZmZ5QGNpc2NvLmNvbTwv YT4mZ3Q7Jmd0OyB3cm90ZTo8YnI+DQo8YnI+DQogwqAgwqAgJmd0Ozxicj4NCiDCoCDCoCAmZ3Q7 IFN0ZXBoYW4sIGFzIG91ciBXRyB0ZWNobmljYWwgYWR2aXNvciBvbiB0b3BpY3MgbGlrZSB0aGlz LCBkbyB5b3U8YnI+DQogwqAgwqBoYXZlIGFkdmljZTxicj4NCiDCoCDCoCAmZ3Q7IG9uIGlmIHdl IHNob3VsZCBzZXQgdXAgYSBzZXBhcmF0ZSBsaXN0IGZvciBwZW9wbGUgdGhhdCB3YW50IHRvPGJy Pg0KIMKgIMKgdGFsayBhYm91dDxicj4NCiDCoCDCoCAmZ3Q7IHBhdGVudCBpc3N1ZXMgcmVsYXRl ZCB0byB0aGUgc3RhbmRhcmRzIGZyb20gdGhpcyBXRz8gRnJvbSBhPGJyPg0KIMKgIMKgcHJhY3Rp Y2FsIHBvaW50IG9mPGJyPg0KIMKgIMKgICZndDsgdmlldywgc29tZW9uZSBjb3VsZCBlYXNpbHkg c2V0IHVwIGEgc2VwYXJhdGUgbGlzdCBlaXRoZXIgb24gdGhlPGJyPjwvZGl2Pg0KIMKgIMKgPGEg aHJlZj0iaHR0cDovL2lldGYub3JnIiB0YXJnZXQ9Il9ibGFuayI+aWV0Zi5vcmc8L2E+ICZsdDs8 YSBocmVmPSJodHRwOi8vaWV0Zi5vcmciIHRhcmdldD0iX2JsYW5rIj5odHRwOi8vaWV0Zi5vcmc8 L2E+Jmd0OzxkaXYgY2xhc3M9ImltIj48YnI+DQogwqAgwqAgJmd0OyBzZXJ2ZXJzIG9yIG5vdCBv biBpZXRmIHNlcnZlcnMuPGJyPg0KPGJyPg0KPGJyPg0KIMKgIMKgX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fX19fX188YnI+DQogwqAgwqBjb2RlYyBtYWlsaW5nIGxp c3Q8YnI+PC9kaXY+DQogwqAgwqA8YSBocmVmPSJtYWlsdG86Y29kZWNAaWV0Zi5vcmciIHRhcmdl dD0iX2JsYW5rIj5jb2RlY0BpZXRmLm9yZzwvYT4gJmx0O21haWx0bzo8YSBocmVmPSJtYWlsdG86 Y29kZWNAaWV0Zi5vcmciIHRhcmdldD0iX2JsYW5rIj5jb2RlY0BpZXRmLm9yZzwvYT4mZ3Q7PGRp diBjbGFzcz0iaW0iPjxicj4NCiDCoCDCoDxhIGhyZWY9Imh0dHBzOi8vd3d3LmlldGYub3JnL21h aWxtYW4vbGlzdGluZm8vY29kZWMiIHRhcmdldD0iX2JsYW5rIj5odHRwczovL3d3dy5pZXRmLm9y Zy9tYWlsbWFuL2xpc3RpbmZvL2NvZGVjPC9hPjxicj4NCjxicj4NCjxicj4NCjxicj4NCjxicj4N Cl9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fPGJyPg0KY29k ZWMgbWFpbGluZyBsaXN0PGJyPg0KPGEgaHJlZj0ibWFpbHRvOmNvZGVjQGlldGYub3JnIiB0YXJn ZXQ9Il9ibGFuayI+Y29kZWNAaWV0Zi5vcmc8L2E+PGJyPg0KPGEgaHJlZj0iaHR0cHM6Ly93d3cu aWV0Zi5vcmcvbWFpbG1hbi9saXN0aW5mby9jb2RlYyIgdGFyZ2V0PSJfYmxhbmsiPmh0dHBzOi8v d3d3LmlldGYub3JnL21haWxtYW4vbGlzdGluZm8vY29kZWM8L2E+PGJyPg0KPC9kaXY+PC9ibG9j a3F1b3RlPg0KPGJyPg0KPC9ibG9ja3F1b3RlPjwvZGl2Pjxicj4NCg== --_000_BCB3F026FAC4C145A4A3330806FEFDA93B74D97C49EMBX01HQjnprn_-- From hoene@uni-tuebingen.de Wed Oct 20 10:55:46 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 2E6183A68FD for ; Wed, 20 Oct 2010 10:55:46 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -5.745 X-Spam-Level: X-Spam-Status: No, score=-5.745 tagged_above=-999 required=5 tests=[AWL=0.503, BAYES_00=-2.599, HELO_EQ_DE=0.35, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id jAymC2eVIGwp for ; Wed, 20 Oct 2010 10:55:44 -0700 (PDT) Received: from mx05.uni-tuebingen.de (mx05.uni-tuebingen.de [134.2.3.4]) by core3.amsl.com (Postfix) with ESMTP id 717553A6879 for ; Wed, 20 Oct 2010 10:55:43 -0700 (PDT) Received: from hoeneT60 ([178.2.222.223]) (authenticated bits=0) by mx05.uni-tuebingen.de (8.13.6/8.13.6) with ESMTP id o9KHupE7021343 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Wed, 20 Oct 2010 19:56:57 +0200 From: "Christian Hoene" To: "'Stephen Botzko'" , "'Jean-Marc Valin'" References: <4CBF269B.8090609@octasic.com> In-Reply-To: Date: Wed, 20 Oct 2010 19:56:51 +0200 Organization: =?UTF-8?Q?Universit=C3=A4t_T=C3=BCbingen?= Message-ID: <000901cb7080$3a26b090$ae7411b0$@de> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_000A_01CB7090.FDAF8090" X-Mailer: Microsoft Office Outlook 12.0 Thread-Index: ActwfjeHQaTwRbNcTxCRVpaBdtzKQAAATshA Content-Language: de X-AntiVirus: NOT checked by Avira MailGate (version: 3.0.0-4; host: mx05) Cc: codec@ietf.org Subject: Re: [codec] Seperate list for IPR discutions? X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 17:55:46 -0000 This is a multi-part message in MIME format. ------=_NextPart_000_000A_01CB7090.FDAF8090 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Sorry guys, =20 For posting patent related language and number on the mailing list. As = a researcher living and working under the German legal system, I am not = subject to the same legal dangers as you. Thus, I was not aware which = implications my citations of patents might have to you.=20 My apologies again. =20 Instead a second mailing list, I would suggest to establish a written = netiquette (or rule) that nobody on this mailing post any "citations of = patent language, including patent numbers and claim language".=20 =20 Yes, I promise, I will not do it again. =20 Christian =20 =20 --------------------------------------------------------------- Dr.-Ing. Christian Hoene Interactive Communication Systems (ICS), University of T=C3=BCbingen=20 Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532=20 http://www.net.uni-tuebingen.de/ =20 From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf = Of Stephen Botzko Sent: Wednesday, October 20, 2010 7:42 PM To: Jean-Marc Valin Cc: codec@ietf.org Subject: Re: [codec] Seperate list for IPR discutions? =20 Stephan's suggestion on the boundary "citations of patent language, = including patent numbers and claim language" is consistent with your = boundary. Detailed analysis of proposed work-arounds, would presumably = include patent language. General discussion of license terms would not. Since I wouldn't subscribe or otherwise access the new list, it doesn't = matter to me if it is publicly archived or not. =20 Based on past IETF experience, it is quite difficult to convince posters = to go off-line for a private discussion. One practical issue is that = the folks who do want to discuss this stuff have no easy way to identify = themselves. I'm thinking that providing the second list might make that = easier. Though perhaps no one but Christian is willing to talk about = patent claims? Stephen Botzko On Wed, Oct 20, 2010 at 1:27 PM, Jean-Marc Valin = wrote: Hi, Just to make things clear, I think there's two sets of issues regarding = IPR. One is discussing the exact terms of IPR disclosure/licensing and = the other is discussion about patent themselves. I think the former = should remain on this list. OTOH, I think the latter is (unfortunately) = best dealt with in private. I don't think there is anything to be gained = in discussing patent claims in a publicly archived list. I don't think = even having that on a separate list is a good idea and I'm not = interested in participating in such a list should it be created. Jean-Marc On 10-10-20 01:03 PM, Stephen Botzko wrote: +1, I'd like to see it handled precisely this way. I think it provides a good way forward, and allows those of us who do not want exposure to patent details to continue to participate on the main list. Stephen Botzko On Wed, Oct 20, 2010 at 12:40 PM, Stephan Wenger > wrote: Hi, Yes, I think a separate mailing list would be an excellent idea. There is IMO no danger to the IETF when using an IETF-hosted list. But, as = you mentioned, there are also other options. Whatever is the easiest for those interested in this work. My preference would be that this list has a subscription process independent from the subscription to codec@ietf.org . As the minimum, there must be an opt-out process. I would recommend that the maintainers of the list, periodically, = make people on codec@ietf.org aware of the existence of the list and the ongoing discussions, in a summarized form. I also don't mind if suggested design choices spill over to codec@ietf; after all, this is the list where technical discussions are being held. The borderline would be citations of patent language, including patent numbers and claim language. Personally, I surely do not want to be involved on this hypothetical = new list. Stephan On 10.20.2010 06:42 , "fluffy@cisco.com " > wrote: > > Stephan, as our WG technical advisor on topics like this, do you have advice > on if we should set up a separate list for people that want to talk about > patent issues related to the standards from this WG? From a practical point of > view, someone could easily set up a separate list either on the ietf.org > servers or not on ietf servers. _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec =20 =20 ------=_NextPart_000_000A_01CB7090.FDAF8090 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable

Sorry guys,

 

For posting patent related language and number on the = mailing list. =C2=A0As a researcher living and working under the German legal = system, I am not subject to the same legal dangers as you. Thus, I was not aware = which implications my citations of patents might have to you. =

My apologies again.

 

Instead a second mailing list, I would suggest to = establish a written netiquette (or rule) that nobody on this mailing post any "citations of patent language, including patent = numbers and claim language".

 

Yes, I promise, I will not do it = again.

 

Christian

 

 

----------------------------------------------------------= -----

Dr.-Ing. Christian Hoene

Interactive Communication Systems (ICS), University of = T=C3=BCbingen

Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 = 2970532

http://www.net.uni-tuebingen.de= /

 

From:= codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of = Stephen Botzko
Sent: Wednesday, October 20, 2010 7:42 PM
To: Jean-Marc Valin
Cc: codec@ietf.org
Subject: Re: [codec] Seperate list for IPR = discutions?

 

Stephan's suggestion = on the boundary  "citations of patent language, including patent = numbers and claim language" is consistent with your boundary.  Detailed = analysis of proposed work-arounds, would presumably include patent = language.  General discussion of license terms would not.

Since I wouldn't subscribe or otherwise access the new list, it doesn't = matter to me if it is publicly archived or not. 

Based on past IETF experience, it is quite difficult to convince posters = to go off-line for a private discussion.  One practical issue is that the = folks who do want to discuss this stuff have no easy way to identify themselves.  I'm thinking that providing the second list might make = that easier.  Though perhaps no one but Christian is willing to talk = about patent claims?

Stephen Botzko

On Wed, Oct 20, 2010 at 1:27 PM, Jean-Marc Valin = <jean-marc.valin@octasic.com> wrote:

Hi,

Just to make things clear, I think there's two sets of issues regarding = IPR. One is discussing the exact terms of IPR disclosure/licensing and the = other is discussion about patent themselves. I think the former should remain on = this list. OTOH, I think the latter is (unfortunately) best dealt with in = private. I don't think there is anything to be gained in discussing patent claims = in a publicly archived list. I don't think even having that on a separate = list is a good idea and I'm not interested in participating in such a list should it be created.

    Jean-Marc



On 10-10-20 01:03 PM, Stephen Botzko wrote:

<mailto:stewe@stewe.org>> wrote:

   Hi,

   Yes, I think a separate mailing list would be an excellent = idea.
     There is
   IMO no danger to the IETF when using an IETF-hosted list.  But, as you
   mentioned, there are also other options.  Whatever is = the easiest
   for those
   interested in this work.

   My preference would be that this list has a subscription = process
   independent

   from the subscription to codec@ietf.org = <mailto:codec@ietf.org>. =  As


   the minimum, there must be an
   opt-out process.

   I would recommend that the maintainers of the list, = periodically, make

   people on codec@ietf.org <mailto:codec@ietf.org> aware of the


   existence of the list and the ongoing
   discussions, in a summarized form.  I also don't mind = if suggested
   design
   choices spill over to codec@ietf; after all, this is the = list where
   technical discussions are being held.  The borderline = would be
   citations of
   patent language, including patent numbers and claim = language.

   Personally, I surely do not want to be involved on this hypothetical new
   list.

   Stephan

   On 10.20.2010 06:42 , "fluffy@cisco.com = <mailto:fluffy@cisco.com>"


   <fluffy@cisco.com <mailto:fluffy@cisco.com>> wrote:

    >
    > Stephan, as our WG technical advisor on topics like = this, do you
   have advice
    > on if we should set up a separate list for people = that want to
   talk about
    > patent issues related to the standards from this WG? = >From a
   practical point of
    > view, someone could easily set up a separate list = either on the

   ietf.org <http://ietf.org>


    > servers or not on ietf servers.


   _______________________________________________
   codec mailing list

   codec@ietf.org <mailto:codec@ietf.org>


   https://www.ietf.org/mailman/listinfo/codec




_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/mailman/listinfo/codec

 

 

------=_NextPart_000_000A_01CB7090.FDAF8090-- From stewe@stewe.org Wed Oct 20 11:38:39 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id DE7233A681E for ; Wed, 20 Oct 2010 11:38:38 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.862 X-Spam-Level: X-Spam-Status: No, score=-1.862 tagged_above=-999 required=5 tests=[AWL=0.737, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id da3NIUYWTUhK for ; Wed, 20 Oct 2010 11:38:37 -0700 (PDT) Received: from stewe.org (stewe.org [85.214.122.234]) by core3.amsl.com (Postfix) with ESMTP id 764473A67D9 for ; Wed, 20 Oct 2010 11:38:34 -0700 (PDT) Received: from [192.168.1.108] (unverified [24.5.132.232]) by stewe.org (SurgeMail 3.9e) with ESMTP id 820578-1743317 for multiple; Wed, 20 Oct 2010 20:40:06 +0200 User-Agent: Microsoft-Entourage/12.27.0.100910 Date: Wed, 20 Oct 2010 11:39:56 -0700 From: Stephan Wenger To: Jean-Marc Valin , Stephen Botzko Message-ID: Thread-Topic: [codec] Seperate list for IPR discutions? Thread-Index: ActwhjD83bzpcHqEtE6hQ+nByXdsCg== In-Reply-To: <4CBF269B.8090609@octasic.com> Mime-version: 1.0 Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit X-Originating-IP: 24.5.132.232 X-Authenticated-User: stewe@stewe.org X-ORBS-Stamp: Your IP (24.5.132.232) was found in the spamhaus database. http://www.spamhaus.net Cc: codec@ietf.org Subject: Re: [codec] Seperate list for IPR discutions? X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 18:38:39 -0000 Hi Jean-Marc, all, I have said this before: discussions about commercial terms, if not conducted very carefully by very knowledgeable people, can have grave consequences to the individuals involved and their employers. I bet the majority of the list participants don't have the authority to discuss anything like this, which may lead to an unbalanced picture. Therefore, I prefer to have those discussions conducted elsewhere as well, and among the willing only. If you insist on discussing commercial terms and licensing declarations, you may wish to reduce the risks for yourself and the community by -never discuss any $$$ amounts, i.e. the monetary compensation part of licensing, -be very careful when discussing other benefits of a licensing model you prefer, and avoid implying that certain industries may benefit from it more than others, -never imply that your statements apply to anyone but yourself, unless you are authorized by that "anyone" (be it community, company, whatever), and -avoiding categorical statements "we can't do this and that because". Rather use "we should not do this and that because", or, better, "I prefer not to do this and that because". Stephan P.s.: I fully understand and appreciate that most or all of our statements in the IETF are made by us as individuals. On 10.20.2010 10:27 , "Jean-Marc Valin" wrote: > Hi, > > Just to make things clear, I think there's two sets of issues regarding > IPR. One is discussing the exact terms of IPR disclosure/licensing and the > other is discussion about patent themselves. I think the former should > remain on this list. OTOH, I think the latter is (unfortunately) best dealt > with in private. I don't think there is anything to be gained in discussing > patent claims in a publicly archived list. I don't think even having that > on a separate list is a good idea and I'm not interested in participating > in such a list should it be created. > > Jean-Marc > > On 10-10-20 01:03 PM, Stephen Botzko wrote: >> +1, I'd like to see it handled precisely this way. >> >> I think it provides a good way forward, and allows those of us who do >> not want exposure to patent details to continue to participate on the >> main list. >> >> Stephen Botzko >> >> >> On Wed, Oct 20, 2010 at 12:40 PM, Stephan Wenger > > wrote: >> >> Hi, >> >> Yes, I think a separate mailing list would be an excellent idea. >> There is >> IMO no danger to the IETF when using an IETF-hosted list. But, as you >> mentioned, there are also other options. Whatever is the easiest >> for those >> interested in this work. >> >> My preference would be that this list has a subscription process >> independent >> from the subscription to codec@ietf.org . As >> the minimum, there must be an >> opt-out process. >> >> I would recommend that the maintainers of the list, periodically, make >> people on codec@ietf.org aware of the >> existence of the list and the ongoing >> discussions, in a summarized form. I also don't mind if suggested >> design >> choices spill over to codec@ietf; after all, this is the list where >> technical discussions are being held. The borderline would be >> citations of >> patent language, including patent numbers and claim language. >> >> Personally, I surely do not want to be involved on this hypothetical new >> list. >> >> Stephan >> >> >> On 10.20.2010 06:42 , "fluffy@cisco.com " >> > wrote: >> >>> >>> Stephan, as our WG technical advisor on topics like this, do you >> have advice >>> on if we should set up a separate list for people that want to >> talk about >>> patent issues related to the standards from this WG? From a >> practical point of >>> view, someone could easily set up a separate list either on the >> ietf.org >>> servers or not on ietf servers. >> >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >> >> >> >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec > From stewe@stewe.org Wed Oct 20 11:43:27 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 6B4593A68BA for ; Wed, 20 Oct 2010 11:43:27 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.194 X-Spam-Level: X-Spam-Status: No, score=-1.194 tagged_above=-999 required=5 tests=[AWL=0.008, BAYES_00=-2.599, HTML_MESSAGE=0.001, MIME_QP_LONG_LINE=1.396] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id MZcJsYBiu1Jx for ; Wed, 20 Oct 2010 11:43:20 -0700 (PDT) Received: from stewe.org (stewe.org [85.214.122.234]) by core3.amsl.com (Postfix) with ESMTP id 84CA63A68EC for ; Wed, 20 Oct 2010 11:42:18 -0700 (PDT) Received: from [192.168.1.108] (unverified [24.5.132.232]) by stewe.org (SurgeMail 3.9e) with ESMTP id 820583-1743317 for multiple; Wed, 20 Oct 2010 20:43:52 +0200 User-Agent: Microsoft-Entourage/12.27.0.100910 Date: Wed, 20 Oct 2010 11:43:38 -0700 From: Stephan Wenger To: Gregory Maxwell , "'stephen.botzko@gmail.com'" , "'jean-marc.valin@octasic.com'" Message-ID: Thread-Topic: [codec] Seperate list for IPR discutions? Thread-Index: Actwfi08ukuKx/BjSx+Q0cCrHKI+EgAAQdGqAAHgM14= In-Reply-To: Mime-version: 1.0 Content-type: multipart/alternative; boundary="B_3370419830_3398599" X-Originating-IP: 24.5.132.232 X-Authenticated-User: stewe@stewe.org X-ORBS-Stamp: Your IP (24.5.132.232) was found in the spamhaus database. http://www.spamhaus.net Cc: "'codec@ietf.org'" Subject: Re: [codec] Seperate list for IPR discutions? X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 18:43:27 -0000 > This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. --B_3370419830_3398599 Content-type: text/plain; charset="ISO-8859-1" Content-transfer-encoding: quoted-printable For the record: Without being interested in any discussion in this venue, I disagree with your statement =B3Being _willfully_ ignorant of your obligation= s which you would otherwise know about is just willfully infringing in a pretty dress.=B2. =20 I support the rest of your email. Stephan On 10.20.2010 10:50 , "Gregory Maxwell" wrote: >=20 > There is, has been, and will be plenty of discussion in _private_. Bein= g > _willfully_ ignorant of your obligations which you would otherwise know a= bout > is just willfully infringing in a pretty dress. >=20 > The goal of the developers here is to not infringe on third part patents.= Full > stop. >=20 > The objection is to a public discussion- because patent discussion must b= e > informed by the relevant experts, because it no one wants to be dragged i= n to > testify on random patents, and because exposing your strategy to a possib= le > attacker makes it weaker, etc. >=20 >=20 >=20 >=20 >=20 >=20 > From: codec-bounces@ietf.org > To: Jean-Marc Valin > Cc: codec@ietf.org > Sent: Wed Oct 20 10:42:27 2010 > Subject: Re: [codec] Seperate list for IPR discutions? > Stephan's suggestion on the boundary=A0 "citations of patent language, incl= uding > patent numbers and claim language" is consistent with your boundary.=A0 Det= ailed > analysis of proposed work-arounds, would presumably include patent langua= ge.=A0 > General discussion of license terms would not. >=20 > Since I wouldn't subscribe or otherwise access the new list, it doesn't m= atter > to me if it is publicly archived or not.=A0 >=20 > Based on past IETF experience, it is quite difficult to convince posters = to go > off-line for a private discussion.=A0 One practical issue is that the folks= who > do want to discuss this stuff have no easy way to identify themselves.=A0 I= 'm > thinking that providing the second list might make that easier.=A0 Though > perhaps no one but Christian is willing to talk about patent claims? >=20 > Stephen Botzko >=20 > On Wed, Oct 20, 2010 at 1:27 PM, Jean-Marc Valin > wrote: >> Hi, >>=20 >> Just to make things clear, I think there's two sets of issues regarding = IPR. >> One is discussing the exact terms of IPR disclosure/licensing and the ot= her >> is discussion about patent themselves. I think the former should remain = on >> this list. OTOH, I think the latter is (unfortunately) best dealt with i= n >> private. I don't think there is anything to be gained in discussing pate= nt >> claims in a publicly archived list. I don't think even having that on a >> separate list is a good idea and I'm not interested in participating in = such >> a list should it be created. >>=20 >> =A0 =A0 Jean-Marc >>=20 >>=20 >> On 10-10-20 01:03 PM, Stephen Botzko wrote: >>> +1, I'd like to see it handled precisely this way. >>>=20 >>> I think it provides a good way forward, and allows those of us who do >>> not want exposure to patent details to continue to participate on the >>> main list. >>>=20 >>> Stephen Botzko >>>=20 >>>=20 >>> On Wed, Oct 20, 2010 at 12:40 PM, Stephan Wenger >> > wrote: >>>=20 >>> =A0 =A0Hi, >>>=20 >>> =A0 =A0Yes, I think a separate mailing list would be an excellent idea. >>> =A0 =A0 =A0There is >>> =A0 =A0IMO no danger to the IETF when using an IETF-hosted list. =A0But, as = you >>> =A0 =A0mentioned, there are also other options. =A0Whatever is the easiest >>> =A0 =A0for those >>> =A0 =A0interested in this work. >>>=20 >>> =A0 =A0My preference would be that this list has a subscription process >>> =A0 =A0independent >>> =A0 =A0from the subscription to codec@ietf.org . =A0A= s >>>=20 >>> =A0 =A0the minimum, there must be an >>> =A0 =A0opt-out process. >>>=20 >>> =A0 =A0I would recommend that the maintainers of the list, periodically, m= ake >>> =A0 =A0people on codec@ietf.org aware of the >>>=20 >>> =A0 =A0existence of the list and the ongoing >>> =A0 =A0discussions, in a summarized form. =A0I also don't mind if suggested >>> =A0 =A0design >>> =A0 =A0choices spill over to codec@ietf; after all, this is the list where >>> =A0 =A0technical discussions are being held. =A0The borderline would be >>> =A0 =A0citations of >>> =A0 =A0patent language, including patent numbers and claim language. >>>=20 >>> =A0 =A0Personally, I surely do not want to be involved on this hypothetica= l new >>> =A0 =A0list. >>>=20 >>> =A0 =A0Stephan >>>=20 >>>=20 >>> =A0 =A0On 10.20.2010 06:42 , "fluffy@cisco.com " >>>=20 >>> =A0 =A0> wrote: >>>=20 >>> =A0 =A0 > >>> =A0 =A0 > Stephan, as our WG technical advisor on topics like this, do you >>> =A0 =A0have advice >>> =A0 =A0 > on if we should set up a separate list for people that want to >>> =A0 =A0talk about >>> =A0 =A0 > patent issues related to the standards from this WG? From a >>> =A0 =A0practical point of >>> =A0 =A0 > view, someone could easily set up a separate list either on the >>> =A0 =A0ietf.org >>>=20 >>> =A0 =A0 > servers or not on ietf servers. >>>=20 >>>=20 >>> =A0 =A0_______________________________________________ >>> =A0 =A0codec mailing list >>> =A0 =A0codec@ietf.org >>>=20 >>> =A0 =A0https://www.ietf.org/mailman/listinfo/codec >>>=20 >>>=20 >>>=20 >>>=20 >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >>=20 >=20 >=20 >=20 > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec --B_3370419830_3398599 Content-type: text/html; charset="ISO-8859-1" Content-transfer-encoding: quoted-printable Re: [codec] Seperate list for IPR discutions? For the record: Without being interested in any discussion in this venue, = I disagree with your statement “Being _willfully_ ignorant of = your obligations which you would otherwise know about is just willfully infr= inging in a pretty dress.”.  

I support the rest of your email.
Stephan


On 10.20.2010 10:50 , "Gregory Maxwell" <gmaxwell@juniper.net> wrote:


There is, has been, and will be plenty of discussion in _private_.  &n= bsp;Being _willfully_ ignorant of your obligations which you would otherwise= know about is just willfully infringing in a pretty dress.

The goal of the developers here is to not infringe on third part patents. F= ull stop.

The objection is to a public discussion- because patent discussion must be = informed by the relevant experts, because it no one wants to be dragged in t= o testify on random patents, and because exposing your strategy to a possibl= e attacker makes it weaker, etc.





From: codec-bounce= s@ietf.org <codec-bounces@ietf.org>
To: Jean-Marc Valin <jean-m= arc.valin@octasic.com>
Cc: codec@ietf.org <codec@ietf.org>
Sent: Wed Oct 20 10:42:27 2010
Subject: Re: [codec] Seperate list for IPR discutions?
Stephan's suggestion= on the boundary=A0 "citations of patent language, including patent numbe= rs and claim language" is consistent with your boundary.=A0 Detailed anal= ysis of proposed work-arounds, would presumably include patent language.=A0 Ge= neral discussion of license terms would not.

Since I wouldn't subscribe or otherwise access the new list, it doesn't mat= ter to me if it is publicly archived or not.=A0

Based on past IETF experience, it is quite difficult to convince posters to= go off-line for a private discussion.=A0 One practical issue is that the folk= s who do want to discuss this stuff have no easy way to identify themselves.= =A0 I'm thinking that providing the second list might make that easier.=A0 Thoug= h perhaps no one but Christian is willing to talk about patent claims?

Stephen Botzko

On Wed, Oct 20, 2010 at 1:27 PM, Jean-Marc Valin <jean-marc.valin@octasic.com> wrote:
Hi,

Just to make things clear, I think there's two sets of issues regarding IPR= . One is discussing the exact terms of IPR disclosure/licensing and the othe= r is discussion about patent themselves. I think the former should remain on= this list. OTOH, I think the latter is (unfortunately) best dealt with in p= rivate. I don't think there is anything to be gained in discussing patent cl= aims in a publicly archived list. I don't think even having that on a separa= te list is a good idea and I'm not interested in participating in such a lis= t should it be created.

 =A0 =A0 Jean-Marc


On 10-10-20 01:03 PM, Stephen Botzko wrote:
+1, I'd like to see it handled precisely this wa= y.

I think it provides a good way forward, and allows those of us who do
not want exposure to patent details to continue to participate on the
main list.

Stephen Botzko


On Wed, Oct 20, 2010 at 12:40 PM, Stephan Wenger <stewe@stewe.org
<mailto:stewe@stewe.org>>= wrote:

 =A0 =A0Hi,

 =A0 =A0Yes, I think a separate mailing list would be an excellent idea.  =A0 =A0 =A0There is
 =A0 =A0IMO no danger to the IETF when using an IETF-hosted list. =A0But, as= you
 =A0 =A0mentioned, there are also other options. =A0Whatever is the easiest<= BR>  =A0 =A0for those
 =A0 =A0interested in this work.

 =A0 =A0My preference would be that this list has a subscription process  =A0 =A0independent
 =A0 =A0from the subscription to codec@ietf.org <mailto:codec@ietf.org>. =A0As
 =A0 =A0the minimum, there must be an
 =A0 =A0opt-out process.

 =A0 =A0I would recommend that the maintainers of the list, periodically, = make
 =A0 =A0people on codec@ietf.org <mailto:codec@ietf.org> aware of the

 =A0 =A0existence of the list and the ongoing
 =A0 =A0discussions, in a summarized form. =A0I also don't mind if suggested=
 =A0 =A0design
 =A0 =A0choices spill over to codec@ietf; after all, this is the list wher= e
 =A0 =A0technical discussions are being held. =A0The borderline would be
 =A0 =A0citations of
 =A0 =A0patent language, including patent numbers and claim language.

 =A0 =A0Personally, I surely do not want to be involved on this hypothetic= al new
 =A0 =A0list.

 =A0 =A0Stephan


 =A0 =A0On 10.20.2010 06:42 , "fluffy@cisc= o.com <mailto:fluffy@cisco.com&= gt;"

 =A0 =A0<fluffy@cisco.com <mailto:fluffy@cisco.com>> wrote:

 =A0 =A0 >
 =A0 =A0 > Stephan, as our WG technical advisor on topics like this, do= you
 =A0 =A0have advice
 =A0 =A0 > on if we should set up a separate list for people that want = to
 =A0 =A0talk about
 =A0 =A0 > patent issues related to the standards from this WG? From a<= BR>  =A0 =A0practical point of
 =A0 =A0 > view, someone could easily set up a separate list either on = the
 =A0 =A0ietf.org <http://ietf.org> &nb= sp;<http://ietf.org>

 =A0 =A0 > servers or not on ietf servers.


 =A0 =A0_______________________________________________
 =A0 =A0codec mailing list
 =A0 =A0codec@ietf.org <mailto:codec@ietf.org>

 =A0 =A0https://www.= ietf.org/mailman/listinfo/codec




_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/= mailman/listinfo/codec



___________= ____________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/= mailman/listinfo/codec
--B_3370419830_3398599-- From gmaxwell@juniper.net Wed Oct 20 15:39:19 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id A62143A696A for ; Wed, 20 Oct 2010 15:39:19 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.206 X-Spam-Level: X-Spam-Status: No, score=-6.206 tagged_above=-999 required=5 tests=[AWL=0.393, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id VbdIjbHjEm4o for ; Wed, 20 Oct 2010 15:39:17 -0700 (PDT) Received: from exprod7og112.obsmtp.com (exprod7og112.obsmtp.com [64.18.2.177]) by core3.amsl.com (Postfix) with ESMTP id 8B3A43A6927 for ; Wed, 20 Oct 2010 15:39:16 -0700 (PDT) Received: from source ([66.129.224.36]) (using TLSv1) by exprod7ob112.postini.com ([64.18.6.12]) with SMTP ID DSNKTL9v61ciJskVnd2gjeGoV5YiRaCbNHS5@postini.com; Wed, 20 Oct 2010 15:40:51 PDT Received: from EMBX01-HQ.jnpr.net ([fe80::c821:7c81:f21f:8bc7]) by P-EMHUB02-HQ.jnpr.net ([fe80::88f9:77fd:dfc:4d51%11]) with mapi; Wed, 20 Oct 2010 15:39:34 -0700 From: Gregory Maxwell To: Stephan Wenger , "'stephen.botzko@gmail.com'" , "'jean-marc.valin@octasic.com'" Date: Wed, 20 Oct 2010 15:39:34 -0700 Thread-Topic: [codec] Seperate list for IPR discutions? Thread-Index: Actwfi08ukuKx/BjSx+Q0cCrHKI+EgAAQdGqAAHgM14ABVoyzg== Message-ID: References: , In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Cc: "'codec@ietf.org'" Subject: Re: [codec] Seperate list for IPR discutions? X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Oct 2010 22:39:19 -0000 Stephan, I think that many of us appreciate your cautionary advice and the benefit o= f your experience. I certainly do. But I think participants also need to be careful to take it only as caution= ary advice and not the gospel truth. I doubt you want your words to create = FUD and thereby make participants here afraid to act, but I think your use = such confident and firm language may sometimes have that effect. This risk = of reasonable caution having the unintended effect of FUD is especially gre= at when your comments are vague. E.g. I understand why you might advise ag= ainst discussing licensing terms in some places=97 due to concerns about cr= eating collusion problems=97 but I can not fathom how an antitrust issues = could arise when discussing free-to-all licensing. Licensing terms are rout= inely discussed in the context of free software projects, even with the inv= olvement of billion dollar corporations. Your comments on the issue of lic= ensing weren't even specific enough for me to ask an attorney if you're rig= ht or not. Responses like that create a cloud of FUD which removes enlight= enment rather than creating it. Some of the advice you've given here directly contradicts the advice I've r= eceived from council. I would encourage people here interested in legal is= sues consult their own council and obtain an analysis specific to their own= circumstances and needs. In particular, the issue of "willful infringement" seems to frequently suff= er from substantial misinformation or, really, out of date information. The= information you've provided here is, to the best of my understanding, inco= rrect. The effective standard for willful infringement applied by US courts change= d a few years ago as a result of In re Seagate[1]. After a brief search I= found this reasonable looking overview which reflects the current state of= the world: http://www.patentbaristas.com/archives/2007/08/24/is-the-tide-o= f-willful-patent-infringement-litigation-receding/ (but there are a great m= any other writeups about this case out there). Things also appear to be sta= tionary on this new precedent (the supreme court declined an appeal of this= decision). The exact nuance of what behaviour is sufficiently reckless to constitute w= illful infringement in the context of the In re Seagate decision is still a= n evolving area. But the old 'you must get a pass from a patent attorney ev= ery time you see a patent number' is simply no longer true. The various new= s and journal articles on this case discuss this subject in depth.=20 The increasingly common advice now given to software developers interested = in creating non-infringing software is "read the patents", even for people = without the benefit of the active involvement of a patent attorney. People = without a lot of background in this area may find this article to be very i= nformative: http://news.swpat.org/2010/03/transcript-tridgell-patents/ Almost nowhere in law does _intentionally_ blinding yourself because you be= lieve that you will likely find something which creates obligations actuall= y relieve you of those obligations (go look up "constructive knowledge"), s= o we shouldn't be surprised to find patent law reflecting this greater patt= ern. In particular I should draw your attention to SEB S.A. v. Montgomery Ward &= Co., Inc.[2] where the court determined that "deliberate indifference" w= as equivalent to specific intent=97 a requirement for a claim of inducement= . The matter is certainly not clear or settled=97 which is why I strongly enc= ourage people to consult their own council=97 but it appears clear that ri= sk of awareness of patent claims is being significantly overstated and the = suggestion that one can get away with intentionally avoiding knowing seems = to be both legally risky and socially reprehensible. Besides=97 there are plenty of good reasons to not spend time looking at pa= tent claims besides believing that you'll learn about something troublesome= : It's somewhat difficult and time consuming. If you believe that parti= cipation here is so dangerous that you can only do so safely by being wilfu= lly blind to the specifics of risks that you believe you exist, then perhap= s you should reconsider your participation. I want to reiterate that in spite of the above that I'm also not interested= in having a public patent discussion here or on another list but I want to= make it clear that my lack of interest has nothing to do with an attempt t= o avoid knowing about patent risk. I know risk exists and I want to be mad= e aware of existing patents which may pose a risk for the usability or roya= lty free status of the codec. I have been involved in the review of patents= for implications on the involved codec(s), and there will be more review a= s the work continues. I just don't believe that a public discussion is a pr= oductive way to conduct this activity. And I believe this statement is tru= e for all the technical contributors to our current codec. Cheers, [1] http://www.fedcir.gov/opinions/M830.pdf [2] http://caselaw.findlaw.com/us-federal-circuit/1506829.html / http://www= .cafc.uscourts.gov/images/stories/opinions-orders/09-1099.pdf (real link se= ems to be down) ________________________________________ From: Stephan Wenger [stewe@stewe.org] Sent: Wednesday, October 20, 2010 7:43 PM To: Gregory Maxwell; 'stephen.botzko@gmail.com'; 'jean-marc.valin@octasic.c= om' Cc: 'codec@ietf.org' Subject: Re: [codec] Seperate list for IPR discutions? For the record: Without being interested in any discussion in this venue, I= disagree with your statement =93Being _willfully_ ignorant of your obligat= ions which you would otherwise know about is just willfully infringing in a= pretty dress.=94. I support the rest of your email. Stephan On 10.20.2010 10:50 , "Gregory Maxwell" wrote: There is, has been, and will be plenty of discussion in _private_. Being = _willfully_ ignorant of your obligations which you would otherwise know abo= ut is just willfully infringing in a pretty dress. The goal of the developers here is to not infringe on third part patents. F= ull stop. The objection is to a public discussion- because patent discussion must be = informed by the relevant experts, because it no one wants to be dragged in = to testify on random patents, and because exposing your strategy to a possi= ble attacker makes it weaker, etc. ________________________________ From: codec-bounces@ietf.org To: Jean-Marc Valin Cc: codec@ietf.org Sent: Wed Oct 20 10:42:27 2010 Subject: Re: [codec] Seperate list for IPR discutions? Stephan's suggestion on the boundary "citations of patent language, includ= ing patent numbers and claim language" is consistent with your boundary. D= etailed analysis of proposed work-arounds, would presumably include patent = language. General discussion of license terms would not. Since I wouldn't subscribe or otherwise access the new list, it doesn't mat= ter to me if it is publicly archived or not. Based on past IETF experience, it is quite difficult to convince posters to= go off-line for a private discussion. One practical issue is that the fol= ks who do want to discuss this stuff have no easy way to identify themselve= s. I'm thinking that providing the second list might make that easier. Th= ough perhaps no one but Christian is willing to talk about patent claims? Stephen Botzko On Wed, Oct 20, 2010 at 1:27 PM, Jean-Marc Valin wrote: Hi, Just to make things clear, I think there's two sets of issues regarding IPR= . One is discussing the exact terms of IPR disclosure/licensing and the oth= er is discussion about patent themselves. I think the former should remain = on this list. OTOH, I think the latter is (unfortunately) best dealt with i= n private. I don't think there is anything to be gained in discussing paten= t claims in a publicly archived list. I don't think even having that on a s= eparate list is a good idea and I'm not interested in participating in such= a list should it be created. Jean-Marc On 10-10-20 01:03 PM, Stephen Botzko wrote: +1, I'd like to see it handled precisely this way. I think it provides a good way forward, and allows those of us who do not want exposure to patent details to continue to participate on the main list. Stephen Botzko On Wed, Oct 20, 2010 at 12:40 PM, Stephan Wenger > wrote: Hi, Yes, I think a separate mailing list would be an excellent idea. There is IMO no danger to the IETF when using an IETF-hosted list. But, as you mentioned, there are also other options. Whatever is the easiest for those interested in this work. My preference would be that this list has a subscription process independent from the subscription to codec@ietf.org . As the minimum, there must be an opt-out process. I would recommend that the maintainers of the list, periodically, make people on codec@ietf.org aware of the existence of the list and the ongoing discussions, in a summarized form. I also don't mind if suggested design choices spill over to codec@ietf; after all, this is the list where technical discussions are being held. The borderline would be citations of patent language, including patent numbers and claim language. Personally, I surely do not want to be involved on this hypothetical ne= w list. Stephan On 10.20.2010 06:42 , "fluffy@cisco.com " > wrote: > > Stephan, as our WG technical advisor on topics like this, do you have advice > on if we should set up a separate list for people that want to talk about > patent issues related to the standards from this WG? From a practical point of > view, someone could easily set up a separate list either on the ietf.org > servers or not on ietf servers. _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec ________________________________ _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec From stewe@stewe.org Fri Oct 22 09:39:38 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 4DB763A66B4 for ; Fri, 22 Oct 2010 09:39:38 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.203 X-Spam-Level: X-Spam-Status: No, score=-1.203 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, MIME_QP_LONG_LINE=1.396] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fca0Ne2omdQG for ; Fri, 22 Oct 2010 09:39:36 -0700 (PDT) Received: from stewe.org (stewe.org [85.214.122.234]) by core3.amsl.com (Postfix) with ESMTP id 4AAB13A690A for ; Fri, 22 Oct 2010 09:39:32 -0700 (PDT) Received: from [10.1.10.108] (unverified [209.17.176.225]) by stewe.org (SurgeMail 3.9e) with ESMTP id 2051-1743317 for multiple; Fri, 22 Oct 2010 18:41:08 +0200 User-Agent: Microsoft-Entourage/12.27.0.100910 Date: Fri, 22 Oct 2010 09:40:57 -0700 From: Stephan Wenger To: Gregory Maxwell , "'stephen.botzko@gmail.com'" , "'jean-marc.valin@octasic.com'" Message-ID: Thread-Topic: [codec] Seperate list for IPR discutions? Thread-Index: Actwfi08ukuKx/BjSx+Q0cCrHKI+EgAAQdGqAAHgM14ABVoyzgBa8iNp In-Reply-To: Mime-version: 1.0 Content-type: text/plain; charset="ISO-8859-1" Content-transfer-encoding: quoted-printable X-Originating-IP: 209.17.176.225 X-Authenticated-User: stewe@stewe.org Cc: "'codec@ietf.org'" Subject: Re: [codec] Seperate list for IPR discutions? X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 22 Oct 2010 16:39:38 -0000 Hi all, Sorry for my late and short reply. I'm currently a bit time pressed. I reviewed my recent emails to the list (over the last week or so), and found nothing that made me jump to post a correction. Which does not mean that I may not do so in the future... One key point of Greg's message I wholeheartedly agree to: when in doubt, consult your own laywer. As of the rest of Greg's email, I agree with many of Greg's points below, but also disagree with a few. However, I'm not prepared to discuss my points on this public list. If Greg is willing, we could probably have a bit of a dialogue between us two (+ whatever advisors we are pulling in), and, if we were agreeing so, could post a summary at a later date. Greg, please let me know if you are interested. Regards, Stephan On 10.20.2010 15:39 , "Gregory Maxwell" wrote: >=20 > Stephan, >=20 > I think that many of us appreciate your cautionary advice and the benefit= of > your experience. I certainly do. >=20 > But I think participants also need to be careful to take it only as cauti= onary > advice and not the gospel truth. I doubt you want your words to create FU= D and > thereby make participants here afraid to act, but I think your use such > confident and firm language may sometimes have that effect. This risk of > reasonable caution having the unintended effect of FUD is especially grea= t > when your comments are vague. E.g. I understand why you might advise aga= inst > discussing licensing terms in some places=8B due to concerns about creating > collusion problems=8B but I can not fathom how an antitrust issues could a= rise > when discussing free-to-all licensing. Licensing terms are routinely disc= ussed > in the context of free software projects, even with the involvement of bi= llion > dollar corporations. Your comments on the issue of licensing weren't eve= n > specific enough for me to ask an attorney if you're right or not. Respon= ses > like that create a cloud of FUD which removes enlightenment rather than > creating it. >=20 > Some of the advice you've given here directly contradicts the advice I've > received from council. I would encourage people here interested in legal > issues consult their own council and obtain an analysis specific to their= own > circumstances and needs. >=20 > In particular, the issue of "willful infringement" seems to frequently su= ffer > from substantial misinformation or, really, out of date information. The > information you've provided here is, to the best of my understanding, > incorrect. >=20 > The effective standard for willful infringement applied by US courts chan= ged a > few years ago as a result of In re Seagate[1]. After a brief search I f= ound > this reasonable looking overview which reflects the current state of the > world:=20 > http://www.patentbaristas.com/archives/2007/08/24/is-the-tide-of-willful-= paten > t-infringement-litigation-receding/ (but there are a great many other wri= teups > about this case out there). Things also appear to be stationary on this n= ew > precedent (the supreme court declined an appeal of this decision). >=20 > The exact nuance of what behaviour is sufficiently reckless to constitute > willful infringement in the context of the In re Seagate decision is stil= l an > evolving area. But the old 'you must get a pass from a patent attorney ev= ery > time you see a patent number' is simply no longer true. The various news = and > journal articles on this case discuss this subject in depth. >=20 > The increasingly common advice now given to software developers intereste= d in > creating non-infringing software is "read the patents", even for people > without the benefit of the active involvement of a patent attorney. Peopl= e > without a lot of background in this area may find this article to be very > informative: http://news.swpat.org/2010/03/transcript-tridgell-patents/ >=20 > Almost nowhere in law does _intentionally_ blinding yourself because you > believe that you will likely find something which creates obligations act= ually > relieve you of those obligations (go look up "constructive knowledge"), s= o we > shouldn't be surprised to find patent law reflecting this greater pattern= . >=20 > In particular I should draw your attention to SEB S.A. v. Montgomery Ward= & > Co., Inc.[2] where the court determined that "deliberate indifference" = was > equivalent to specific intent=8B a requirement for a claim of inducement. >=20 > The matter is certainly not clear or settled=8B which is why I strongly > encourage people to consult their own council=8B but it appears clear that= risk > of awareness of patent claims is being significantly overstated and the > suggestion that one can get away with intentionally avoiding knowing seem= s to > be both legally risky and socially reprehensible. >=20 > Besides=8B there are plenty of good reasons to not spend time looking at pa= tent > claims besides believing that you'll learn about something troublesome: = It's > somewhat difficult and time consuming. If you believe that participati= on > here is so dangerous that you can only do so safely by being wilfully bli= nd to > the specifics of risks that you believe you exist, then perhaps you shoul= d > reconsider your participation. >=20 > I want to reiterate that in spite of the above that I'm also not interest= ed in > having a public patent discussion here or on another list but I want to m= ake > it clear that my lack of interest has nothing to do with an attempt to av= oid > knowing about patent risk. I know risk exists and I want to be made awar= e of > existing patents which may pose a risk for the usability or royalty free > status of the codec. I have been involved in the review of patents for > implications on the involved codec(s), and there will be more review as t= he > work continues. I just don't believe that a public discussion is a produc= tive > way to conduct this activity. And I believe this statement is true for a= ll > the technical contributors to our current codec. >=20 > Cheers, >=20 >=20 > [1] http://www.fedcir.gov/opinions/M830.pdf > [2] http://caselaw.findlaw.com/us-federal-circuit/1506829.html / > http://www.cafc.uscourts.gov/images/stories/opinions-orders/09-1099.pdf (= real > link seems to be down) >=20 >=20 > ________________________________________ > From: Stephan Wenger [stewe@stewe.org] > Sent: Wednesday, October 20, 2010 7:43 PM > To: Gregory Maxwell; 'stephen.botzko@gmail.com'; 'jean-marc.valin@octasic= .com' > Cc: 'codec@ietf.org' > Subject: Re: [codec] Seperate list for IPR discutions? >=20 > For the record: Without being interested in any discussion in this venue,= I > disagree with your statement =B3Being _willfully_ ignorant of your obligati= ons > which you would otherwise know about is just willfully infringing in a pr= etty > dress.=B2. >=20 > I support the rest of your email. > Stephan >=20 >=20 > On 10.20.2010 10:50 , "Gregory Maxwell" wrote: >=20 >=20 > There is, has been, and will be plenty of discussion in _private_. Bein= g > _willfully_ ignorant of your obligations which you would otherwise know a= bout > is just willfully infringing in a pretty dress. >=20 > The goal of the developers here is to not infringe on third part patents.= Full > stop. >=20 > The objection is to a public discussion- because patent discussion must b= e > informed by the relevant experts, because it no one wants to be dragged i= n to > testify on random patents, and because exposing your strategy to a possib= le > attacker makes it weaker, etc. >=20 >=20 >=20 >=20 >=20 > ________________________________ > From: codec-bounces@ietf.org > To: Jean-Marc Valin > Cc: codec@ietf.org > Sent: Wed Oct 20 10:42:27 2010 > Subject: Re: [codec] Seperate list for IPR discutions? > Stephan's suggestion on the boundary "citations of patent language, incl= uding > patent numbers and claim language" is consistent with your boundary. Det= ailed > analysis of proposed work-arounds, would presumably include patent langua= ge. > General discussion of license terms would not. >=20 > Since I wouldn't subscribe or otherwise access the new list, it doesn't m= atter > to me if it is publicly archived or not. >=20 > Based on past IETF experience, it is quite difficult to convince posters = to go > off-line for a private discussion. One practical issue is that the folks= who > do want to discuss this stuff have no easy way to identify themselves. I= 'm > thinking that providing the second list might make that easier. Though > perhaps no one but Christian is willing to talk about patent claims? >=20 > Stephen Botzko >=20 > On Wed, Oct 20, 2010 at 1:27 PM, Jean-Marc Valin > wrote: > Hi, >=20 > Just to make things clear, I think there's two sets of issues regarding I= PR. > One is discussing the exact terms of IPR disclosure/licensing and the oth= er is > discussion about patent themselves. I think the former should remain on t= his > list. OTOH, I think the latter is (unfortunately) best dealt with in priv= ate. > I don't think there is anything to be gained in discussing patent claims = in a > publicly archived list. I don't think even having that on a separate list= is a > good idea and I'm not interested in participating in such a list should i= t be > created. >=20 > Jean-Marc >=20 >=20 > On 10-10-20 01:03 PM, Stephen Botzko wrote: > +1, I'd like to see it handled precisely this way. >=20 > I think it provides a good way forward, and allows those of us who do > not want exposure to patent details to continue to participate on the > main list. >=20 > Stephen Botzko >=20 >=20 > On Wed, Oct 20, 2010 at 12:40 PM, Stephan Wenger > wrote: >=20 > Hi, >=20 > Yes, I think a separate mailing list would be an excellent idea. > There is > IMO no danger to the IETF when using an IETF-hosted list. But, as yo= u > mentioned, there are also other options. Whatever is the easiest > for those > interested in this work. >=20 > My preference would be that this list has a subscription process > independent > from the subscription to codec@ietf.org . As >=20 > the minimum, there must be an > opt-out process. >=20 > I would recommend that the maintainers of the list, periodically, mak= e > people on codec@ietf.org aware of the >=20 > existence of the list and the ongoing > discussions, in a summarized form. I also don't mind if suggested > design > choices spill over to codec@ietf; after all, this is the list where > technical discussions are being held. The borderline would be > citations of > patent language, including patent numbers and claim language. >=20 > Personally, I surely do not want to be involved on this hypothetical = new > list. >=20 > Stephan >=20 >=20 > On 10.20.2010 06:42 , "fluffy@cisco.com " >=20 > > wrote: >=20 >>=20 >> Stephan, as our WG technical advisor on topics like this, do you > have advice >> on if we should set up a separate list for people that want to > talk about >> patent issues related to the standards from this WG? From a > practical point of >> view, someone could easily set up a separate list either on the > ietf.org >=20 >> servers or not on ietf servers. >=20 >=20 > _______________________________________________ > codec mailing list > codec@ietf.org >=20 > https://www.ietf.org/mailman/listinfo/codec >=20 >=20 >=20 >=20 > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec >=20 >=20 >=20 > ________________________________ > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From rchen@broadcom.com Fri Oct 22 22:11:58 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id E086D3A69D5 for ; Fri, 22 Oct 2010 22:11:57 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: 0 X-Spam-Level: X-Spam-Status: No, score=x tagged_above=-999 required=5 tests=[] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id BTUv5QrSxerC for ; Fri, 22 Oct 2010 22:11:57 -0700 (PDT) Received: from mms1.broadcom.com (mms1.broadcom.com [216.31.210.17]) by core3.amsl.com (Postfix) with ESMTP id F190F3A69D3 for ; Fri, 22 Oct 2010 22:11:56 -0700 (PDT) Received: from [10.9.200.131] by mms1.broadcom.com with ESMTP (Broadcom SMTP Relay (Email Firewall v6.3.2)); Fri, 22 Oct 2010 22:13:27 -0700 X-Server-Uuid: 02CED230-5797-4B57-9875-D5D2FEE4708A Received: from IRVEXCHCCR01.corp.ad.broadcom.com ([10.252.49.30]) by IRVEXCHHUB01.corp.ad.broadcom.com ([10.9.200.131]) with mapi; Fri, 22 Oct 2010 22:13:27 -0700 From: "Raymond (Juin-Hwey) Chen" To: "'codec@ietf.org'" Importance: low X-Priority: 5 Date: Fri, 22 Oct 2010 22:13:24 -0700 Thread-Topic: A pitch filtering method to improve the mode 1 CELT codec Thread-Index: Actwfi08ukuKx/BjSx+Q0cCrHKI+EgAAQdGqAAHgM14ABVoyzgBa8iNpABf4WUA= Message-ID: References: In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: x-cr-hashedpuzzle: AAU2 CN8I Cfwz Cz8j C32n D6sb FINh FXng FlCF F1UY Gr92 IVEe Ij66 I76Y JLZi Jzl+; 1; YwBvAGQAZQBjAEAAaQBlAHQAZgAuAG8AcgBnAA==; Sosha1_v1; 7; {F1F323E1-B9D5-431C-8064-EDC34277C9DC}; cgBjAGgAZQBuAEAAYgByAG8AYQBkAGMAbwBtAC4AYwBvAG0A; Sat, 23 Oct 2010 05:13:24 GMT; QQAgAHAAaQB0AGMAaAAgAGYAaQBsAHQAZQByAGkAbgBnACAAbQBlAHQAaABvAGQAIAB0AG8AIABpAG0AcAByAG8AdgBlACAAdABoAGUAIABtAG8AZABlACAAMQAgAEMARQBMAFQAIABjAG8AZABlAGMA x-cr-puzzleid: {F1F323E1-B9D5-431C-8064-EDC34277C9DC} acceptlanguage: en-US MIME-Version: 1.0 X-WSS-ID: 60DCB17D47832346648-01-01 Content-Type: multipart/mixed; boundary=_002_CB68DF4CFBEF4942881AD37AE1A7E8C75F264FFF8CIRVEXCHCCR01c_ Subject: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 23 Oct 2010 05:11:58 -0000 --_002_CB68DF4CFBEF4942881AD37AE1A7E8C75F264FFF8CIRVEXCHCCR01c_ Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable All, In the last few weeks, I have been working on a pitch prefiltering/postfilt= ering method to try to=20 improve the output audio quality of the IETF mode 1 CELT codec for certain = audio signals that CELT=20 low-delay modes give more audible coding distortion than usual (typically n= early periodic solo=20 music instrument signals). I am happy to report that this method has achie= ved fairly significant=20 audio quality improvement for many of these kinds of audio signals that I t= ried. This is achieved=20 without increasing the coding delay and with only a slight increase in the = codec complexity and=20 encoding bit-rate (a potential increase of around 2 kb/s or so; less if a r= ange coder is used). I have been in discussion with Jean-Marc about this and have sent him CELT-= coded audio signals at=20 different bit-rates obtained with or without this method. After listening,= he agreed that the=20 method made a significant improvement when compared with the current CELT i= mplementation, and he=20 would like to try to incorporate this method into his CELT C code to evalua= te and verify the=20 performance gain and possibly further develop the method. =20 Jean-Marc asked me to write a detailed description of the method in a docum= ent and send the=20 document to the IETF codec WG email reflector so he can read it, understand= it, and then implement=20 it himself in his CELT C code. I am attaching such a technical description= of the method. Note that although this document is 7 pages long, the pitch prefiltering/po= stfiltering method is=20 actually quite simple. The document is long only because (1) I tried to de= scribe the technical=20 details as much as I can so Jean-Marc can implement it exactly as I had it = without any ambiguity,=20 and (2) I also described/discussed a few alternative approaches for some of= the algorithm=20 components in case we need to go those routes. Note also that this is "fresh out of the oven" as I only had time to finish= a very rough first cut=20 for a proof-of-concept simulation and did not have time yet to refine or tu= ne it. Thus, I am sure=20 there are still many rough edges that need to be smoothed out and many area= s that can be improved=20 upon. However, I am very much encouraged that even such a rough first cut = showed a lot of promise=20 as it made a very large audio quality improvement at several different bit-= rates for a trumpet=20 test signal, which is a signal that CELT low-delay modes have problems with= . Therefore, after=20 this pitch prefiltering/postfiltering method is integrated into CELT and fu= rther refined/tuned, I=20 think it has the potential to make the IETF mode 1 CELT codec more robust t= o different audio=20 signals. That is, it can potentially improve CELT performance for many sig= nals that are currently=20 difficult for CELT to code well, and thus ensure a more uniform audio quali= ty level across a wider=20 variety of different audio signals. If Jean-Marc's integration, testing, a= nd further development=20 of this pitch filtering method confirms this potential, then it is proposed= that this pitch=20 prefiltering/postfiltering method be incorporated into Mode 1 of the IETF O= pus codec to make the=20 codec more robust to different input signal types. Cheers, Raymond --_002_CB68DF4CFBEF4942881AD37AE1A7E8C75F264FFF8CIRVEXCHCCR01c_ Content-Type: application/msword; name="Pitch filtering for CELT.doc" Content-Description: Pitch filtering for CELT.doc Content-Disposition: attachment; filename="Pitch filtering for CELT.doc"; size=274432; creation-date="Fri, 22 Oct 2010 21:29:34 GMT"; modification-date="Fri, 22 Oct 2010 22:12:45 GMT" Content-Transfer-Encoding: base64 0M8R4KGxGuEAAAAAAAAAAAAAAAAAAAAAPgADAP7/CQAGAAAAAAAAAAAAAAAFAAAADgIAAAAAAAAA EAAATQAAAAMAAAD+////AAAAABECAABrAAAAsgAAABACAAAPAgAA//////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// ///////////////////////////////////////////////////////////////////////////s pcEAA4AJBAAA+BK/AAAAAAAAEAAAAAAACAAAkU8AAA4AYmpiaoDOgM4AAAAAAAAAAAAAAAAAAAAA AAAJBBYANIIAAOKkAADipAAAkUYAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD//w8AAAAA AAAAAAD//w8AAAAAAAAAAAD//w8AAAAAAAAAAAAAAAAAAAAAALcAAAAAAAIIAAAAAAAAAggAAEUV AAAAAAAARRUAAAAAAABFFQAAAAAAAEUVAAAAAAAARRUAABQAAAAAAAAAAAAAAP////8AAAAAWRUA AAAAAABZFQAAAAAAAFkVAAAAAAAAWRUAAKwAAAAFFgAAJAAAAFkVAAAAAAAA5CQAAKgBAAApFgAA 1AIAAP0YAAAAAAAA/RgAAAAAAAD9GAAAAAAAAP0YAAAAAAAA2BkAAAAAAADYGQAAAAAAANgZAAAA AAAAbyQAAAIAAABxJAAAAAAAAHEkAAAAAAAAcSQAAAAAAABxJAAAAAAAAHEkAAAAAAAAcSQAAAAA AACMJgAAogIAAC4pAACKAAAAcSQAAC0AAAAAAAAAAAAAAAAAAAAAAAAARRUAAAAAAADYGQAAAAAA AAAAAAAAAAAAAAAAAAAAAADYGQAAAAAAANgZAAAAAAAA2BkAAAAAAADYGQAAAAAAAHEkAAAAAAAA AAAAAAAAAABFFQAAAAAAAEUVAAAAAAAA/RgAAAAAAAAAAAAAAAAAAP0YAADbAAAAniQAABYAAAD8 GwAAAAAAAPwbAAAAAAAA/BsAAAAAAADYGQAAUgAAAEUVAAAAAAAA/RgAAAAAAABFFQAAAAAAAP0Y AAAAAAAAbyQAAAAAAAAAAAAAAAAAAPwbAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAA2BkAAAAAAABvJAAAAAAAAAAAAAAAAAAA/BsAAAAAAAD8GwAA VgAAAP8iAABAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAwyMAAAAAAAD9GAAAAAAAAP////8AAAAA0E27vGty ywEAAAAAAAAAAFkVAAAAAAAAKhoAAAAAAAA/IwAADAAAAAAAAAAAAAAAWyQAABQAAAC0JAAAMAAA AOQkAAAAAAAASyMAAHgAAAC4KQAAAAAAACoaAADSAQAAuCkAABgAAADDIwAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAALgpAAAAAAAAAAAAAAAAAABFFQAAAAAAAMMjAACYAAAA2BkAAAAAAADYGQAAAAAAAPwb AAAAAAAA2BkAAAAAAADYGQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA2BkA AAAAAADYGQAAAAAAANgZAAAAAAAAcSQAAAAAAABxJAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAA/BsAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAANgZAAAA AAAA2BkAAAAAAADYGQAAAAAAAOQkAAAAAAAA2BkAAAAAAADYGQAAAAAAANgZAAAAAAAA2BkAAAAA AAAAAAAAAAAAAP////8AAAAA/////wAAAAD/////AAAAAAAAAAAAAAAA/////wAAAAD/////AAAA AP////8AAAAA/////wAAAAD/////AAAAAP////8AAAAA/////wAAAAD/////AAAAAP////8AAAAA /////wAAAAD/////AAAAAP////8AAAAA/////wAAAAD/////AAAAALgpAAAAAAAA2BkAAAAAAADY GQAAAAAAANgZAAAAAAAA2BkAAAAAAADYGQAAAAAAANgZAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADYGQAAAAAAANgZAAAAAAAA2BkA AAAAAAACCAAACQwAAAsUAAA6AQAABQASAQAACQQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAFBpdGNo IFByZWZpbHRlcmluZyBhbmQgUG9zdGZpbHRlcmluZyBUZWNobmlxdWVzIGZvciBJbXByb3Zpbmcg dGhlIEF1ZGlvIFF1YWxpdHkgb2YgdGhlIElFVEYgT3B1cyBNb2RlIDEgQ29kZWMgKHRoZSBPcmln aW5hbCBDRUxUIENvZGVjKQ0NSnVpbi1Id2V5IChSYXltb25kKSBDaGVuDUJyb2FkY29tIENvcnBv cmF0aW9uDTUzMDAgQ2FsaWZvcm5pYSBBdmVudWUNSXJ2aW5lLCBDYWxpZm9ybmlhIDkyNjEyDVVT QQ0NSW50cm9kdWN0aW9uDQ1UaGlzIGRvY3VtZW50IGRlc2NyaWJlcyBwaXRjaCBwcmVmaWx0ZXJp bmcgYW5kIHBvc3RmaWx0ZXJpbmcgdGVjaG5pcXVlcyBmb3IgaW1wcm92aW5nIHRoZSBvdXRwdXQg YXVkaW8gcXVhbGl0eSBvZiB0aGUgSUVURiBPcHVzIG1vZGUgMSBjb2RlYyAodGhlIG9yaWdpbmFs IENFTFQgY29kZWMpIHdoZW4gdGhlIGlucHV0IGF1ZGlvIHNpZ25hbCBpcyBwZXJpb2RpYyBvciBu ZWFybHkgcGVyaW9kaWMuICBTdWNoIHRlY2huaXF1ZXMgYXJlIGZvdW5kIHRvIHByb3ZpZGUgYSBm YWlybHkgc3Vic3RhbnRpYWwgaW1wcm92ZW1lbnQgdG8gdGhlIG91dHB1dCBhdWRpbyBxdWFsaXR5 IG9mIHRoZSBDRUxUIGNvZGVjIGZvciBuZWFybHkgcGVyaW9kaWMgYXVkaW8gc2lnbmFscyBwcm9k dWNlZCBieSBjZXJ0YWluIHNvbG8gbXVzaWMgaW5zdHJ1bWVudHMgc3VjaCBhcyBhIHRydW1wZXQs IGVzcGVjaWFsbHkgZm9yIGxvdy1kZWxheSBtb2RlcyB3aXRoIHNtYWxsZXIgY29kZWMgZnJhbWUg c2l6ZXMuICANQW4gaW5oZXJlbnQgbGltaXRhdGlvbiBvZiBsb3ctZGVsYXkgdHJhbnNmb3JtIGF1 ZGlvIGNvZGVjcyBlbXBsb3lpbmcgc21hbGwgdHJhbnNmb3JtIHdpbmRvdyBzaXplcyBpcyB0aGF0 IHRoZSBmcmVxdWVuY3kgcmVzb2x1dGlvbiBvZiBzdWNoIHRyYW5zZm9ybXMgaXMgaW5zdWZmaWNp ZW50IHRvIHJlc29sdmUgdGhlIHBpdGNoIGhhcm1vbmljcyBvZiBzb21lIG9mIHRoZSBuZWFybHkg cGVyaW9kaWMgc2VnbWVudHMgb2YgbXVzaWMgYW5kIHNwZWVjaCBzaWduYWxzLiAgQXMgYSByZXN1 bHQsIHN1Y2ggbG93LWRlbGF5IHRyYW5zZm9ybSBjb2RlY3MgdGVuZHMgdG8gZ2l2ZSBtb3JlIGF1 ZGlibGUgY29kaW5nIGRpc3RvcnRpb24gd2hlbiBlbmNvZGluZyBuZWFybHkgcGVyaW9kaWMgbXVz aWMgYW5kIHNwZWVjaCBzaWduYWxzLCBldmVuIHRob3VnaCB0aGUgY29kaW5nIHBlcmZvcm1hbmNl IG1heSBiZSBmaW5lIGZvciBvdGhlciBub24tcGVyaW9kaWMgc2lnbmFscy4gIEluY3JlYXNpbmcg dGhlIHRyYW5zZm9ybSB3aW5kb3cgc2l6ZSB3aWxsIHJlZHVjZSBzdWNoIGNvZGluZyBkaXN0b3J0 aW9uIGZvciBuZWFybHkgcGVyaW9kaWMgbXVzaWMgYW5kIHNwZWVjaCBzaWduYWxzLCBidXQgd2ls bCBhbHNvIGhhdmUgdGhlIHVuZGVzaXJhYmxlIGVmZmVjdCBvZiAgaW5jcmVhc2luZyB0aGUgY29k aW5nIGRlbGF5LiAgVGhlIHBpdGNoIHByZWZpbHRlcmluZyBhbmQgcG9zdGZpbHRlcmluZyB0ZWNo bmlxdWVzIGludHJvZHVjZWQgaW4gdGhpcyBkb2N1bWVudCBoYXMgdGhlIHBvdGVudGlhbCB0byBy ZWR1Y2Ugc3VjaCBjb2RpbmcgZGlzdG9ydGlvbiBzaWduaWZpY2FudGx5IGZvciBuZWFybHkgcGVy aW9kaWMgbXVzaWMgYW5kIHNwZWVjaCBzaWduYWxzIHdpdGhvdXQgaW5jcmVhc2luZyB0aGUgY29k aW5nIGRlbGF5IGFuZCB3aXRoIG9ubHkgYSBzbGlnaHQgaW5jcmVhc2UgaW4gdGhlIGNvZGVjIGNv bXBsZXhpdHkgYW5kIGVuY29kaW5nIGJpdC1yYXRlICAoYXJvdW5kIDIga2Ivcywgb3IgbGVzcyB3 aXRoIGEgcmFuZ2UgY29kZXIpLg1UaGlzIGRvY3VtZW50IGdpdmVzIGEgZGV0YWlsZWQgdGVjaG5p Y2FsIGRlc2NyaXB0aW9uIG9mIHN1Y2ggcGl0Y2ggcHJlZmlsdGVyaW5nIGFuZCBwb3N0ZmlsdGVy aW5nIHRlY2huaXF1ZXMsIHdpdGggdGhlIGdvYWwgb2YgZW5hYmxpbmcgQ0VMVCBjb2RlYyBkZXZl bG9wZXJzIHRvIGltcGxlbWVudCBhbmQgaW50ZWdyYXRlIHRoZXNlIHRlY2huaXF1ZXMgaW4gdGhl IGN1cnJlbnQgQyBjb2RlIGZvciBDRUxUIHRvIGV2YWx1YXRlIHRoZSB0ZWNobmljYWwgbWVyaXRz IG9mIHRoZXNlIHRlY2huaXF1ZXMuICAgSWYgc3VjaCBldmFsdWF0aW9uIGNvbmZpcm1zIG91ciBw cmVsaW1pbmFyeSBmaW5kaW5ncyBvZiBzdWJzdGFudGlhbCBhdWRpbyBxdWFsaXR5IGltcHJvdmVt ZW50IGFzIGV4cGVjdGVkLCB0aGVuIGl0IGlzIHByb3Bvc2VkIHRoYXQgc3VjaCBwaXRjaCBwcmVm aWx0ZXJpbmcgYW5kIHBvc3RmaWx0ZXJpbmcgdGVjaG5pcXVlcyBiZSBpbmNvcnBvcmF0ZWQgaW50 byBNb2RlIDEgb2YgdGhlIElFVEYgT3B1cyBjb2RlYyB0byBlbnN1cmUgYSBtb3JlIHVuaWZvcm0g YXVkaW8gcXVhbGl0eSBsZXZlbCBhY3Jvc3MgYSB3aWRlciB2YXJpZXR5IG9mIGRpZmZlcmVudCBh dWRpbyBzaWduYWxzLg0NRGV0YWlsZWQgRGVzY3JpcHRpb24gb2YgdGhlIFRlY2huaXF1ZXMNDVRo ZSBwaXRjaCBwcmVmaWx0ZXJpbmcgaXMgYSBwcmUtcHJvY2Vzc2luZyB0ZWNobmlxdWUgYXBwbGll ZCB0byB0aGUgaW5wdXQgYXVkaW8gc2lnbmFsIGJlZm9yZSB0aGUgQ0VMVCBlbmNvZGVyLiAgVGhl IHBpdGNoIHBvc3RmaWx0ZXJpbmcgaXMgYSBwb3N0LXByb2Nlc3NpbmcgdGVjaG5pcXVlIGFwcGxp ZWQgdG8gdGhlIG91dHB1dCBkZWNvZGVkIGF1ZGlvIHNpZ25hbCBvZiB0aGUgQ0VMVCBkZWNvZGVy LiAgRmlndXJlIDEgc2hvd3MgYSBzaW1wbGlmaWVkIGhpZ2gtbGV2ZWwgYmxvY2sgZGlhZ3JhbSBv ZiBzdWNoIGFuIGFycmFuZ2VtZW50Lg0TIEVNQkVEIFZpc2lvLkRyYXdpbmcuMTEgIBQBFQ1GaWcu IDEgU2ltcGxpZmllZCBoaWdoLWxldmVsIGJsb2NrIGRpYWdyYW0gb2YgYSBzeXN0ZW0gaW5jb3Jw b3JhdGluZyAgcGl0Y2ggcHJlZmlsdGVyaW5nIGFuZCBwb3N0ZmlsdGVyaW5nIA0gDVRoZSBwaXRj aCBwcmVmaWx0ZXIgYWRhcHRpdmVseSBib29zdHMgdGhlIGZyZXF1ZW5jeSBjb21wb25lbnRzIGlu IHRoZSBzcGVjdHJhbCB2YWxsZXlzIGJldHdlZW4gcGl0Y2ggaGFybW9uaWNzIHdoZW4gdGhlIGlu cHV0IHNpZ25hbCBleGhpYml0cyBzaWduaWZpY2FudCBwaXRjaCBwZXJpb2RpY2l0eS4gVGhlIGVm ZmVjdCBpcyBlc3NlbnRpYWxseSBhZGFwdGl2ZSBjb21iIGZpbHRlcmluZy4gVGhlIHByZWZpbHRl cmVkIHZlcnNpb24gb2YgdGhlIGlucHV0IGF1ZGlvIHNpZ25hbCBpcyB0aGVuIGVuY29kZWQgYnkg dGhlIENFTFQgZW5jb2RlciBhbmQgZGVjb2RlZCBieSB0aGUgQ0VMVCBkZWNvZGVyIGFzIHVzdWFs LiAgVGhlIGRlY29kZWQgYXVkaW8gc2lnbmFsIGlzIHRoZW4gcGFzc2VkIHRocm91Z2ggYSBjb3Jy ZXNwb25kaW5nIHBpdGNoIHBvc3RmaWx0ZXIsIHdoaWNoIGluIHRoZSBpZGVhbCBjYXNlIGlzIGFu IGV4YWN0IGludmVyc2UgZmlsdGVyIG9mIHRoZSBwaXRjaCBwcmVmaWx0ZXIgZm9yIHRoYXQgZnJh bWUgb2YgdGhlIGF1ZGlvIHNpZ25hbC4gIFRodXMsIHRoZSBwaXRjaCBwb3N0ZmlsdGVyIGF0dGVu dWF0ZXMgdGhlIGludGVyLWhhcm1vbmljIHNwZWN0cmFsIHZhbGxleXMuICBUaGUgY29tYmluZWQg ZWZmZWN0IG9mIHRoZSBwaXRjaCBwcmVmaWx0ZXIgYW5kIHBpdGNoIHBvc3RmaWx0ZXIgaW4gdGhl IHN5c3RlbSBvZiBGaWcuIDEgaXMgdG8gc2hhcGUgdGhlIHNwZWN0cnVtIG9mIHRoZSBjb2Rpbmcg bm9pc2Ugc3VjaCB0aGF0IHRoZSBub2lzZSBzcGVjdHJ1bSBpbiB0aGUgaW50ZXItaGFybW9uaWMg c3BlY3RyYWwgdmFsbGV5IHJlZ2lvbnMgaXMgYXR0ZW51YXRlZCByZWxhdGl2ZSB0byB0aGF0IG9m IGEgc3lzdGVtIHdpdGhvdXQgdGhlIHBpdGNoIHByZWZpbHRlciBhbmQgcG9zdGZpbHRlci4NVG8g YWxsb3cgdGhlIHBpdGNoIHBvc3RmaWx0ZXIgdG8gYmUgYW4gZXhhY3QgaW52ZXJzZSBmaWx0ZXIg b2YgdGhlIHBpdGNoIHByZWZpbHRlciwgaXQgaXMgbmVjZXNzYXJ5IHRvIHRyYW5zbWl0IHRoZSBw aXRjaCBwZXJpb2QgYW5kIHRoZSBwaXRjaCBmaWx0ZXIgdGFwKHMpIG9mIHRoZSBwaXRjaCBwcmVm aWx0ZXIgZnJvbSB0aGUgZW5jb2RlciBzaWRlIHRvIHRoZSBkZWNvZGVyIHNpZGUuICBGaWcuIDIg ZGVwaWN0cyBhIG1vcmUgZGV0YWlsZWQgYmxvY2sgZGlhZ3JhbSBzaG93aW5nIHBpdGNoIHBhcmFt ZXRlciBlc3RpbWF0aW9uLCBxdWFudGl6YXRpb24sIHRyYW5zbWlzc2lvbiwgYW5kIGRlY29kaW5n LiAgVGhlIG1haW4gYXVkaW8gc2lnbmFsIHBhdGggYW5kIHRoZSBtYWluIGJsb2NrcyBpbiBGaWcu IDEgYXJlIHNob3duIHdpdGggdGhpY2sgbGluZXMuDRMgRU1CRUQgVmlzaW8uRHJhd2luZy4xMSAg FAEVDUZpZy4gMiAgQSBtb3JlIGRldGFpbGVkIGJsb2NrIGRpYWdyYW0gb2YgYSBzeXN0ZW0gaW5j b3Jwb3JhdGluZyAgcGl0Y2ggcHJlZmlsdGVyaW5nIGFuZCBwb3N0ZmlsdGVyaW5nIA0gDUl0IHNo b3VsZCBiZSBub3RlZCB0aGF0IHRvIGFjaGlldmUgdGhlIGRlc2lyZWQgbm9pc2Ugc3BlY3RyYWwg c2hhcGluZyBlZmZlY3Qgb2YgYXR0ZW51YXRpbmcgdGhlIHNwZWN0cmFsIHZhbGxleXMgYmV0d2Vl biBwaXRjaCBoYXJtb25pY3MgYWRlcXVhdGVseSwgaXQgaXMgZXNzZW50aWFsIGZvciB0aGUgcGl0 Y2ggcGFyYW1ldGVyIGVzdGltYXRvciB0byBleHRyYWN0IHRoZSB0cnVlIHBpdGNoIHBlcmlvZCBy YXRoZXIgdGhhbiB0aGUgaW50ZWdlciBtdWx0aXBsZXMgb2YgdGhlIHRydWUgcGl0Y2ggcGVyaW9k LiAgQW55IGxvdy1jb21wbGV4aXR5IHBpdGNoIGVzdGltYXRvciBjYW4gYmUgdXNlZCBoZXJlIHRv IGV4dHJhY3QgdGhlIHBpdGNoIHBlcmlvZCBhdCBhIHJhdGUgb2YgYWJvdXQgb25lIHBpdGNoIHBl cmlvZCBldmVyeSA1IG1zIG9yIHNvLCBhcyBsb25nIGFzIGl0IGNhbiBleHRyYWN0IHRoZSB0cnVl IHBpdGNoIHBlcmlvZCAocmF0aGVyIHRoYW4gaW50ZWdlciBtdWx0aXBsZXMgb2YgaXQpIGF0IGxl YXN0IHRoZSB2YXN0IG1ham9yaXR5IG9mIHRoZSB0aW1lLiAgDVRoZSBwaXRjaCBwYXJhbWV0ZXIg ZXN0aW1hdG9yIGluIEZpZy4gMiBhbHNvIG5lZWRzIHRvIGNhbGN1bGF0ZSB0aGUgcGl0Y2ggZmls dGVyIHRhcChzKSB0byBiZSB1c2VkIGJ5IHRoZSBwaXRjaCBwcmVmaWx0ZXIgYW5kIHBvc3RmaWx0 ZXIuICBJbiBpdHMgc2ltcGxlc3QgZm9ybSwgYSBzaW5nbGUgbm9uLXplcm8gZmlsdGVyIHRhcCBh dCBhIGJ1bGsgZGVsYXkgb2YgdGhlIHBpdGNoIHBlcmlvZCBpcyB1c2VkIGZvciBib3RoIHRoZSBw aXRjaCBwcmVmaWx0ZXIgYW5kIHBvc3RmaWx0ZXIuICBBbHRob3VnaCBpdCBpcyBzaW1wbGUsIHRo ZSBkb3duIHNpZGUgaXMgdGhhdCB0aGUgZGlmZmVyZW5jZSBiZXR3ZWVuIHBlYWtzIGFuZCB2YWxs ZXlzIGluIHRoZSBmcmVxdWVuY3kgcmVzcG9uc2VzIG9mIHN1Y2ggc2luZ2xlLXRhcCBwaXRjaCBm aWx0ZXJzIHJlbWFpbiBjb25zdGFudCBhY3Jvc3MgdGhlIGZyZXF1ZW5jeSByYW5nZS4gIFRodXMs IGZvciB0aG9zZSBhdWRpbyBzaWduYWxzIHdoZXJlIHRoZSBwaXRjaCBoYXJtb25pYyBwZWFrcyBh cmUgcHJvbWluZW50IG9ubHkgaW4gbG93IGZyZXF1ZW5jaWVzLCBzdWNoIHNpbmdsZS10YXAgcGl0 Y2ggZmlsdGVycyBtYXkgaW50cm9kdWNlIG1vcmUgcGVyaW9kaWNpdHkgaW4gdGhlIGhpZ2gtZnJl cXVlbmN5IHJlZ2lvbiB0aGFuIGluIHRoZSBvcmlnaW5hbCBpbnB1dCBhdWRpbyBzaWduYWwuICBU byBhdm9pZCB0aGlzIHByb2JsZW0sIG9uZSBjYW4gdXNlIG11bHRpcGxlIGZpbHRlciB0YXBzIChm b3IgZXhhbXBsZSwgMiwgMywgNCwgb3IgNSB0YXBzKSBhcm91bmQgdGhlIGJ1bGsgZGVsYXkgb2Yg dGhlIHBpdGNoIHBlcmlvZC4gIEJ5IHByb3Blcmx5IGNob29zaW5nIHRoZSBmaWx0ZXIgdGFwcywg b25lIGNhbiByZWR1Y2UgdGhlIGRpZmZlcmVuY2UgYmV0d2VlbiB0aGUgcGVha3MgYW5kIHZhbGxl eXMgb2YgdGhlIGZyZXF1ZW5jeSByZXNwb25zZSBhcyB0aGUgZnJlcXVlbmN5IGluY3JlYXNlcy4g IEl0IGlzIGNlcnRhaW5seSBhbHNvIHBvc3NpYmxlIHRvIHVzZSBzdWItYmFuZCBkZWNvbXBvc2l0 aW9uIGFuZCBhcHBseSBwaXRjaCBwcmVmaWx0ZXJpbmcgYW5kIHBvc3RmaWx0ZXJpbmcgb25seSB0 byB0aGUgbG93ZXIgYmFuZCwgYWx0aG91Z2ggZG9pbmcgc28gaW5ldml0YWJseSBpbmN1cnMgYWRk aXRpb25hbCBmaWx0ZXJpbmcgZGVsYXkgYW5kIGNvbXBsZXhpdHkuDSBJbiBvdXIgcHJlbGltaW5h cnkgc2ltdWxhdGlvbnMsIGV2ZW4gc2ltcGxlIHNpbmdsZS10YXAgcGl0Y2ggcHJlZmlsdGVyIGFu ZCBwb3N0ZmlsdGVyIGNhbiBhY2hpZXZlIHNpZ25pZmljYW50IGF1ZGlvIHF1YWxpdHkgaW1wcm92 ZW1lbnQgZm9yIG1hbnkgYXVkaW8gc2lnbmFscy4gIFRoZSBmb2xsb3dpbmcgZGVzY3JpcHRpb24g d2lsbCBjb25jZW50cmF0ZSBvbiBzdWNoIHNpbmdsZS10YXAgcGl0Y2ggcHJlZmlsdGVyIGFuZCBw b3N0ZmlsdGVyLiAgIFRoZSBkZXNjcmlwdGlvbiBjYW4gZWFzaWx5IGJlIGdlbmVyYWxpemVkIHRv IHRoZSBtdWx0aS10YXAgZmlsdGVyIGNhc2UuDVRoZSBwaXRjaCBwcmVmaWx0ZXIgY2FuIHRha2Ug c2V2ZXJhbCBwb3NzaWJsZSBmb3Jtcy4gIEl0IGNhbiBiZSBhbiBhbGwtemVybyBmaWx0ZXIsIGFu IGFsbC1wb2xlIGZpbHRlciwgb3IgYSBwb2xlLXplcm8gZmlsdGVyLiAgQXQgaXRzIHNpbXBsZXN0 LCB0aGUgcGl0Y2ggcHJlZmlsdGVyIGNhbiBiZSBpbXBsZW1lbnRlZCBhcyBhbiBhbGwtemVybyBG aW5pdGUgSW1wdWxzZSBSZXNwb25zZSAoRklSKSBmaWx0ZXIgd2l0aCBhIHNpbmdsZSBmaWx0ZXIg dGFwIGF0IGEgYnVsayBkZWxheSBvZiB0aGUgcGl0Y2ggcGVyaW9kLiAgTW9yZSBzcGVjaWZpY2Fs bHksIGxldCBiIGRlbm90ZSB0aGUgZmlsdGVyIHRhcCB3ZWlnaHQgYW5kIGxldCAgcCBkZW5vdGUg dGhlIHBpdGNoIHBlcmlvZCBpbiBzYW1wbGVzLCBpLmUuLCB0aGUgdGltZSBwZXJpb2QgYnkgd2hp Y2ggdGhlIG5lYXJseSBwZXJpb2RpYyBpbnB1dCBhdWRpbyBzaWduYWwgcmVwZWF0cyBpdHMgd2F2 ZWZvcm0gYXBwcm94aW1hdGVseSwgdGhlbiwgdGhlIHJlbGF0aW9uc2hpcCBiZXR3ZWVuIHRoZSBp bnB1dCBzaWduYWwgc2FtcGxlIHMobikgYW5kIG91dHB1dCBzaWduYWwgc2FtcGxlIGQobikgYXQg dGltZSBpbmRleCBuIGlzIGRlZmluZWQgYnkgdGhlIGZvbGxvd2luZyBkaWZmZXJlbmNlIGVxdWF0 aW9uLg0TIEVNQkVEIEVxdWF0aW9uLjMgIBQBFQ1TdWNoIGFuIGFsbC16ZXJvIEZJUiBmaWx0ZXIg aGFzIGEgdHJhbnNmZXIgZnVuY3Rpb24gb2YNEyBFTUJFRCBFcXVhdGlvbi4zICAUARUuDVR5cGlj YWxseSwgdGhlIHRhcCB3ZWlnaHQgaXMgY2hvc2VuIHRvIGJlIBMgRU1CRUQgRXF1YXRpb24uMyAg FAEVLCB3aXRoIGIgPSAwIHdoZW4gdGhlcmUgaXMgbm90IHN1ZmZpY2llbnQgcGVyaW9kaWNpdHkg ZGV0ZWN0ZWQgaW4gdGhlIGlucHV0IGF1ZGlvIHNpZ25hbC4gIFRoZSBtb3JlIHBlcmlvZGljIHRo ZSBpbnB1dCBhdWRpbyBzaWduYWwsIHRoZSBjbG9zZXIgYiBpcyB0byAxLiAgVGhlIGZyZXF1ZW5j eSByZXNwb25zZSBvZiBzdWNoIGEgZmlsdGVyIEgoeikgaGFzIGVxdWFsbHktc3BhY2VkIGRvd253 YXJkIHNwaWtlcyBsb2NhdGVkIGF0IHRoZSBoYXJtb25pYyBmcmVxdWVuY2llcyBvZiB0aGUgcGl0 Y2ggZnJlcXVlbmN5IBMgRU1CRUQgRXF1YXRpb24uMyAgFAEVIEh6LCB3aGVyZSATIEVNQkVEIEVx dWF0aW9uLjMgIBQBFSBpcyB0aGUgc2FtcGxpbmcgcmF0ZSBvZiB0aGUgaW5wdXQgYXVkaW8gc2ln bmFsIGluIEh6LiAgDVRoZSBwaXRjaCBwb3N0ZmlsdGVyIGlzIHRoZSBleGFjdCBpbnZlcnNlIGZp bHRlciBvZiB0aGUgcGl0Y2ggcHJlZmlsdGVyLiAgRGVub3RlIGl0cyBpbnB1dCBzaWduYWwgYXMg EyBFTUJFRCBFcXVhdGlvbi4zICAUARVhbmQgb3V0cHV0IHNpZ25hbCBhcyATIEVNQkVEIEVxdWF0 aW9uLjMgIBQBFSBhdCB0aW1lIGluZGV4IG4sIHRoZW4gdGhlIGlucHV0LW91dHB1dCByZWxhdGlv bnNoaXAgb2YgdGhlIHBpdGNoIHBvc3RmaWx0ZXIgaXMgZ2l2ZW4gYnkNEyBFTUJFRCBFcXVhdGlv bi4zICAUARUNU3VjaCBhIHBpdGNoIHBvc3RmaWx0ZXIgaGFzIGEgdHJhbnNmZXIgZnVuY3Rpb24g b2YgDRMgRU1CRUQgRXF1YXRpb24uMyAgFAEVDVRoaXMgZmlsdGVyIGhhcyBhIGZyZXF1ZW5jeSBy ZXNwb25zZSB0aGF0IGlzIGEgbWlycm9yIGltYWdlIG9mIHRoZSBob3Jpem9udGFsIGF4aXMsIHdp dGggdXB3YXJkIHNwaWtlcyBsb2NhdGVkIGF0IHRoZSBoYXJtb25pYyBmcmVxdWVuY2llcyBvZiB0 aGUgcGl0Y2ggZnJlcXVlbmN5IBMgRU1CRUQgRXF1YXRpb24uMyAgFAEVIEh6LiAgDU9uZSBjYW4g dXNlIGFuIGFsbC1wb2xlIHBpdGNoIHByZWZpbHRlciBpbiB0aGUgZm9ybSBvZiANEyBFTUJFRCBF cXVhdGlvbi4zICAUARUNYW5kIGEgY29ycmVzcG9uZGluZyBhbGwtemVybyBwaXRjaCBwb3N0Zmls dGVyIGluIHRoZSBmb3JtIG9mIA0TIEVNQkVEIEVxdWF0aW9uLjMgIBQBFQ1GdXJ0aGVybW9yZSwg b25lIGNhbiBldmVuIHVzZSBwb2xlLXplcm8gZmlsdGVycyBmb3IgYm90aCB0aGUgcGl0Y2ggcHJl ZmlsdGVyIGFuZCB0aGUgcGl0Y2ggcG9zdGZpbHRlciwgaW4gdGhlIGZvcm1zIG9mIA0TIEVNQkVE IEVxdWF0aW9uLjMgIBQBFQ1hbmQNEyBFTUJFRCBFcXVhdGlvbi4zICAUARUNcmVzcGVjdGl2ZWx5 LiAgVGhlc2UgbGFzdCBmb3JtcyBvZiBwb2xlLXplcm8gZmlsdGVycyBnaXZlcyBtb3JlIGNvbnRy b2wgb2YgdGhlIHNoYXBlIG9mIHRoZSBmcmVxdWVuY3kgcmVzcG9uc2UgYXJvdW5kIGVhY2ggcGl0 Y2ggaGFybW9uaWMsIGFsdGhvdWdoIGF0IGEgY29zdCBvZiBtb3JlIGNvbXB1dGF0aW9uYWwgY29t cGxleGl0eS4NVG8ga2VlcCB0aGUgcGl0Y2ggcHJlZmlsdGVyIGFuZCBwb3N0ZmlsdGVyIGxvdyBj b21wbGV4aXR5IGFuZCBjb25jZXB0dWFsbHkgc2ltcGxlLCB0aGUgcmVzdCBvZiB0aGlzIGRvY3Vt ZW50IHdpbGwgY29uY2VudHJhdGUgb24gdGhlIGZpcnN0IGV4YW1wbGUgYWJvdmUgd2hlcmUgEyBF TUJFRCBFcXVhdGlvbi4zICAUARUgYW5kIBMgRU1CRUQgRXF1YXRpb24uMyAgFAEVLiAgSXQgaXMg YWdhaW4gcmVsYXRpdmVseSBlYXN5IHRvIGdlbmVyYWxpemUgb3IgZXh0ZW5kIHRvIG90aGVyIGZv cm1zIG9mIHBpdGNoIGZpbHRlcnMuDU5vdGUgdGhhdCBpZiB0aGVyZSBpcyBubyBxdWFudGl6YXRp b24gYXBwbGllZCBvbiB0aGUgcGl0Y2ggcHJlZmlsdGVyIG91dHB1dCBzaWduYWwgZChuKSwgdGhl biATIEVNQkVEIEVxdWF0aW9uLjMgIBQBFSwgYW5kIHRodXMgZnJvbSB0aGUgdHdvIGRpZmZlcmVu Y2UgZXF1YXRpb25zIGFib3ZlLCBpdCBmb2xsb3dzIHRoYXQNEyBFTUJFRCBFcXVhdGlvbi4zICAU ARUNQnkgZW5zdXJpbmcgdGhhdCB0aGUgc2V0IG9mIHtwLCBifSBpcyBpZGVudGljYWwgaW4gdGhl IHBpdGNoIHByZWZpbHRlciBhbmQgdGhlIHBpdGNoIHBvc3RmaWx0ZXIsIGFuZCBieSBlbnN1cmlu ZyB0aGF0IHRoZSBzaWduYWwgYXJyYXlzIBMgRU1CRUQgRXF1YXRpb24uMyAgFAEVIGFuZCATIEVN QkVEIEVxdWF0aW9uLjMgIBQBFSBzdGFydCB3aXRoIHRoZSBzYW1lIGluaXRpYWwgY29uZGl0aW9u LCB0aGUgc2Vjb25kIGFuZCB0aGUgdGhpcmQgdGVybSBvbiB0aGUgcmlnaHQgc2lkZSBvZiB0aGUg bGFzdCBlcXVhbCBzaWduIGluIHRoZSBlcXVhdGlvbiBhYm92ZSB3aWxsIGV4YWN0bHkgY2FuY2Vs IGVhY2ggb3RoZXIgb3V0LCByZXN1bHRpbmcgaW4gEyBFTUJFRCBFcXVhdGlvbi4zICAUARUsIHRo YXQgaXMsIHBlcmZlY3QgcmVjb25zdHJ1Y3Rpb24uDU9mIGNvdXJzZSwgd2l0aCB0aGUgcXVhbnRp emF0aW9uIGVmZmVjdCBpbnRyb2R1Y2VkIGJ5IHRoZSBDRUxUIGVuY29kZXIgYW5kIENFTFQgZGVj b2Rlciwgc3VjaCBwZXJmZWN0IHJlY29uc3RydWN0aW9uIHByb3BlcnR5IGlzIGxvc3QuICBIb3dl dmVyLCBpZiB0aGUgcXVhbnRpemF0aW9uIGVycm9yIGlzIHJlbGF0aXZlbHkgc21hbGwsIGkuZS4s IHRoZSBzaWduYWwtdG8tY29kaW5nLW5vaXNlIHJhdGlvIGlzIHJlYXNvbmFibHkgaGlnaCwgdGhl biB0aGUgb3V0cHV0IHNpZ25hbCBvZiB0aGUgcGl0Y2ggcG9zdGZpbHRlciB3aWxsIHN0aWxsIGJl IHJlYXNvbmFibHkgY2xvc2UgdG8gdGhlIGlucHV0IHNpZ25hbCBvZiB0aGUgcGl0Y2ggcHJlZmls dGVyLg1Gb3IgdGhlIHBpdGNoIGZpbHRlciB0YXAgYiwgaXQgaXMgcmVhc29uYWJsZSB0byBtYWtl IGl0IHByb3BvcnRpb25hbCB0byBhIHBhcmFtZXRlciB0aGF0IG1lYXN1cmVzIHRoZSBjb3JyZWxh dGlvbiBiZXR3ZWVuIHRoZSBhZGphY2VudCBwaXRjaCBjeWNsZSB3YXZlZm9ybXMsIHN1Y2ggYXMg dGhlIGNvc2luZSBvZiB0aGUgYW5nbGUgYmV0d2VlbiB0aGUgdmVjdG9yIG9mIHRoZSBjdXJyZW50 IGZyYW1lIG9mIGF1ZGlvIHNpZ25hbCBzYW1wbGVzIGFuZCB0aGUgdmVjdG9yIG9mIHRoZSBhdWRp byBzYW1wbGVzIHRoYXQgYXJlIG9uZSBwaXRjaCBwZXJpb2QgZWFybGllci4gIFNwZWNpZmljYWxs eSwgbGV0IEwgYmUgdGhlIGxlbmd0aCBvZiB0aGUgZnJhbWUgYW5kIGxldCB0aW1lIGluZGV4IG4g PSAxLCAyLCCFLCBMICBjb3JyZXNwb25kIHRvIHRoZSBjdXJyZW50IGZyYW1lLCB0aGVuLCB0aGUg bm9ybWFsaXplZCBjb3JyZWxhdGlvbiBjLCB3aGljaCBpcyB0aGUgY29zaW5lIG9mIHRoZSBhbmds ZSBkZXNjcmliZWQgYWJvdmUsIGlzIGNhbGN1bGF0ZWQgYXMgDRMgRU1CRUQgRXF1YXRpb24uMyAg FAEVDVRvIHJlZHVjZSB0aGUgY29tcGxleGl0eSwgIHRoZSBub3JtYWxpemVkIGNvcnJlbGF0aW9u IG1heSBiZSBhcHByb3hpbWF0ZWQgYnkgdGhlIG9wdGltYWwgdGFwIHdlaWdodCBvZiB0aGUgc2lu Z2xlLXRhcCBwaXRjaCBwcmVkaWN0b3IsIGNhbGN1bGF0ZWQgYXMNEyBFTUJFRCBFcXVhdGlvbi4z ICAUARUNVGhlIHBpdGNoIHByZWZpbHRlciBhbmQgcG9zdGZpbHRlciB0YXAgYiBjYW4gdGhlbiBi ZSBvYnRhaW5lZCBhcw0TIEVNQkVEIEVxdWF0aW9uLjMgIBQBFSAgLg13aGVyZSBhIHJlYXNvbmFi bGUgdmFsdWUgb2YgEyBFTUJFRCBFcXVhdGlvbi4zICAUARUgbWF5IGJlIGluIHRoZSByYW5nZSBv ZiAwLjQgdG8gMC45LCB3aGlsZSBhIHJlYXNvbmFibGUgdmFsdWUgb2YgdGhlIHRocmVzaG9sZCBU IG1heSBiZSBhcm91bmQgMC42LCBhbHRob3VnaCBhIHRocmVzaG9sZCBvZiAwIHdpbGwgd29yaywg dG9vLg1Ob3RlIHRoYXQgaWYgdGhlIGZyYW1lIHNpemUgaXMgdmVyeSBzbWFsbCAoc3VjaCBhcyAy LjUgbXMpLCB0aGUgc3VtbWF0aW9uIG92ZXIgc3VjaCBhIHNob3J0IHBlcmlvZCBtYXkgbm90IGdp dmUgcmVsaWFibGUgcmVzdWx0LiAgSW4gdGhpcyBjYXNlLCBpdCBtYXkgYmUgYmVuZWZpY2lhbCB0 byB1c2UgYSBsb25nZXIgc3VtbWF0aW9uIHdpbmRvdyBieSBpbmNsdWRpbmcgc29tZSBvZiB0aGUg c2lnbmFsIHNhbXBsZXMgaW4gcHJldmlvdXMgZnJhbWVzLg1PbmUgcHJhY3RpY2FsIHByb2JsZW0g d2hlbiBhcHBseWluZyBzdWNoIHBpdGNoIHByZWZpbHRlciBhbmQgcGl0Y2ggcG9zdGZpbHRlciBp cyB0aGF0IHdoZW4gdGhlIHBpdGNoIHBlcmlvZCBwIGFuZCB0aGUgZmlsdGVyIHRhcCBiIGNoYW5n ZSBhdCB0aGUgZnJhbWUgYm91bmRhcnksIHRoZXJlIGlzIG9mdGVuIGEgd2F2ZWZvcm0gZGlzY29u dGludWl0eSBpbiB0aGUgb3V0cHV0IHNpZ25hbHMgb2Ygc3VjaCBmaWx0ZXJzLCBhbmQgdGhpcyB3 aWxsIGNhdXNlIGFuIGF1ZGlibGUgY2xpY2sgYW5kIGNhbiBjYXVzZSB1bmRlc2lyYWJsZSBlZmZl Y3QgaW4gdGhlIGF1ZGlvIGVuY29kZXIuICBUaGlzIHByb2JsZW0gY2FuIGJlIGF2b2lkZWQgYnkg YXBwbHlpbmcgYW4gb3ZlcmxhcC1hZGQgbWV0aG9kIGFzIGRlc2NyaWJlZCBpbiB0aGUgVS5TLiBw YXRlbnQgTm8uIDcsMzUzLDE2OC4gIFNwZWNpZmljYWxseSwgYXQgdGhlIGJlZ2lubmluZyBvZiB0 aGUgY3VycmVudCBmcmFtZSBhbmQgd2l0aCB0aGUgZmlsdGVyIG1lbW9yeSBzZXQgYXQgdGhlIHZh bHVlIGxlZnQgYWZ0ZXIgZmlsdGVyaW5nIHRoZSBsYXN0IHNhbXBsZSBvZiB0aGUgbGFzdCBmcmFt ZSwgdHdvIGZpbHRlcmluZyBvcGVyYXRpb25zIGFyZSBwZXJmb3JtZWQgZm9yIHRoZSBmaXJzdCBL IHNhbXBsZXMgb2YgdGhlIGN1cnJlbnQgZnJhbWUsIHdoZXJlIHRoZSAgdmFsdWUgb2YgSyBpcyB1 c3VhbGx5IGNob3NlbiB0byBjb3JyZXNwb25kIHRvIDIuNSBtcyBvciBsb25nZXIuIFRoZSBmaXJz dCBmaWx0ZXJpbmcgb3BlcmF0aW9uIGlzIHBlcmZvcm1lZCB3aXRoIHRoZSBmaWx0ZXIgcGFyYW1l dGVycyAodGhlIHBpdGNoIHBlcmlvZCBwIGFuZCBwaXRjaCBmaWx0ZXIgdGFwIGIpIG9mIHRoZSBs YXN0IGZyYW1lLCBhbmQgdGhlIHNlY29uZCBmaWx0ZXJpbmcgb3BlcmF0aW9uIGlzIHBlcmZvcm1l ZCB3aXRoIHRoZSBmaWx0ZXIgcGFyYW1ldGVycyBvZiB0aGUgY3VycmVudCBmcmFtZS4gIE5vdGUg dGhhdCBib3RoIGZpbHRlcmluZyBvcGVyYXRpb25zIHNob3VsZCBzdGFydCB3aXRoIHRoZSBzYW1l IGZpbHRlciBtZW1vcnkgdGhhdCB3YXMgbGVmdCBhZnRlciBmaWx0ZXJpbmcgdGhlIGxhc3Qgc2Ft cGxlIG9mIHRoZSBsYXN0IGZyYW1lLiAgQSBmYWRlLW91dCB3aW5kb3cgKGNhbiBiZSBhIGRvd253 YXJkLXNsb3BpbmcgdHJpYW5ndWxhciB3aW5kb3cpIG9mIEsgc2FtcGxlcyBpcyBhcHBsaWVkIHRv IHRoZSBvdXRwdXQgc2lnbmFsIG9mIHRoZSBmaXJzdCBmaWx0ZXJpbmcgb3BlcmF0aW9uLCB3aGls ZSBhIGZhZGUtaW4gd2luZG93IChjYW4gYmUgYW4gdXB3YXJkLXNsb3BpbmcgdHJpYW5ndWxhciB3 aW5kb3cpIG9mIEsgc2FtcGxlcyBpcyBhcHBsaWVkIHRvIHRoZSBvdXRwdXQgc2lnbmFsIG9mIHRo ZSBzZWNvbmQgZmlsdGVyaW5nIG9wZXJhdGlvbgAuACAAVABoAGUAIABzAHUAbQAgAG8AZgAgAHQA aABlACAAZgBhAGQAZQAtAGkAbgAgAGEAbgBkACAAZgBhAGQAZQAtAG8AdQB0ACAAdwBpAG4AZABv AHcAcwAgAGkAcwAgAHUAbgBpAHQAeQAgAGEAdAAgAGUAdgBlAHIAeQAgAG8AbgBlACAAbwBmACAA dABoAGUAIABLACAAcwBhAG0AcABsAGUAcwAuACAAIABUAGgAZQAgAHQAdwBvACAAdwBpAG4AZABv AHcAZQBkACAAZgBpAGwAdABlAHIAIABvAHUAdABwAHUAdAAgAHMAaQBnAG4AYQBsAHMAIABhAHIA ZQAgAGEAZABkAGUAZAAgAHQAbwBnAGUAdABoAGUAcgAgAGEAbgBkACAAdQBzAGUAZAAgAGEAcwAg AHQAaABlACAAZgBpAG4AYQBsACAAZgBpAGwAdABlAHIAIABvAHUAdABwAHUAdAAgAHMAaQBnAG4A YQBsAC4AIABJAHQAIABpAHMAIABhAHMAcwB1AG0AZQBkACAAdABoAGEAdAAgAEsAIABkIiAATAAu ACAAIABJAGYAIABLACAAPAAgAEwALAAgAHQAaABlAG4AIABmAHIAbwBtACAAKABLACsAMQApAC0A dABoACAAcwBhAG0AcABsAGUAIAB0AG8AIAB0aGUgbGFzdCAoTC10aCkgc2FtcGxlIG9mIHRoZSBj dXJyZW50IGZyYW1lLCBvbmx5IG9uZSBmaWx0ZXJpbmcgb3BlcmF0aW9uIGlzIHBlcmZvcm1lZCB1 c2luZyB0aGUgZmlsdGVyIHBhcmFtZXRlcnMgb2YgdGhlIGN1cnJlbnQgZnJhbWUuICBTdWNoIGFu IG92ZXJsYXAtYWRkIGZpbHRlcmluZyBtZXRob2QgZW5zdXJlcyBzbW9vdGggd2F2ZWZvcm0gdHJh bnNpdGlvbiBhbmQgZWxpbWluYXRlIHdhdmVmb3JtIGRpc2NvbnRpbnVpdGllcyBhdCBmcmFtZSBi b3VuZGFyaWVzLiANTm90ZSB0aGF0IHdpdGggc3VjaCBhbiBvdmVybGFwLWFkZCBmaWx0ZXJpbmcg YXBwcm9hY2gsIHRoZSBwZXJmZWN0IHJlY29uc3RydWN0aW9uIHByb3BlcnR5IGZvciB0aGUgbm9u LW92ZXJsYXAtYWRkIHZlcnNpb24gb2YgdGhlIHNpbXBsZSBwaXRjaCBwcmVmaWx0ZXIgYW5kIHBp dGNoIHBvc3RmaWx0ZXIgYXMgZGVzY3JpYmVkIGVhcmxpZXIgbm8gbG9uZ2VyIGhvbGRzIHRydWUu ICBJbiBmYWN0LCBpdCBjYW4gYmUgc2hvd24gdGhhdCB0byBtYWludGFpbiB0aGUgcGVyZmVjdCBy ZWNvbnN0cnVjdGlvbiBwcm9wZXJ0eSwgdGhlIHBhcmFsbGVsIGZpbHRlcmluZyBhbmQgb3Zlcmxh cC1hZGQgb2YgdGhlIHR3byBmaWx0ZXJlZCBvdXRwdXQgc2lnbmFscyBzaG91bGQgYmUgcGVyZm9y bWVkIG5vdCBmb3IgdGhlIGVudGlyZSBhbGwtcG9sZSBwaXRjaCBwb3N0ZmlsdGVyIBMgRU1CRUQg RXF1YXRpb24uMyAgFAEVLCBidXQgb25seSBmb3IgdGhlIGFsbC16ZXJvIEZJUiBmaWx0ZXIgEyBF TUJFRCBFcXVhdGlvbi4zICAUARUgaW4gdGhlIGZlZWRiYWNrIGJyYW5jaCBvZiB0aGUgYWxsLXBv bGUgZmlsdGVyIBMgRU1CRUQgRXF1YXRpb24uMyAgFAEVLiAgRm9yIHRoZSBwaXRjaCBwcmVmaWx0 ZXIgEyBFTUJFRCBFcXVhdGlvbi4zICAUARUsIGFwcGx5aW5nIHRoZSBvdmVybGFwLWFkZCBmaWx0 ZXJpbmcgYXBwcm9hY2ggdG8gdGhlIGVudGlyZSATIEVNQkVEIEVxdWF0aW9uLjMgIBQBFSBmaWx0 ZXIgaXMgbWF0aGVtYXRpY2FsbHkgZXF1aXZhbGVudCB0byBhcHBseWluZyB0aGUgb3ZlcmxhcC1h ZGQgZmlsdGVyaW5nIGFwcHJvYWNoIG9ubHkgdG8gdGhlIGFsbC16ZXJvIEZJUiBmaWx0ZXIgEyBF TUJFRCBFcXVhdGlvbi4zICAUARUgaW4gdGhlIGZlZWQtZm9yd2FyZCBicmFuY2ggb2YgdGhlIGFs bC16ZXJvIGZpbHRlciATIEVNQkVEIEVxdWF0aW9uLjMgIBQBFS4NVGhlIGFsbC16ZXJvIHBpdGNo IHByZWZpbHRlciB3aXRoIG92ZXJsYXAtYWRkIGlzIHJlbGF0aXZlbHkgc3RyYWlnaHRmb3J3YXJk IHRvIGltcGxlbWVudC4gIE9uIHRoZSBvdGhlciBoYW5kLCBkdWUgdG8gdGhlIHJlY3Vyc2l2ZSBu YXR1cmUgb2YgYWxsLXBvbGUgZmlsdGVyaW5nLCB0aGUgYWxsLXBvbGUgcGl0Y2ggcG9zdGZpbHRl ciBuZWVkcyB0byBiZSBoYW5kbGVkIHdpdGggY2FyZSwgZXNwZWNpYWxseSB3aGVuIHRoZSBwaXRj aCBwZXJpb2QgaXMgc21hbGxlciB0aGFuIHRoZSBvdmVybGFwLWFkZCBsZW5ndGggSy4gICBUaGUg bW9zdCBpbXBvcnRhbnQgdGhpbmcgdG8gbm90ZSBpcyB0aGF0IHRoZSB0d28gZmlsdGVyaW5nIG9w ZXJhdGlvbnMgY2Fubm90IGJlIGltcGxlbWVudGVkIGluZGVwZW5kZW50IG9mIGVhY2ggb3RoZXIg Zm9yIHRoZSBlbnRpcmUgSyBzYW1wbGVzIGFuZCB0aGVuIHdpbmRvd2VkIGFuZCBvdmVybGFwLWFk ZGVkIHRvZ2V0aGVyLiAgVGhpcyBpcyBiZWNhdXNlIHRoZSB3YXZlZm9ybSBkaXNjb250aW51aXR5 IGF0IHRoZSBiZWdpbm5pbmcgb2YgdGhlIGN1cnJlbnQgZnJhbWUgcmVzdWx0aW5nIGZyb20gc3Vj aCBpbmRlcGVuZGVudCBmaWx0ZXJpbmcgd2lsbCBiZSByZXBlYXRlZCBiZWZvcmUgdGhlIEsgc2Ft cGxlcyBvZiBvdmVybGFwLWFkZCBwZXJpb2QgaXMgb3ZlciwgYW5kIHRoZXJlZm9yZSB0aGUgb3Zl cmxhcC1hZGQgb3BlcmF0aW9uIHdpbGwgbm90IGJlIGFibGUgdG8gc21vb3RoIG91dCBzdWNoIHJl cGVhdGVkIHdhdmVmb3JtIGRpc2NvbnRpbnVpdGllcyBhZnRlciB0aGUgYmVnaW5uaW5nIG9mIHRo ZSBjdXJyZW50IGZyYW1lLg1UaGUgY29ycmVjdCB3YXkgdG8gaW1wbGVtZW50IHRoZSBvdmVybGFw LWFkZCBhcHByb2FjaCBpcyB0byBwZXJmb3JtIHRoZSBvdmVybGFwLWFkZCBvZiB0aGUgdHdvIGZp bHRlcmluZyBvdXRwdXQgc2lnbmFsIHNhbXBsZS1ieS1zYW1wbGUuICBUaGVuLCB0aGUgd2F2ZWZv cm0gZGlzY29udGludWl0eSBhdCB0aGUgYmVnaW5uaW5nIG9mIHRoZSBmcmFtZSBpcyBhbHJlYWR5 IHNtb290aGVkIG91dCBieSB0aGUgb3ZlcmxhcC1hZGQgb3BlcmF0aW9uIGJ5IHRoZSB0aW1lIHRo ZSBmaWx0ZXJpbmcgb3BlcmF0aW9uIHJlYWNoZWQgb25lIHBpdGNoIHBlcmlvZCBpbnRvIHRoZSBm cmFtZSwgc28gdGhlcmUgd2lsbCBub3QgYmUgYSByZXBlYXRlZCB3YXZlZm9ybSBkaXNjb250aW51 aXR5IHRoZXJlLg1TcGVjaWZpY2FsbHksIGxldCB0aGUgdGltZSBpbmRleCBuIGZvciB0aGUgY3Vy cmVudCBmcmFtZSBiZSBmcm9tIDEgdG8gTCwgYW5kIGxldCATIEVNQkVEIEVxdWF0aW9uLjMgIBQB FSBhbmQgEyBFTUJFRCBFcXVhdGlvbi4zICAUARUgYmUgdGhlIGZhZGUtaW4gd2luZG93IHNhbXBs ZSBhbmQgZmFkZS1vdXQgd2luZG93IHNhbXBsZSBhdCB0aW1lIGluZGV4IG4sIHJlc3BlY3RpdmVs eS4gIEluIGFkZGl0aW9uLCBsZXQgEyBFTUJFRCBFcXVhdGlvbi4zICAUARUgYW5kIBMgRU1CRUQg RXF1YXRpb24uMyAgFAEVIGJlIHRoZSBwaXRjaCBwZXJpb2QgYW5kIHRoZSBwaXRjaCBwb3N0Zmls dGVyIHRhcCBvZiB0aGUgcHJldmlvdXMgZnJhbWUsIHJlc3BlY3RpdmVseS4gIFRoZW4sIHRoZSBh bGwtcG9sZSBwaXRjaCBwb3N0ZmlsdGVyaW5nIHdpdGggb3ZlcmxhcC1hZGQgc2hvdWxkIGJlIHBl cmZvcm1lZCBzYW1wbGUtYnktc2FtcGxlIGZvciB0aGUgZmlyc3QgSyBzYW1wbGVzIG9mIHRoZSBj dXJyZW50IGZyYW1lIGJ5IHRoZSBmb2xsb3dpbmcgcHNldWRvLWNvZGUuDSAgICBGb3IgbiBmcm9t IDEgdG8gSywgY2FsY3VsYXRlIHRoZSBwaXRjaCBwb3N0ZmlsdGVyIG91dHB1dCBzYW1wbGUgYXMN ICATIEVNQkVEIEVxdWF0aW9uLjMgIBQBFQ0gICAgZW5kIA1BZnRlciBmaWx0ZXJpbmcgdGhlIGZp cnN0IEsgc2FtcGxlcywgaWYgTCA+IEssIHRoZW4gdGhlIGZpbHRlcmluZyBmcm9tIHRoZSAoSysx KS10aCBzYW1wbGUgdG8gdGhlIEwtdGggc2FtcGxlIGlzIGp1c3Qgc2ltcGxlIGFsbC1wb2xlIGZp bHRlcmluZyB1c2luZyAgdGhlIGRpZmZlcmVuY2UgZXF1YXRpb24gDRMgRU1CRUQgRXF1YXRpb24u MyAgFAEVLg1JbiBvdXIgc2ltdWxhdGlvbnMsIEsgaXMgY2hvc2VuIHRvIGJlIGNvcnJlc3BvbmRp bmcgdG8gMi41IG1zLCBvciAxMjAgc2FtcGxlcyBhdCA0OCBrSHogc2FtcGxpbmcgcmF0ZS4gIEJ5 IGRvaW5nIHRoaXMsIHdlIGRvbpJ0IG5lZWQgdG8gY2hhbmdlIHRoZSBvdmVybGFwLWFkZCBsZW5n dGggSyBmb3IgYWxsIGZvdXIgcG9zc2libGUgZnJhbWUgc2l6ZXMgb2YgdGhlIE9wdXMgbW9kZSAx IENFTFQgY29kZWMsIHNpbmNlIHRoZSBzbWFsbGVzdCBmcmFtZSBzaXplIGlzIDIuNSBtcy4NSW4g b3VyIHNpbXVsYXRpb25zLCB3ZSBmb3VuZCB0aGF0IGV2ZW4gaWYgd2UgZG8gbm90IGFwcGx5IGEg cGl0Y2ggcHJlZmlsdGVyIGFuZCBvbmx5IGFwcGx5IGEgcGl0Y2ggcG9zdGZpbHRlciwgdGhlIHJl c3VsdGluZyBvdXRwdXQgYXVkaW8gcXVhbGl0eSBpcyBzdGlsbCBpbXByb3ZlZCBub3RpY2VhYmx5 IGNvbXBhcmVkIHdpdGggbm8gcG9zdGZpbHRlcmluZywgYWx0aG91Z2ggdGhlIGltcHJvdmVtZW50 IGlzIHNvbWV3aGF0IGxlc3MgdGhhbiB0aGUgY2FzZSB3aGVuIGJvdGggdGhlIHBpdGNoIHByZWZp bHRlciBhbmQgcGl0Y2ggcG9zdGZpbHRlciBhcmUgdXNlZC4gIFRoZXJlZm9yZSwgaXQgaXMgcG9z c2libGUgdG8gYWNoaWV2ZSBhdWRpbyBxdWFsaXR5IGVuaGFuY2VtZW50IJNmb3IgZnJlZZQgd2l0 aG91dCBzcGVuZGluZyBhZGRpdGlvbmFsIGJpdHMgdG8gdHJhbnNtaXQgdGhlIHBpdGNoIHBlcmlv ZCBhbmQgcGl0Y2ggZmlsdGVyIHRhcChzKSBmcm9tIHRoZSBlbmNvZGVyIHNpZGUgdG8gdGhlIGRl Y29kZXIgc2lkZS4gIFRoZSBvbmx5IHByaWNlIHBhaWQgaXMgc29tZSBhZGRpdGlvbmFsIGNvbXBs ZXhpdHkuICBJbiB0aGlzIGNhc2UsIHRoZSBwaXRjaCBwZXJpb2QgYW5kIHRoZSBwaXRjaCBwb3N0 ZmlsdGVyIHRhcChzKSB3aWxsIG5lZWQgdG8gYmUgZGVyaXZlZCBsb2NhbGx5IGZyb20gdGhlIGRl Y29kZWQgYXVkaW8gc2lnbmFsLg0NQ29uY2x1c2lvbg0NVGhpcyBkb2N1bWVudCBnaXZlcyAgc3Vm ZmljaWVudCB0ZWNobmljYWwgZGV0YWlscyB0byBhbGxvdyBvbmUgdG8gaW1wbGVtZW50IHRoZSBw aXRjaCBwcmVmaWx0ZXJpbmcgYW5kIHBvc3RmaWx0ZXJpbmcgdGVjaG5pcXVlcyBmb3IgdGhlIElF VEYgT3B1cyBtb2RlIDEgQ0VMVCBjb2RlYyB0byBpbXByb3ZlIGl0cyBvdXRwdXQgYXVkaW8gcXVh bGl0eSBmb3IgY2VydGFpbiBhdWRpbyBzaWduYWxzIHdpdGggYSBoaWdoIGRlZ3JlZSBvZiBwZXJp b2RpY2l0eS4gIFNwZWNpYWwgY2FyZSBuZWVkcyB0byBiZSB0YWtlbiB3aGVuIGltcGxlbWVudGlu ZyBhbiBhbGwtcG9sZSBwaXRjaCBmaWx0ZXIgd2l0aCBvdmVybGFwLWFkZC4gVGhlc2UgcGl0Y2gg cHJlZmlsdGVyaW5nIGFuZCBwb3N0ZmlsdGVyaW5nIHRlY2huaXF1ZXMgZ2l2ZSBzdWJzdGFudGlh bCBhdWRpbyBxdWFsaXR5IGltcHJvdmVtZW50IGZvciBtYW55IGF1ZGlvIHNpZ25hbHMgd2UgdHJp ZWQgdGhhdCB3ZXJlIHByb2JsZW1hdGljIGZvciB0aGUgQ0VMVCBjb2RlYyB0byBjb2RlIHdlbGws IHRodXMgdGhlc2UgcGl0Y2ggZmlsdGVyaW5nIHRlY2huaXF1ZXMgaGF2ZSB0aGUgcG90ZW50aWFs IHRvIG1ha2UgQ0VMVCBtb3JlIHJvYnVzdCB0byBkaWZmZXJlbnQgdHlwZXMgb2YgYXVkaW8gc2ln bmFscy4gIFRoZSBwcmljZSBwYWlkIGlzIGEgc21hbGwgaW5jcmVhc2UgaW4gYml0LXJhdGUgYW5k IGNvZGVjIGNvbXBsZXhpdHkuDQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAIAABCCAAAVwgAAIcIAACICAAAiQgAAKEIAACiCAAA6ggAAOsIAADs CAAA+QgAAPsJAABDCgAARwoAAEwKAABSCgAAdAoAAH8KAACACgAAiAoAAI0KAACTCgAApwoAAKkK AACwCgAA8goAAPMKAAAVDQAAHA0AACwNAADw5PDV8MSzp5jwjeSBdYF1gWmBdYF1gWmBaeRaTloA ABcWaNEPbwBCKgFPSgAAUUoAAHBoAAAAAB0VaLAMigAWaLAMigBCKgFPSgAAUUoAAHBoAAAAABcW aLAMigBCKgFPSgAAUUoAAHBoAAAAABcWaJFBHgBCKgFPSgAAUUoAAHBoAAAAABcWaCgVVQBCKgFP SgAAUUoAAHBoAAAAABUVaL1fBgAWaL1fBgBCKgFwaAAAAAAdFWi9XwYAFmiUZnoAQioBT0oAAFFK AABwaAAAAAAXFmiUZnoAQioBT0oAAFFKAABwaAAAAAAgFWiUZnoAFmiUZnoANgiBQioBT0oAAFFK AABwaAAAAAAAIBVolGZ6ABZovV8GADYIgUIqAU9KAABRSgAAcGgAAAAAAB0VaL1fBgAWaNdE2ABC KgFPSgAAUUoAAHBoAAAAABcWaL1fBgBCKgFPSgAAUUoAAHBoAAAAAB0VaL1fBgAWaL1fBgBCKgFP SgAAUUoAAHBoAAAAAAAeAAgAAIgIAACJCAAAoggAALcIAADOCAAA5wgAAOsIAADsCAAA+QgAAPoI AADzCgAAAA8AAG0RAABuEQAAlREAAJYRAADMEgAA6hIAAFYTAABYEwAA7RYAAJ4YAAD3AAAAAAAA AAAAAAAA8gAAAAAAAAAAAAAAAOgAAAAAAAAAAAAAAADoAAAAAAAAAAAAAAAA6AAAAAAAAAAAAAAA AOgAAAAAAAAAAAAAAADoAAAAAAAAAAAAAAAA4AAAAAAAAAAAAAAAANgAAAAAAAAAAAAAAADRAAAA AAAAAAAAAAAA8gAAAAAAAAAAAAAAAPIAAAAAAAAAAAAAAADKAAAAAAAAAAAAAAAAygAAAAAAAAAA AAAAAMIAAAAAAAAAAAAAAADRAAAAAAAAAAAAAAAA8gAAAAAAAAAAAAAAAPIAAAAAAAAAAAAAAADy AAAAAAAAAAAAAAAA8gAAAAAAAAAAAAAAAPIAAAAAAAAAAAAAAADyAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAgCAAomAAtGAQBnZFlF3QAABgAAFKQAAGdkSDd2AAAGAAAUpAAAZ2SUZnoACAIACiYAC0YB AGdkvV8GAAAHAAADJAFhJAFnZL1fBgAACQAAAyQBFKQUAGEkAWdklGZ6AAAEAABnZL1fBgAABwEA AyQBYSQBZ2S9XwYAABYsDQAAMw0AAGMNAACODQAAnw0AAKANAAAaDgAAJA4AADIOAAA3DgAAPg4A AF8OAACrDgAAwA4AANIOAADwDgAA8g4AAP4OAAD/DgAAAA8AAHcPAACHDwAAqw8AALkPAAC8DwAA vg8AAOkPAABsEQAAbREAAG4RAACVEQAAlhEAAKkRAACtEQAA+BEAAPwRAAAGEgAAzBIAAPPk2OTY 88m9yfPJ87HzsaWx89jzvfO9873zvfOxmo5/c3/zf2cAAAAAAAAAAAAAAAAAAAAAABcWaKInLgBC KgFPSgAAUUoAAHBoAAAAABcWaBlsPwBCKgFPSgAAUUoAAHBoAAAAAB0VaNEPbwAWaNEPbwBCKgFP SgAAUUoAAHBoAAAAABcWaJRmegBCKgFPSgAAUUoAAHBoAAAAABUVaFlF3QAWaFlF3QBCKgFwaAAA AAAXFmg6AzYAQioBT0oAAFFKAABwaAAAAAAXFmhIN3YAQioBT0oAAFFKAABwaAAAAAAXFmhZRd0A QioBT0oAAFFKAABwaAAAAAAdFWiwDIoAFmjRD28AQioBT0oAAFFKAABwaAAAAAAXFmiwDIoAQioB T0oAAFFKAABwaAAAAAAdFWiwDIoAFmiwDIoAQioBT0oAAFFKAABwaAAAAAAXFmjRD28AQioBT0oA AFFKAABwaAAAAAAAJcwSAADNEgAA5hIAAOcSAADpEgAA6hIAAFUTAABWEwAAVxMAAGsTAAB9FAAA hRQAAJ0UAAClFAAACRUAAAsVAABnFQAAaxUAAMQVAADGFQAA7RYAAMsXAADUFwAA2xcAAPkXAACd GAAAnhgAAJ8YAAC4GAAAuRgAALoYAAC7GAAAvBgAAMEYAADCGAAAwxgAANMYAAAiGQAA9/Po9/Pg z8C0wKjAqMCcwLTAnLSctJC0kKiIhHlxiIRpZGlkaQAAAAAAAAAAAAAACRZonXNRADUIgQ8VaKIn LgAWaJ1zUQA1CIEPA2oPDwAAFmidc1EAVQgBFQNqV1ZsUAoIARZonXNRAFUIAVYIAQYWaJ1zUQAA DwNqAAAAABZonXNRAFUIARcWaJ1zUQBCKgFPSgAAUUoAAHBoAAAAABcWaBlsPwBCKgFPSgAAUUoA AHBoAAAAABcWaNEPbwBCKgFPSgAAUUoAAHBoAAAAABcWaKInLgBCKgFPSgAAUUoAAHBoAAAAAB0V aNEPbwAWaNEPbwBCKgFPSgAAUUoAAHBoAAAAACAVaKInLgAWaKInLgA1CIFCKgFPSgAAUUoAAHBo AAAAAAAPFWiiJy4AFmiiJy4ANQiBFQNqkVNsUAoIARZooicuAFUIAVYIAQYWaKInLgAADwNqAAAA ABZooicuAFUIAQAlnhgAALwYAAAjGQAAJRkAAFAbAAAAIAAASiEAAN8jAAD3IwAALiQAAEckAAAh JgAAIicAADonAABuJwAAhicAAEYoAAB+KAAAligAANQoAADsKAAAZCkAAPoAAAAAAAAAAAAAAAD1 AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA9QAAAAAA AAAAAAAAAPUAAAAAAAAAAAAAAADtAAAAAAAAAAAAAAAA6AAAAAAAAAAAAAAAAO0AAAAAAAAAAAAA AAD1AAAAAAAAAAAAAAAA6AAAAAAAAAAAAAAAAO0AAAAAAAAAAAAAAADgAAAAAAAAAAAAAAAA7QAA AAAAAAAAAAAAAPUAAAAAAAAAAAAAAADYAAAAAAAAAAAAAAAAzQAAAAAAAAAAAAAAAMUAAAAAAAAA AAAAAAC9AAAAAAAAAAAAAAAAuAAAAAAAAAAAAAAAAAAAAAAAAAAABAAAZ2RQdH0AAAcAAAMkAWEk AWdkUHR9AAgRAAomAAtGAABnZFB0fQALEgADJAEKJgALRgAAYSQBZ2RQdH0ACBIACiYAC0YAAGdk UHR9AAgRAAomAAtGAABnZFdlJwAABAAAZ2RXZScAAAcAAAMkAWEkAWdkV2UnAAAEAABnZJ1zUQAA BAAAZ2S9XwYAABUiGQAAIxkAACQZAAAlGQAATxsAAFAbAAD6GwAAChwAAA0cAAAxHAAAJx0AAP8f AAABIAAAASEAAEkhAABKIQAAyiEAANohAADeIQAA3yEAAO4hAACMIgAAjSIAALQiAAC1IgAAeyMA AHwjAAB9IwAAfiMAAJkjAACaIwAAmyMAAJwjAACsIwAArSMAAN4jAADfIwAA4CMAAPMjAAD0IwAA 9SMAAPYjAAD3IwAA7+DUyLzIsMiwyLDIsLywvKG8obyhkKGQoZChkKGQoZChkKG8iIR5bIiEAAAZ A2qkNAAAFWg1GQ0AFmhXZScARUj2/1UIARUDakKkaFAKCAEWaFdlJwBVCAFWCAEGFmhXZScAAA8D agAAAAAWaFdlJwBVCAEgFWhXZScAFmhXZScANgiBQioBT0oAAFFKAABwaAAAAAAAHRVoV2UnABZo V2UnAEIqAU9KAABRSgAAcGgAAAAAFxZo1jEtAEIqAU9KAABRSgAAcGgAAAAAFxZoV2UnAEIqAU9K AABRSgAAcGgAAAAAFxZoimSgAEIqAU9KAABRSgAAcGgAAAAAFxZonXNRAEIqAU9KAABRSgAAcGgA AAAAHRVo0Q9vABZonXNRAEIqAU9KAABRSgAAcGgAAAAAIBVooicuABZonXNRADUIgUIqAU9KAABR SgAAcGgAAAAAKvcjAAAtJAAALiQAAC8kAABCJAAAQyQAAEQkAABFJAAARiQAAEckAABxJAAAciQA AIUkAACGJAAAhyQAAIgkAACPJAAAkSQAABglAAAaJQAATCUAAE0lAABOJQAATyUAAK8lAACwJQAA wyUAAMQlAADFJQAAxiUAANElAADSJQAA5SUAAPXt5eHWyeXhvfWu9ZyLrvV/9X/1f/V/9a71bVyu 9a71AAAAAAAAAAAAAAAAAAAAAAAAAAAhA2pfPAAAFWhXZScAFmhXZScARUj0/09KAABRSgAAVQgB IwNqIqFoUAoIARVoV2UnABZoV2UnAE9KAABRSgAAVQgBVggBFxVoV2UnABZoV2UnADYIgU9KAABR SgAAIQNqIzoAABVoV2UnABZoV2UnAEVI+v9PSgAAUUoAAFUIASMDaqWRaFAKCAEVaFdlJwAWaFdl JwBPSgAAUUoAAFUIAVYIAR0DagAAAAAVaFdlJwAWaFdlJwBPSgAAUUoAAFUIARcWaFdlJwBCKgFP SgAAUUoAAHBoAAAAABkDai43AAAVaMF8IAAWaFdlJwBFSOT/VQgBFQNqXKNoUAoIARZoV2UnAFUI AVYIAQYWaFdlJwAADwNqAAAAABZoV2UnAFUIAQ4WaFdlJwBPSgAAUUoAAAAUFWhXZScAFmhXZScA T0oAAFFKAAAg5SUAAOYlAADnJQAA6CUAACAmAACGJgAAhyYAAJomAACbJgAAnCYAAJ0mAACyJgAA syYAAMYmAADHJgAA7dzNwrOfs4hyn7NdTDQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAv A2oIX2xQCggBFWhXZScAFmhXZScANgiBQioBT0oAAFFKAABVCAFWCAFwaAAAAAAgFWhXZScAFmhX ZScANgiBQioBT0oAAFFKAABwaAAAAAAAKQNqAAAAABVoV2UnABZoV2UnADYIgUIqAU9KAABRSgAA VQgBcGgAAAAAKgNqtEAAABVoV2UnABZoV2UnAEIqAUVI9v9PSgAAUUoAAFUIAXBoAAAAAAAsA2oD X2xQCggBFWhXZScAFmhXZScAQioBT0oAAFFKAABVCAFWCAFwaAAAAAAAJgNqAAAAABVoV2UnABZo V2UnAEIqAU9KAABRSgAAVQgBcGgAAAAAAB0VaFdlJwAWaFdlJwBCKgFPSgAAUUoAAHBoAAAAABQV aFdlJwAWaFdlJwBPSgAAUUoAAAAdA2oAAAAAFWhXZScAFmhXZScAT0oAAFFKAABVCAEhA2qiPgAA FWhXZScAFmhXZScARUj0/09KAABRSgAAVQgBIwNqOqFoUAoIARVoV2UnABZoV2UnAE9KAABRSgAA VQgBVggBAA7HJgAAyCYAAMkmAADYJgAA2SYAACEnAAAiJwAAIycAADYnAAA3JwAAOCcAADknAABu JwAAbycAAIInAACDJwAAhCcAAIUnAAAoKAAAKSgAADwoAAA9KAAAPigAAD8oAABFKAAA6NTFtMWo oJyRhKCcoJx5bKDF1MVVP9TFAAAqA2qySgAAFWhXZScAFmhXZScAQioBRUj0/09KAABRSgAAVQgB cGgAAAAAACwDaoRfbFAKCAEVaFdlJwAWaFdlJwBCKgFPSgAAUUoAAFUIAVYIAXBoAAAAAAAZA2ql RwAAFWjBfCAAFmhXZScARUji/1UIARUDauHlaFAKCAEWaFdlJwBVCAFWCAEZA2oBRQAAFWg1GQ0A FmhXZScARUj2/1UIARUDajujaFAKCAEWaFdlJwBVCAFWCAEGFmhXZScAAA8DagAAAAAWaFdlJwBV CAEXFmhXZScAQioBT0oAAFFKAABwaAAAAAAgFWhXZScAFmhXZScANgiBQioBT0oAAFFKAABwaAAA AAAAHRVoV2UnABZoV2UnAEIqAU9KAABRSgAAcGgAAAAAJgNqAAAAABVoV2UnABZoV2UnAEIqAU9K AABRSgAAVQgBcGgAAAAAAC0DattCAAAVaFdlJwAWaFdlJwA2CIFCKgFFSPb/T0oAAFFKAABVCAFw aAAAAAAAGEUoAABGKAAASigAAH4oAAB/KAAAkigAAJMoAACUKAAAlSgAANQoAADVKAAA6CgAAOko AADqKAAA6ygAAOwoAABkKQAAZSkAAHgpAAB5KQAAeikAAHspAAB8KQAAgCkAAIEpAACUKQAAlSkA AJYpAACXKQAAmCkAAFgqAAC8KgAA8+vn3+fUx9/n3+e8r9/noN/nlYjf533f53Jl3+dUoAAAAAAA AAAAACAVaJRmegAWaJRmegBDShgAT0oAAFBKAABRSgAAYUoYAAAZA2ovVQAAFWh/EWIAFmhQdH0A RUjo/1UIARUDam7kaFAKCAEWaFB0fQBVCAFWCAEUFWhQdH0AFmhQdH0AT0oAAFFKAAAAGQNqTVIA ABVofxFiABZoUHR9AEVI6P9VCAEVA2oc5GhQCggBFmhQdH0AVQgBVggBHRVoUHR9ABZoUHR9AEIq AU9KAABRSgAAcGgAAAAAGQNqtE8AABVofxFiABZoUHR9AEVI8v9VCAEVA2rM42hQCggBFmhQdH0A VQgBVggBGQNq80wAABVofxFiABZoUHR9AEVI6P9VCAEVA2qu42hQCggBFmhQdH0AVQgBVggBDwNq AAAAABZoUHR9AFUIAQYWaFB0fQAADxZoUHR9AEIqAXBoAAAAABcWaFdlJwBCKgFPSgAAUUoAAHBo AAAAAAAfZCkAAHwpAACAKQAAmCkAAFgqAACCKwAAOywAAFMsAAD+LQAAcy8AAJQxAACsMQAAQTIA AFkyAACaMgAAtTIAAHUzAAB3NAAADz0AALpAAAD7QwAAhEUAAKVHAADsRwAA9wAAAAAAAAAAAAAA APIAAAAAAAAAAAAAAAD3AAAAAAAAAAAAAAAA7QAAAAAAAAAAAAAAAPIAAAAAAAAAAAAAAADoAAAA AAAAAAAAAAAA4AAAAAAAAAAAAAAAAOgAAAAAAAAAAAAAAADyAAAAAAAAAAAAAAAA8gAAAAAAAAAA AAAAANgAAAAAAAAAAAAAAADyAAAAAAAAAAAAAAAA2AAAAAAAAAAAAAAAANMAAAAAAAAAAAAAAADY AAAAAAAAAAAAAAAA0wAAAAAAAAAAAAAAANMAAAAAAAAAAAAAAADtAAAAAAAAAAAAAAAAzgAAAAAA AAAAAAAAAM4AAAAAAAAAAAAAAADOAAAAAAAAAAAAAAAAzgAAAAAAAAAAAAAAAM4AAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAEAABnZG8MpwAABAAAZ2QAQYgAAAcAAAMkAWEkAWdkAEGIAAAH AAADJAFhJAFnZDVGxQAABAAAZ2SpY8EAAAQAAGdklGZ6AAAEAABnZJ1zUQAABwAAAyQBYSQBZ2RQ dH0AABe8KgAAxCoAAPcqAAD4KgAACysAAAwrAAANKwAADisAABMrAAAUKwAAJysAACgrAAApKwAA KisAAC0rAACCKwAA1SsAANYrAADXKwAA2CsAAOArAADhKwAA9CsAAPPk0OS5o9Dk0OSMdtDk82dW Z1ZnQmcAAAAAAAAAAAAAAAAAAAAAAAAmA2oAAAAAFWipY8EAFmipY8EAQioBT0oAAFFKAABVCAFw aAAAAAAAIBVoqWPBABZoqWPBADYIgUIqAU9KAABRSgAAcGgAAAAAAB0VaKljwQAWaKljwQBCKgFP SgAAUUoAAHBoAAAAACoDaqRaAAAVaFB0fQAWaFB0fQBCKgFFSOj/T0oAAFFKAABVCAFwaAAAAAAA LANqoWBsUAoIARVoUHR9ABZoUHR9AEIqAU9KAABRSgAAVQgBVggBcGgAAAAAACoDag9YAAAVaFB0 fQAWaFB0fQBCKgFFSPL/T0oAAFFKAABVCAFwaAAAAAAALANqnGBsUAoIARVoUHR9ABZoUHR9AEIq AU9KAABRSgAAVQgBVggBcGgAAAAAACYDagAAAAAVaFB0fQAWaFB0fQBCKgFPSgAAUUoAAFUIAXBo AAAAAAAdFWhQdH0AFmhQdH0AQioBT0oAAFFKAABwaAAAAAAXFmhQdH0AQioBT0oAAFFKAABwaAAA AAAAFvQrAAD1KwAA9isAAPcrAAA6LAAAOywAADwsAABPLAAAUCwAAFEsAABSLAAAUywAAFQsAABw LAAAcSwAAHMsAAB0LAAA2ywAANwsAADvLAAA6dO/sKScmI2AnJh0ZVRlVGVAZQAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAJgNqAAAAABVoNUbFABZoNUbFAEIqAU9KAABRSgAAVQgBcGgAAAAA ACAVaDVGxQAWaDVGxQA2CIFCKgFPSgAAUUoAAHBoAAAAAAAdFWg1RsUAFmg1RsUAQioBT0oAAFFK AABwaAAAAAAXFmg1RsUAQioBT0oAAFFKAABwaAAAAAAZA2rDXwAAFWg1GQ0AFmg1RsUARUj2/1UI ARUDapSkaFAKCAEWaDVGxQBVCAFWCAEGFmg1RsUAAA8DagAAAAAWaDVGxQBVCAEXFmipY8EAQioB T0oAAFFKAABwaAAAAAAdFWipY8EAFmipY8EAQioBT0oAAFFKAABwaAAAAAAmA2oAAAAAFWipY8EA FmipY8EAQioBT0oAAFFKAABVCAFwaAAAAAAAKgNqZF0AABVoqWPBABZoqWPBAEIqAUVI9v9PSgAA UUoAAFUIAXBoAAAAAAAsA2rGZmxQCggBFWipY8EAFmipY8EAQioBT0oAAFFKAABVCAFWCAFwaAAA AAAT7ywAAPAsAADxLAAA8iwAAPcsAAD4LAAACy0AAAwtAAANLQAADi0AAMQtAADFLQAA2C0AANkt AADaLQAA2y0AAP0tAAD+LQAAOC4AADwuAABJLgAATS4AAHIvAADp07+wv7CZg7+wv7BsVr+wR7A7 sDuwAAAAABcWaDVGxQBCKgFPSgAAUUoAAHBoAAAAAB0VaKljwQAWaDVGxQBCKgFPSgAAUUoAAHBo AAAAACoDaipnAAAVaDVGxQAWaDVGxQBCKgFFSPb/T0oAAFFKAABVCAFwaAAAAAAALANqCWdsUAoI ARVoNUbFABZoNUbFAEIqAU9KAABRSgAAVQgBVggBcGgAAAAAACoDavxkAAAVaDVGxQAWaDVGxQBC KgFFSPb/T0oAAFFKAABVCAFwaAAAAAAALANqBWdsUAoIARVoNUbFABZoNUbFAEIqAU9KAABRSgAA VQgBVggBcGgAAAAAAB0VaDVGxQAWaDVGxQBCKgFPSgAAUUoAAHBoAAAAACYDagAAAAAVaDVGxQAW aDVGxQBCKgFPSgAAUUoAAFUIAXBoAAAAAAAqA2rWYgAAFWg1RsUAFmg1RsUAQioBRUj2/09KAABR SgAAVQgBcGgAAAAAACwDagNnbFAKCAEVaDVGxQAWaDVGxQBCKgFPSgAAUUoAAFUIAVYIAXBoAAAA ABZyLwAAcy8AAIwvAACNLwAAyzAAAM0wAAD7MAAA/TAAAAgxAAALMQAATTEAAE4xAACBMQAAgjEA AJIxAACTMQAAlDEAAJUxAACoMQAAqTEAAKoxAACrMQAArDEAAEAyAABBMgAAQjIAAFUyAABWMgAA VzIAAFgyAABZMgAAgDIAAIIyAACaMgAAmzIAAK4yAADz5NPk0+TT5NPk0+TT5Me7s6+kl7Ov5Iuz r4Bzs6/k0+Rf5AAAAAAAAAAAAAAAJgNqAAAAABVoAEGIABZoAEGIAEIqAU9KAABRSgAAVQgBcGgA AAAAABkDaidtAAAVaK4XrgAWaABBiABFSMT/VQgBFQNqfb9oUAoIARZoAEGIAFUIAVYIARcWaABB iABCKgFPSgAAUUoAAHBoAAAAABkDaoppAAAVaI930wAWaABBiABFSLz/VQgBFQNqtGFsUAoIARZo AEGIAFUIAVYIAQYWaABBiAAADwNqAAAAABZoAEGIAFUIARcWaJ1zUQBCKgFPSgAAUUoAAHBoAAAA ABcWaFdlJwBCKgFPSgAAUUoAAHBoAAAAACAVaABBiAAWaABBiAA2CIFCKgFPSgAAUUoAAHBoAAAA AAAdFWgAQYgAFmgAQYgAQioBT0oAAFFKAABwaAAAAAAXFmipY8EAQioBT0oAAFFKAABwaAAAAAAA I64yAACvMgAAsDIAALEyAADRMgAA0jIAAOUyAADmMgAA5zIAAOgyAAA2MwAAODMAAHQzAAB3NAAA 4zQAAOU0AAD4NAAA+jQAAP82AAABNwAAMzcAADU3AADCNwAAxDcAANk3AADaNwAA2zcAANw3AABR OAAA6dXBssGym4XBsnSyaFlIWUhZSFlIWUhZSFlIWSAVaG8MpwAWaG8MpwA2CIFCKgFPSgAAUUoA AHBoAAAAAAAdFWhvDKcAFmhvDKcAQioBT0oAAFFKAABwaAAAAAAXFmi9ShoAQioBT0oAAFFKAABw aAAAAAAgFWgAQYgAFmgAQYgANgiBQioBT0oAAFFKAABwaAAAAAAAKgNqVXMAABVoNUbFABZoNUbF AEIqAUVI9P9PSgAAUUoAAFUIAXBoAAAAAAAsA2qOZ2xQCggBFWgAQYgAFmg1RsUAQioBT0oAAFFK AABVCAFWCAFwaAAAAAAAHRVoAEGIABZoAEGIAEIqAU9KAABRSgAAcGgAAAAAJgNqAAAAABVoAEGI ABZoAEGIAEIqAU9KAABRSgAAVQgBcGgAAAAAACYDandwAAAVaABBiAAWaABBiABCKgFPSgAAUUoA AFUIAXBoAAAAAAAsA2pEwWhQCggBFWgAQYgAFmgAQYgAQioBT0oAAFFKAABVCAFWCAFwaAAAAAAc UTgAAOA4AAAjOQAAJDkAALY5AAC3OQAAmDoAAJo6AACiOwAApjsAAKo7AACsOwAAuDsAAMI7AADc OwAA3jsAAAo8AAAMPAAADjwAAA09AAAOPQAAwz4AAMQ+AADXPgAA2D4AANk+AADaPgAAAT8AAAI/ AAAVPwAAFj8AAPPk0+TT5NPk0+TT5NPk0+TTxOS4qZWpfmiVqZWpUQAAAAAAAAAALANq6GJsUAoI ARVobwynABZolGZ6AEIqAU9KAABRSgAAVQgBVggBcGgAAAAAACoDanJ1AAAVaG8MpwAWaJRmegBC KgFFSOj/T0oAAFFKAABVCAFwaAAAAAAALANq22JsUAoIARVobwynABZolGZ6AEIqAU9KAABRSgAA VQgBVggBcGgAAAAAACYDagAAAAAVaG8MpwAWaJRmegBCKgFPSgAAUUoAAFUIAXBoAAAAAAAdFWhv DKcAFmiUZnoAQioBT0oAAFFKAABwaAAAAAAXFmgqFUwAQioBT0oAAFFKAABwaAAAAAAdFWhULC0A FmhvDKcAQioBT0oAAFFKAABwaAAAAAAgFWhvDKcAFmhvDKcANgiBQioBT0oAAFFKAABwaAAAAAAA HRVobwynABZobwynAEIqAU9KAABRSgAAcGgAAAAAFxZo4GxWAEIqAU9KAABRSgAAcGgAAAAAAB4W PwAAFz8AABg/AABHPwAASD8AAFs/AABcPwAAXT8AAF4/AAB5PwAAej8AAI0/AACOPwAAjz8AAJA/ AADMPwAAzT8AAOA/AADhPwAA4j8AAOM/AABXQAAAWEAAAGtAAADq1sfWx7Ca1sfWx4Nt1sfWx1ZA 1sfWxwAAAAAAAAAAACoDalJ/AAAVaG8MpwAWaJRmegBCKgFFSPL/T0oAAFFKAABVCAFwaAAAAAAA LANq4WJsUAoIARVobwynABZolGZ6AEIqAU9KAABRSgAAVQgBVggBcGgAAAAAACoDar18AAAVaG8M pwAWaJRmegBCKgFFSPL/T0oAAFFKAABVCAFwaAAAAAAALANq5GJsUAoIARVobwynABZolGZ6AEIq AU9KAABRSgAAVQgBVggBcGgAAAAAACoDaoJ6AAAVaG8MpwAWaJRmegBCKgFFSPL/T0oAAFFKAABV CAFwaAAAAAAALANq3mJsUAoIARVobwynABZolGZ6AEIqAU9KAABRSgAAVQgBVggBcGgAAAAAAB0V aG8MpwAWaJRmegBCKgFPSgAAUUoAAHBoAAAAACYDagAAAAAVaG8MpwAWaJRmegBCKgFPSgAAUUoA AFUIAXBoAAAAAAAqA2oyeAAAFWhvDKcAFmiUZnoAQioBRUj6/09KAABRSgAAVQgBcGgAAAAAF2tA AABsQAAAbUAAAG5AAAChQAAAokAAALVAAAC2QAAAt0AAALhAAAC5QAAAukAAAOBBAADhQQAA5UEA AGpCAABsQgAAOkMAADxDAAD6QwAA+0MAAJJFAADp07+wv7CZg7+wd2tda1FDUUNRa1EAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAGhZodDp/ADYIgUIqAU9KAABRSgAAcGgAAAAAABcWaHQ6fwBCKgFP SgAAUUoAAHBoAAAAABoWaOBsVgA2CIFCKgFPSgAAUUoAAHBoAAAAAAAXFmjgbFYAQioBT0oAAFFK AABwaAAAAAAXFmhvDKcAQioBT0oAAFFKAABwaAAAAAAqA2rbgwAAFWhvDKcAFmiUZnoAQioBRUjy /09KAABRSgAAVQgBcGgAAAAAACwDauxibFAKCAEVaG8MpwAWaJRmegBCKgFPSgAAUUoAAFUIAVYI AXBoAAAAAAAdFWhvDKcAFmiUZnoAQioBT0oAAFFKAABwaAAAAAAmA2oAAAAAFWhvDKcAFmiUZnoA QioBT0oAAFFKAABVCAFwaAAAAAAAKgNqjYEAABVobwynABZolGZ6AEIqAUVI+v9PSgAAUUoAAFUI AXBoAAAAAAAsA2rwYmxQCggBFWhvDKcAFmiUZnoAQioBT0oAAFFKAABVCAFWCAFwaAAAAAAVkkUA AKVFAACnRQAAykUAAMtFAADRRQAA1EUAANVFAADWRQAA6UUAAOpFAADrRQAA7EUAAPFFAADyRQAA BUYAAAZGAAAHRgAACEYAAE9GAABQRgAAckYAAPPl89TzxbmonIhyqJyonF5IqJw68wAAAAAAAAAa FmgqFUwANgiBQioBT0oAAFFKAABwaAAAAAAAKgNqTYgAABVoKhVMABZovUoaAEIqAUVI9P9PSgAA UUoAAFUIAXBoAAAAAAAmA2pga2xQCggBFmi9ShoAQioBT0oAAFFKAABVCAFWCAFwaAAAAAAAKgNq FoYAABVoKhVMABZoKhVMAEIqAUVI9P9PSgAAUUoAAFUIAXBoAAAAAAAmA2pSa2xQCggBFmgqFUwA QioBT0oAAFFKAABVCAFWCAFwaAAAAAAAFxZoKhVMAEIqAU9KAABRSgAAcGgAAAAAIANqAAAAABZo KhVMAEIqAU9KAABRSgAAVQgBcGgAAAAAABcWaHQ6fwBCKgFPSgAAUUoAAHBoAAAAAB0VaL1KGgAW aHQ6fwBCKgFPSgAAUUoAAHBoAAAAACAVaL1KGgAWaL1KGgA2CIFCKgFPSgAAUUoAAHBoAAAAAAAa Fmi9ShoANgiBQioBT0oAAFFKAABwaAAAAAAAFxZovUoaAEIqAU9KAABRSgAAcGgAAAAAABVyRgAA c0YAAIZGAACHRgAAiEYAAIlGAACORgAAj0YAAKJGAACjRgAApEYAAKVGAABoRwAAaUcAAKNHAACk RwAApUcAAKhHAACpRwAArUcAAK5HAAC5RwAAukcAAO/jz7nv4+/jpY/v44HjdeNnWUs6SzoAACAV aI4McAAWaL1KGgA2CIFCKgFPSgAAUUoAAHBoAAAAAAAbFmi9ShoAQioBT0oDAFFKAwBeSgMAcGgA AAAAGxZoVCwtAEIqAU9KAwBRSgMAXkoDAHBoAAAAABsWaI4McABCKgFPSgMAUUoDAF5KAwBwaAAA AAAXFmiODHAAQioBT0oAAFFKAABwaAAAAAAaFmi9ShoANgiBQioBT0oAAFFKAABwaAAAAAAAKgNq nIwAABVoKhVMABZovUoaAEIqAUVI9P9PSgAAUUoAAFUIAXBoAAAAAAAmA2qXbGxQCggBFmi9ShoA QioBT0oAAFFKAABVCAFWCAFwaAAAAAAAKgNqhYoAABVoKhVMABZovUoaAEIqAUVI9P9PSgAAUUoA AFUIAXBoAAAAAAAmA2qjbGxQCggBFmi9ShoAQioBT0oAAFFKAABVCAFWCAFwaAAAAAAAFxZovUoa AEIqAU9KAABRSgAAcGgAAAAAIANqAAAAABZovUoaAEIqAU9KAABRSgAAVQgBcGgAAAAAFrpHAADs RwAA7kcAAO9HAAACSAAAA0gAAARIAAAFSAAABkgAAAhIAAAJSAAADkgAAA9IAAApSAAAKkgAADdI AAA5SAAAO0gAADxIAABbSAAAXEgAAHFIAABySAAAv0gAAMBIAADTSAAA8ePQ8bqi0PGU45SDd2Z3 WHdYd1h3WHdQTAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABhZojgxwAAAPA2oAAAAAFmiODHAAVQgB GhZojgxwADYIgUIqAU9KAABRSgAAcGgAAAAAACAVaI4McAAWaI4McAA2CIFCKgFPSgAAUUoAAHBo AAAAAAAXFmiODHAAQioBT0oAAFFKAABwaAAAAAAhFWi9ShoAFmiODHAAQioBT0oDAFFKAwBeSgMA cGgAAAAAGxZojgxwAEIqAU9KAwBRSgMAXkoDAHBoAAAAAC4DarOOAAAVaL1KGgAWaI4McABCKgFF SPT/T0oDAFFKAwBVCAFeSgMAcGgAAAAAACoDaixvbFAKCAEWaI4McABCKgFPSgMAUUoDAFUIAVYI AV5KAwBwaAAAAAAAJANqAAAAABZovUoaAEIqAU9KAwBRSgMAVQgBXkoDAHBoAAAAAAAbFmhULC0A QioBT0oDAFFKAwBeSgMAcGgAAAAAGxZovUoaAEIqAU9KAwBRSgMAXkoDAHBoAAAAAAAZ7EcAAAZI AAAPSAAAv0gAANhIAADqSQAArUwAAK5MAAC5TAAAukwAAJFPAAD2AAAAAAAAAAAAAAAA8QAAAAAA AAAAAAAAAOwAAAAAAAAAAAAAAADkAAAAAAAAAAAAAAAA3wAAAAAAAAAAAAAAANgAAAAAAAAAAAAA AADYAAAAAAAAAAAAAAAA0AAAAAAAAAAAAAAAAMkAAAAAAAAAAAAAAADEAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAAAZ2Sdc1EAAAYA ABSkAABnZLRY8gAIAgAKJgALRgEAZ2S0WPIAAAYAABSkAABnZEg3dgAABAAAZ2SUZnoAAAcAAAMk AWEkAWdkjgxwAAAEAABnZG8MpwAABAAAZ2SODHAAAAgAABGE0AJghNACZ2S9ShoAAArTSAAA1EgA ANVIAADWSAAA10gAANhIAADaSAAA7EgAAO1IAAB+SQAAf0kAAOpJAAAMSgAArEwAAK1MAACuTAAA uEwAALlMAAC6TAAACk0AAMBNAAAcTgAAIk4AAFZOAACBTgAA9Off28/AtKa0prSajoJzaF1Wgko+ Sj5KFxZo9iAoAEIqAU9KAABRSgAAcGgAAAAAFxZotECHAEIqAU9KAABRSgAAcGgAAAAADBVotFjy ABZotFjyAAAVFWhyf44AFmhvDKcAQioBcGgAAAAAFRVocn+OABZotFjyAEIqAXBoAAAAAB0VaG8M pwAWaEg3dgBCKgFPSgAAUUoAAHBoAAAAABcWaLRY8gBCKgFPSgAAUUoAAHBoAAAAABcWaKljwQBC KgFPSgAAUUoAAHBoAAAAABcWaCxtaABCKgFPSgAAUUoAAHBoAAAAABoWaG8MpwA2CIFCKgFPSgAA UUoAAHBoAAAAAAAXFmhvDKcAQioBT0oAAFFKAABwaAAAAAAdFWiODHAAFmhvDKcAQioBT0oAAFFK AABwaAAAAAAXFmiODHAAQioBT0oAAFFKAABwaAAAAAAGFmhULC0AAA8DagAAAAAWaI4McABVCAEZ A2r1kQAAFWg1GQ0AFmiODHAARUj2/1UIARUDajujaFAKCAEWaI4McABVCAFWCAEAGIFOAACGTgAA kE8AAJFPAADz59gAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAB0VaL1fBgAWaABBiABCKgFPSgAAUUoAAHBoAAAAABcW aEg3dgBCKgFPSgAAUUoAAHBoAAAAABcWaPYgKABCKgFPSgAAUUoAAHBoAAAAAAADMgAxkGgBOnDX RNgAH7DQLyCw4D0hsKAFIrCgBSOQoAUkkKAFJbAAABew0AIYsNACDJDQAgAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAPDwAARABkAAAA AAAAAAoAAAAAAAAAAAAAAAAALTR5BrwCvAIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AA8ABPAwAAAAsgQK8AgAAAABBAAAAAoAACMAC/AMAAAABEEBAAAA/wEAAAgAAAAQ8AQAAAAAAACA IgAH8IsOAAACBC/hfwbrNlc8yJpdoOXZ35v/AGcOAAABAAAARAAAAAAAEgBAPRrwXw4AAC/hfwbr Nlc8yJpdoOXZ35vQNAAAAAAAAAAAAAB7AwAAbwAAAONrgQAIDRAALQ4AAAD+eAHVmw1sldUZx0/r 2/aKVbuC2CJfdQhtAVexEmbh9u0HiFgDY1CZdoh6UcyKNjpmF5vwOnUyg7N+JSCNZIoZc8YR0cUY jdUsflRncIKyuUTn0KgZ4sw2SQx2v/9577n35dLbllJge8j/Puc85+M953+e8/GevuQYY1pBISgA 150A0E7OW27M704yZuLci+cZk2N2VBtTl2OM5zIkdUDhbRhjucbMJT0q/r4Cs+7fnqECMw1MBFQ3 NcfPMWMJF4Hcou73VGxFEsq7B+wGylvl59p84XOD2gm+Z9tMshnv56XC43xjRmGLAZpiVivsm1wf XQ30HJR5Uz/IiHcfiStfFLK7fAR7Y37OCfMJzE7a89CSluW3Ntw848/1Lcu3p/RtV31TJ7vTSjc5 OTm3kP80+luEDvtgTBlhtfNKIP7V/0lJGBPMIZiUdFjtvBzMAsovLXFaVte+MCX9exZB9xwXJm/Q jL0K5FHhePQqtHxA9bQW9eS3jhYO5LQWCT35rixDHrgwWYNI2+uIJyVIhdX2cuDaoDHuBk8BjXGl H/a/hDgp9VYlw47TmJ+bX4VNz7Jj9NvKuM2S/FGb68xicyG/TYRn+rlBC3oRQHrpmuXNQzeYueRZ YswX91JHEDedO6wOnvdqFffbx1pNGO0kHS7G5PqiNissGQ0Ulm0xaOOheu4Xvb29qJR0yogUGI/W 3mCuNVfYuRhaD/d3h7dlT7F5ecaummnrdnhtQDU8h63xxlHtnXPf8tSIky7fWbMKe+e8tzzZPiC9 06Vj19jHVBCpBs98VGw++MUOr536ygl3/HJXzUjs6tsX1FGILZR7Or5ofMvbTbzO3NNRMr1xyddl J/p795W0TyTDfuxF6Mvu2lWzn3LreLbia1lv9tg6SttLi0rb3bP/+nGxmUp87en49k07vM+eervm wdKykWNv/fUJ4lOitUY8txNP/CRstvNHZWFkTQKcSaQCvQB9Plppjz22u05Q2iziIqsJRIXhsqJ6 nK/oeSyH5sRkRve8k3ES+fNW0AXkz1P9tE9UEp6EvQRk82356ipSW2yetK+GPr2GvlxrrjfB+/fG g6s/jgc9r1tt/K+t7r70tFpr3zXeahrUp9caMxi/df0SV+p/KxBXs9Hi8WK00qI8XkRcbW0CUXE8 LsNYZ1w/0jyKS/Evcc91fD6HbRsYKp8dlG0DSGrua5+bywp3FWwmzEpmnqmsh8PJ8PWg1cGa10I9 8V+h/fVvwenk2u7bK6wO/Dqrh5fjdtpVCqkL0UvQK9BRjpW2HJvan43jFtIy+pby1xhp4tqB4CF8 b8C2HmTyfSR76As3zbN76DTqnQi0bjN1rXx8zuK6J1aur59x/T/qpbVXZu6TmjcCbA+4J6reWTZv WotF+aVk3k0vhIHkr/M3lXNh8gaXENczC8FcoLGZD5TvbZPwEgUJz+VnPIa8D6o+rRsat0VAvOu8 MZtwFRBXbm6Ub/is/vYTb69f3GwaxJXi7754n9Wyw90to/ycoIhyKiPe1QehP+6mkFoC1BaJ48j1 T3at+eJC/ujsMSptJj4NrAH14BawAOj5SwsqvPxYwltasBlUHMRXX/VRJIi0t454UoJU2PG1hZQN QHxV+tnX1k0LGxpLdp9cp7W1jfwJgKTWgnwiC2n9j1kl9GvuHRM3t06Pm9/cbXUw/oUwvvMdq7sn nVCr9OD9MVbT5D5X2WLqdXyJW4Ulo4HCsjkexekS0AbGsh+OQC9An4FWmltjlVZKXG1uAlGJrrHR /uhZbn3tb95vJV8XyJz3g+V2FWVbAJLiNo9IZL3v8OOm/Pm4qdhodfDQq2F81H6ru2/Lq7XpWRgd +r41Ad7G0ZYL0Gejo5wqbRo2tbUJRCXKabofvb1u/3e8qowbS7dvbcMmHx0qn22U7ctXv88Z4Br2 LntSvGN63H+FnemJu602Wx4O46V/CvUrplbp/s9KrB5uXx0Pd+fQzgvRmvtRXpXmYxvIV6P9Ea8e cH4qLcnkdje2N8DhcDtz88sNW9rG+VoHNlB2PUBSvqo518B5YDWz8AbOBDfyb6VGYM9Dc8wfm+Lm yQVx6eDN+602e96zuvtxvFbpH460urtrqtX+l34Yv6DZ6iD3OquHcwzW0eYHIF3ngc/RJejoGChN 6436ls23V5DWV78LsYt//NlqzQ/FY0lkG5utpHeBoY7NKsq2ACQ1Nnp2ev6Znz4O82/ETeIjq7tb R8DsG/FgyplW+w9rBEgf9nXkCfgcQ1veQU9FR7lWWgU2tTUb18tIS/fj+tS5TFzK7yWZvn4/tnVg qHwuoex8gKT4ZGoyX6/VbnfN2fHujaW1wfhOq82oR0I9rNyt4HkPwc909DY0Q3MQd0qbjU3tysad +pFs8xS3TmA6hK8AG/0aMl8+ZasBkuKL5tH24LVnjwI36pc4ETfb++BGaeJGbcjGzSLSVF6S6T/b sG0BQ/WfNsomAJLiQ2v6jbCsdfIK3bBVvxt60T5TK2/yewqtDnrHhPFnZ1ht1iwIdRbvGuqZ6fdJ /vb0wZ/SxN9A+1C0PzHye0BzWXNzONe6d6ueqnPn0VXU3QKQFLd65oXs78nT6Jct4Sn07J+Hp9At T4fxUtY+nUIvYO3TaTULo0M/M2l9075xMTgDyP/cOVRppUBtzeaTy0iL9MPy6bg81nym11xz3M6g FfAxDnwPZJ5BlTYNDMRnpB+D2ju2Uefhzv2of7ZRvq+5Hz2zmeN8BtW76zlgMcg8gyrNBwPN/Wh/ 3N4ykK+q3mqQua6O8MN32BLSvvr8Ezuv9b6uvwcY7rQf3Lixvpy0QpALJiXBDJ5DMCnpsPJkvp9/ 8vlXNt9Z/Lp6XDhGRc3Yq0AAJoC/AVZp61/5sQM5+bGe/PxTBIUP5LiyBRRxYbIP+l18K5m7gLiI +SbXJyxudHeBGpa/tYhD/S1lcfOKlBanjluXnu1vLYPh+XLaOgvoHCItcVorYF5oOuTXcaZyLkze 1DisxT4efAiuA6oncwwUd2WHMg7l1ClfUBvmAY1HJ9CYVPrZ70ccf/3dPS+yZ8WrzCq696t4cO72 uCn6i9VB3t5Q/zC31tq3nx7qYd+PNtOXGaAbNIAc4PYjpTG5LK9N6KhE3+Ej/bDrp0dGIZYs4Pg/ mbEThzvAH0DmHB8sn52UvQ0gqf1d68oie3K6GlZbOUXZ++feH/AuU1gbfPao1WbpS2G8559hvJu/ SJHeff4pVgdPn2613/ltq81FNaEeVt7V9geAuHgdXAKivCttKVCfsvG+grQ++mv5z0uW1RhEofok ZwH5tBuPDYTXg8zxOJI1p7+7aa019k76oY5QH+Hd9EDrirt3pYtWXP8z1xWNwyTwPpgLvgHzgfaK /FMSXs+pw3M37daSFupdBML1ve+76ZufvLlBd9D6O604U1x3087e1910OXVqfNVu9Udg6g16H3R8 OZ5Uz0igOuWnzh6j0mbi08DXoB4UkGEBWn733KkVXmtRwnvu1M1geO+p11N/AELuTK5PuBoc7t7Y n5/q+wNxvvcVr9H6K38HOJK/oQy3n66Ea/npaQyQ89PW0QlvRtmx99PCu7xG+emmhQWWK8Xlp87e l5/KL4X+fNP52mB8sJhM8sHxaOeDk8oqvJ3wMalsMxheH9zKs7rAkfogVWT9FkZ86ny2aeGElNb5 THan+/sWZjAcX04DZoGB/JMsB4kbG5Vz4bzkmlCFbSzjoPPZWrQ7n+0s68nfOUU4kLOzTDg+5zPH X3/nM/dtwsQBv2Tx182speNxgHYSDRdjdOtnSTKsfKOTYdkch1pjl4BW8F24q0BfhT4frTR3PlOa xg3Os54TlpEW6Ydtg97FdBYYASTuue48sBWb8+upfthu5askLH8qUSTjO6LB8Jl+zz5+31qIs9m0 /gr0xehMPi/CNhCf6X4cvW8tonx20KY2gKTOu7zLmEbOuJFvLdbMCb+tuLQr/LZih75j4duLm/9j tb+/OPy2Yntl+M3F6vqj9K3Fd+B2Ie37EXoFOsqx0pZjU/ubQFTcO0ULxoy+Wb+Vz8aAtAPBQ/zX x1YNMs+zI/xD7xDO+7TcrqPH+g7hEXiYQBs9Fs/VaPlcTeWBnJrKnvyaKkHh/487BHGoPei8T/2U 1t7kuHXp/4t3CA8zDtqj8hgHt0dljoHibo3EbwMXplgQ2V/riCclSIWpPvVtpvZJvfN1gU4g/+xv TXX89bdHRd69j9sdwkt0cgb9+Tu6AR2d70oTGQOtqZF+DOoOYTd1vgEy5/hg+dxA2fUASa2p2g8X 8bdq/c3ioFuEE6vDW4N9fM/KbYFZGd4imKK9Nt59ZW54ezAnvEXwnwlvEcy5ZdbeffL0MP3Vxn5v E4p5fiEQfyXJMKrfM8I60p+GY/nVl2i9w0b5V9pSbOpbE4iKW29XYOyr3+6cECPd6wNagyVuPrjz g+M2c2xivsn1ya+1eTjf03Q21vvZHVdMdu9pR/Stm+bpLCBxWqzKhyXu/TiMpfuvco4L8qa+detg DOaStgk9H40yO6ckvI0Vx/497dHJk+17mtYWcaa43tOcva/3tHLaK79UuyPr3RyiSUnfLSjPFFAC FJY4vhw3so8EztedPQZn7j5hA5nqybMFvQAt/7uzosKrqUx4d1ZsBsP7Lree+gMgnz1afqr9UZy/ +OYy56f/U/cJuTiw/HQl2vlpTVXCq5557P30/fuWWT+9vvkyy5Xi8lNn78tPB+ObztcG44NXwoN8 UP93x/ngWTMrvHfEx8zNYPh8UGtzAmhOMA/Map6pc6GLqy1XE68wj/HrJPqGGQ279IP1mURVn9Z9 7RdjQRGQz+f6YRpRcwoYpQBSBBT+L42m8FWVJQAARABkAAAAAAAAAAoAAAAAAAAAAAAAAAAAITVa F7ACsAIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA8ABPAwAAAAsgQK8AgAAAACBAAA AAoAACMAC/AMAAAABEECAAAA/wEAAAgAAAAQ8AQAAAABAACAIgAH8BElAAACBM11uzNVhAkiMbCD 1hzZoy7/AO0kAAABAAAAUw8AAAAAEgBAPRrw5SQAAM11uzNVhAkiMbCD1hzZoy64nQAAAAAAAAAA AACLAwAAjwEAAPzGgwBn6zkAsyQAAAD+eAHdnQ98FdWd9oeQwCUGAQAAAAIAAAADAAAABAAAAAUA AAAGAAAABwAAAAgAAAAJAAAACgAAAAsAAAAMAAAADQAAAA4AAAAPAAAAEAAAABEAAAASAAAAEwAA ABQAAAAVAAAAFgAAABcAAAAYAAAAGQAAABoAAAAbAAAAHAAAAB0AAAAeAAAAHwAAACAAAAAhAAAA IgAAACMAAAAkAAAAJQAAACYAAAAnAAAAKAAAACkAAAAqAAAAKwAAACwAAAAtAAAALgAAAC8AAAAw AAAAMQAAADIAAAAzAAAANAAAADUAAAA2AAAANwAAADgAAAA5AAAAOgAAADsAAAA8AAAAPQAAAD4A AAA/AAAAQAAAAEEAAAD+////QwAAAEQAAABFAAAARgAAAEcAAABIAAAASQAAAKIBAAD9////TAAA AGoAAACGAQAAsQAAAFAAAABRAAAAUgAAAFMAAABUAAAAVQAAAFYAAABXAAAAWAAAAFkAAABaAAAA WwAAAFwAAABdAAAAXgAAAF8AAABgAAAAYQAAAGIAAABjAAAAZAAAAGUAAABmAAAAZwAAAGgAAABp AAAA/v///7AAAAD9////bQAAAG4AAABvAAAAcAAAAHEAAAByAAAAcwAAAHQAAAB1AAAAdgAAAHcA AAB4AAAAeQAAAHoAAAB7AAAAfAAAAH0AAAB+AAAAfwAAAIAAAABSAG8AbwB0ACAARQBuAHQAcgB5 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAFgAFAf////////// 0AAAAAYJAgAAAAAAwAAAAAAAAEYAAAAAAAAAAAAAAADA6sS8a3LLAU4AAACARAAAAAAAAEQAYQB0 AGEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAKAAIB////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQgAA AJmUAAAAAAAAVwBvAHIAZABEAG8AYwB1AG0AZQBuAHQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAABoAAgHPAAAA//////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAANIIAAAAAAABPAGIAagBlAGMAdABQAG8AbwBsAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAFgABAdIAAAD/////BAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAA0Ny4vGtyywHA6sS8a3LLAQAAAAAAAAAAAAAAAF8AMQAzADQAOQAyADcANgA1ADYA MQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAYAAEBGwAAAJMAAAAI AAAAFBoCAAAAAADAAAAAAAAARgAAAADQ3Li8a3LLAdBNu7xrcssBAAAAAAAAAAAAAAAAAQBPAGwA ZQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAoAAgH///////////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA FAAAAAAAAAADAEUAUABSAEkATgBUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAEAACAAUAAAAHAAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAE8AAADQNAAAAAAAAAEAQwBvAG0AcABPAGIAagAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIB////////////////AAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQAAAHMAAAAAAAAA/v///wIAAAD+/////v////7///8G AAAABwAAAAgAAAAJAAAACgAAAAsAAAAMAAAA/v////7///8PAAAA/v////7////+////EwAAABQA AAAVAAAAFgAAABcAAAAYAAAAGQAAAP7////+////HAAAAP7////+////HwAAAP7////+////IgAA AP7////+////JQAAACYAAAD+/////v///ykAAAD+/////v////7////+////LgAAAP7////+//// MQAAAP7////+////NAAAAP7////+/////v////7///85AAAA/v////7////+/////v///z4AAAD+ /////v////7////+////QwAAAP7////+////RgAAAP7////+////SQAAAP7////+////TAAAAE0A AAD+/////v///1AAAAD+/////v///1MAAAD+/////v///1YAAAD+/////v///1kAAABaAAAA/v// //7///9dAAAA/v////7///9gAAAA/v////7///9jAAAA/v////7///9mAAAAZwAAAP7////+//// agAAAP7////+////bQAAAG4AAAD+/////v///3EAAAD+/////v///3QAAAD+/////v///3cAAAD+ /////v///3oAAAB7AAAA/v////7///9+AAAA/v////7///8BAAACAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQD+/wMKAAD/////FBoC AAAAAADAAAAAAAAARhgAAABNaWNyb3NvZnQgVmlzaW8gRHJhd2luZwASAAAAVmlzaW8gMTEuMCBT aGFwZXMAEQAAAFZpc2lvLkRyYXdpbmcuMTEA9DmycQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAMADQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAA/v8AAAQAAgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAP7/AAAFAQIAAAAAAAAAAAAAAAAAAAAAAAIAAAAC1c3VnC4bEJOXCAArLPmu RAAAAAXVzdWcLhsQk5cIACss+a6AAQAAPAEAAAsAAAABAAAAYAAAAAIAAABoAAAADgAAAHQAAAAP AAAAgAAAABcAAACgAAAACwAAAKgAAAAQAAAAsAAAABMAAAC4AAAAFgAAAMAAAAANAAAAyAAAAAwA AAAEAQAAAgAAAOQEAAAeAAAABAAAAAAAAAAeAAAABAAAAAEAAABsAAAADAAAAAcAAABuAwAAbgAA AAAAAAAAAAAANV4AAK8LAAAgRU1GAAABANA0AABBAQAABAAAAAAAAAAAAAAAAAAAAIAHAACwBAAA CAIAAEUBAAAAAAAAAAAAAAAAAABA7wcAiPUEAEYAAAAsAAAAIAAAAEVNRisBQAEAHAAAABAAAAAC EMDbAQAAAGAAAABgAAAARgAAAOQAAADYAAAARU1GKzBAAgAQAAAABAAAAAAAgD8fQAQADAAAAAAA AAAeQAUADAAAAAAAAAAdQAAAFAAAAAgAAAACAAAAbQAAAAhAAAJAAAAANAAAAAIQwNsAAAAAzgAA AAAAAAAK16M+AgAAAAIAAAACAAAAAgAAAAAAAAACEMDbAAAAAAAAAP8IQAEDSAAAADwAAAACEMDb BQAAAAAAAABcXoNDezLZQlxes0N7MtlCXF6zQ4Rj/kFcXoNDhGP+QVxeg0N7MtlCAAEBAYEAAAAV QAEAEAAAAAQAAAAAAAAAIQAAAAgAAABiAAAADAAAAAEAAAAkAAAAJAAAAAAAgD0AAAAAAAAAAAAA gD0AAAAAAAAAAAIAAABfAAAAOAAAAAEAAAA4AAAAAAAAADgAAAAAAAAAAAABAAUAAAAAAAAAAAAA AAAAAAAAAAAAAAAAACUAAAAMAAAAAQAAACUAAAAMAAAABQAAgFYAAAAwAAAABQEAAB4AAABoAQAA bgAAAAUAAABsEMoGbBbKBmwW/QFsEP0BbBDKBiUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAk AAAAAACAQQAAAAAAAAAAAACAQQAAAAAAAAAAAgAAACgAAAAMAAAAAQAAAEYAAADAAAAAtAAAAEVN RisqQAAAJAAAABgAAAAAAMBCAAAAAAAAAAAAAMBCXF6DQ4Rj/kEIQAIGMAAAACQAAAACEMDbAKsq PgAAAAAAAAAAAAAAAAUAAABBAFIASQBBAEwAAAA2QAKAXAAAAFAAAAAAAAD/AQAAAAEAAAAEAAAA QwBFAEwAVAAA8JI+AACAPgCQ0D4AAIA+gL0EPwAAgD5AeBw/AACAPgAAgD8AAAAAAAAAAAAAgD8A AAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAAFIAAABwAQAAAQAAAPD/ //8AAAAAAAAAAAAAAACQAQAAAAAAAAcABABBAHIAaQBhAGwAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAJ4ZIQIZALABliXyd5Yl8ndo5BIA IJPxd/uS8Xco5BIAZAEAAAAAAAAAAAAAoHEaAAYAAAAAAAAAoRkKAdDlEgDkk/F37JPxd/D///8A AAAAAAAAAAAAAACQAQAAWAOnAwcABAB45BIAAACnAyICkXwPAAAAWAmnAwAApwMAANAEUOQSABgt RFT7IQlA7e8YeCAAAABQ5RIAEAAAAJsBkXzbAZF8POoSABAAAAB/AwAAPOUSAPcPGXgZEBl4AAAA AAAAAAA05hIAKxAZeH8XAAB4GRAvDAAAAOjnEgDqtNM7mhkhGByDpwMMAAAA6OcSAHiuHACdjfF3 IOmQfGR2AAgAAAAAJQAAAAwAAAABAAAAVAAAAGQAAAAiAQAAKQAAAEoBAAA6AAAAAQAAAKqq2EGq qthBIgEAADgAAAAEAAAATAAAAAAAAAAAAAAAAAAAAP//////////VAAAAEMARQBMAFQADAAAAAsA AAAJAAAAAAAAACUAAAAMAAAADQAAgEYAAACoAAAAnAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAA ABgAAAAAAMBCAAAAAAAAAAAAAMBCXF6DQ4Rj/kE2QAKAaAAAAFwAAAAAAAD/AQAAAAEAAAAFAAAA QQB1AGQAaQBvAIDfkj6AZuY+gMrLPoBm5j4AQPs+gGbmPsBaFT+AZuY+gNUeP4Bm5j4AAIA/AAAA AAAAAAAAAIA/AAAAAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUA AAAMAAAAAQAAAFQAAABsAAAAIgEAADwAAABKAQAATQAAAAEAAACqqthBqqrYQSIBAABLAAAABQAA AEwAAAAAAAAAAAAAAAAAAAD//////////1gAAABBAHUAZABpAG8AAP8LAAAACQAAAAkAAAAEAAAA AAAAACUAAAAMAAAADQAAgEYAAAC8AAAAsAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAAABgAAAAA AMBCAAAAAAAAAAAAAMBCXF6DQ4Rj/kE2QAKAfAAAAHAAAAAAAAD/AQAAAAEAAAAHAAAARQBuAGMA bwBkAGUAcgAAKkI+gGYmPwAAmj6AZiY/gHXJPoBmJj+AIPQ+gGYmPwDLET+AZiY/wIUpP4BmJj+A QEE/gGYmPwAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAA FgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAHgAAAAZAQAATwAAAFQBAABgAAAAAQAAAKqq2EGq qthBGQEAAF4AAAAHAAAATAAAAAAAAAAAAAAAAAAAAP//////////XAAAAEUAbgBjAG8AZABlAHIA AAALAAAACQAAAAgAAAAJAAAACQAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAAJgAAACMAAAARU1G KytAAAAMAAAAAAAAAAhAAAJAAAAANAAAAAIQwNsAAAAAzgAAAAAAAAAK16M+AgAAAAIAAAACAAAA AgAAAAAAAAACEMDbAAAAAL93Rv8IQAEDLAAAACAAAAACEMDbAgAAAAAAAADmMVJBrmWMQjJv7EKu ZYxCAAEBARVAAQAQAAAABAAAAAAAAAAkAAAAJAAAAAAAgD0AAAAAAAAAAAAAgD0AAAAAAAAAAAIA AABfAAAAOAAAAAIAAAA4AAAAAAAAADgAAAAAAAAAAAABAAUAAAAAAAAARne/AAAAAAAAAAAAAAAA ACUAAAAMAAAAAgAAACUAAAAMAAAABQAAgFcAAAAkAAAADAAAAEUAAAB4AAAASAAAAAIAAADTAGQE ZAdkBCUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAkAAAAAACAQQAAAAAAAAAAAACAQQAAAAAA AAAAAgAAACgAAAAMAAAAAgAAAEYAAABcAAAAUAAAAEVNRisIQAEDPAAAADAAAAACEMDbBAAAAAAA AAAomOpChQmFQlJWAEOuZYxCKJjqQtfBk0IomOpChQmFQgABAYEUQAGAEAAAAAQAAAC/d0b/JAAA ACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAJwAAABgAAAACAAAAAAAAAEZ3vwAAAAAA JQAAAAwAAAACAAAAEwAAAAwAAAABAAAAJQAAAAwAAAAIAACAVgAAACwAAAB1AAAAQgAAAIEAAABK AAAABAAAAFUHKQQGCGQEVQefBFUHKQQlAAAADAAAAAcAAIATAAAADAAAAAEAAAAlAAAADAAAAAAA AIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIAAABGAAAApAAAAJgAAABFTUYrKkAA ACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQptPQ0QY2A1BNkACgHAAAABkAAAAAAAA/wEAAAABAAAA BgAAAE8AdQB0AHAAdQB0AACSGj4Agy0+AKmPPgCDLT6AHr8+AIMtPgDU1j4Agy0+wCQDPwCDLT6A 3xo/AIMtPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYA AAAMAAAAGAAAACUAAAAMAAAAAQAAAFQAAABwAAAAHAMAAAoAAABKAwAAGwAAAAEAAACqqthBqqrY QRwDAAAZAAAABgAAAEwAAAAAAAAAAAAAAAAAAAD//////////1gAAABPAHUAdABwAHUAdAAMAAAA CQAAAAQAAAAJAAAACQAAAAAAAAAlAAAADAAAAA0AAIBGAAAAqAAAAJwAAABFTUYrK0AAAAwAAAAA AAAAKkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQptPQ0QY2A1BNkACgGgAAABcAAAAAAAA/wEA AAABAAAABQAAAEEAdQBkAGkAbwAAfEA+ACi9PgApmT4AKL0+gJ7IPgAovT4AFPg+ACi9PsCEBT8A KL0+AACAPwAAAAAAAAAAAACAPwAAAAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAA DAAAABgAAAAlAAAADAAAAAEAAABUAAAAbAAAAB8DAAAdAAAARwMAAC4AAAABAAAAqqrYQaqq2EEf AwAALAAAAAUAAABMAAAAAAAAAAAAAAAAAAAA//////////9YAAAAQQB1AGQAaQBvAIA/CwAAAAkA AAAJAAAABAAAAAAAAAAlAAAADAAAAA0AAIBGAAAAsAAAAKQAAABFTUYrK0AAAAwAAAAAAAAAKkAA ACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQptPQ0QY2A1BNkACgHAAAABkAAAAAAAA/wEAAAABAAAA BgAAAFMAaQBnAG4AYQBsAACHLT5AxxE/gK6PPkDHET8ApKI+QMcRP4AZ0j5AxxE/gMcAP0DHET9A ghg/QMcRPwAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYA AAAMAAAAGAAAACUAAAAMAAAAAQAAAFQAAABwAAAAHgMAADEAAABJAwAAQgAAAAEAAACqqthBqqrY QR4DAABAAAAABgAAAEwAAAAAAAAAAAAAAAAAAAD//////////1gAAABTAGkAZwBuAGEAbAALAAAA BAAAAAkAAAAJAAAACQAAAAAAAAAlAAAADAAAAA0AAIBGAAAA2AAAAMwAAABFTUYrK0AAAAwAAAAA AAAAKkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQjafxkOkcB1ANkACgJgAAACMAAAAAAAA/wEA AAABAAAACgAAAEMAbwBtAHAAcgBlAHMAcwBlAGQAAOSePQDNTD4Asko+AM1MPoDOlD4AzUw+AOTb PgDNTD7ArAU/AM1MPgDiEz8AzUw+wJwrPwDNTD5A8kA/AM1MPsBHVj8AzUw+gAJuPwDNTD4AAIA/ AAAAAAAAAAAAAIA/AAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAl AAAADAAAAAEAAABUAAAAiAAAAJUBAAAHAAAA7gEAABgAAAABAAAAqqrYQaqq2EGVAQAAFgAAAAoA AABMAAAAAAAAAAAAAAAAAAAA//////////9gAAAAQwBvAG0AcAByAGUAcwBzAGUAZAAMAAAACQAA AA0AAAAJAAAABQAAAAkAAAAIAAAACAAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAAKgAAACcAAAA RU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEI2n8ZDpHAdQDZAAoBo AAAAXAAAAAAAAP8BAAAAAQAAAAUAAABBAHUAZABpAG8AAHmsPgDNzD4AZOU+AM3MPsBsCj8Azcw+ gCciPwDNzD5Aois/AM3MPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAA AAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAGwAAACuAQAAGgAAANYBAAArAAAA AQAAAKqq2EGqqthBrgEAACkAAAAFAAAATAAAAAAAAAAAAAAAAAAAAP//////////WAAAAEEAdQBk AGkAbwAAAAsAAAAJAAAACQAAAAQAAAAAAAAAJQAAAAwAAAANAACARgAAAJQAAACIAAAARU1GKytA AAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEI2n8ZDpHAdQDZAAoBUAAAASAAA AAAAAP8BAAAAAQAAAAMAAABCAGkAdAAAZy4+wJkZP4AekD7AmRk/ABSjPsCZGT8AAIA/AAAAAAAA AAAAAIA/AAAAAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAM AAAAAQAAAFQAAABgAAAAngEAAC0AAACwAQAAPgAAAAEAAACqqthBqqrYQZ4BAAA8AAAAAwAAAEwA AAAAAAAAAAAAAAAAAAD//////////1QAAABCAGkAdAAAAAsAAAAEAAAAAAAAACUAAAAMAAAADQAA gEYAAACAAAAAdAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAAABgAAAAAAMBCAAAAAAAAAAAAAMBC Np/GQ6RwHUA2QAKAQAAAADQAAAAAAAD/AQAAAAEAAAABAAAALQCAybo+wJkZPwAAgD8AAAAAAAAA AAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwA AAABAAAAVAAAAFQAAACwAQAALQAAALMBAAA+AAAAAQAAAKqq2EGqqthBsAEAADwAAAABAAAATAAA AAAAAAAAAAAAAAAAAP//////////UAAAAC0AhUIAAAAAJQAAAAwAAAANAACARgAAALAAAACkAAAA RU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEI2n8ZDpHAdQDZAAoBw AAAAZAAAAAAAAP8BAAAAAQAAAAYAAABzAHQAcgBlAGEAbQAANNc+wJkZP4DvAD/AmRk/QMoMP8CZ GT+A/xo/wJkZP0C6Mj/AmRk/AHVKP8CZGT8AAIA/AAAAAAAAAAAAAIA/AAAAAAAAAAASAAAADAAA AAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAcAAAALYBAAAtAAAA 5AEAAD4AAAABAAAAqqrYQaqq2EG2AQAAPAAAAAYAAABMAAAAAAAAAAAAAAAAAAAA//////////9Y AAAAcwB0AHIAZQBhAG0ACAAAAAQAAAAFAAAACQAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAAKgA AACcAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwELXMLRBGNgN QTZAAoBoAAAAXAAAAAAAAP8BAAAAAQAAAAUAAABJAG4AcAB1AHQAAPJcPgCDLT6ALoY+AIMtPgCk tT4Agy0+gBnlPgCDLT6ARwo/AIMtPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAAB AAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAGwAAAArAAAACgAAAE0A AAAbAAAAAQAAAKqq2EGqqthBKwAAABkAAAAFAAAATAAAAAAAAAAAAAAAAAAAAP//////////WAAA AEkAbgBwAHUAdABpAAQAAAAJAAAACQAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAAKgAAACcAAAA RU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwELXMLRBGNgNQTZAAoBo AAAAXAAAAAAAAP8BAAAAAQAAAAUAAABBAHUAZABpAG8AAHxAPgAovT4AKZk+ACi9PoCeyD4AKL0+ ABT4PgAovT7AhAU/ACi9PgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAA AAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAGwAAAApAAAAHQAAAFEAAAAuAAAA AQAAAKqq2EGqqthBKQAAACwAAAAFAAAATAAAAAAAAAAAAAAAAAAAAP//////////WAAAAEEAdQBk AGkAbwBuAAsAAAAJAAAACQAAAAQAAAAAAAAAJQAAAAwAAAANAACARgAAALAAAACkAAAARU1GKytA AAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwELXMLRBGNgNQTZAAoBwAAAAZAAA AAAAAP8BAAAAAQAAAAYAAABTAGkAZwBuAGEAbAAAhy0+QMcRP4Cujz5AxxE/AKSiPkDHET+AGdI+ QMcRP4DHAD9AxxE/QIIYP0DHET8AAIA/AAAAAAAAAAAAAIA/AAAAAAAAAAASAAAADAAAAAEAAAAY AAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAcAAAACcAAAAxAAAAUgAAAEIA AAABAAAAqqrYQaqq2EEnAAAAQAAAAAYAAABMAAAAAAAAAAAAAAAAAAAA//////////9YAAAAUwBp AGcAbgBhAGwACwAAAAQAAAAJAAAACQAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAAEAAAAA0AAAA RU1GKytAAAAMAAAAAAAAAApAAIAkAAAAGAAAAPfu6P8BAAAAUlYAQ4Rj/kEAAMBCmpmZQigAAAAM AAAAAgAAACQAAAAkAAAAAACAPQAAAAAAAAAAAACAPQAAAAAAAAAAAgAAACcAAAAYAAAAAgAAAAAA AADo7vcAAAAAACUAAAAMAAAAAgAAACUAAAAMAAAACAAAgFYAAAAwAAAAgAAAAB8AAADhAAAAbQAA AAUAAAAGCP0BBgjKBgYOygYGDv0BBgj9ASUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAkAAAA AACAQQAAAAAAAAAAAACAQQAAAAAAAAAAAgAAAEYAAACoAAAAnAAAAEVNRisIQAACQAAAADQAAAAC EMDbAAAAAM4AAAAAAAAACtejPgIAAAACAAAAAgAAAAIAAAAAAAAAAhDA2wAAAAAAAAD/CEABA0gA AAA8AAAAAhDA2wUAAAAAAAAAUlYAQ3sy2UJSVmBDezLZQlJWYEOEY/5BUlYAQ4Rj/kFSVgBDezLZ QgABAQGBAAAAFUABABAAAAAEAAAAAAAAACQAAAAkAAAAAACAPQAAAAAAAAAAAACAPQAAAAAAAAAA AgAAAF8AAAA4AAAAAwAAADgAAAAAAAAAOAAAAAAAAAAAAAEABQAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAJQAAAAwAAAADAAAAJQAAAAwAAAAFAACAVgAAADAAAAB/AAAAHgAAAOIAAABuAAAABQAAAAYI ygYGDsoGBg79AQYI/QEGCMoGJQAAAAwAAAAHAACAJQAAAAwAAAAAAACAJAAAACQAAAAAAIBBAAAA AAAAAAAAAIBBAAAAAAAAAAACAAAAKAAAAAwAAAADAAAARgAAAJwAAACQAAAARU1GKypAAAAkAAAA GAAAAAAAwEIAAAAAAAAAAAAAwEJSVgBDhGP+QTZAAoBoAAAAXAAAAAAAAP8BAAAAAQAAAAUAAABQ AGkAdABjAGgAACWhPoAzsz4AENo+gDOzPoAF7T6AM7M+gF0CP4Azsz4Asxc/gDOzPgAAgD8AAAAA AAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAA AAwAAAABAAAAVAAAAGwAAACfAAAAMgAAAMAAAABDAAAAAQAAAKqq2EGqqthBnwAAAEEAAAAFAAAA TAAAAAAAAAAAAAAAAAAAAP//////////WAAAAFAAaQB0AGMAaADAQgsAAAAEAAAABAAAAAgAAAAA AAAAJQAAAAwAAAANAACARgAAANAAAADEAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAA wEIAAAAAAAAAAAAAwEJSVgBDhGP+QTZAAoCQAAAAhAAAAAAAAP8BAAAAAQAAAAkAAABQAHIAZQBm AGkAbAB0AGUAcgAA/1k+AM0MP4DqpT4AzQw/AFXCPgDNDD+AyvE+AM0MPwDABD8AzQw/wDoOPwDN DD+AtRc/AM0MP0CQIz8AzQw/AEs7PwDNDD8AAIA/AAAAAAAAAAAAAIA/AAAAAAAAAAAAABIAAAAM AAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAMAAAAAQAAAFQAAACEAAAAlQAAAEYA AADLAAAAVwAAAAEAAACqqthBqqrYQZUAAABVAAAACQAAAEwAAAAAAAAAAAAAAAAAAAD///////// /2AAAABQAHIAZQBmAGkAbAB0AGUAcgCAPwsAAAAFAAAACQAAAAQAAAAEAAAABAAAAAQAAAAJAAAA AAAAACUAAAAMAAAADQAAgEYAAACYAAAAjAAAAEVNRisrQAAADAAAAAAAAAAIQAACQAAAADQAAAAC EMDbAAAAAM4AAAAAAAAACtejPgIAAAACAAAAAgAAAAIAAAAAAAAAAhDA2wAAAAC/d0b/CEABAywA AAAgAAAAAhDA2wIAAAAAAAAAUlZgQ65ljEIAnnxDrmWMQgABAQEVQAEAEAAAAAQAAAAAAAAAJAAA ACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAXwAAADgAAAADAAAAOAAAAAAAAAA4AAAA AAAAAAAAAQAFAAAAAAAAAEZ3vwAAAAAAAAAAAAAAAAAlAAAADAAAAAMAAAAlAAAADAAAAAUAAIBX AAAAJAAAAN8AAABFAAAA/gAAAEgAAAACAAAABg5kBMoPZAQlAAAADAAAAAcAAIAlAAAADAAAAAAA AIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIAAAAoAAAADAAAAAMAAABGAAAAXAAA AFAAAABFTUYrCEABAzwAAAAwAAAAAhDA2wQAAAAAAAAAe7J7Q4UJhUJcXoNDrmWMQnuye0PXwZNC e7J7Q4UJhUIAAQGBFEABgBAAAAAEAAAAv3dG/ygAAAAMAAAAAgAAACQAAAAkAAAAAACAPQAAAAAA AAAAAACAPQAAAAAAAAAAAgAAACcAAAAYAAAAAgAAAAAAAABGd78AAAAAACUAAAAMAAAAAgAAABMA AAAMAAAAAQAAACUAAAAMAAAACAAAgFYAAAAsAAAA+wAAAEIAAAAHAQAASgAAAAQAAAC8DykEbBBk BLwPnwS8DykEJQAAAAwAAAAHAACAEwAAAAwAAAABAAAAJQAAAAwAAAAAAACAJAAAACQAAAAAAIBB AAAAAAAAAAAAAIBBAAAAAAAAAAACAAAARgAAAIwAAACAAAAARU1GKwhAAAJAAAAANAAAAAIQwNsA AAAAzgAAAAAAAAAK16M+AgAAAAIAAAACAAAAAgAAAAAAAAACEMDbAAAAAL93Rv8IQAEDLAAAACAA AAACEMDbAgAAAAAAAABcXrNDrmWMQu3HBESuZYxCAAEBgRVAAQAQAAAABAAAAAAAAAAkAAAAJAAA AAAAgD0AAAAAAAAAAAAAgD0AAAAAAAAAAAIAAABfAAAAOAAAAAMAAAA4AAAAAAAAADgAAAAAAAAA AAABAAUAAAAAAAAARne/AAAAAAAAAAAAAAAAACUAAAAMAAAAAwAAACUAAAAMAAAABQAAgFcAAAAk AAAAZQEAAEUAAAAVAgAASAAAAAIAAABsFmQEMiFkBCUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQA AAAkAAAAAACAQQAAAAAAAAAAAACAQQAAAAAAAAAAAgAAACgAAAAMAAAAAwAAAEYAAABcAAAAUAAA AEVNRisIQAEDPAAAADAAAAACEMDbBAAAAAAAAAAMjQREhQmFQptPB0SuZYxCDI0ERNfBk0IMjQRE hQmFQgABAYEUQAGAEAAAAAQAAAC/d0b/JAAAACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAAC AAAAJQAAAAwAAAACAAAAEwAAAAwAAAABAAAAJQAAAAwAAAAIAACAVgAAACwAAAASAgAAQgAAAB4C AABKAAAABAAAACQhKQTUIWQEJCGfBCQhKQQlAAAADAAAAAcAAIATAAAADAAAAAEAAAAlAAAADAAA AAAAAIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIAAABGAAAAqAAAAJwAAABFTUYr CEAAAkAAAAA0AAAAAhDA2wAAAADOAAAAAAAAAArXoz4CAAAAAgAAAAIAAAACAAAAAAAAAAIQwNsA AAAAAAAA/whAAQNIAAAAPAAAAAIQwNsFAAAAAAAAAJtPB0R7MtlCm08fRHsy2UKbTx9EhGP+QZtP B0SEY/5Bm08HRHsy2UIAAQEBgQAAABVAAQAQAAAABAAAAAAAAAAkAAAAJAAAAAAAgD0AAAAAAAAA AAAAgD0AAAAAAAAAAAIAAABfAAAAOAAAAAMAAAA4AAAAAAAAADgAAAAAAAAAAAABAAUAAAAAAAAA AAAAAAAAAAAAAAAAAAAAACUAAAAMAAAAAwAAACUAAAAMAAAABQAAgFYAAAAwAAAAHAIAAB4AAAB/ AgAAbgAAAAUAAADUIcoG1CfKBtQn/QHUIf0B1CHKBiUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQA AAAkAAAAAACAQQAAAAAAAAAAAACAQQAAAAAAAAAAAgAAACgAAAAMAAAAAwAAAEYAAACcAAAAkAAA AEVNRisqQAAAJAAAABgAAAAAAMBCAAAAAAAAAAAAAMBCm08HRIRj/kE2QAKAaAAAAFwAAAAAAAD/ AQAAAAEAAAAFAAAAQwBFAEwAVAAgAADwkj4AAIA+AJDQPgAAgD6AvQQ/AACAPkB4HD8AAIA+QIg2 PwAAgD4AAIA/AAAAAAAAAAAAAIA/AAAAAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYA AAAMAAAAGAAAACUAAAAMAAAAAQAAAFQAAABsAAAAOQIAACkAAABjAgAAOgAAAAEAAACqqthBqqrY QTkCAAA4AAAABQAAAEwAAAAAAAAAAAAAAAAAAAD//////////1gAAABDAEUATABUACAAAAAMAAAA CwAAAAkAAAAKAAAAAAAAACUAAAAMAAAADQAAgEYAAACoAAAAnAAAAEVNRisrQAAADAAAAAAAAAAq QAAAJAAAABgAAAAAAMBCAAAAAAAAAAAAAMBCm08HRIRj/kE2QAKAaAAAAFwAAAAAAAD/AQAAAAEA AAAFAAAAQQB1AGQAaQBvAIDfkj6AZuY+gMrLPoBm5j4AQPs+gGbmPsBaFT+AZuY+gNUeP4Bm5j4A AIA/AAAAAAAAAAAAAIA/AAAAAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAA GAAAACUAAAAMAAAAAQAAAFQAAABsAAAAOQIAADwAAABhAgAATQAAAAEAAACqqthBqqrYQTkCAABL AAAABQAAAEwAAAAAAAAAAAAAAAAAAAD//////////1gAAABBAHUAZABpAG8AAAALAAAACQAAAAkA AAAEAAAAAAAAACUAAAAMAAAADQAAgEYAAAC8AAAAsAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAA ABgAAAAAAMBCAAAAAAAAAAAAAMBCm08HRIRj/kE2QAKAfAAAAHAAAAAAAAD/AQAAAAEAAAAHAAAA RABlAGMAbwBkAGUAcgAAdT0+gGYmP4BanD6AZiY/ANDLPoBmJj8Ae/Y+gGYmP0D4Ej+AZiY/ALMq P4BmJj/AbUI/gGYmPwAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwA AAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAHgAAAAvAgAATwAAAGsCAABgAAAAAQAA AKqq2EGqqthBLwIAAF4AAAAHAAAATAAAAAAAAAAAAAAAAAAAAP//////////XAAAAEQAZQBjAG8A ZABlAHIAAAAMAAAACQAAAAgAAAAJAAAACQAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAAEAAAAA0 AAAARU1GKytAAAAMAAAAAAAAAApAAIAkAAAAGAAAAPfu6P8BAAAANekoRIRj/kEAAMBCmpmZQigA AAAMAAAAAgAAACQAAAAkAAAAAACAPQAAAAAAAAAAAACAPQAAAAAAAAAAAgAAACcAAAAYAAAAAgAA AAAAAADo7vcAAAAAACUAAAAMAAAAAgAAACUAAAAMAAAACAAAgFYAAAAwAAAAowIAAB8AAAAEAwAA bQAAAAUAAAA7Kv0BOyrKBjswygY7MP0BOyr9ASUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAk AAAAAACAQQAAAAAAAAAAAACAQQAAAAAAAAAAAgAAAEYAAACoAAAAnAAAAEVNRisIQAACQAAAADQA AAACEMDbAAAAAM4AAAAAAAAACtejPgIAAAACAAAAAgAAAAIAAAAAAAAAAhDA2wAAAAAAAAD/CEAB A0gAAAA8AAAAAhDA2wUAAAAAAAAANekoRHsy2UI16UBEezLZQjXpQESEY/5BNekoRIRj/kE16ShE ezLZQgABAQGBAAAAFUABABAAAAAEAAAAAAAAACQAAAAkAAAAAACAPQAAAAAAAAAAAACAPQAAAAAA AAAAAgAAAF8AAAA4AAAAAwAAADgAAAAAAAAAOAAAAAAAAAAAAAEABQAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAJQAAAAwAAAADAAAAJQAAAAwAAAAFAACAVgAAADAAAACiAgAAHgAAAAUDAABuAAAABQAA ADsqygY7MMoGOzD9ATsq/QE7KsoGJQAAAAwAAAAHAACAJQAAAAwAAAAAAACAJAAAACQAAAAAAIBB AAAAAAAAAAAAAIBBAAAAAAAAAAACAAAAKAAAAAwAAAADAAAARgAAAJwAAACQAAAARU1GKypAAAAk AAAAGAAAAAAAwEIAAAAAAAAAAAAAwEI16ShEhGP+QTZAAoBoAAAAXAAAAAAAAP8BAAAAAQAAAAUA AABQAGkAdABjAGgAACWhPoAzsz4AENo+gDOzPoAF7T6AM7M+gF0CP4Azsz4Asxc/gDOzPgAAgD8A AAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAA JQAAAAwAAAABAAAAVAAAAGwAAADCAgAAMgAAAOMCAABDAAAAAQAAAKqq2EGqqthBwgIAAEEAAAAF AAAATAAAAAAAAAAAAAAAAAAAAP//////////WAAAAFAAaQB0AGMAaACAPQsAAAAEAAAABAAAAAgA AAAAAAAAJQAAAAwAAAANAACARgAAANgAAADMAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAA AAAAwEIAAAAAAAAAAAAAwEI16ShEhGP+QTZAAoCYAAAAjAAAAAAAAP8BAAAAAQAAAAoAAABQAG8A cwB0AGYAaQBsAHQAZQByAAAJND4AzQw/gO+SPgDNDD8AZcI+AM0MPwAQ7T4AzQw/wGICPwDNDD+A PQ4/AM0MP0C4Fz8AzQw/ADMhPwDNDD/ADS0/AM0MP4DIRD8AzQw/AACAPwAAAAAAAAAAAACAPwAA AAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAA AIgAAAC1AgAARgAAAPICAABXAAAAAQAAAKqq2EGqqthBtQIAAFUAAAAKAAAATAAAAAAAAAAAAAAA AAAAAP//////////YAAAAFAAbwBzAHQAZgBpAGwAdABlAHIACwAAAAkAAAAIAAAABAAAAAQAAAAE AAAABAAAAAQAAAAJAAAAAAAAACUAAAAMAAAADQAAgEYAAACYAAAAjAAAAEVNRisrQAAADAAAAAAA AAAIQAACQAAAADQAAAACEMDbAAAAAM4AAAAAAAAACtejPgIAAAACAAAAAgAAAAIAAAAAAAAAAhDA 2wAAAAC/d0b/CEABAywAAAAgAAAAAhDA2wIAAAAAAAAAm08fRK5ljEKHYSZErmWMQgABAQEVQAEA EAAAAAQAAAAAAAAAJAAAACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAXwAAADgAAAAD AAAAOAAAAAAAAAA4AAAAAAAAAAAAAQAFAAAAAAAAAEZ3vwAAAAAAAAAAAAAAAAAlAAAADAAAAAMA AAAlAAAADAAAAAUAAIBXAAAAJAAAAHwCAABFAAAAmwIAAEgAAAACAAAA1CdkBJkpZAQlAAAADAAA AAcAAIAlAAAADAAAAAAAAIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIAAAAoAAAA DAAAAAMAAABGAAAAXAAAAFAAAABFTUYrCEABAzwAAAAwAAAAAhDA2wQAAAAAAAAApSYmRIUJhUI1 6ShErmWMQqUmJkTXwZNCpSYmRIUJhUIAAQGBFEABgBAAAAAEAAAAv3dG/ygAAAAMAAAAAgAAACQA AAAkAAAAAACAPQAAAAAAAAAAAACAPQAAAAAAAAAAAgAAACcAAAAYAAAAAgAAAAAAAABGd78AAAAA ACUAAAAMAAAAAgAAABMAAAAMAAAAAQAAACUAAAAMAAAACAAAgFYAAAAsAAAAmAIAAEIAAACkAgAA SgAAAAQAAACKKSkEOypkBIopnwSKKSkEJQAAAAwAAAAHAACAEwAAAAwAAAABAAAAJQAAAAwAAAAA AACAJAAAACQAAAAAAIBBAAAAAAAAAAAAAIBBAAAAAAAAAAACAAAARgAAAIwAAACAAAAARU1GKwhA AAJAAAAANAAAAAIQwNsAAAAAzgAAAAAAAAAK16M+AgAAAAIAAAACAAAAAgAAAAAAAAACEMDbAAAA AL93Rv8IQAEDLAAAACAAAAACEMDbAgAAAAAAAAA16UBErmWMQsHOWESuZYxCAAEBgRVAAQAQAAAA BAAAAAAAAAAkAAAAJAAAAAAAgD0AAAAAAAAAAAAAgD0AAAAAAAAAAAIAAABfAAAAOAAAAAMAAAA4 AAAAAAAAADgAAAAAAAAAAAABAAUAAAAAAAAARne/AAAAAAAAAAAAAAAAACUAAAAMAAAAAwAAACUA AAAMAAAABQAAgFcAAAAkAAAAAgMAAEUAAABlAwAASAAAAAIAAAA7MGQENDZkBCUAAAAMAAAABwAA gCUAAAAMAAAAAAAAgCQAAAAkAAAAAACAQQAAAAAAAAAAAACAQQAAAAAAAAAAAgAAACgAAAAMAAAA AwAAAEYAAABcAAAAUAAAAEVNRisIQAEDPAAAADAAAAACEMDbBAAAAAAAAADfk1hEhQmFQm9WW0Su ZYxC35NYRNfBk0Lfk1hEhQmFQgABAYEUQAGAEAAAAAQAAAC/d0b/JAAAACQAAAAAAIA9AAAAAAAA AAAAAIA9AAAAAAAAAAACAAAAJQAAAAwAAAACAAAAEwAAAAwAAAABAAAAJQAAAAwAAAAIAACAVgAA ACwAAABiAwAAQgAAAG4DAABKAAAABAAAACU2KQTWNmQEJTafBCU2KQQlAAAADAAAAAcAAIATAAAA DAAAAAEAAAAlAAAADAAAAAAAAIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIAAABM AAAAZAAAAAwAAAAIAAAAbQMAAG0AAAAMAAAACAAAAGIDAABmAAAAKQCqAAAAAAAAAAAAAACAPwAA AAAAAAAAAACAPwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIgAAAAwAAAD/////RgAA ABwAAAAQAAAARU1GKwJAAAAMAAAAAAAAAA4AAAAUAAAAAAAAABAAAAAUAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAMATwBiAGoASQBuAGYAbwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAASAAIBBgAAAAoAAAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAwAAAAYAAAAAAAAAVgBpAHMAaQBvAEQAbwBjAHUAbQBlAG4AdAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABwAAgH///////////////8AAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAABsAAAAu08AAAAAAABWAGkAcwBpAG8ASQBuAGYAbwByAG0A YQB0AGkAbwBuAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIgACAAkAAAALAAAA//// /wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQAAAAcAAAAAAAAAAUAUwB1AG0A bQBhAHIAeQBJAG4AZgBvAHIAbQBhAHQAaQBvAG4AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAo AAIB/////wwAAAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAlAAAACA2 AAAAAAAAgQAAAIIAAACDAAAAhAAAAIUAAACGAAAAhwAAAIgAAACJAAAAigAAAIsAAACMAAAAjQAA AI4AAACPAAAAkAAAAJEAAACSAAAAkwAAAP7///+VAAAAlgAAAJcAAACYAAAAmQAAAJoAAACbAAAA nAAAAJ0AAACeAAAAnwAAAKAAAAChAAAAogAAAKMAAACkAAAApQAAAKYAAACnAAAAqAAAAKkAAACq AAAAqwAAAKwAAACtAAAArgAAAK8AAAD+////AgEAAAMBAAD9////tAAAALUAAAC2AAAAtwAAALgA AAC5AAAAugAAALsAAAC8AAAAvQAAAL4AAAC/AAAAwAAAAMEAAADCAAAAwwAAAMQAAADFAAAAxgAA AMcAAADIAAAAyQAAAMoAAADLAAAAzAAAAM0AAADOAAAAzwAAANAAAADRAAAA0gAAANMAAADUAAAA 1QAAANYAAADXAAAA2AAAANkAAADaAAAA2wAAANwAAADdAAAA3gAAAN8AAADgAAAA4QAAAOIAAADj AAAA5AAAAOUAAADmAAAA5wAAAOgAAADpAAAA6gAAAOsAAADsAAAA7QAAAO4AAADvAAAA8AAAAPEA AADyAAAA8wAAAPQAAAD1AAAA9gAAAPcAAAD4AAAA+QAAAPoAAAD7AAAA/AAAAP0AAAD+AAAA/wAA AAABAABWaXNpbyAoVE0pIERyYXdpbmcNCgAAAAAAAAsAu08AAACEAQAUAAAAhCYyAlNOAABoAQAA UgAAAAAAAAAAAAAA0N4SAAAAAAAAAAAAAAAAAAAAAAAAAAAAyN4SAAAAAAAAAAAAfoV1MgAAAAAA SMAEAAAAANwAAAAAAAAAAAAAAO3k6vEDOOny////gwD/9vL/8/vwBwH48QCJgOrxFgSAFgAgABkA gP8AwMDAAObm5v8Azc3NALOzs+8AmpqaIQCAAGb/ZmYATU1NADO/MzMAGhoa+fJJ/3nAAEZ3vwDo v+73ALfJ4+ny8v/07ADQ3O4AsP/F4wDt8fgA3f/izQDv8eQAxf/NqQDa38gA7P/v4gDi6fMArfvB 4oEA5wB8nM3/AODo8wDb4/L/AKi/4QCtwuK7APNvAOjs3aEA2f8A6u3gAPT38PsA88MA+vr6AOl/ 7N8ASHi/ABgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABgAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAFUa6/D/8vAd6/Ac6POg+vQHD+vw9vHm9Urn9AQ66/BE6/AB AFQ1AejzgPLxSQ9bD/Lx4PsxAeb1aOrn9BDr8Fjr8AIAVIUY6/BA3/yUAZQB5vUBrpkCAAADLgQF 6/AGquvwB+vwCOvwCevwCqrr8Avr8Azr8A3r8A6q6/AP5/ST4/gvyAJV5wEBAcIB5wECAGJXAQD+ HRIBIxYCIxZfA2IAAP7q8YXABPrq8ZkQEzB6FK5Hj+F6hD++A6MGxwECDEwTlAECAFwWJhVCEDEX cn8SA6MGJhMEYgIlFFUFlRYGlRYHlRYIuREmfxIJIaoW6vGGxATq8QEDwwATEW8GTRMMJGQTICkN UOD78D8aH5QcOxezFEgkFcISfxIG6xd/EgeLLFEIiyzqGB0SCoEmCyMmSqITDNUsDTUmohMO6xQL 8D9EEofIBPAVExAjLzgvP00TODTgPyBVMPLwQ+9BCw+6Cn4U1SwB1SwFAtUsA9UncS+DJVA2lCSI WTbaFZYVCZUWyijrG6ACfQjgEBNrP8EDHRSWFZQZAJYVbCmWFcIf1Bv9P8oiHRJVC5UWDJUWDZUW DpUWVQ+VFhCVFhGVFhKVFiUTQRWk0ATq8TMGI+L5Df1vAgkEPlNAI0I64wFfAgBhAP5fUgFlVVUC ZVUDZVUEZVUFZVUFBmVVB2VQo0+1TtpP7ERO6wECEGCvEV9SEWVVVRJlVRNlVRRlVRVlVVUWZVUX ZVUYpFUZlRYvGmIJBOJTG+lSfxIlHAU7hNQE6vERNU9NFQMAo9gEJQUTED5RpGlxO1LjUgDpUuNS AchnAshnCQPpUurxp9wEcWg+UcBq0urxt+AE6vH3EBMgIBijBi9/owYICEF/Xn9GJ4EgJRaDcKIz qRejHX8SBDqWfAWpHAZiCJdG13BUpSOWfAmWfAqWfAupF9Lq8b3kBOrxsXZp8D8PiBOIEz1Pczwd F5IzVOsU4zUCeIwDiywEiyxBBYssii+qG6xS6vG+6AQK6vFREBQCbWVZjyAbuhBCcV0BglVAlJFR 6vG/7AQC6vH6dm92n0wVX19xX4NfAJVfp1+1T8dPyV/rT/1PD1kqHRIUlRYVQRXA8ARgklgsWMMB owTAP4Wt0JauWdioroyoc1XOoeXMpgB4n+6vhp+Yn6qfvJ/On+CfQPKfBKEkFRCvIqR/EhC7pkqi ExGqvBKXpqITE8q8qRSppqITFeq8Fqq8FyaqvBggzaWiExkqx0BkqpYVG5UWHJUWHZUWHjJBFcj0 BBs/0D+Nz5/GAEpbRCHDABcQlWXDAYI60bUAlzXRtac10bW3NdG1p4UMMJCyigwwj0QkFQcjEQyy CKoStQkStQoStQtRtRC6IxYRIxYSYgdDE5IC+AgeywBon7Dfwt/U3+bfaPjfCu/l9prHAAMN1wGW ywEBASzmA9cBxwEBQQHfATDi3gIw40rmA98BIETj4wFO43LmcecD5wFO44qQ5gGQ5gPrAU7jruYB oq7mA+8BTuPM5gHM5gMo8wFO4+rmAermA/cBTuOKCPYBCPYD+wFO4yb2ASIm9gOUAU7jRPYDdWM3 AZpY9gFY9gMSS+Z29gGidvYDRoFO45T2AZT2AwkUS+ay9gGy9rHyROOy8SIBIxVL5uT24/cD5PFE 46Tk8QEjFi3mFgYAFgYDEBYBROMWAQEjF0vmSAZHBwEDSAFE40gBQgOfAU7jegZJA4QBROMZS+aY BgOYAZySBwElAABpejR9MRyC1sJUhAEsRXU5iGb3ZJTytWhULFfbBxAnADBmzqJVxdyuBTW/MPSu EQnbBohmSSQEW09tT39DgCXfkUWVIyMWSN9a3xSzDCoStQ0StQ4StQ8SsKSzqEUWs7S7LDNZJjRZ JjWSgSY2PBZzEzlGYfphalCPBc0P3w+DK5W1aDgMFyw4P8UmITPYIvNEF0UWAQE9VDsX3K8VP3c/ iT+bPxCtONcmr3QULAUUJzXe4FVBCOlSXr8STFkgBdIMWTeEhrf6YWtL5nQvhi9CSogAtWh1aZjj gHBzxJ1hLjGSAQVE6fJUY0H4YwZwdk8GcCCmAYhNDUuSAUt1aPdkWBEBKAFHTvG0aVgRC0lbRZxF BZO1aC79hMbAeXcjlyOVUQJF1iyx+mGFhWgtNYP/MHoUrkfheoSIuMRwdpIBAmKIgxL6YYZCq0hH /YSxbbFjxS9QDUsUv8F7XIf2SLCHyrJfC2jOB2rgPyCNgAZw70EAHD8lPzc/ST9bO9w8lxJ0EgEF 7DVqIcgPCk8cT6QqRlEQtS/HL9kv6Eb+9S9Lf11/ABSasRJTRW4lY0+jf4dPmU8I12y6RaYBJLxn XJXZRaYBAMt5W0X9Tw9fIV8zX0VfV18AaV97X41fn18Hf8Nf1V9WQgD4NZJ/MJ82kiyVSJnPf7JN AJIBLmC+ZV7hok2uSWEVLI+gPo9Qj2KM6FFhFZzCgzI3HMdxDqCsPw2vH6YwYRMlb1ifSW8CAB+m ZmQKb6wCb6wDb6dTRVgRX08ALZ+yf8R/p0+5RjbhRRHKRECEAcppNuE24elPFBMFrp+gwJ/Sn+Sf bY3BhDLZSP+hABnFmopppIo7AXAWAgNiF6+PwY/Tj+WP94/eXwD+tQqvHKAszwSzN69Sr1uv6wIC FCwDiTwEYgIKidOkExg2D8MEuTlzYyDiFsbwhcZSYu2SAWEBIu8yv1S4rsGf0Ae4z16fIjzTAEPQ PdbtkgJD1MQPKMFv02/6DnI8RwS/sx0UAjVDwCsfPR9PH2Eb77K2FBQC5e8yyVS4PFJgvaAnuQA2 4je1O7Jd0ZfljJKk5TzSCLHld3K+5QCnMcvl/3LY5QEA/uHl5RHy8uUv8v/lTfIADPVh8hn1f/Im 9Z3yM/W78gBA9e3yTfUpAlr1UQJn9YMC2CO2mAGB9QAai/gAGwaY+AAipfSu0VZCO7G1rwBaYNCv 3K/urwC/63/9fw+OBIp1W70vbLZzsHi/irqIFKjqkP61YRVxD8gYJsEBDxlJecAkzzbPSMx6AnhV 8XgSARHwAHqEkYD++gMDYpAAYnQAo2J9CRRthQwFOH6Ee36GhBpGd78AAca/eEvzeBAXEHUAYnse GACi37Tfxt/Y3ljA7NfZRf7RJVXYEsUJ7xvu6RjFNe8AR+VQSF3Rvf/P/08mO7FdLQj+/7JNV7Es x0dI8jy4V7EAV7UcBYp1DCVbD3G/fw+RBgSuiGqVTa4PwA/SD+QK1jFJA5bipzEBV7IpGTN+hH9P G+i0gU5rRB+I0byXNuNebG3P5D8rpAKgRDGdz5MZAiZMbQXpkHUGIfBgA0XWAdyRH2VPVWQBFYKR SsJzH4Uflx+pHxQFCsMfJ26lvN8f8R8DL2qWAFLj/UaBdo2RNC9GLz+fUZ8AfC8XDykPOw9ND94v cQ8CP/SVD84TfA/DG+ju9x8AHLfJ4zw/Tj/hDcr7AgAWFOEJGWQSagBDYsEJHxsfLR+Hgx1gogim P10WAhMAbR+lT7dPyU8AtR/HH9kfEV8jXw8vISyD0wBiX3Rfhl+YX6pfvF/OX+BfwPJfBG8Wbyhv Om9MaB7yf/TsAB/Q3O5gb/hyb4Rv8wIyAGJAAHti4qJucQBi0rdvBMlvMYUtNhxfo/lvyn0mfwA4 f0p/XH9uf4B/kn+kf7Z/AIufraHxA+R/9n8IjxqPLI8APo9Qj2KPdI+Gj5iPqo+8j/5MaCCwxeMA Ie3D8fjij/SPBp+WZXIAK2K+omuRsWLkOZ9LnwBdn2+fgZ+Tn6Wft5/Jn9ufAO2f/58RryOvNa8i K3oxWa8Aa699r4+voa+zr8Wv16/prwD7rw2/H78xv0O6ODJXYlq/AGy/fr/0AQIfYxOebby/zr8A 4L/yvwTPFs8ozzrPTM9ezwBwz4LPlM+mz7jPQNPQz+LPAPTPBt8Y3yrfPN9O32Dfct/whN+W36jf Q7oi3eLNHwAj7/Hk0N/i3/Tf+hiYy6JrLgBiTAADYt0n7znvS+9d72/vge8Ak++l77fvye/b7+3v /+8R/wQj/yIrDUT/Vv9o/3r/jP8Anv+w/8L/1P/m//j/Cg8cDzzJj9uAJMXNqUcPWQ8oaw8Yn6hl MfoisJ4PsA8Awg/UD+YP+A8KHxwfLh9AH0BSH2Qfdh+IH5of0HsOux8AzR/fH/EfAy8VLycvOS9L L+BdL28vgS+TL0dtJdrfP8gAJuzv4r4v0C944i/vQg4wPwBixwA+D0UAYtsVPyc/OT9LPwBdP28/ gT+TP6U/tz/JP9s/EO0//z8RT9B7DzJPRE9WTwBoT3pPjE+eT7BPwk/UT+ZP+PhPCl+33Sfi6fMA hyitwTRfRl9YX2VzkucAYmSbDKeycwBiAbyMX55fsF/CX9Rf5l/4XwAKbxxvLm9Ab1JvZG92b4hv AkerEKlvu2/Nb99v8W8DfwAVfyd/OX9Lf11/b3+Bf5GGPXHOkynv8eeyc61/KL9/0X/ekTBxUt4D hOWXAAqPHI8uj0CPUo9kj3aPiI8Amo+sj76P0I/ij/SPBp/FxAERIJ8yn0SfVp9on3qfjJ+Anp+w n8Kf1J/mn/ift90q/3yczQAr4Ojz4COvNa9Hr0nCfaBrAGI9m+5+agBi3AOPga8Ak6+lr7evya/b r+2v/68Rv0AjvzW/R79Zv2u/NfsSjL8Anr+wv8K/1L/mv/i/Cs8cz+Auz0DPUs9kz6UtLNvjP/IA Lai/4Y/Poc88s8/idXEAYtl3W/TAD3UAYrnmz/jPCt8c3wAu30DfUt9k33bfiN+a36zfEL7f0N/i 36wbEwPvFe8n7wA570vvXe9v74Hvk++l77fv+Mnv2++lLRvo7vcABy6twqt/F/8p/8TGbvAB4Uj+ /39r/33/j/+h/7P/AMX/1//p//v/DQ8fDzEPQw8EVQ9nDxR6D4wPng+wD8IPANQP5g/4DwofHB8u H0AfUh/+ZB0v8/TsADDow+zdfR+PH6EfrjItAN9iQABi4ndbMwAfYkQAYtfUH+YfQzU9OAEsMeDo 2RQvkCH+flQ9AGI7AGLToGBIOC9KL1wvbi5tgCkYgC40lC+mL7gvyi+fRn0iYgHwYEikYVej9i8I Pxo/LD8APj9QP2I/dD+GP5g/qj+8PzzOP2kYMurt4A1T9T8IB08ZTzxhMTBCR/xnW1JPPGRPdk4z 9Pfwi0+dSB5EQEkAYuWvT8FP008A5U/3TwlfG18tXz9fUV/NbwIXlRZ+X5Bfol+0X8Zf2F8A6l/8 Xw5vIG8yb0RvVm9kHXs08xZwIt3izYFvSJNvpW/Igjk2eu10MjBCAcvYb+pv/G8KI31jGn/rfwA+ f1B/Yn90f4Z/mH+qf7x/EM5/4H/An4WzFwuPHY8vjwBBj1OPZY93j4mPm4+tj7+P+NGP449kHTX6 +voADzbp7N8OnyCfMp9VsiKBo+vRy7xgqEDYZZ93nwCJn5ufrZ9+r9Gf45/1nwevABmvK689r0+v Ya9zr4WvdMEAm6+tr7+v0a/jr/WvB78Zv0Arvz2/T79hv3O/jiNx09MPGkZ3v6GznL+uv8C/BLQU WfB78rTq1/m/C892TgAf4zTPn0Zj613Pb8+Bz5PPgHwvji/Jz9vP7c/WL3EFGQAl3zffSd9b323f f9+R36PfwLXfx9/Z3+vf/d8P7TdIIXgi7zTvRu+tHnxq73zvBI7vd00dES9Bzynx2e/r7wD97w// n8+xz0X/V/9p//nPJAvf6SQabhSj/ASz/zwkAQjN9E4pcxtxGfr/DA8eDwBhEu71PQ9PAWERVw9p D4/19AsRNQUzeERPG+i0J4FOayYfmEYQnPIYNup6sofr+JxyZBzHcYZjMKw/Yj90NjUDtgTgMz8g kTCnAO9BdRnkFqh6+DszdDb+OzIBxDwCAsQ8A8Q3Vh9oH3ofjB95JAou/yfDNbzBH9Mf5R8tRhI7 MgdhRnqyai4IDk8gT4rYSJXYSHtKR60PtgYwBTMqUvOCR4NGfACnA3pGEGlKqjxxXzryayIExE/W T4qhXIilWAdKRzj0BoEbAB4vMC8FaXUBE21iL3Qvhi8AmC+qL7wvzi/gL/IvBD8WOAB78yc/OT9L P10/L3+BP5M/AF5ftz/JP9s/7T//P5hfql8ANU9HT1lPa099T49PoU+zTwDUf+Z/6U/7TxZfH18x X0NfAFVfL595X4tfmY+rj8Ff018AXZLqX/xfzpZcYdydK289bwBPb2Fvc2+Fb5dvqW+7b81vAN9v 8W8DfxV/J3/5r0t/XX8AKJ+Bf5N/pX+3f8l/Yp90nwD/fxGPI481j0ePWY9rj32PAJ6/sL+zj8WP 4I/pj/uPDZ8AH5/5z0OfVZ9jz3XPi5+dmxUS6fIV4/gB3PAAABvEYAFVRuvw//LwA+vwROjzZvr0 AAPx8g4P///q8VAfDeD79vHm9Wjn9ATr8L0o6/ABAFQY6/AQUN/8WQFZAeb1AevwAkIItZLj+ISJ AlVB6PMIbUCrDcA/vQW/Sejzo/A/zwZ8CeP4UOD78Kk/5g/h+gH18SDp8rd0hQTq8U+lAxAQQOb1 zEgfWhYEBFofdx8Bw5Lp8siJBOrxSKUDCBTQcT+sH74WARvwPwE2FFUHIRYP6BKkQghepQM64vn+ IAIJBP0wBgcX9ogBBDGJARtgE2L/AG8AeAAsAHL/AGUAYwB0AGG/AG4AZwBsSSAsOwBiTyBzAGlL IBUgEurxZYUEiAEcXgfhDXElKuL5g+P4CqUDBaIU6POhaYkEdS+HL9/8lOP4WHwFJ3wHECcAMFUp MrXFCB4h0AS/MN3+BwB8BjcVnRJ1/enyIPPwAADv6fL/mP9Q8AAAB+DwBA8WDygPOg8AAA/6//+M iASP///we4wAVwKL/wXwXw+AcQ+DD5UPpw+5D8sP1QOMBwDi//2q6/BEAHIAYQD/ZwAgAHQAaAB1 Zfnwc/3wYQBw//JXbwBu+/Bv+fhk8/KndwBpEQD48XD18mX7AC7r8CAAYwB1VAMAFAFtKQB6//Bk +fCXYgB5HwZn9/AqBWFUEQBYAXP/8Gz/8GP78BVpDwIgBQJuIQB6AToBAwAAAAAAAG1i4/jov+jz 4D/09P4DBOC/G+i0gU5vG+i/HBUC4D8cA2voPxQD4Pv0AMA0Cw3w+/Twv5VI6/D/8vAF6/Tr8EQi 6/AEAgTw8xENGu/0IwlC5PcB6/D28T0F6vFo5/Q1BuvwMD4CVBjr8GgBUOL5XQFdATkJAuvwAwIE akUFm+P49Y0CVUHo81P4P7cPwATwvwXovwV34D9RNgnPAxaNAdsEdZAAYCDlBQX+pvwCBXWUAAYZ QpEBAkz28UEBAQT28QUBB0QCgJQBQBOQAUATQQFAE+rxB1UAKBIDLhYFOBYBQhaAgxJQEoMSWhKD EmQSTQKEskoIHbEDNQ0ADI0BAr+ATAAAQP7q8aOUjQTq8daxA/3y8PcR/vT8FDULiY0BAWA/Q/8A bwBwAHkAcu8AaQBnVAB0ACC/ACgAYwApLiAyrmAAMAAzLiBNJiBjLiQgbwBzHiBmLCIdIalyICBZ IWEsIGkeIG77AC4uICAAQQBsqnQgICQoc3giZUwgZeokIHaKIGRsIAAA/t0vjQJgElYmIHMAX18A UwBiYiAuMiD/aABtACEAIwB/NAA1ADcAOL8gBJsg6vGgkQTq8WgBwBzHGjrr8KQGBgAALd4vAyOb CQQDJ/A/OQkJAr+iQggmDT9MBjcKEOf0bAQ+BF0HVWgPeg+MD+jzFSFY5vWUAcARAQfq8YpKCMVU sQoAciA6BnkxAgAqAhJiQwD+EUIBGBIcRikDRAL28Qp4MgODEjtCQk0Ci+UYyCG0DwxOAR5PAN8S VUHVJWFP1AR6TyFJiEEAUUUENaxLBEbDTR1N40lFBaky+E87BQ+5MnJBAQrXAMD+XVIBY1MBwFLf EnG5NPbxLGIHFN/8KDdFsD8CN5nj+FatSuIGKNwQ51wxMfwEIgHUEPwCgBdTBxiMT8hTTQXWW8cG 4AU/+FW/71YBacdN/AImQwAHGMZV6BXWX9QE715EMANvABVvJl3KUdUlsGsERlpnyGcgbm8YV4pv nGkENVyxb1ZqAPZf4m8JG2l/6/AoEuf0eTH+ZQA929ZK/Bow/0iGqh3+gZoIA/tnAgiUMWUEKo8C N+rxFcnj+BG5MlRBAbwRvBHS1SK00ihdU1QHFCZAArdxAWQy1djr8Ejr8ITr8D0A9y0AAunywcuE H/UP6/D/DAAAAJ30/25an1XkowAA/QH99AAAIkqPTK9N8NAl4vkG5/RG/uvwpJAzAi4UAP8AQQEA ANYAI/7r8ESqEAJvFQDrAEo/Agrr8PRIDusCuV0AemEGHI8fvwIzFgAABFUBAH0r6/A8TjICN4EA +TM/AvLx9JEzAmoGgQAXA1ABrQGVHuvwAuvwBPP06fIfBAAE8vEAVUbr8P/y8ATr8ETo82b69AAC 8fIOD///6vFQHw3g+/bx5vVo5/QD6/C9JOvwAQBUGOvwDJDf/FkBWQHm9QEGAujzktrj+JUGAlU/ 6PMIQLanDcA/uQW/Sejz8FE/ywZFCeP4UOD78KcEdAofgQX3QwAgABEGAS0DuQYT/urxyIUE6vEl SKEK0MEFWh/QpwQDFdfwPwFDFAeBBf8PSo8SpIkIfKED4vn+IALHCQT9MAYDF4gBBE/+BgEbYCJs AGkA/24AZQAsAGMAuW/qEOsRYwB07h5vKwBy7hBy8hB1/BDtEV9iAGEAc+gQY7sREurxb4UEiAEc XgfdDTYlCuL5qOP4ZqEDxhHGETIDbQGIAAEBxhUB/FkAnR0GAmAJQwIuvBALJgYBAWC7EaEiCqcv uSJC6vFliQQ6L0wv3/yD4/glCqEDBUkU6PNpVgTyL5QEP9/8lOP4WKwXRQcQrycAMFWmMsUEHiEW zAS/MN3+B0UGRBWMAuukAerxIPPwAACW/u/yjf9egH////+/f//+A3///PoLAPgLAPAAAAf0/wN/ 9/QHf/f3tZ8gAB8gBIAALQP/fXw1DHAX9/9gSQLeFgH//2AfVgA//9//fYP/AWABAcH//wH4g3gB f4v3/wH3YQCPiv8G//M4MzP/f4r//4MAAwP/j4n//wfzD///A/9/74n/BYeOAAeHh/8Bj4P/B38P +d6QAYX/AX+gAI8An5mQD/+PqgCfAv+Z+V0AqQb/AI4Atwfw/sIHA4//8IMzAc+DgzOCuQITAYYA kwQDwwB1AYXLAZAA/4KfAAGpAL4ADBECHxQfj9//M5MzPywW8An3mQA/Hhbw/5/w/kcRh3gGcP// 8OV4kwAGXhDcAIr/An9/8IMAi/8EJhDvj4z/BAQAf4z/DYOdALT//cDr8EQAcgBhAP9nACAAbwBu AHV0+/Ag//BoAGX58F1w9fJlACwDBm758HFk8/YGBR4BZABzAwK+AgFiAGwAdQkCeKcAGSA8Afzx IDsAaFL18HAJADwBKPPwZSMAdyAAaTcCaQBj9fBVdGUEY/vybgkAY//wXWn78ikALuvwbHUCDWWf AgAArWLj+OC/9PQ//Py//xvotIFOG+C/BRwVAz8cBRQF9P8GC5VI6/D/8vAF6/Tr8ESK6/AE6/AC 6fLy8RENBwjv9CMJ5PcB6/D28T0F6vHVaOf0CevwPD4CVBiC6/Ak3/xdAV0BOQkJAQOoAgT28SMB BuvwCOf0m3rj+EZBAAIAVUDo86v4P8MN8NQGv8ME4G7LBeC/UDYKAhwGAn91CgAAYHUMEBD/A+Ri AgAGgEHvAABA/ggSAXULKhARDRUcFwYBAhMSDhJFBDoYAy8SKhJQFBqRARkC9vFBAQEE9vEUEEAA bRsGAQR1CQBgIOgFVQVoGAN0FgV+FgGIElMFdZQAkh4VBgEG+QZaHhMOBgEHYeYZCPMViurxnUoI jgYCwgXcBgA9QCkt8D8BD24SRAIkeBJLJwJuHFUtAXQWfycRA6MciSjq8ZwGBp4AvQr+7ggychzH cRzHv+E/QJ/0SeEgz7U/1yTR3yW/P/kKQFMAFgkTjxv+CjIBwxyr/ipBE0wQEIFBAS3eIBBiBQBy 6vEDBwfABYEJAV0AbBmCEqgSaoISBAoyAzwydR8QEJOALiARCjIEhTIVOBA0pBaCEgUKMgV1kxAU OQLq8aCRBOrxaAHAAEwGSAiXAS9N6fKjAgTq8daSGiP98vASQf4XRDULL/4GAmASVgBpAHP/AF8A UwBiAGHvAC4AY1QAbQAh/wAjADQANQA4ewAwV0AAAP6JJRL/YD9DAG8AcADdeUwwaQBnVAB0AFcg AChHQCl4QDJXQlUzeEBNOUBjTDBvO0AnbwBmdkJnQXJqQKNBVWF2QGloQG5FQCB4QFdBAGy+QCBu SHPCQlVlO0BlTDB21EBkRUDEXEDq8aRCCEYwvwEIAngJAQkDGkECAAkEHke78D85BggjBANCpzSh BOrxEwxHAQAwEDNBSQFaQurxhJ0E6vEr3TSg3v16J4ISZTjq8b+lCCYAZ19LBzgJkxS+QD8DlAFs AQcBAFVoAYIR4vmUAdQ14IgJ6vFfQfhZwAABAQ+K6vGKSgg0GiogEOAVAQkApjEONGJDAKUyJDRo YgLq8YuzKEJr3Ahba1QgbGoEtGHq8WkGBEEBCBFlBANFhDkJvV2U4/hVAWpU42cQNycAMtkicbwe Vub1USHfBVpB6fLpIUUJ420QJnBPMoRX6vFC0TimMWUA/zKQVRA2l6ZP/5Wj+7ZMk0C8gWcCBAUB aAFlBL9/4WvJKuP4EQYCVEEBHfR0rSLhtNE4MWHwcJMUJkD8AERRvVLV2OvwSOvwhOvwuAD3LQAC 6fJi04Qf9R/r8P8MAAAArzj/QvhRX9fqAAD3AQAd/vMAhIdyX0I2KeBx4vkG5/T9Ruvw7JMzAi8a /wAAdQEAANYA/SPr8EyrEAKkG3cAADdPAEYACuvw/wxgBALbHAAA0Yc/AmYBAAId6PNEAF0rZwFP MgKAASI/Ar7y8TyVMwKEgQCUAQNQAa0BAAAEAAAAHQAAAAAAAAAAAAQAAACldOvwNO/06vEC+/QB /22rF4lEfSJA/2gyT3Yx5/E/6OP4/vXq8Qvr8IQJK78CsiEAAArr8EE+LQK0ESsCvDcG6vEBAdz0 AAAdAAAAHwAAAAAAAAAAAAQAAACldOvwNO/06vEC+/QB/22rF4lEfSJA/2gyT3Yx5/E/6OP4/vXq 8Qvr8MQSK78CDCIAAArr8EE+LQIUEysCFjcG6vEBAdz0FAAAAEIAAAAAAAAAAAAAAAAAAAADAAAA AAAAAEAAAADEjzMCxiEAAEYAAABSAAAAAAAAAAAAAAAAAAAAAAAAAEAAAAAcqzMCICIAAEYAAABS AAAAAAAFSOvwPNz/BA8WDygOBUjr8Dzc/wQPFg8oDhQAAABCAAAAAAAAAAAAAAAAAAAAAwAAAAAA AABFAAAArKozArwiAAANAAAAUgAAAAAAAAAAAAAAAAAAAAAAAABFAAAAVKszAskiAAANAAAAUgAA AAAAlarr8Cjr8Hbc/+H6Aurr8AXn9B7r8DSQM78CgRkAAJvr8FP7AMnr8GS2KgIcrxoAABPr8EIh Anz/kzMCGCEAAJr6LwI/6/CMjzMCZu8iAABS6/BQAEQK6/D8XQDWYQYSBYMBFAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAYAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAF NOvwKNz/BA8WDFUV6/D/8vAc6/Bg6PP8+vTx8gAAhxkCMAbg+/C/3P/g+/LxCQNMD1MAAPbx5vVG 5/QE6/CdROvwAQBUeQHq8QOAQgaQDQUHpAfg+3UB5vVo7oEGAAAk6/ACAFSFGOvwDN/82AHYAeb1 AdTdAujzkuP4hIUCVUH/basXiUR9IkD/QWgyT3Yx5/GrP0Ho88A3Fr84FPBmNxXwPwcRXx8AUBMM +T9hH+H6Af8DAASlPOnymwQU6vGRIBNAjOb1wR3wP9MdyBhzGQHHzwMQhQELEgcRAQRtFt0BBHUL EGAg5vWjBf4GIvsCECIFFiIFIXXYACApOxIIGJAgEwUSpAcR5RTA2xV5JtCDLtgslS55KHNVuyHl uSaRHxjbLwMTCxUAbw0lXgTgBKjECQwm3/yo4/hmIBP9KPLwSzGxAwFrIQFLMQMS3/wLAAAd3QJg CfVD/CBudzBlAGMA1XT8IHLr8P5pMAACFQH68f5sMgpyP4Qy6vEJZQgU5wFI3QcSDOYC5wmoBxXY AXUBBevwBuvwByrr8Ajr8Anr8Arr8IkxAurxg+P4EEEjEAcRwTEWRwBcFShECxEzSwsVKER1ATNL APsFKET8MTNLxAUoRARBM0tA/DEkSIkxM0sAQSRIDzBOQARBJEgGITNLCEEkSBEwTkAMQSRIFiEz SxBBJEgaME5UFEUoRBswSsn4CIbdAmFQDEF1AXUBBxERH+vwxMxRCxER3AHcAYkxETcY6/DmUfVB ESzr8PNRBiEjER7r8ABhN1EREgESAcYWIREz6/AaYXlRETQA6/AnYZpRN1Hk98ExBxGzRQB9BHUB B1NcFfLx/DGQD0UNIEVl6vHmRdAFdQEo3QcGIaC4DcQJ7DmqFeb1RbsUVP8QLVQPxQlAQP/QaQOd NtDZP1tADtUVmJkfcekTdR3gHHXZP1HUafVBHRMyXBVTYyXVawBCLSL1QWLDUgJ1dnRyN1EBDAYh RMNTk3YB+1J6ARRCpAgYaS1YeVwT/goCCQTUc0Z+GwEAISS0UfsFPUpE/rEBHLfJ4wAAGQdJecCi ELACySbJJnMekQHFQxVxXBUu3QPm9UP6+mBMXoAKAEEAda8AZABp/CAK+mBuKn0wb3WAZYMwCneA CRMoBT8XP9/8lOP4WFh3xAdvECcAMrsiVcXJLqkhGwUSBumxBQnEBgGFgOP4cdUE/DHzUeAEFCDf /PwxxUXoP+jzmeP4EgGAvhEndsIffBfcd3+VXBVkgJCaG3YScBsFn5a6mXlRAwCcd5x4DaKmcpd9 f5ULFY+eAB52yx0iALyf+wXWlaCW552YnpcBr3V/AQMno9WvAYD7AoORxAVRr+ibr5/l9mcAwFjj AeAEZL+3i6hR5vU3UQcDAFQHEWwxbDH7ArRhoMVFqBHgAFwf5vX4ySTwLT94b///BxFs/TQAQQ4S AQIAVeMBYb38tWefJPY36vGJ7A5ddYqvGE+RBL4f5faLCBhtxRV2fs/7BQCYzh51hsfEBW3O1M/F RW3PADRuCxGSRVFo+wMLEQsRam8Ab9u/ZwcRxUULFdAFiTEnYQDgBGEJ6vEEQRRFkGkAQQhJ/Ott OJNVQDtBtFCvPRTnPwt3Mxji81c/QJofcskW5eMf5bu5P/caAAAX3QEC7XXnAGB1EEBgBIA2OBBA /u7xAgMg5lvj6urxna8YmBrUQHAI7+xQCL8N6De68+9BuCP1DecBAynCS+JwiTHYAQRAuecBAG2A aTBAehpgQP6z4mEDue/L7gYjkoUBAgYiqpGGAQsSp+NIE0U3AQsADW0zARIBUSIfEQdVETVYf+ny 3AEDCiJ5c332KAoiq3soVVcgEwgK8IXz4tJzAtl/XWdh4gt0Tiz/8AsRenNd5gzl8okxJO7zv8Gc Rlj7ZTQr5R5z/7k/MnIcx3Ecf8fhP0Cf9Ek8AJXPMQXROgW/Ouv7A4X6Z1gjRTMwTxvotP+BTms/ GkZ3vwsAASDWAkHwtPAg1bRhYLy1tbT0Jmi+ghXwv2/fAMDx9rPwNdlRCMTgOU/HJc8gQ88qQ0Rz Yc+errmw4IjPABsJHuBsG06/YL9yv4S/lr8AwDK/VUTfZt/vtjEpeVEfGwDTZby1qGUtEfJxnoQo VYwH6CsXXmnvaZH8ZFpnDO+0AAwhjeCYutz7/iAM4Fo0EvDNNaskcB81AXHkyiUScHLVI9Q65iyR Ac8DWFHXAgR1LRBg8nTgPxsF/v4iBXXz4Ag5BfMWw1UBBBXyAzA2xUJLeXigs11/vTECAoBMcuUA +LEP8QDPEs8eHzbPNh9DdCBPGSDRHd/tcv4iAAQ5PtACEzMBGjkkQXYd9zXMJghPAgY38BE1LU+0 zE5O1SVnTwB5T8xH3s33PrZPGk/aTwndgPc/YE8gX4RPSoRJqV2IT05zgHQAcLpScItTd4AFZ3+A YfewjIPz4ZaPqI8Auo/Mj96P8I8CnxSfJp840QBl0QgvGi8sLz4vUC9iL3Qv+IYvmC+qLaHeSrx3 /7ASQEBQeUw0v0RJ6D9AlWOi8S8/QGhmYXHmVXU5AAJgc9buLwA/Ej8kOWXRNDkAu3ZJP1s/bT9/ P5E/oz+1PwDHP9k/6z9ZX2tffV+KT0VOAFd2k4/IT7ePlU/bjGB18o8ABJ85X/RPjIRBn6WPZZ9D X4CEj+2PrZ8Rn51a5ZGuWEP6zlBtvlByAGUAc9JPoGWXoMVbQtRQdAAVLU+gdEuiYUeg31/xXwAD bxVvJ285b0tvXW9vby7bANrRk2+lb7dvyW/bb+1v/2/4EX8jf6otqMBE8u8B1zDgwy/VL+cvhH+W f6h/AQLa0TQ5VsbUf+Z/+H8KjwAcjy6PQI9Sj2SPdo/kn/afAAivFZ9FT1dPON9dnxifbt8AsU+S 36Tfcp8a3wtfPt/+30DMnx3fht9G76rfm1wuq1sBSdhQv1/RX36vkK+ir7SvAMav2K/qr/yvDr+F Z//xKrEAMrwqsbHzQ7WPYUm/5wPVywDK89HHjbX6B5+yCR8bHy0V+sy5Rdm0PrrzQbj/I/0/QNBp A50DNtAKwJ5R9tFrFILQfwMd4JUV2T9RQgkM0nPKiVOBzxXSQjsyLRFAMgIY3wbeAugRAQwM0UAz /QYoC9L3AtDypGnYLcIJFdMN/vXyCQQ+E/TTJPef5FSeURXVKKtbUGeiY4ywqQp9EEyhZvjgbLxQ ZQBLoAPz0YEN/x//Mf9D/1X/kGf/ef+L/53/ceTEj2EsgqsXFMDMj2HvxdvP9BOZAMTItMHcsZEG ge6vJFARRhcEjyUV1WSgKoUGCsAXJa8mBMopa7EDBhcGGB0yC9IBHSCPJaJZoyqIBrktv8ovPwfA 5iWPlvctg9cRP98PAQMENzPlPwFBspMhtRVhP/grBL8v8RZnPja9w64UdE/HGznJxMj3AQMAVMUR gMGIgMNsFO/FPKsT0PXTT/g2BSTwP+b///+z8ajBAM/FtM/Gz3gv6s/8zPQTEdkAXyF97/EWaNl9 VX8Gjl9BBQCoXogFlle1FX1e5F9PVX1fACJaMrEM0Y9hMr9JZNGB2P+Af2vFcb8XT1XRhY21s/E0 AKsXqhEjWSqxzEkCBS9RpLHlCfQAClIPRyRVQLl/dpO6dSsEQHUHFXD4JZpqcskmdTdwMXOpuRTG wRQXmcJ1mMBgvXX6YGAEgEGgwRWiqxEDMHZrc/YRnRLY6v7ZtCLd+SDckQL+HXhQEC1UD8UFLh13 AQApcnJwxRFHUdwqsMl3AQCAs/BAehkzoMHDcgBws/ECBPJ3At9/AfJ/BI+Zwsl/LY5Qw4QM0cUS AlDCBhJxg/YRp6owWBPZs/1gsg2rEmABARRRaOJ0wQYFBSHFDMvmAwMDPT8PE+MD34ZBsh8RKqS1 V9mzCOgAAugCPBNxAkMfv7dxcgt0Tu7QFrPxenNtdgxRksURWpOS9hGc82RhCWAydYgDufc/MnL+ sBzH4T+vQJ/0SaiQz52V0VKmlb9Ke0GzhfdoI9mz/zBPG+i0gU5rfz8aRne/AAEwZgUCo4AC9Zdv ERXVxUSvA2wnr04X8L9/b///DNEAqMFDVSIDGlItDz8PbNVXXzFVDxNwXwNvuT8RQ5xfAIuVCsDY q15PcE+CT5RPpk8AU2JQwVBvYmRQwW1vor+RbwSjb7HzOKsXn+bzFPJh2m/47G9RDwx3zRmXGeb7 rxIdeAqNBHzzE2r+L3hQw+5Bf1N/ZX+od3+JfyfgV7FyDR14bj+rF4lEfRa3fk+DCCCF8sYxjwHx zwPfP4NUswhLg0TWMN8DQ99Wj2iCd4EgcYmV1/YRizBYNibCj8cfAJHgtbaLj52Pr4/Bj9OJd4EE 44NE5gFE5naCAZ8TnyWfgDefSZprl2CY9wFvn2EJDgBjxZafqJ+6n8yf3p/wnwKvAL5Ltbcsrz6v q7ter3Cvgq8AlK+mr7ivyq/cr89aUcUDvwAVv4sPOb9Lv12/9gIe4Xi8AB7jFRH2EZW/CBsAzytV z7kAS1FvFLERh/1LUbUZP1kmzYrq5XDZwhgdeNteKk9RkGQZIqMa4ROlUyfvUDJCMD/rYeh3gfIB AQwdMeODQCMmNjIdIv6ycOFZNS1kWaATo5fjle+n4+q0huv4PfrzowJvIRy3yeMAfwAZSXnAAAFt IxCi5qLmJ/lJQgFPU9tRE6Wo2YGGAPK1Q+ngTO4gIO8AQQB1pzBpAG++HcBEAGUAYwkwZAoPMHId wGlWNFAxxmGGBChjGafl1N2UJFhYZFdjF18QJwAyVX4yxaLmovK1IaP1FFGIsulTRQkmYxYBgCvO TVMsgwcpYSAjWU9RT1U7X9WzmSRYFFEA11e9EDFuL0Sh8WwnD0UTpYFk1F8qdKP1uFhsJyMhAwQs JywsAzYip00PRUklH04ArhY5TbPQTE/u9WZFBmx9Rxi3WZNNBS8BA7dDZV9FAwATQWMV4U94Sz9P 8bZ2D4gPAJoPrA++D3Pyw7UbpPIFU281+KL1Pwgf///yAcPxADARMBGfMdP08V2MZfc/XRMAw7UG D+0VIA8uWOE7Sw9sSAAOf+71KH6uFRZ3YxUsD2V+EMO1/W/iOnixFtYP6AfMcQD2D/5/Gh8sH8n5 8FLa/+z//CLP6OcKRbEi3hYBHpofrB++H9Af4h/0Hr1GkQK9RnhSHSMb31YWlgGAFpE7L00vXy9x L9Wz3SkqquorUAcwdBEwaB3AUCoJMHOakGYHMGyakBg/ACo/PD9OP2A/cj+EP5Y/qD8Auj/MP94/ 8D8CTxRPJk+8XwAMX1xPbk+AT5JOFp+1RT6/AM9Po1/zTwVfF19/fztfJr8AnUwBnLtIAZiTX36/ t1/grwDbX+1f/18RbyNvNW+Hwul1AO7PZm8N2cxxiG+ab6xvvm8A0G/ib/Rvsn8Yfyp/PH++3wBg f3J/hH+Wfxjvun+9QY6hANR8NPO9QTyR8X+x/0uXjqUIPJU4FTORNIbHxZF+qR+BCDOSFIwwEQhJ oTv4coljo/9VQD145FURSlUbjYhgE7WaE7LJQfVMUvBM87k/xSnAABeGwdsCdUaAYHUV8GAEW4BB J+D+FYbBA0v2VIbz2SGd3ijqWARylBLVGo2ICIeCHI2HAQCFKY3ycMxxoqEfgOT3ATsAgDOQQHoz iPHe8gMBcBCS6vQOBvr/GAgXDyIACQLk/0gOTxEDAZLgklBuUqFDjAPZIackiBPv0939qeABAA2G wmABADqhoPLOgR+FAtHejOQm9oigLCwezzaTDfHZIVfv0wgjAgJ48DyTX5ECXJ9Ml86M8gt0Tn3w TGF6c0KI9gxsEsxxdRPZIZwO9PJ6iVBN9bODuT8ycv8cx3Ecx+E/QFef9EnDEM+4FdHBFam/Zft+ 44US+CPv0zD/TxvotIFOaz+/GkZ3vwABS+YCAr4AAhAn1MHhJUVk+gFALzZlmfC/mu////oBI9GA XtU7gzXSRo9YjzyVct9VMQHHg4zfyb65PyATt98HVUBAE7XzK3nPi8+dzwCvz8HP03JYsWvvurRY sYjvAL0/rO++79Dv4u/07wb/GP/+79SKHkx8USUh90BAyJCFOMA3rwcmnvFK+GdDwBBe/3D/6IL/ lP99mJjv1IaiWO8RbwsgjYiNmj9/5zM/IkBAwNX1AQwHBMxxOwUJVkwPE1gSXwB7ClixjAlaV6IP tA/GD3iRMTXdD3eSO4EDA1i/KJgAkCImHzgTclNEH1YfaBaGFxB7GMxRih96iUt6RbEfwx8A1R/n H/kfCy8dLy8rzFEIfwBTLyt9zFF5L4svnS+vL8EvANMv5S/3Lwk6aEUePzA/CY8AVD9mP8XMAQAA AAIAAAACAAAAAAAAABFVLunyAuvwBOvwHevwVR7r8B/r8CTr8Czr8FUw6/Az6/A06/A33PAAAAAA AAAEAAAApXTr8DTv9OrxC+vwAv7r8AFtqxeJRH3/IkBoMk92MefD8T/j+AIBAgH69UwmfzICXTYA ACfr8H1D/fJUCysChDcAFQrr8EHp8gHc9BQAAAAWAAAAAAAAAAAAAAAAAAAAAQAAAAAAAABAAAAA ZC4yAo42AABMAAAAUgAAAAAABUjr8Dzc/wQPFg8oDhQAAAAWAAAAAAAAAAAAAAAAAAAAAQAAAAAA AABFAAAA1C4yAgw3AAANAAAAUgAAAAAAFAAAAFwAAAAAAAAAAAAAAAEAAAAEAAAAAAAAABUAAABM LTIC3iMAAG4SAADTAMkAAAAc/CMCTDYAABEAAABAAD8AAAAsLjIC2jYAAC4AAABQAEQAAACcLjIC GTcAAC4AAABQAAAAAAAAAAAAVTHr8P/y8APr8Cjo8xUc6/AM3P8A9vH28eb1VQHr8ALn9E8iCEQm AmFVQQHo8/LxVQ///04FkGoJ4PspAeb1aCIIAQECFwBUGOvwBH0Piwrq8VW84/grFgJV6fIJrw2u fA4fBgjp8smJCB5CJgJRLQG/ChkFLQJIKgQAPQ8oAFgPaRP28XcdsQ+QD6iiD7QP4fqb4/hFFgJR EPoPCC8aLywvAP1m6/BlAQAAQQD/cgBpAGEAbADfIABVAG758GMA328AZABl//BNAGlT3P/f/P82 A+k/LwX3AT9gNAECCwYEBQJSAQQkAf1m6/AFAQAAUwD/eQBtAGIAbwChbNz/Eg8kDzYPgOrxBf8F AQIBBwYCBQEHNwH9ZuvwBQEAAFcAf2kAbgBnAGT39KFz3P8YDyoP4vmA6vEFADcK/Wbr8EUBAABB AH9yAGkAYQBs3P/cEA8iDwCHeunygAje5/T/AQBARAD/Ai8LBgQCUgEEJAH9ZuvwpQEAAFMAd2kA bfXwdQBu3P+0Eg/d/gPp8g4I5vUB+wAE6fICAQYAAwEBUwEkAf1m6/ClAQAAUAD/TQBpAG4AZwBF TPnwVdz/Fg/h+gPp8ncuCBbn9AEAEOnyAwICNgNVA/1m6/ClAQAATQD/UwAgAFAARwD/bwB0AGgA aQDxY9z/Gg/l9r8CAKDf+/zHaBDn9J8A/wJAAADX3wIL/wYABwIFCAIEACQB/Wbr8KUBAABEAH9v AHQAdQBt3P/8EA8iDwCvAgCw++9812kw5/SfAAj/QAAA198CCwYHAAABUwEkAf1m6/BFAQAAUwD3 eQBs7fBhAGUA8W7c/xQP3/yHBgAE3uL5nwAAIOrxAQr/BQIFAwYDAwMAJAHtZuvwRQHw8ABzAP90 AHIAYQBuAP9nAGUAbABvAFUg9fBkAwBz9/Bh3P+24fqHehUCgAjn9P//AQEgAAAoIAL/CwYEAwUE BAIBBCQB/Wbr8EUBAABWAP9yAGkAbgBkADFh3P8SD93+AwDz8OP4DjgFAQEGTQBSASQB/Wbr8EUB AABTAP9oAHIAdQB0ADFp3P8SDyQPAAQmD+vwBwIABTwI/Wbr8EUBAABNAF9hAG4AZ/fwbNz/JBIP JA+AJQ/o8wQ+CP1m6/BFAQAAVAB/dQBuAGcAYdz/SBAPIg/r8EAmD+nyBD4I/Wbr8EcBAABTAN9l AG4AZPnweQCxYdz/FA/f/N97AQKA/Qjn9P8BASAAAH8oIAILBgQCUgEBBCQB/Wbr8EUBAABSAB1h 9/B2AGnc/xAPIg8S6/ACJg81AgU8CP1m6/BHAQAARAB/aABlAG4Addz/fBAPIg8A33sAYevw+4AI 5/T/AQEgAP8AKCACCwYEAgJSAQQkAf1m6/BFAQAATAAfYQB0AGj38P//EQ/OIw8AABAmD+vwAgAB BDwI/Wbr8EUBAABHAN9hAHUAdPfwbQCRadz/FA/d/iAmD+vwAgMABTwI/Wbr8EUBAABDAP9vAHIA ZABpAP9hACAATgBlAHF33P8aD+X2AwAA8/D65PcBOAQCCwMEAgJSAQQkAf1m6/BHAQAATQD/UwAg AEYAYQAfcgBzAGnc/xYP4fp733v98AAAgAjn9P//AQEgAAAoIF8CCwYEAlIBBCQB/Wbr8KUBAABH AH91AGwAaQBt3P/8EA8iDwCvAgCw++9812kw5/SfAAj/QAAA198CCwYHAAABUwEkAf1m6/BFAQAA VAD/aQBtAGUAcwDXIABO+/B3//BSAJ1v+fBhAG7c/93+h+166fKACOf0/wEA/UBEAP8CAgYDBR8E BQIDBCQBGAAAABACAAAAAAAAAAAAAAAAAAAAAAAAGAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADX AAAATD8VAlo4AABFAAAAQgDXAAAA1D4VAp84AAAuAAAAQgDXAAAAXD4VAs04AAAlAAAAQgDXAAAA 5D0VAvI4AAA1AAAAQgDXAAAAbD0VAic5AAAyAAAAQgDXAAAA9DwVAlk5AAAzAAAAQgDXAAAAfDwV Aow5AABHAAAAQgDXAAAABDwVAtM5AAA6AAAAQgDXAAAAjDsVAg06AAA3AAAAQgDXAAAAFDsVAkQ6 AABMAAAAQgDXAAAAnDoVApA6AAAtAAAAQgDXAAAAJDoVAr06AAAnAAAAQgDXAAAArDkVAuQ6AAAi AAAAQgDXAAAANDkVAgY7AAAjAAAAQgDXAAAAvDgVAik7AAA5AAAAQgDXAAAARDgVAmI7AAAiAAAA QgDXAAAAzDcVAoQ7AAA5AAAAQgDXAAAAVDcVAr07AAAmAAAAQgDXAAAA3DYVAuM7AAAnAAAAQgDX AAAAZDYVAgo8AAA6AAAAQgDXAAAA7DUVAkQ8AAA+AAAAQgDXAAAAdDUVAoI8AAA6AAAAQgDXAAAA BAAzArw8AABHAAAAQgAAAAAAAQAAAEcAdQBpAGQAZQAAAPUS6/AB6/BUAGgAf2UARABvAGPc8PUS 6/AB6/BQAGEAf2cAZQAtADHc8PUi6/AB6/BHAGUAf3MAdAB1AHL38N8gAEYAb//wbQANYfvwAAD1 GOvwAuvwUgBlAP9jAHQAYQBuADdnAGz38AAAAAAAAPUW6/AB6/BWAGkA/XP38G8AIAA5AAEw3PD1 GOvwAevwQwBvAH1u+fBlAGMAdPfwAXLc8PUW6/AB6/BWAGkAfXP38G8AIAAwAQADAAD1FuvwAevw VgBpAP1z9/BvACAAMAABMdzw9Rbr8AHr8FYAaQD9c/fwbwAgADAAATLc8PUW6/AB6/BWAGkA/XP3 8G8AIAAwAAEz3PD1FuvwAevwVgBpAP1z9/BvACAAMQABMNzw9Rbr8AHr8FYAaQB9c/fwbwAgADEB AAMAAPUW6/AB6/BWAGkA/XP38G8AIAAxAAEy3PD1FuvwAevwVgBpAP1z9/BvACAAMQABM9zw9Rbr 8AHr8FYAaQD9c/fwbwAgADIAATDc8PUW6/AB6/BWAGkA/XP38G8AIAAyAAEx3PD1FuvwAevwVgBp AH1z9/BvACAAMgEAAwAA9Rbr8AHr8FYAaQD9c/fwbwAgADIAATPc8PUW6/AB6/BWAGkA/XP38G8A IAA1AAEw3PD1FuvwAevwVgBpAP1z9/BvACAANQABMdzw9Rbr8AHr8FYAaQD9c/fwbwAgADUAATLc 8PUW6/AB6/BWAGkA/XP38G8AIAA1AAEz3PD1FuvwAevwVgBpAP1z9/BvACAANwABMNzw9Rbr8AHr 8FYAaQD9c/fwbwAgADgAATDc8AEAAABCAGEAcwBpAGMAAAD1HuvwAevwQgBhAP9zAGkAYwAgAPdT AGj38GQAbwABd9zwAAAAAPUa6/AD6/B2AGkAf3MAVgBlAHL58B9pAG8Abtzw9Rzr8AHr8FIAZQD/ YwB0AGEAbgB3ZwBs9/AuADLc8PUo6/AC6/BEAHkA/24AYQBtAGkA12MAIAEAb/nwbgAVZQEAdAcA ctzwAAAAAAAAAAD1HuvwAevwQgBsAP9vAGMAawAgAH1O+fByAG0AYffwAwAAAAAAAPUc6/AB6/BS AGUA/2MAdABhAG4Ad2cAbPfwLgA53PAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA9S7r8AHr 8EQAeQD/bgBhAG0AaQDXYwAgAQBv+fBuANVlAQB0BwBy7fAzAAE53PAAAAAAAAAAAAAAAAD1Huvw AevwUgBlAP9jAHQAYQBuAPdnAGz38C4AMQABONzwAAAAAAAAAAD1LuvwAevwRAB5AP9uAGEAbQBp ANdjACABAG/58G4A1WUBAHQHAHLt8DUAATTc8PUu6/AB6/BEAHkA/24AYQBtAGkA12MAIAEAb/nw bgBVZQEAdAcAcu3wNRkAAwAAAAAAAAAAAAD1LuvwAevwRAB5AP9uAGEAbQBpANdjACABAG/58G4A 1WUBAHQHAHLt8DEAATXc8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABgA AAAgCAAABQAAAAAAAAAAAAAAUAAAAFAAAAAAAAAAMwAAAHz8IwLXPgAAEAAAAEUAMwAAAGT8IwLn PgAAFQAAAEcAMwAAAEz8IwL8PgAAFQAAAEcAMwAAAASyKgIRPwAAJQAAAEcAMwAAAGR+LAI2PwAA HAAAAEcAMwAAAMSOHwJSPwAABAAAAEUAMwAAAKR+LAJWPwAAGQAAAEcAMwAAAMR+LAJvPwAAGgAA AEcAMwAAAOR+LAKJPwAAGQAAAEcAMwAAAAR/LAKiPwAAGQAAAEcAMwAAACR/LAK7PwAAGQAAAEcA MwAAAER/LALUPwAAGQAAAEcAMwAAAGR/LALtPwAAGQAAAEcAMwAAAIR/LAIGQAAAGQAAAEcAMwAA AKR/LAIfQAAAGQAAAEcAMwAAAMR/LAI4QAAAGQAAAEcAMwAAAOR/LAJRQAAAGQAAAEcAMwAAAARA MwJqQAAAGQAAAEcAMwAAACRAMwKDQAAAGQAAAEcAMwAAAERAMwKcQAAAGQAAAEcAMwAAAGRAMwK1 QAAAGQAAAEcAMwAAAIRAMwLOQAAAGQAAAEcAMwAAAKRAMwLnQAAAGQAAAEcAMwAAAMRAMwIAQQAA GQAAAEcAMwAAAORAMwIZQQAAGQAAAEcAMwAAAARBMwIyQQAAGQAAAEcAMwAAAMT8IwJLQQAAEAAA AEUAMwAAAKSyKgJbQQAAIgAAAEcAMwAAAEyUHwJ9QQAABAAAAEUAMwAAAERBMwKBQQAAHQAAAEcA MwAAAGRBMwKeQQAAHwAAAEcAMwAAAKQvKgK9QQAAKQAAAEcAMwAAAFSUHwLmQQAABAAAAEUAMwAA AFyUHwLqQQAABAAAAEUAMwAAAMyyKgLuQQAAIgAAAEcAMwAAADSUHwIQQgAABAAAAEUAMwAAAIRB MwIUQgAAHwAAAEcAMwAAAOyTHwIzQgAABAAAAEUAMwAAAPSTHwI3QgAABAAAAEUAMwAAAByUHwI7 QgAABAAAAEUAMwAAAAyUHwI/QgAABAAAAEUAMwAAAIyUHwJDQgAABAAAAEUAMwAAAJSUHwJHQgAA BAAAAEUAMwAAAJyUHwJLQgAABAAAAEUAMwAAAEw0MgJPQgAALwAAAEcAMwAAAKSUHwJ+QgAABAAA AEUAMwAAAKyUHwKCQgAABAAAAEUAMwAAALSUHwKGQgAABAAAAEUAMwAAAJSzKgKKQgAAIgAAAEcA MwAAALyUHwKsQgAABAAAAEUAMwAAAMSUHwKwQgAABAAAAEUAMwAAACw1MgK0QgAALwAAAEcAMwAA AGQ1MgLjQgAALwAAAEcAMwAAAMyUHwISQwAABAAAAEUAMwAAANSUHwIWQwAABAAAAEUAMwAAANQ1 MgIaQwAALwAAAEcAMwAAANyUHwJJQwAABAAAAEUAMwAAAOSUHwJNQwAABAAAAEUAMwAAAOyUHwJR QwAABAAAAEUAMwAAAPSUHwJVQwAABAAAAEUAMwAAAPyUHwJZQwAABAAAAEUAMwAAAASVHwJdQwAA BAAAAEUAMwAAAAyVHwJhQwAABAAAAEUAMwAAABSVHwJlQwAABAAAAEUAMwAAAByVHwJpQwAABAAA AEUAMwAAACSVHwJtQwAABAAAAEUAMwAAACyVHwJxQwAABAAAAEUAMwAAAOSMHwJ1QwAABAAAAEUA MwAAADSVHwJ5QwAABAAAAEUAMwAAADyVHwJ9QwAABAAAAEUAMwAAAESVHwKBQwAABAAAAEUAMwAA AEyVHwKFQwAABAAAAEUAMwAAAFSVHwKJQwAABAAAAEUAMwAAAOyMHwKNQwAABAAAAEUAMwAAAFyV HwKRQwAABAAAAEUAMwAAAGSVHwKVQwAABAAAAEUAMwAAAGyVHwKZQwAABAAAAEUAMwAAAHSVHwKd QwAABAAAAEUAMwAAAHyVHwKhQwAABAAAAEUAMwAAAISVHwKlQwAABAAAAEUAAAAAAAEAAAACAAAA AwAAAAQAAAAFAAAABgAAAAcAAAAIAAAACQAAAAoAAAALAAAADAAAAA0AAAAOAAAADwAAABAAAAAR AAAAEgAAABMAAAAUAAAAFQAAABYAAAAXAAAAGAAAABkAAAAaAAAAGwAAABwAAAAdAAAAHgAAAB8A AAAgAAAAIQAAACIAAAAjAAAAJAAAACUAAAAmAAAAJwAAACgAAAApAAAAKgAAACsAAAAsAAAALQAA AC4AAAAvAAAAMAAAADEAAAAyAAAAMwAAADQAAAA1AAAANgAAADcAAAA4AAAAOQAAADoAAAA7AAAA PAAAAD0AAAA+AAAAPwAAAEAAAABBAAAAQgAAAEMAAABEAAAARQAAAEYAAABHAAAASAAAAEkAAABK AAAASwAAAEwAAABNAAAATgAAAE8AAAAAAAAAVWLl9gbr8Afr8Ajr8FUJ6/AK6/AL6/AM6/BVDevw DuvwD+vwEOvwVRHr8BLr8BPr8BTr8FUV6/AW6/AX6/AY6/BVGevwGuvwG+vwItzwAAAAAAAABAAA AKV06/A07/Tq8Rjr8AL+6/ABbasXiUR9/yJAaDJPdjHnQ/E/4/gCAQIB6vEL6/D/tFIyAq1KAAD1 TuvwQy0C5AsrAlX7NwAK6/BB6fIB3PQAAAEAAAADAAAAAAAldOvwNO/06vEC3P/d/vT+8f71C+vw BAwrAl9USwAACuvwQS0Cn9SMHwJeNwD+8UEC6fIBEAQUAAAASgEAAAAAAAAAAAAAAAAAAA8AAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQAAAAHQxMgIFSwAATwAAAFIAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQAAAAKwxMgJgSwAANgAAAFIAAAAAAAVI6/A83P8E DxYPKA4FSOvwPNz/BA8WDygOFAAAAEoBAAAAAAAAAAAAAAAAAAAPAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAEUAAAAcMjICxEwAAA0AAABSAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAEUAAABUMjIC0UwAAA0AAABSAAAAAABVEuf0AfXwUPnwVNzyGAAAABoAAAAB AAAAAAAAAAAAAAABAAAAAQAAAAAAAAAvAAAArPwjAgxOAAANAAAAQwAAAAAAAAAAAHvwAuvwAQAA 5PPwTuD7AQAD6/AKBQTr8H0J6/B3AQEAItz/4DQPRg9YD/n+5PeHGQLxMIYBaA/o8yjv4y3/TTjD 40AoDmnjGV+rAqcGpw5AHP+dBPf8iLIq7/Hp8ha+6wjkChICgOvw2O7r8EIAF+vwvCYyqwJY8/Ag 6/BQCxL0ohEQfBUWLx9BHxrr8CzvJzICoPPwfhIA7wDSADzr8FQBNL8CHhQAAArr8Eb7AEPr8KQR KwIoum8QBuvwRAAd6/A0/ywyAiwjAABjruvwUgAk6/BsjxCPkpMQHCsf6fIheRGPEK9UkxAYEynr 8NyPENOTEPULlxIn6/AULTICX0s3AABoGRIx6/BfDC8yArftEKPr8PnSrh8FEswwMgID7z0AANDz 8FAAMl4TATEyAql4EQdoEOsAP+vwPDEglksAywAqJyJE+wExIN5Mukgl/+vw/DIy6QZg+wAuixEz MgIZTgcAADapFgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAP7/AAAFAQIAAAAAAAAAAAAAAAAAAAAAAAEAAADghZ/y+U9o EKuRCAArJ7PZMAAAAPA1AAALAAAAAQAAAGAAAAACAAAAaAAAAAMAAAB0AAAABAAAAIAAAAAFAAAA pAAAAAYAAACwAAAABwAAALwAAAAIAAAAyAAAABIAAADsAAAADQAAAAQBAAARAAAAEAEAAAIAAADk BAAAHgAAAAQAAAAAAAAAHgAAAAQAAAAAAAAAHgAAABwAAABKdWluLUh3ZXkgKFJheW1vbmQpIENo ZW4AAAAAHgAAAAQAAAAAAAAAHgAAAAQAAAAAAAAAHgAAAAQAAAAAAAAAHgAAABwAAABKdWluLUh3 ZXkgKFJheW1vbmQpIENoZW4AAAAAHgAAABAAAABNaWNyb3NvZnQgVmlzaW8AQAAAAGCgCPRPcssB RwAAANg0AAD/////DgAAAAEAAABsAAAADAAAAAcAAABuAwAAbgAAAAAAAAAAAAAANV4AAK8LAAAg RU1GAAABANA0AABBAQAABAAAAAAAAAAAAAAAAAAAAIAHAACwBAAACAIAAEUBAAAAAAAAAAAAAAAA AABA7wcAiPUEAEYAAAAsAAAAIAAAAEVNRisBQAEAHAAAABAAAAACEMDbAQAAAGAAAABgAAAARgAA AOQAAADYAAAARU1GKzBAAgAQAAAABAAAAAAAgD8fQAQADAAAAAAAAAAeQAUADAAAAAAAAAAdQAAA FAAAAAgAAAACAAAAbQAAAAhAAAJAAAAANAAAAAIQwNsAAAAAzgAAAAAAAAAK16M+AgAAAAIAAAAC AAAAAgAAAAAAAAACEMDbAAAAAAAAAP8IQAEDSAAAADwAAAACEMDbBQAAAAAAAABcXoNDezLZQlxe s0N7MtlCXF6zQ4Rj/kFcXoNDhGP+QVxeg0N7MtlCAAEBAYEBAYEVQAEAEAAAAAQAAAAAAAAAIQAA AAgAAABiAAAADAAAAAEAAAAkAAAAJAAAAAAAgD0AAAAAAAAAAAAAgD0AAAAAAAAAAAIAAABfAAAA OAAAAAEAAAA4AAAAAAAAADgAAAAAAAAAAAABAAUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACUAAAAM AAAAAQAAACUAAAAMAAAABQAAgFYAAAAwAAAABQEAAB4AAABoAQAAbgAAAAUAAABsEMoGbBbKBmwW /QFsEP0BbBDKBiUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAkAAAAAACAQQAAAAAAAAAAAACA QQAAAAAAAAAAAgAAACgAAAAMAAAAAQAAAEYAAADAAAAAtAAAAEVNRisqQAAAJAAAABgAAAAAAMBC AAAAAAAAAAAAAMBCXF6DQ4Rj/kEIQAIGMAAAACQAAAACEMDbAKsqPgAAAAAAAAAAAAAAAAUAAABB AFIASQBBAEwAAAA2QAKAXAAAAFAAAAAAAAD/AQAAAAEAAAAEAAAAQwBFAEwAVAAA8JI+AACAPgCQ 0D4AAIA+gL0EPwAAgD5AeBw/AACAPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAABIAAAAMAAAAAQAA ABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAAFIAAABwAQAAAQAAAPD///8AAAAAAAAAAAAAAACQAQAA AAAAAAcABABBAHIAaQBhAGwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAADuBITOEgDGMtU7TPfuBJD37gQAAAAAnM4SAERzFHiwjOkEAAEAAAtf 1DtoAAAAsI3pBLCM6QTAzhIAkHMUeLCM6QQAAAAAaAEAAAgAAAAAAAAANAAAAJjPEgAA+e4EmPfu BAjPEgB8jtU7EwAAABgAAAAQjukE7M4SAAAAAAAAAJF8EIzpBLjPEgBBAJF8GC1EVPshCUDt7xh4 IAAAANjPEgAQAAAAW43VOxiO6QSojOkEEAAAAH8DAADEzxIAAAAZeBkQGXgIAAAAAAAAALzQEgAr EBl4fxcAAFjm7gTqtNM79hghqRyD7wFwAQAAAAAAAAgCAAAMAAAAeAEAAGR2AAgAAAAAJQAAAAwA AAABAAAAVAAAAGQAAAAiAQAAKQAAAEoBAAA6AAAAAQAAAKqq2EGqqthBIgEAADgAAAAEAAAATAAA AAAAAAAAAAAAAAAAAP//////////VAAAAEMARQBMAFQADAAAAAsAAAAJAAAAAAAAACUAAAAMAAAA DQAAgEYAAACoAAAAnAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAAABgAAAAAAMBCAAAAAAAAAAAA AMBCXF6DQ4Rj/kE2QAKAaAAAAFwAAAAAAAD/AQAAAAEAAAAFAAAAQQB1AGQAaQBvAIDfkj6AZuY+ gMrLPoBm5j4AQPs+gGbmPsBaFT+AZuY+gNUeP4Bm5j4AAIA/AAAAAAAAAAAAAIA/AAAAAAAAAAAA ABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAMAAAAAQAAAFQAAABsAAAA IgEAADwAAABKAQAATQAAAAEAAACqqthBqqrYQSIBAABLAAAABQAAAEwAAAAAAAAAAAAAAAAAAAD/ /////////1gAAABBAHUAZABpAG8AAAALAAAACQAAAAkAAAAEAAAAAAAAACUAAAAMAAAADQAAgEYA AAC8AAAAsAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAAABgAAAAAAMBCAAAAAAAAAAAAAMBCXF6D Q4Rj/kE2QAKAfAAAAHAAAAAAAAD/AQAAAAEAAAAHAAAARQBuAGMAbwBkAGUAcgAAKkI+gGYmPwAA mj6AZiY/gHXJPoBmJj+AIPQ+gGYmPwDLET+AZiY/wIUpP4BmJj+AQEE/gGYmPwAAgD8AAAAAAAAA AAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwA AAABAAAAVAAAAHgAAAAZAQAATwAAAFQBAABgAAAAAQAAAKqq2EGqqthBGQEAAF4AAAAHAAAATAAA AAAAAAAAAAAAAAAAAP//////////XAAAAEUAbgBjAG8AZABlAHIAAAALAAAACQAAAAgAAAAJAAAA CQAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAAJgAAACMAAAARU1GKytAAAAMAAAAAAAAAAhAAAJA AAAANAAAAAIQwNsAAAAAzgAAAAAAAAAK16M+AgAAAAIAAAACAAAAAgAAAAAAAAACEMDbAAAAAL93 Rv8IQAEDLAAAACAAAAACEMDbAgAAAAAAAADmMVJBrmWMQjJv7EKuZYxCAAEBARVAAQAQAAAABAAA AAAAAAAkAAAAJAAAAAAAgD0AAAAAAAAAAAAAgD0AAAAAAAAAAAIAAABfAAAAOAAAAAIAAAA4AAAA AAAAADgAAAAAAAAAAAABAAUAAAAAAAAARne/AAAAAAAAAAAAAAAAACUAAAAMAAAAAgAAACUAAAAM AAAABQAAgFcAAAAkAAAADAAAAEUAAAB4AAAASAAAAAIAAADTAGQEZAdkBCUAAAAMAAAABwAAgCUA AAAMAAAAAAAAgCQAAAAkAAAAAACAQQAAAAAAAAAAAACAQQAAAAAAAAAAAgAAACgAAAAMAAAAAgAA AEYAAABcAAAAUAAAAEVNRisIQAEDPAAAADAAAAACEMDbBAAAAAAAAAAomOpChQmFQlJWAEOuZYxC KJjqQtfBk0IomOpChQmFQgABAYEUQAGAEAAAAAQAAAC/d0b/JAAAACQAAAAAAIA9AAAAAAAAAAAA AIA9AAAAAAAAAAACAAAAJwAAABgAAAACAAAAAAAAAEZ3vwAAAAAAJQAAAAwAAAACAAAAEwAAAAwA AAABAAAAJQAAAAwAAAAIAACAVgAAACwAAAB1AAAAQgAAAIEAAABKAAAABAAAAFUHKQQGCGQEVQef BFUHKQQlAAAADAAAAAcAAIATAAAADAAAAAEAAAAlAAAADAAAAAAAAIAkAAAAJAAAAAAAgEEAAAAA AAAAAAAAgEEAAAAAAAAAAAIAAABGAAAApAAAAJgAAABFTUYrKkAAACQAAAAYAAAAAADAQgAAAAAA AAAAAADAQptPQ0QY2A1BNkACgHAAAABkAAAAAAAA/wEAAAABAAAABgAAAE8AdQB0AHAAdQB0AACS Gj4Agy0+AKmPPgCDLT6AHr8+AIMtPgDU1j4Agy0+wCQDPwCDLT6A3xo/AIMtPgAAgD8AAAAAAAAA AAAAgD8AAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAMAAAA AQAAAFQAAABwAAAAHAMAAAoAAABKAwAAGwAAAAEAAACqqthBqqrYQRwDAAAZAAAABgAAAEwAAAAA AAAAAAAAAAAAAAD//////////1gAAABPAHUAdABwAHUAdAAMAAAACQAAAAQAAAAJAAAACQAAAAAA AAAlAAAADAAAAA0AAIBGAAAAqAAAAJwAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAYAAAAAADA QgAAAAAAAAAAAADAQptPQ0QY2A1BNkACgGgAAABcAAAAAAAA/wEAAAABAAAABQAAAEEAdQBkAGkA bwAAfEA+ACi9PgApmT4AKL0+gJ7IPgAovT4AFPg+ACi9PsCEBT8AKL0+AACAPwAAAAAAAAAAAACA PwAAAAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEA AABUAAAAbAAAAB8DAAAdAAAARwMAAC4AAAABAAAAqqrYQaqq2EEfAwAALAAAAAUAAABMAAAAAAAA AAAAAAAAAAAA//////////9YAAAAQQB1AGQAaQBvAAAACwAAAAkAAAAJAAAABAAAAAAAAAAlAAAA DAAAAA0AAIBGAAAAsAAAAKQAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAYAAAAAADAQgAAAAAA AAAAAADAQptPQ0QY2A1BNkACgHAAAABkAAAAAAAA/wEAAAABAAAABgAAAFMAaQBnAG4AYQBsAACH LT5AxxE/gK6PPkDHET8ApKI+QMcRP4AZ0j5AxxE/gMcAP0DHET9Aghg/QMcRPwAAgD8AAAAAAAAA AAAAgD8AAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAMAAAA AQAAAFQAAABwAAAAHgMAADEAAABJAwAAQgAAAAEAAACqqthBqqrYQR4DAABAAAAABgAAAEwAAAAA AAAAAAAAAAAAAAD//////////1gAAABTAGkAZwBuAGEAbAALAAAABAAAAAkAAAAJAAAACQAAAAAA AAAlAAAADAAAAA0AAIBGAAAA2AAAAMwAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAYAAAAAADA QgAAAAAAAAAAAADAQjafxkOkcB1ANkACgJgAAACMAAAAAAAA/wEAAAABAAAACgAAAEMAbwBtAHAA cgBlAHMAcwBlAGQAAOSePQDNTD4Asko+AM1MPoDOlD4AzUw+AOTbPgDNTD7ArAU/AM1MPgDiEz8A zUw+wJwrPwDNTD5A8kA/AM1MPsBHVj8AzUw+gAJuPwDNTD4AAIA/AAAAAAAAAAAAAIA/AAAAAAAA AAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAiAAA AJUBAAAHAAAA7gEAABgAAAABAAAAqqrYQaqq2EGVAQAAFgAAAAoAAABMAAAAAAAAAAAAAAAAAAAA //////////9gAAAAQwBvAG0AcAByAGUAcwBzAGUAZAAMAAAACQAAAA0AAAAJAAAABQAAAAkAAAAI AAAACAAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAAKgAAACcAAAARU1GKytAAAAMAAAAAAAAACpA AAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEI2n8ZDpHAdQDZAAoBoAAAAXAAAAAAAAP8BAAAAAQAA AAUAAABBAHUAZABpAG8AAHmsPgDNzD4AZOU+AM3MPsBsCj8Azcw+gCciPwDNzD5Aois/AM3MPgAA gD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAY AAAAJQAAAAwAAAABAAAAVAAAAGwAAACuAQAAGgAAANYBAAArAAAAAQAAAKqq2EGqqthBrgEAACkA AAAFAAAATAAAAAAAAAAAAAAAAAAAAP//////////WAAAAEEAdQBkAGkAbwAAAAsAAAAJAAAACQAA AAQAAAAAAAAAJQAAAAwAAAANAACARgAAAJQAAACIAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAA GAAAAAAAwEIAAAAAAAAAAAAAwEI2n8ZDpHAdQDZAAoBUAAAASAAAAAAAAP8BAAAAAQAAAAMAAABC AGkAdAAAZy4+wJkZP4AekD7AmRk/ABSjPsCZGT8AAIA/AAAAAAAAAAAAAIA/AAAAAAAAAAAAABIA AAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAMAAAAAQAAAFQAAABgAAAAngEA AC0AAACwAQAAPgAAAAEAAACqqthBqqrYQZ4BAAA8AAAAAwAAAEwAAAAAAAAAAAAAAAAAAAD///// /////1QAAABCAGkAdAAAAAsAAAAEAAAAAAAAACUAAAAMAAAADQAAgEYAAACAAAAAdAAAAEVNRisr QAAADAAAAAAAAAAqQAAAJAAAABgAAAAAAMBCAAAAAAAAAAAAAMBCNp/GQ6RwHUA2QAKAQAAAADQA AAAAAAD/AQAAAAEAAAABAAAALQCAybo+wJkZPwAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAA AAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAFQAAACwAQAA LQAAALMBAAA+AAAAAQAAAKqq2EGqqthBsAEAADwAAAABAAAATAAAAAAAAAAAAAAAAAAAAP////// ////UAAAAC0AAAAAAAAAJQAAAAwAAAANAACARgAAALAAAACkAAAARU1GKytAAAAMAAAAAAAAACpA AAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEI2n8ZDpHAdQDZAAoBwAAAAZAAAAAAAAP8BAAAAAQAA AAYAAABzAHQAcgBlAGEAbQAANNc+wJkZP4DvAD/AmRk/QMoMP8CZGT+A/xo/wJkZP0C6Mj/AmRk/ AHVKP8CZGT8AAIA/AAAAAAAAAAAAAIA/AAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAW AAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAcAAAALYBAAAtAAAA5AEAAD4AAAABAAAAqqrYQaqq 2EG2AQAAPAAAAAYAAABMAAAAAAAAAAAAAAAAAAAA//////////9YAAAAcwB0AHIAZQBhAG0ACAAA AAQAAAAFAAAACQAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAAKgAAACcAAAARU1GKytAAAAMAAAA AAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwELXMLRBGNgNQTZAAoBoAAAAXAAAAAAAAP8B AAAAAQAAAAUAAABJAG4AcAB1AHQAAPJcPgCDLT6ALoY+AIMtPgCktT4Agy0+gBnlPgCDLT6ARwo/ AIMtPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAA AAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAGwAAAArAAAACgAAAE0AAAAbAAAAAQAAAKqq2EGqqthB KwAAABkAAAAFAAAATAAAAAAAAAAAAAAAAAAAAP//////////WAAAAEkAbgBwAHUAdAAAAAQAAAAJ AAAACQAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAAKgAAACcAAAARU1GKytAAAAMAAAAAAAAACpA AAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwELXMLRBGNgNQTZAAoBoAAAAXAAAAAAAAP8BAAAAAQAA AAUAAABBAHUAZABpAG8AAHxAPgAovT4AKZk+ACi9PoCeyD4AKL0+ABT4PgAovT7AhAU/ACi9PgAA gD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAY AAAAJQAAAAwAAAABAAAAVAAAAGwAAAApAAAAHQAAAFEAAAAuAAAAAQAAAKqq2EGqqthBKQAAACwA AAAFAAAATAAAAAAAAAAAAAAAAAAAAP//////////WAAAAEEAdQBkAGkAbwAAAAsAAAAJAAAACQAA AAQAAAAAAAAAJQAAAAwAAAANAACARgAAALAAAACkAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAA GAAAAAAAwEIAAAAAAAAAAAAAwELXMLRBGNgNQTZAAoBwAAAAZAAAAAAAAP8BAAAAAQAAAAYAAABT AGkAZwBuAGEAbAAAhy0+QMcRP4Cujz5AxxE/AKSiPkDHET+AGdI+QMcRP4DHAD9AxxE/QIIYP0DH ET8AAIA/AAAAAAAAAAAAAIA/AAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAA ABgAAAAlAAAADAAAAAEAAABUAAAAcAAAACcAAAAxAAAAUgAAAEIAAAABAAAAqqrYQaqq2EEnAAAA QAAAAAYAAABMAAAAAAAAAAAAAAAAAAAA//////////9YAAAAUwBpAGcAbgBhAGwACwAAAAQAAAAJ AAAACQAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAAEAAAAA0AAAARU1GKytAAAAMAAAAAAAAAApA AIAkAAAAGAAAAPfu6P8BAAAAUlYAQ4Rj/kEAAMBCmpmZQigAAAAMAAAAAgAAACQAAAAkAAAAAACA PQAAAAAAAAAAAACAPQAAAAAAAAAAAgAAACcAAAAYAAAAAgAAAAAAAADo7vcAAAAAACUAAAAMAAAA AgAAACUAAAAMAAAACAAAgFYAAAAwAAAAgAAAAB8AAADhAAAAbQAAAAUAAAAGCP0BBgjKBgYOygYG Dv0BBgj9ASUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAkAAAAAACAQQAAAAAAAAAAAACAQQAA AAAAAAAAAgAAAEYAAACoAAAAnAAAAEVNRisIQAACQAAAADQAAAACEMDbAAAAAM4AAAAAAAAACtej PgIAAAACAAAAAgAAAAIAAAAAAAAAAhDA2wAAAAAAAAD/CEABA0gAAAA8AAAAAhDA2wUAAAAAAAAA UlYAQ3sy2UJSVmBDezLZQlJWYEOEY/5BUlYAQ4Rj/kFSVgBDezLZQgABAQGBAQGBFUABABAAAAAE AAAAAAAAACQAAAAkAAAAAACAPQAAAAAAAAAAAACAPQAAAAAAAAAAAgAAAF8AAAA4AAAAAwAAADgA AAAAAAAAOAAAAAAAAAAAAAEABQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAJQAAAAwAAAADAAAAJQAA AAwAAAAFAACAVgAAADAAAAB/AAAAHgAAAOIAAABuAAAABQAAAAYIygYGDsoGBg79AQYI/QEGCMoG JQAAAAwAAAAHAACAJQAAAAwAAAAAAACAJAAAACQAAAAAAIBBAAAAAAAAAAAAAIBBAAAAAAAAAAAC AAAAKAAAAAwAAAADAAAARgAAAJwAAACQAAAARU1GKypAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAA wEJSVgBDhGP+QTZAAoBoAAAAXAAAAAAAAP8BAAAAAQAAAAUAAABQAGkAdABjAGgAACWhPoAzsz4A ENo+gDOzPoAF7T6AM7M+gF0CP4Azsz4Asxc/gDOzPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAA EgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAGwAAACf AAAAMgAAAMAAAABDAAAAAQAAAKqq2EGqqthBnwAAAEEAAAAFAAAATAAAAAAAAAAAAAAAAAAAAP// ////////WAAAAFAAaQB0AGMAaAAAAAsAAAAEAAAABAAAAAgAAAAAAAAAJQAAAAwAAAANAACARgAA ANAAAADEAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEJSVgBD hGP+QTZAAoCQAAAAhAAAAAAAAP8BAAAAAQAAAAkAAABQAHIAZQBmAGkAbAB0AGUAcgAA/1k+AM0M P4DqpT4AzQw/AFXCPgDNDD+AyvE+AM0MPwDABD8AzQw/wDoOPwDNDD+AtRc/AM0MP0CQIz8AzQw/ AEs7PwDNDD8AAIA/AAAAAAAAAAAAAIA/AAAAAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAA ABYAAAAMAAAAGAAAACUAAAAMAAAAAQAAAFQAAACEAAAAlQAAAEYAAADLAAAAVwAAAAEAAACqqthB qqrYQZUAAABVAAAACQAAAEwAAAAAAAAAAAAAAAAAAAD//////////2AAAABQAHIAZQBmAGkAbAB0 AGUAcgAAAAsAAAAFAAAACQAAAAQAAAAEAAAABAAAAAQAAAAJAAAAAAAAACUAAAAMAAAADQAAgEYA AACYAAAAjAAAAEVNRisrQAAADAAAAAAAAAAIQAACQAAAADQAAAACEMDbAAAAAM4AAAAAAAAACtej PgIAAAACAAAAAgAAAAIAAAAAAAAAAhDA2wAAAAC/d0b/CEABAywAAAAgAAAAAhDA2wIAAAAAAAAA UlZgQ65ljEIAnnxDrmWMQgABAQEVQAEAEAAAAAQAAAAAAAAAJAAAACQAAAAAAIA9AAAAAAAAAAAA AIA9AAAAAAAAAAACAAAAXwAAADgAAAADAAAAOAAAAAAAAAA4AAAAAAAAAAAAAQAFAAAAAAAAAEZ3 vwAAAAAAAAAAAAAAAAAlAAAADAAAAAMAAAAlAAAADAAAAAUAAIBXAAAAJAAAAN8AAABFAAAA/gAA AEgAAAACAAAABg5kBMoPZAQlAAAADAAAAAcAAIAlAAAADAAAAAAAAIAkAAAAJAAAAAAAgEEAAAAA AAAAAAAAgEEAAAAAAAAAAAIAAAAoAAAADAAAAAMAAABGAAAAXAAAAFAAAABFTUYrCEABAzwAAAAw AAAAAhDA2wQAAAAAAAAAe7J7Q4UJhUJcXoNDrmWMQnuye0PXwZNCe7J7Q4UJhUIAAQGBFEABgBAA AAAEAAAAv3dG/ygAAAAMAAAAAgAAACQAAAAkAAAAAACAPQAAAAAAAAAAAACAPQAAAAAAAAAAAgAA ACcAAAAYAAAAAgAAAAAAAABGd78AAAAAACUAAAAMAAAAAgAAABMAAAAMAAAAAQAAACUAAAAMAAAA CAAAgFYAAAAsAAAA+wAAAEIAAAAHAQAASgAAAAQAAAC8DykEbBBkBLwPnwS8DykEJQAAAAwAAAAH AACAEwAAAAwAAAABAAAAJQAAAAwAAAAAAACAJAAAACQAAAAAAIBBAAAAAAAAAAAAAIBBAAAAAAAA AAACAAAARgAAAIwAAACAAAAARU1GKwhAAAJAAAAANAAAAAIQwNsAAAAAzgAAAAAAAAAK16M+AgAA AAIAAAACAAAAAgAAAAAAAAACEMDbAAAAAL93Rv8IQAEDLAAAACAAAAACEMDbAgAAAAAAAABcXrND rmWMQu3HBESuZYxCAAEBgRVAAQAQAAAABAAAAAAAAAAkAAAAJAAAAAAAgD0AAAAAAAAAAAAAgD0A AAAAAAAAAAIAAABfAAAAOAAAAAMAAAA4AAAAAAAAADgAAAAAAAAAAAABAAUAAAAAAAAARne/AAAA AAAAAAAAAAAAACUAAAAMAAAAAwAAACUAAAAMAAAABQAAgFcAAAAkAAAAZQEAAEUAAAAVAgAASAAA AAIAAABsFmQEMiFkBCUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAkAAAAAACAQQAAAAAAAAAA AACAQQAAAAAAAAAAAgAAACgAAAAMAAAAAwAAAEYAAABcAAAAUAAAAEVNRisIQAEDPAAAADAAAAAC EMDbBAAAAAAAAAAMjQREhQmFQptPB0SuZYxCDI0ERNfBk0IMjQREhQmFQgABAYEUQAGAEAAAAAQA AAC/d0b/JAAAACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAJQAAAAwAAAACAAAAEwAA AAwAAAABAAAAJQAAAAwAAAAIAACAVgAAACwAAAASAgAAQgAAAB4CAABKAAAABAAAACQhKQTUIWQE JCGfBCQhKQQlAAAADAAAAAcAAIATAAAADAAAAAEAAAAlAAAADAAAAAAAAIAkAAAAJAAAAAAAgEEA AAAAAAAAAAAAgEEAAAAAAAAAAAIAAABGAAAAqAAAAJwAAABFTUYrCEAAAkAAAAA0AAAAAhDA2wAA AADOAAAAAAAAAArXoz4CAAAAAgAAAAIAAAACAAAAAAAAAAIQwNsAAAAAAAAA/whAAQNIAAAAPAAA AAIQwNsFAAAAAAAAAJtPB0R7MtlCm08fRHsy2UKbTx9EhGP+QZtPB0SEY/5Bm08HRHsy2UIAAQEB gQEBgRVAAQAQAAAABAAAAAAAAAAkAAAAJAAAAAAAgD0AAAAAAAAAAAAAgD0AAAAAAAAAAAIAAABf AAAAOAAAAAMAAAA4AAAAAAAAADgAAAAAAAAAAAABAAUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACUA AAAMAAAAAwAAACUAAAAMAAAABQAAgFYAAAAwAAAAHAIAAB4AAAB/AgAAbgAAAAUAAADUIcoG1CfK BtQn/QHUIf0B1CHKBiUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAkAAAAAACAQQAAAAAAAAAA AACAQQAAAAAAAAAAAgAAACgAAAAMAAAAAwAAAEYAAACcAAAAkAAAAEVNRisqQAAAJAAAABgAAAAA AMBCAAAAAAAAAAAAAMBCm08HRIRj/kE2QAKAaAAAAFwAAAAAAAD/AQAAAAEAAAAFAAAAQwBFAEwA VAAgAADwkj4AAIA+AJDQPgAAgD6AvQQ/AACAPkB4HD8AAIA+QIg2PwAAgD4AAIA/AAAAAAAAAAAA AIA/AAAAAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAMAAAA AQAAAFQAAABsAAAAOQIAACkAAABjAgAAOgAAAAEAAACqqthBqqrYQTkCAAA4AAAABQAAAEwAAAAA AAAAAAAAAAAAAAD//////////1gAAABDAEUATABUACAAAAAMAAAACwAAAAkAAAAKAAAAAAAAACUA AAAMAAAADQAAgEYAAACoAAAAnAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAAABgAAAAAAMBCAAAA AAAAAAAAAMBCm08HRIRj/kE2QAKAaAAAAFwAAAAAAAD/AQAAAAEAAAAFAAAAQQB1AGQAaQBvAIDf kj6AZuY+gMrLPoBm5j4AQPs+gGbmPsBaFT+AZuY+gNUeP4Bm5j4AAIA/AAAAAAAAAAAAAIA/AAAA AAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAMAAAAAQAAAFQA AABsAAAAOQIAADwAAABhAgAATQAAAAEAAACqqthBqqrYQTkCAABLAAAABQAAAEwAAAAAAAAAAAAA AAAAAAD//////////1gAAABBAHUAZABpAG8AAAALAAAACQAAAAkAAAAEAAAAAAAAACUAAAAMAAAA DQAAgEYAAAC8AAAAsAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAAABgAAAAAAMBCAAAAAAAAAAAA AMBCm08HRIRj/kE2QAKAfAAAAHAAAAAAAAD/AQAAAAEAAAAHAAAARABlAGMAbwBkAGUAcgAAdT0+ gGYmP4BanD6AZiY/ANDLPoBmJj8Ae/Y+gGYmP0D4Ej+AZiY/ALMqP4BmJj/AbUI/gGYmPwAAgD8A AAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAA JQAAAAwAAAABAAAAVAAAAHgAAAAvAgAATwAAAGsCAABgAAAAAQAAAKqq2EGqqthBLwIAAF4AAAAH AAAATAAAAAAAAAAAAAAAAAAAAP//////////XAAAAEQAZQBjAG8AZABlAHIAAAAMAAAACQAAAAgA AAAJAAAACQAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAAEAAAAA0AAAARU1GKytAAAAMAAAAAAAA AApAAIAkAAAAGAAAAPfu6P8BAAAANekoRIRj/kEAAMBCmpmZQigAAAAMAAAAAgAAACQAAAAkAAAA AACAPQAAAAAAAAAAAACAPQAAAAAAAAAAAgAAACcAAAAYAAAAAgAAAAAAAADo7vcAAAAAACUAAAAM AAAAAgAAACUAAAAMAAAACAAAgFYAAAAwAAAAowIAAB8AAAAEAwAAbQAAAAUAAAA7Kv0BOyrKBjsw ygY7MP0BOyr9ASUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAkAAAAAACAQQAAAAAAAAAAAACA QQAAAAAAAAAAAgAAAEYAAACoAAAAnAAAAEVNRisIQAACQAAAADQAAAACEMDbAAAAAM4AAAAAAAAA CtejPgIAAAACAAAAAgAAAAIAAAAAAAAAAhDA2wAAAAAAAAD/CEABA0gAAAA8AAAAAhDA2wUAAAAA AAAANekoRHsy2UI16UBEezLZQjXpQESEY/5BNekoRIRj/kE16ShEezLZQgABAQGBAQGBFUABABAA AAAEAAAAAAAAACQAAAAkAAAAAACAPQAAAAAAAAAAAACAPQAAAAAAAAAAAgAAAF8AAAA4AAAAAwAA ADgAAAAAAAAAOAAAAAAAAAAAAAEABQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAJQAAAAwAAAADAAAA JQAAAAwAAAAFAACAVgAAADAAAACiAgAAHgAAAAUDAABuAAAABQAAADsqygY7MMoGOzD9ATsq/QE7 KsoGJQAAAAwAAAAHAACAJQAAAAwAAAAAAACAJAAAACQAAAAAAIBBAAAAAAAAAAAAAIBBAAAAAAAA AAACAAAAKAAAAAwAAAADAAAARgAAAJwAAACQAAAARU1GKypAAAAkAAAAGAAAAAAAwEIAAAAAAAAA AAAAwEI16ShEhGP+QTZAAoBoAAAAXAAAAAAAAP8BAAAAAQAAAAUAAABQAGkAdABjAGgAACWhPoAz sz4AENo+gDOzPoAF7T6AM7M+gF0CP4Azsz4Asxc/gDOzPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAA AAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAGwA AADCAgAAMgAAAOMCAABDAAAAAQAAAKqq2EGqqthBwgIAAEEAAAAFAAAATAAAAAAAAAAAAAAAAAAA AP//////////WAAAAFAAaQB0AGMAaAAAAAsAAAAEAAAABAAAAAgAAAAAAAAAJQAAAAwAAAANAACA RgAAANgAAADMAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEI1 6ShEhGP+QTZAAoCYAAAAjAAAAAAAAP8BAAAAAQAAAAoAAABQAG8AcwB0AGYAaQBsAHQAZQByAAAJ ND4AzQw/gO+SPgDNDD8AZcI+AM0MPwAQ7T4AzQw/wGICPwDNDD+APQ4/AM0MP0C4Fz8AzQw/ADMh PwDNDD/ADS0/AM0MP4DIRD8AzQw/AACAPwAAAAAAAAAAAACAPwAAAAAAAAAAEgAAAAwAAAABAAAA GAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAIgAAAC1AgAARgAAAPICAABX AAAAAQAAAKqq2EGqqthBtQIAAFUAAAAKAAAATAAAAAAAAAAAAAAAAAAAAP//////////YAAAAFAA bwBzAHQAZgBpAGwAdABlAHIACwAAAAkAAAAIAAAABAAAAAQAAAAEAAAABAAAAAQAAAAJAAAAAAAA ACUAAAAMAAAADQAAgEYAAACYAAAAjAAAAEVNRisrQAAADAAAAAAAAAAIQAACQAAAADQAAAACEMDb AAAAAM4AAAAAAAAACtejPgIAAAACAAAAAgAAAAIAAAAAAAAAAhDA2wAAAAC/d0b/CEABAywAAAAg AAAAAhDA2wIAAAAAAAAAm08fRK5ljEKHYSZErmWMQgABAQEVQAEAEAAAAAQAAAAAAAAAJAAAACQA AAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAXwAAADgAAAADAAAAOAAAAAAAAAA4AAAAAAAA AAAAAQAFAAAAAAAAAEZ3vwAAAAAAAAAAAAAAAAAlAAAADAAAAAMAAAAlAAAADAAAAAUAAIBXAAAA JAAAAHwCAABFAAAAmwIAAEgAAAACAAAA1CdkBJkpZAQlAAAADAAAAAcAAIAlAAAADAAAAAAAAIAk AAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIAAAAoAAAADAAAAAMAAABGAAAAXAAAAFAA AABFTUYrCEABAzwAAAAwAAAAAhDA2wQAAAAAAAAApSYmRIUJhUI16ShErmWMQqUmJkTXwZNCpSYm RIUJhUIAAQGBFEABgBAAAAAEAAAAv3dG/ygAAAAMAAAAAgAAACQAAAAkAAAAAACAPQAAAAAAAAAA AACAPQAAAAAAAAAAAgAAACcAAAAYAAAAAgAAAAAAAABGd78AAAAAACUAAAAMAAAAAgAAABMAAAAM AAAAAQAAACUAAAAMAAAACAAAgFYAAAAsAAAAmAIAAEIAAACkAgAASgAAAAQAAACKKSkEOypkBIop nwSKKSkEJQAAAAwAAAAHAACAEwAAAAwAAAABAAAAJQAAAAwAAAAAAACAJAAAACQAAAAAAIBBAAAA AAAAAAAAAIBBAAAAAAAAAAACAAAARgAAAIwAAACAAAAARU1GKwhAAAJAAAAANAAAAAIQwNsAAAAA zgAAAAAAAAAK16M+AgAAAAIAAAACAAAAAgAAAAAAAAACEMDbAAAAAL93Rv8IQAEDLAAAACAAAAAC EMDbAgAAAAAAAAA16UBErmWMQsHOWESuZYxCAAEBgRVAAQAQAAAABAAAAAAAAAAkAAAAJAAAAAAA gD0AAAAAAAAAAAAAgD0AAAAAAAAAAAIAAABfAAAAOAAAAAMAAAA4AAAAAAAAADgAAAAAAAAAAAAB AAUAAAAAAAAARne/AAAAAAAAAAAAAAAAACUAAAAMAAAAAwAAACUAAAAMAAAABQAAgFcAAAAkAAAA AgMAAEUAAABlAwAASAAAAAIAAAA7MGQENDZkBCUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAk AAAAAACAQQAAAAAAAAAAAACAQQAAAAAAAAAAAgAAACgAAAAMAAAAAwAAAEYAAABcAAAAUAAAAEVN RisIQAEDPAAAADAAAAACEMDbBAAAAAAAAADfk1hEhQmFQm9WW0SuZYxC35NYRNfBk0Lfk1hEhQmF QgABAYEUQAGAEAAAAAQAAAC/d0b/JAAAACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAA JQAAAAwAAAACAAAAEwAAAAwAAAABAAAAJQAAAAwAAAAIAACAVgAAACwAAABiAwAAQgAAAG4DAABK AAAABAAAACU2KQTWNmQEJTafBCU2KQQlAAAADAAAAAcAAIATAAAADAAAAAEAAAAlAAAADAAAAAAA AIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIAAABMAAAAZAAAAAwAAAAIAAAAbQMA AG0AAAAMAAAACAAAAGIDAABmAAAAKQCqAAAAAAAAAAAAAACAPwAAAAAAAAAAAACAPwAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIgAAAAwAAAD/////RgAAABwAAAAQAAAARU1GKwJAAAAM AAAAAAAAAA4AAAAUAAAAAAAAABAAAAAUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABQBEAG8A YwB1AG0AZQBuAHQAUwB1AG0AbQBhAHIAeQBJAG4AZgBvAHIAbQBhAHQAaQBvAG4AAAAAAAAAAAAA ADgAAgD///////////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAFAAAA AAIAAAAAAABfADEAMwA0ADkAMgA3ADcAMgA3ADEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAGAABAf//////////EQAAABQaAgAAAAAAwAAAAAAAAEYAAAAA0E27vGty ywHQTbu8a3LLAQAAAAAAAAAAAAAAAAEATwBsAGUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAKAAIB////////////////AAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADQAAABQAAAAAAAAAAwBFAFAAUgBJAE4AVAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAAAgAOAAAAEAAAAP// //8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACzAAAAuJ0AAAAAAAAAAAAAHgAA ABgAAABCcm9hZGNvbSBDb3Jwb3JhdGlvbgAAAAADAAAAAAAMAAsAAAAAAAAACwAAAAAAAAALAAAA AAAAAAsAAAAAAAAAHhAAAAMAAAAIAAAAUGFnZS0xAE4MAAAAUmVjdGFuZ2xlAEVOFAAAAER5bmFt aWMgY29ubmVjdG9yAGVuDBAAAAQAAAAeAAAACAAAAFBhZ2VzAAAAAwAAAAEAAAAeAAAACAAAAE1h c3RlcnMAAwAAAAIAAACAAAAABAAAAAAAAAAoAAAAAQAAAGAAAAACAAAAaAAAAAMAAAB0AAAAAgAA AAIAAAAOAAAAX1BJRF9MSU5LQkFTRQADAAAAFQAAAF9WUElEX0FMVEVSTkFURU5BTUVTAAACAAAA 5AQAAEEAAAACAAAAAAAAAB4AAAAEAAAAAAAAAAEAAAIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAP7/AwoAAP////8UGgIAAAAAAMAA AAAAAABGGAAAAE1pY3Jvc29mdCBWaXNpbyBEcmF3aW5nABIAAABWaXNpbyAxMS4wIFNoYXBlcwAR AAAAVmlzaW8uRHJhd2luZy4xMQD0ObJxAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEBAAD+//// OQEAAGMBAAAFAQAABgEAAAcBAAAIAQAACQEAAAoBAAALAQAADAEAAA0BAAAOAQAADwEAABABAAAR AQAAEgEAABMBAAAUAQAAFQEAABYBAAAXAQAAGAEAABkBAAAaAQAAGwEAABwBAAAdAQAAHgEAAB8B AAAgAQAAIQEAACIBAAAjAQAAJAEAACUBAAAmAQAAJwEAACgBAAApAQAAKgEAACsBAAAsAQAALQEA AC4BAAAvAQAAMAEAADEBAAAyAQAAMwEAADQBAAA1AQAANgEAADcBAAA4AQAA/v///2QBAAA7AQAA PAEAAD0BAAA+AQAAPwEAAEABAABBAQAAQgEAAEMBAABEAQAARQEAAEYBAABHAQAASAEAAEkBAABK AQAASwEAAEwBAABNAQAATgEAAE8BAABQAQAAUQEAAFIBAABTAQAAVAEAAFUBAABWAQAAVwEAAFgB AABZAQAAWgEAAFsBAABcAQAAXQEAAF4BAABfAQAAYAEAAGEBAABiAQAA/v///2YBAABlAQAAZwEA AGgBAABpAQAAawEAAGoBAABsAQAAbgEAAG0BAABvAQAAcQEAAHABAAByAQAAdAEAAHMBAAB1AQAA dgEAAHcBAAB5AQAAeAEAAHoBAAB7AQAAfAEAAH4BAAB9AQAAfwEAAIEBAACCAQAAAQAAAGwAAAAS AAAABwAAAHYDAACFAQAAAAAAAAAAAADsXwAAKSoAACBFTUYAAAEAuJ0AALwDAAAEAAAAAAAAAAAA AAAAAAAAgAcAALAEAAAIAgAARQEAAAAAAAAAAAAAAAAAAEDvBwCI9QQARgAAACwAAAAgAAAARU1G KwFAAQAcAAAAEAAAAAIQwNsBAAAAYAAAAGAAAABGAAAA5AAAANgAAABFTUYrMEACABAAAAAEAAAA AACAPx9ABAAMAAAAAAAAAB5ABQAMAAAAAAAAAB1AAAAUAAAACAAAAAIAAACNAQAACEAAAkAAAAA0 AAAAAhDA2wAAAADOAAAAAAAAAOxROEACAAAAAgAAAAIAAAACAAAAAAAAAAIQwNsAAAAAAAAA/whA AQNIAAAAPAAAAAIQwNsFAAAAAAAAAJNqh0NIobdDk2q3Q0iht0OTardD4TqRQ5Nqh0PhOpFDk2qH Q0iht0MAAQEBgQAAABVAAQAQAAAABAAAAAAAAAAhAAAACAAAAGIAAAAMAAAAAQAAACQAAAAkAAAA AACAPQAAAAAAAAAAAACAPQAAAAAAAAAAAgAAAF8AAAA4AAAAAQAAADgAAAAAAAAAOAAAAAAAAAAA AAEALgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAJQAAAAwAAAABAAAAJQAAAAwAAAAFAACAVgAAADAA AAAMAQAAIAEAAHIBAAByAQAABQAAAO4Q9RbuFvUW7hYoEu4QKBLuEPUWJQAAAAwAAAAHAACAJQAA AAwAAAAAAACAJAAAACQAAAAAAIBBAAAAAAAAAAAAAIBBAAAAAAAAAAACAAAAKAAAAAwAAAABAAAA RgAAAMwAAADAAAAARU1GKypAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEKTaodD4TqRQwhAAgYw AAAAJAAAAAIQwNsAqyo+AAAAAAAAAAAAAAAABQAAAEEAUgBJAEEATAAAADZAAoBoAAAAXAAAAAAA AP8BAAAAAQAAAAUAAABDAEUATABUACAAAPCSPgAAgD4AkNA+AACAPoC9BD8AAIA+QHgcPwAAgD5A iDY/AACAPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAA FgAAAAwAAAAYAAAAUgAAAHABAAABAAAA8P///wAAAAAAAAAAAAAAAJABAAAAAAAABwAEAEEAcgBp AGEAbAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAyxkhAhkAsAGWJfJ3liXyd2jkEgAgk/F3+5LxdyjkEgBkAQAAAAAAAAAAAABwfBoABgAA AAAAAADOGQoB0OUSAOST8Xfsk/F38P///wAAAAAAAAAAAAAAAJABAAAAAAAABwAEAHjkEgAAAKcD IgKRfBMAAAAYCqcDAACnAwAAzwRQ5BIAGC1EVPshCUDt7xh4IAAAAFDlEgAQAAAAmwGRfNsBkXw8 6hIAEAAAAH8DAAA85RIA9w8ZeBkQGXgAAAAAAAAAADTmEgArEBl4fxcAAMhrUAEAAAAABgAAAOq0 0zvHGSE4HIOnAwwAAAAAEAAAAAAAAAAAAQAMAAAAZHYACAAAAAAlAAAADAAAAAEAAABUAAAAbAAA ACoBAAArAQAAVAEAADwBAAABAAAAqqrYQaqq2EEqAQAAOgEAAAUAAABMAAAAAAAAAAAAAAAAAAAA //////////9YAAAAQwBFAEwAVAAgAAB0DAAAAAsAAAAJAAAACgAAAAAAAAAlAAAADAAAAA0AAIBG AAAAqAAAAJwAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQpNq h0PhOpFDNkACgGgAAABcAAAAAAAA/wEAAAABAAAABQAAAEEAdQBkAGkAbwCA35I+gGbmPoDKyz6A ZuY+AED7PoBm5j7AWhU/gGbmPoDVHj+AZuY+AACAPwAAAAAAAAAAAACAPwAAAAAAAAAAAAASAAAA DAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAbAAAACoBAAA/ AQAAUgEAAFABAAABAAAAqqrYQaqq2EEqAQAATgEAAAUAAABMAAAAAAAAAAAAAAAAAAAA//////// //9YAAAAQQB1AGQAaQBvAABECwAAAAkAAAAJAAAABAAAAAAAAAAlAAAADAAAAA0AAIBGAAAAvAAA ALAAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQpNqh0PhOpFD NkACgHwAAABwAAAAAAAA/wEAAAABAAAABwAAAEUAbgBjAG8AZABlAHIAACpCPoBmJj8AAJo+gGYm P4B1yT6AZiY/gCD0PoBmJj8AyxE/gGYmP8CFKT+AZiY/gEBBP4BmJj8AAIA/AAAAAAAAAAAAAIA/ AAAAAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAMAAAAAQAA AFQAAAB4AAAAIQEAAFIBAABcAQAAYwEAAAEAAACqqthBqqrYQSEBAABhAQAABwAAAEwAAAAAAAAA AAAAAAAAAAD//////////1wAAABFAG4AYwBvAGQAZQByAAAACwAAAAkAAAAIAAAACQAAAAkAAAAJ AAAAAAAAACUAAAAMAAAADQAAgEYAAACYAAAAjAAAAEVNRisrQAAADAAAAAAAAAAIQAACQAAAADQA AAACEMDbAAAAAM4AAAAAAAAA7FE4QAIAAAACAAAAAgAAAAIAAAAAAAAAAhDA2wAAAAAAAAD/CEAB AywAAAAgAAAAAhDA2wIAAAAAAAAAZNypQRRupEOUi+5CFG6kQwABAQEVQAEAEAAAAAQAAAAAAAAA JAAAACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAXwAAADgAAAACAAAAOAAAAAAAAAA4 AAAAAAAAAAAAAQAuAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAlAAAADAAAAAIAAAAlAAAADAAAAAUA AIBXAAAAJAAAABIAAABGAQAAegAAAEwBAAACAAAAVAGOFHUHjhQlAAAADAAAAAcAAIAlAAAADAAA AAAAAIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIAAAAoAAAADAAAAAIAAABGAAAA XAAAAFAAAABFTUYrCEABAzwAAAAwAAAAAhDA2wQAAAAAAAAA3GzrQlxPoUPAbghDFG6kQ9xs60LN jKdD3GzrQlxPoUMAAQGBFEABgBAAAAAEAAAAAAAA/yQAAAAkAAAAAACAPQAAAAAAAAAAAACAPQAA AAAAAAAAAgAAACcAAAAYAAAAAgAAAAAAAAAAAAAAAAAAACUAAAAMAAAAAgAAABMAAAAMAAAAAQAA ACUAAAAMAAAACAAAgFYAAAAsAAAAdQAAAEIBAACJAAAAUAEAAAQAAABcByoUhwiOFFwH8hRcByoU JQAAAAwAAAAHAACAEwAAAAwAAAABAAAAJQAAAAwAAAAAAACAJAAAACQAAAAAAIBBAAAAAAAAAAAA AIBBAAAAAAAAAAACAAAARgAAAIwAAACAAAAARU1GKwhAAAJAAAAANAAAAAIQwNsAAAAAzgAAAAAA AAAK16M+AgAAAAIAAAACAAAAAgAAAAAAAAACEMDbAAAAAL93Rv8IQAEDLAAAACAAAAACEMDbAgAA AAAAAACzEMRCFG6kQ7MQxEJ7lHxDAAEBgRVAAQAQAAAABAAAAAAAAAAkAAAAJAAAAAAAgD0AAAAA AAAAAAAAgD0AAAAAAAAAAAIAAABfAAAAOAAAAAMAAAA4AAAAAAAAADgAAAAAAAAAAAABAAUAAAAA AAAARne/AAAAAAAAAAAAAAAAACUAAAAMAAAAAwAAACUAAAAMAAAABQAAgFcAAAAkAAAAYAAAAPsA AABkAAAASgEAAAIAAAAhBo4UIQbKDyUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAkAAAAAACA QQAAAAAAAAAAAACAQQAAAAAAAAAAAgAAACgAAAAMAAAAAwAAAEYAAABcAAAAUAAAAEVNRisIQAED PAAAADAAAAACEMDbBAAAAAAAAACKtLxCAIB9Q7MQxELDdXJD3GzLQgCAfUOKtLxCAIB9QwABAYEU QAGAEAAAAAQAAAC/d0b/KAAAAAwAAAACAAAAJAAAACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAA AAACAAAAJwAAABgAAAACAAAAAAAAAEZ3vwAAAAAAJQAAAAwAAAACAAAAEwAAAAwAAAABAAAAJQAA AAwAAAAIAACAVgAAACwAAABeAAAA8gAAAGYAAAD+AAAABAAAAOYF2A8hBigPXAbYD+YF2A8lAAAA DAAAAAcAAIATAAAADAAAAAEAAAAlAAAADAAAAAAAAIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEA AAAAAAAAAAIAAABGAAAApAAAAJgAAABFTUYrKkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQhaC TERcD3xDNkACgHAAAABkAAAAAAAA/wEAAAABAAAABgAAAE8AdQB0AHAAdQB0AAB2zD0AAIA+APtq PgAAgD4A86Q+AACAPoCovD4AAIA+AB7sPgAAgD7AyQ0/AACAPgAAgD8AAAAAAAAAAAAAgD8AAAAA AAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAMAAAAAQAAAFQAAABw AAAAPAMAAAUBAABqAwAAFgEAAAEAAACqqthBqqrYQTwDAAAUAQAABgAAAEwAAAAAAAAAAAAAAAAA AAD//////////1gAAABPAHUAdABwAHUAdAAMAAAACQAAAAQAAAAJAAAACQAAAAAAAAAlAAAADAAA AA0AAIBGAAAAqAAAAJwAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAYAAAAAADAQgAAAAAAAAAA AADAQhaCTERcD3xDNkACgGgAAABcAAAAAAAA/wEAAAABAAAABQAAAEEAdQBkAGkAbwAAJgw+gGbm PgD8fT6AZuY+gHOuPoBm5j4A6d0+gGbmPoDe8D6AZuY+AACAPwAAAAAAAAAAAACAPwAAAAAAAAAA AAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAbAAA AD8DAAAYAQAAZwMAACkBAAABAAAAqqrYQaqq2EE/AwAAJwEAAAUAAABMAAAAAAAAAAAAAAAAAAAA //////////9YAAAAQQB1AGQAaQBvAAAACwAAAAkAAAAJAAAABAAAAAAAAAAlAAAADAAAAA0AAIBG AAAAsAAAAKQAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQhaC TERcD3xDNkACgHAAAABkAAAAAAAA/wEAAAABAAAABgAAAFMAaQBnAG4AYQBsAABg8j2AZiY/AAZr PoBmJj+AeIg+gGYmPwDutz6AZiY/gGPnPoBmJj+AbAs/gGYmPwAAgD8AAAAAAAAAAAAAgD8AAAAA AAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAMAAAAAQAAAFQAAABw AAAAPQMAACsBAABoAwAAPAEAAAEAAACqqthBqqrYQT0DAAA6AQAABgAAAEwAAAAAAAAAAAAAAAAA AAD//////////1gAAABTAGkAZwBuAGEAbAALAAAABAAAAAkAAAAJAAAACQAAAAAAAAAlAAAADAAA AA0AAIBGAAAA2AAAAMwAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAYAAAAAADAQgAAAAAAAAAA AADAQpNqt0MUbqRDNkACgJgAAACMAAAAAAAA/wEAAAABAAAACgAAAEMAbwBtAHAAcgBlAHMAcwBl AGQAAOSePQDNTD4Asko+AM1MPoDOlD4AzUw+AOTbPgDNTD7ArAU/AM1MPgDiEz8AzUw+wJwrPwDN TD5A8kA/AM1MPsBHVj8AzUw+gAJuPwDNTD4AAIA/AAAAAAAAAAAAAIA/AAAAAAAAAAASAAAADAAA AAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAiAAAAHYBAABNAQAA zwEAAF4BAAABAAAAqqrYQaqq2EF2AQAAXAEAAAoAAABMAAAAAAAAAAAAAAAAAAAA//////////9g AAAAQwBvAG0AcAByAGUAcwBzAGUAZAAMAAAACQAAAA0AAAAJAAAABQAAAAkAAAAIAAAACAAAAAkA AAAAAAAAJQAAAAwAAAANAACARgAAAKgAAACcAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAA AAAAwEIAAAAAAAAAAAAAwEKTardDFG6kQzZAAoBoAAAAXAAAAAAAAP8BAAAAAQAAAAUAAABBAHUA ZABpAG8AAHmsPgDNzD4AZOU+AM3MPsBsCj8Azcw+gCciPwDNzD5Aois/AM3MPgAAgD8AAAAAAAAA AAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwA AAABAAAAVAAAAGwAAACPAQAAYAEAALcBAABxAQAAAQAAAKqq2EGqqthBjwEAAG8BAAAFAAAATAAA AAAAAAAAAAAAAAAAAP//////////WAAAAEEAdQBkAGkAbwAAAAsAAAAJAAAACQAAAAQAAAAAAAAA JQAAAAwAAAANAACARgAAAJQAAACIAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIA AAAAAAAAAAAAwEKTardDFG6kQzZAAoBUAAAASAAAAAAAAP8BAAAAAQAAAAMAAABCAGkAdAAAZy4+ wJkZP4AekD7AmRk/ABSjPsCZGT8AAIA/AAAAAAAAAAAAAIA/AAAAAAAAAAAAABIAAAAMAAAAAQAA ABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAMAAAAAQAAAFQAAABgAAAAfwEAAHMBAACRAQAA hAEAAAEAAACqqthBqqrYQX8BAACCAQAAAwAAAEwAAAAAAAAAAAAAAAAAAAD//////////1QAAABC AGkAdAAAAAsAAAAEAAAAAAAAACUAAAAMAAAADQAAgEYAAACAAAAAdAAAAEVNRisrQAAADAAAAAAA AAAqQAAAJAAAABgAAAAAAMBCAAAAAAAAAAAAAMBCk2q3QxRupEM2QAKAQAAAADQAAAAAAAD/AQAA AAEAAAABAAAALQCAybo+wJkZPwAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAA GAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAFQAAACSAQAAcwEAAJUBAACE AQAAAQAAAKqq2EGqqthBkgEAAIIBAAABAAAATAAAAAAAAAAAAAAAAAAAAP//////////UAAAAC0A AAAAAAAAJQAAAAwAAAANAACARgAAALAAAACkAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAA AAAAwEIAAAAAAAAAAAAAwEKTardDFG6kQzZAAoBwAAAAZAAAAAAAAP8BAAAAAQAAAAYAAABzAHQA cgBlAGEAbQAANNc+wJkZP4DvAD/AmRk/QMoMP8CZGT+A/xo/wJkZP0C6Mj/AmRk/AHVKP8CZGT8A AIA/AAAAAAAAAAAAAIA/AAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgA AAAlAAAADAAAAAEAAABUAAAAcAAAAJcBAABzAQAAxQEAAIQBAAABAAAAqqrYQaqq2EGXAQAAggEA AAYAAABMAAAAAAAAAAAAAAAAAAAA//////////9YAAAAcwB0AHIAZQBhAG0ACAAAAAQAAAAFAAAA CQAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAAKgAAACcAAAARU1GKytAAAAMAAAAAAAAACpAAAAk AAAAGAAAAAAAwEIAAAAAAAAAAAAAwEJJ9PRBe9SCQzZAAoBoAAAAXAAAAAAAAP8BAAAAAQAAAAUA AABJAG4AcAB1AHQAAPJcPgDNTD6ALoY+AM1MPgCktT4AzUw+gBnlPgDNTD6ARwo/AM1MPgAAgD8A AAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAA JQAAAAwAAAABAAAAVAAAAGwAAAAzAAAACgEAAFUAAAAbAQAAAQAAAKqq2EGqqthBMwAAABkBAAAF AAAATAAAAAAAAAAAAAAAAAAAAP//////////WAAAAEkAbgBwAHUAdAAAAAQAAAAJAAAACQAAAAkA AAAAAAAAJQAAAAwAAAANAACARgAAAKgAAACcAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAA AAAAwEIAAAAAAAAAAAAAwEJJ9PRBe9SCQzZAAoBoAAAAXAAAAAAAAP8BAAAAAQAAAAUAAABBAHUA ZABpAG8AAHxAPgDNzD4AKZk+AM3MPoCeyD4Azcw+ABT4PgDNzD7AhAU/AM3MPgAAgD8AAAAAAAAA AAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwA AAABAAAAVAAAAGwAAAAxAAAAHQEAAFkAAAAuAQAAAQAAAKqq2EGqqthBMQAAACwBAAAFAAAATAAA AAAAAAAAAAAAAAAAAP//////////WAAAAEEAdQBkAGkAbwAAAAsAAAAJAAAACQAAAAQAAAAAAAAA JQAAAAwAAAANAACARgAAALAAAACkAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIA AAAAAAAAAAAAwEJJ9PRBe9SCQzZAAoBwAAAAZAAAAAAAAP8BAAAAAQAAAAYAAABTAGkAZwBuAGEA bAAAhy0+wJkZP4Cujz7AmRk/AKSiPsCZGT+AGdI+wJkZP4DHAD/AmRk/QIIYP8CZGT8AAIA/AAAA AAAAAAAAAIA/AAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAA DAAAAAEAAABUAAAAcAAAAC8AAAAwAQAAWgAAAEEBAAABAAAAqqrYQaqq2EEvAAAAPwEAAAYAAABM AAAAAAAAAAAAAAAAAAAA//////////9YAAAAUwBpAGcAbgBhAGwACwAAAAQAAAAJAAAACQAAAAkA AAAAAAAAJQAAAAwAAAANAACARgAAAEAAAAA0AAAARU1GKytAAAAMAAAAAAAAAApAAIAkAAAAGAAA APfu6P8BAAAAwG4IQ+E6kUMAAMBCnJmZQigAAAAMAAAAAgAAACQAAAAkAAAAAACAPQAAAAAAAAAA AACAPQAAAAAAAAAAAgAAACcAAAAYAAAAAgAAAAAAAADo7vcAAAAAACUAAAAMAAAAAgAAACUAAAAM AAAACAAAgFYAAAAwAAAAiAAAACIBAADpAAAAcAEAAAUAAACHCCgShwj1FocO9RaHDigShwgoEiUA AAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAkAAAAAACAQQAAAAAAAAAAAACAQQAAAAAAAAAAAgAA AEYAAACoAAAAnAAAAEVNRisIQAACQAAAADQAAAACEMDbAAAAAM4AAAAAAAAA7FE4QAIAAAACAAAA AgAAAAIAAAAAAAAAAhDA2wAAAAAAAAD/CEABA0gAAAA8AAAAAhDA2wUAAAAAAAAAwG4IQ0iht0PA bmhDSKG3Q8BuaEPhOpFDwG4IQ+E6kUPAbghDSKG3QwABAQGBAAAAFUABABAAAAAEAAAAAAAAACQA AAAkAAAAAACAPQAAAAAAAAAAAACAPQAAAAAAAAAAAgAAAF8AAAA4AAAAAwAAADgAAAAAAAAAOAAA AAAAAAAAAAEALgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAJQAAAAwAAAADAAAAJQAAAAwAAAAFAACA VgAAADAAAACGAAAAIAEAAOsAAAByAQAABQAAAIcI9RaHDvUWhw4oEocIKBKHCPUWJQAAAAwAAAAH AACAJQAAAAwAAAAAAACAJAAAACQAAAAAAIBBAAAAAAAAAAAAAIBBAAAAAAAAAAACAAAAKAAAAAwA AAADAAAARgAAAJwAAACQAAAARU1GKypAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwELAbghD4TqR QzZAAoBoAAAAXAAAAAAAAP8BAAAAAQAAAAUAAABQAGkAdABjAGgAACWhPoAzsz4AENo+gDOzPoAF 7T6AM7M+gF0CP4Azsz4Asxc/gDOzPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAAB AAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAGwAAACnAAAANQEAAMgA AABGAQAAAQAAAKqq2EGqqthBpwAAAEQBAAAFAAAATAAAAAAAAAAAAAAAAAAAAP//////////WAAA AFAAaQB0AGMAaAAAAAsAAAAEAAAABAAAAAgAAAAAAAAAJQAAAAwAAAANAACARgAAAJQAAACIAAAA RU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwELAbghD4TqRQzZAAoBU AAAASAAAAAAAAP8BAAAAAQAAAAMAAABQAHIAZQAAlT0+AM0MP4C1lz4AzQw/ACC0PgDNDD8AAIA/ AAAAAAAAAAAAAIA/AAAAAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAA ACUAAAAMAAAAAQAAAFQAAABgAAAAmgAAAEgBAACxAAAAWQEAAAEAAACqqthBqqrYQZoAAABXAQAA AwAAAEwAAAAAAAAAAAAAAAAAAAD//////////1QAAABQAHIAZQAAAAsAAAAFAAAAAAAAACUAAAAM AAAADQAAgEYAAACAAAAAdAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAAABgAAAAAAMBCAAAAAAAA AAAAAMBCwG4IQ+E6kUM2QAKAQAAAADQAAAAAAAD/AQAAAAEAAAABAAAALQCAleM+AM0MPwAAgD8A AAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAA JQAAAAwAAAABAAAAVAAAAFQAAACzAAAASAEAALYAAABZAQAAAQAAAKqq2EGqqthBswAAAFcBAAAB AAAATAAAAAAAAAAAAAAAAAAAAP//////////UAAAAC0AAAAAAAAAJQAAAAwAAAANAACARgAAALAA AACkAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwELAbghD4TqR QzZAAoBwAAAAZAAAAAAAAP8BAAAAAQAAAAYAAABmAGkAbAB0AGUAcgAAAAA/AM0MP8DaCz8AzQw/ gFUVPwDNDD9A0B4/AM0MPwCrKj8AzQw/wGVCPwDNDD8AAIA/AAAAAAAAAAAAAIA/AAAAAAAAAAAS AAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAcAAAALgA AABIAQAA1QAAAFkBAAABAAAAqqrYQaqq2EG4AAAAVwEAAAYAAABMAAAAAAAAAAAAAAAAAAAA//// //////9YAAAAZgBpAGwAdABlAHIABAAAAAQAAAAEAAAABAAAAAkAAAAAAAAAJQAAAAwAAAANAACA RgAAAEAAAAA0AAAARU1GKytAAAAMAAAAAAAAAApAAIAkAAAAGAAAAPfu6P8BAAAAZSFIQvaoJUP9 /79CmpmZQiQAAAAkAAAAAACAPQAAAAAAAAAAAACAPQAAAAAAAAAAAgAAACUAAAAMAAAAAgAAACUA AAAMAAAACAAAgFYAAAAwAAAAMgAAAKUAAACTAAAA8wAAAAUAAAAhA1sKIQMoDyEJKA8hCVsKIQNb CiUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAkAAAAAACAQQAAAAAAAAAAAACAQQAAAAAAAAAA AgAAAEYAAACoAAAAnAAAAEVNRisIQAACQAAAADQAAAACEMDbAAAAAM4AAAAAAAAACtejPgIAAAAC AAAAAgAAAAIAAAAAAAAAAhDA2wAAAAAAAAD/CEABA0gAAAA8AAAAAhDA2wUAAAAAAAAAZSFIQsN1 ckNZCBJDw3VyQ1kIEkP2qCVDZSFIQvaoJUNlIUhCw3VyQwABAQGBAAAAFUABABAAAAAEAAAAAAAA ACQAAAAkAAAAAACAPQAAAAAAAAAAAACAPQAAAAAAAAAAAgAAAF8AAAA4AAAAAwAAADgAAAAAAAAA OAAAAAAAAAAAAAEABQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAJQAAAAwAAAADAAAAJQAAAAwAAAAF AACAVgAAADAAAAAwAAAApAAAAJQAAAD0AAAABQAAACEDKA8hCSgPIQlbCiEDWwohAygPJQAAAAwA AAAHAACAJQAAAAwAAAAAAACAJAAAACQAAAAAAIBBAAAAAAAAAAAAAIBBAAAAAAAAAAACAAAAKAAA AAwAAAADAAAARgAAAJwAAACQAAAARU1GKypAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEJlIUhC 9qglQzZAAoBoAAAAXAAAAAAAAP8BAAAAAQAAAAUAAABQAGkAdABjAGgAACWhPgAAgD4AENo+AACA PoAF7T4AAIA+gF0CPwAAgD4Asxc/AACAPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwA AAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAGwAAABQAAAArwAA AHEAAADAAAAAAQAAAKqq2EGqqthBUAAAAL4AAAAFAAAATAAAAAAAAAAAAAAAAAAAAP////////// WAAAAFAAaQB0AGMAaAAAAAsAAAAEAAAABAAAAAgAAAAAAAAAJQAAAAwAAAANAACARgAAANAAAADE AAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEJlIUhC9qglQzZA AoCQAAAAhAAAAAAAAP8BAAAAAQAAAAkAAABQAGEAcgBhAG0AZQB0AGUAcgAAPuM9gGbmPgB1Yz6A ZuY+ADChPoBm5j6Amr0+gGbmPgAQ7T6AZuY+wBIaP4Bm5j6AzTE/gGbmPkCoPT+AZuY+AGNVP4Bm 5j4AAIA/AAAAAAAAAAAAAIA/AAAAAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAM AAAAGAAAACUAAAAMAAAAAQAAAFQAAACEAAAAPQAAAMIAAACGAAAA0wAAAAEAAACqqthBqqrYQT0A AADRAAAACQAAAEwAAAAAAAAAAAAAAAAAAAD//////////2AAAABQAGEAcgBhAG0AZQB0AGUAcgAA AAsAAAAJAAAABQAAAAkAAAANAAAACQAAAAQAAAAJAAAAAAAAACUAAAAMAAAADQAAgEYAAADQAAAA xAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAAABgAAAAAAMBCAAAAAAAAAAAAAMBCZSFIQvaoJUM2 QAKAkAAAAIQAAAAAAAD/AQAAAAEAAAAJAAAARQBzAHQAaQBtAGEAdABvAHIAAJ8XPoBmJj+AuoQ+ gGYmP4Blrz6AZiY/ABvHPoBmJj+AENo+gGYmPwCTED+AZiY/wE0oP4BmJj+AKDQ/gGYmP0DjSz+A ZiY/AACAPwAAAAAAAAAAAACAPwAAAAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAA DAAAABgAAAAlAAAADAAAAAEAAABUAAAAhAAAAEAAAADVAAAAggAAAOYAAAABAAAAqqrYQaqq2EFA AAAA5AAAAAkAAABMAAAAAAAAAAAAAAAAAAAA//////////9gAAAARQBzAHQAaQBtAGEAdABvAHIA AAALAAAACAAAAAQAAAAEAAAADQAAAAkAAAAEAAAACQAAAAAAAAAlAAAADAAAAA0AAIBGAAAAmAAA AIwAAABFTUYrK0AAAAwAAAAAAAAACEAAAkAAAAA0AAAAAhDA2wAAAADOAAAAAAAAAOxROEACAAAA AgAAAAIAAAACAAAAAAAAAAIQwNsAAAAAAAAA/whAAQMsAAAAIAAAAAIQwNsCAAAAAAAAAMBuaEMU bqRDMKx9QxRupEMAAQEBFUABABAAAAAEAAAAAAAAACQAAAAkAAAAAACAPQAAAAAAAAAAAACAPQAA AAAAAAAAAgAAAF8AAAA4AAAAAwAAADgAAAAAAAAAOAAAAAAAAAAAAAEALgAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAJQAAAAwAAAADAAAAJQAAAAwAAAAFAACAVwAAACQAAADmAAAARgEAAAEBAABMAQAA AgAAAIcOjhTbD44UJQAAAAwAAAAHAACAJQAAAAwAAAAAAACAJAAAACQAAAAAAIBBAAAAAAAAAAAA AIBBAAAAAAAAAAACAAAAKAAAAAwAAAADAAAARgAAAFwAAABQAAAARU1GKwhAAQM8AAAAMAAAAAIQ wNsEAAAAAAAAANQcfENcT6FDk2qHQxRupEPUHHxDzYynQ9QcfENcT6FDAAEBgRRAAYAQAAAABAAA AAAAAP8oAAAADAAAAAIAAAAkAAAAJAAAAAAAgD0AAAAAAAAAAAAAgD0AAAAAAAAAAAIAAAAnAAAA GAAAAAIAAAAAAAAAAAAAAAAAAAAlAAAADAAAAAIAAAATAAAADAAAAAEAAAAlAAAADAAAAAgAAIBW AAAALAAAAPwAAABCAQAADwEAAFABAAAEAAAAwg8qFO4QjhTCD/IUwg8qFCUAAAAMAAAABwAAgBMA AAAMAAAAAQAAACUAAAAMAAAAAAAAgCQAAAAkAAAAAACAQQAAAAAAAAAAAACAQQAAAAAAAAAAAgAA AEYAAAA0AAAAKAAAAEVNRisKQACAJAAAABgAAAD37uj/AQAAAMBuCEMK10lCAADAQpqZmUIoAAAA DAAAAAIAAAAkAAAAJAAAAAAAgD0AAAAAAAAAAAAAgD0AAAAAAAAAAAIAAAAnAAAAGAAAAAIAAAAA AAAA6O73AAAAAAAlAAAADAAAAAIAAAAlAAAADAAAAAgAAIBWAAAAMAAAAIgAAAAyAAAA6QAAAIAA AAAFAAAAhwgoA4cI9QeHDvUHhw4oA4cIKAMlAAAADAAAAAcAAIAlAAAADAAAAAAAAIAkAAAAJAAA AAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIAAABGAAAAqAAAAJwAAABFTUYrCEAAAkAAAAA0AAAA AhDA2wAAAADOAAAAAAAAAArXoz4CAAAAAgAAAAIAAAACAAAAAAAAAAIQwNsAAAAAAAAA/whAAQNI AAAAPAAAAAIQwNsFAAAAAAAAAMBuCEMfhf5CwG5oQx+F/kLAbmhDCtdJQsBuCEMK10lCwG4IQx+F /kIAAQEBgQAAABVAAQAQAAAABAAAAAAAAAAkAAAAJAAAAAAAgD0AAAAAAAAAAAAAgD0AAAAAAAAA AAIAAABfAAAAOAAAAAMAAAA4AAAAAAAAADgAAAAAAAAAAAABAAUAAAAAAAAAAAAAAAAAAAAAAAAA AAAAACUAAAAMAAAAAwAAACUAAAAMAAAABQAAgFYAAAAwAAAAhwAAADEAAADqAAAAgQAAAAUAAACH CPUHhw71B4cOKAOHCCgDhwj1ByUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAkAAAAAACAQQAA AAAAAAAAAACAQQAAAAAAAAAAAgAAACgAAAAMAAAAAwAAAEYAAACcAAAAkAAAAEVNRisqQAAAJAAA ABgAAAAAAMBCAAAAAAAAAAAAAMBCwG4IQwrXSUI2QAKAaAAAAFwAAAAAAAD/AQAAAAEAAAAFAAAA UABpAHQAYwBoAAAloT4AAIA+ABDaPgAAgD6ABe0+AACAPoBdAj8AAIA+ALMXPwAAgD4AAIA/AAAA AAAAAAAAAIA/AAAAAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUA AAAMAAAAAQAAAFQAAABsAAAApwAAADsAAADIAAAATAAAAAEAAACqqthBqqrYQacAAABKAAAABQAA AEwAAAAAAAAAAAAAAAAAAAD//////////1gAAABQAGkAdABjAGgAAAALAAAABAAAAAQAAAAIAAAA AAAAACUAAAAMAAAADQAAgEYAAADQAAAAxAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAAABgAAAAA AMBCAAAAAAAAAAAAAMBCwG4IQwrXSUI2QAKAkAAAAIQAAAAAAAD/AQAAAAEAAAAJAAAAUABhAHIA YQBtAGUAdABlAHIAAD7jPYBm5j4AdWM+gGbmPgAwoT6AZuY+gJq9PoBm5j4AEO0+gGbmPsASGj+A ZuY+gM0xP4Bm5j5AqD0/gGbmPgBjVT+AZuY+AACAPwAAAAAAAAAAAACAPwAAAAAAAAAAAAASAAAA DAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAhAAAAJMAAABP AAAA3AAAAGAAAAABAAAAqqrYQaqq2EGTAAAAXgAAAAkAAABMAAAAAAAAAAAAAAAAAAAA//////// //9gAAAAUABhAHIAYQBtAGUAdABlAHIAAAALAAAACQAAAAUAAAAJAAAADQAAAAkAAAAEAAAACQAA AAAAAAAlAAAADAAAAA0AAIBGAAAA0AAAAMQAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAYAAAA AADAQgAAAAAAAAAAAADAQsBuCEMK10lCNkACgJAAAACEAAAAAAAA/wEAAAABAAAACQAAAFEAdQBh AG4AdABpAHoAZQByAAAKDj6AZiY/AGWJPoBmJj+A2rg+gGYmPwBQ6D6AZiY/wOILP4BmJj+AvRc/ gGYmP0A4IT+AZiY/wI02P4BmJj+ASE4/gGYmPwAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAA AAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAIQAAACWAAAA YgAAANoAAABzAAAAAQAAAKqq2EGqqthBlgAAAHEAAAAJAAAATAAAAAAAAAAAAAAAAAAAAP////// ////YAAAAFEAdQBhAG4AdABpAHoAZQByAAAADAAAAAkAAAAJAAAACQAAAAQAAAAEAAAACAAAAAkA AAAAAAAAJQAAAAwAAAANAACARgAAAKAAAACUAAAARU1GKytAAAAMAAAAAAAAAAhAAAJAAAAANAAA AAIQwNsAAAAAzgAAAAAAAAAK16M+AgAAAAIAAAACAAAAAgAAAAAAAAACEMDbAAAAAL93Rv8IQAED NAAAACgAAAACEMDbAwAAAAAAAACzEMRC9qglQ7MQxEJSuLFCD6D8QlK4sUIAAQEBFUABABAAAAAE AAAAAAAAACQAAAAkAAAAAACAPQAAAAAAAAAAAACAPQAAAAAAAAAAAgAAAF8AAAA4AAAAAwAAADgA AAAAAAAAOAAAAAAAAAAAAAEABQAAAAAAAABGd78AAAAAAAAAAAAAAAAAJQAAAAwAAAADAAAAJQAA AAwAAAAFAACAVwAAACgAAABgAAAAVwAAAIAAAACnAAAAAwAAACEGWwohBo4F5QeOBSUAAAAMAAAA BwAAgCUAAAAMAAAAAAAAgCQAAAAkAAAAAACAQQAAAAAAAAAAAACAQQAAAAAAAAAAAgAAACgAAAAM AAAAAwAAAEYAAABcAAAAUAAAAEVNRisIQAEDPAAAADAAAAACEMDbBAAAAAAAAAAFyfpCKVyqQsBu CENSuLFCBcn6QnsUuUIFyfpCKVyqQgABAYEUQAGAEAAAAAQAAAC/d0b/KAAAAAwAAAACAAAAJAAA ACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAJwAAABgAAAACAAAAAAAAAEZ3vwAAAAAA JQAAAAwAAAACAAAAEwAAAAwAAAABAAAAJQAAAAwAAAAIAACAVgAAACwAAAB9AAAAVQAAAIkAAABd AAAABAAAANcHUwWHCI4F1wfJBdcHUwUlAAAADAAAAAcAAIATAAAADAAAAAEAAAAlAAAADAAAAAAA AIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIAAABGAAAANAAAACgAAABFTUYrCkAA gCQAAAAYAAAA9+7o/wEAAADGnbJD9qglQwAAwELMzKxCKAAAAAwAAAACAAAAJAAAACQAAAAAAIA9 AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAJwAAABgAAAACAAAAAAAAAOju9wAAAAAAJQAAAAwAAAAC AAAAJQAAAAwAAAAIAACAVgAAADAAAABlAQAApQAAAMYBAAD9AAAABQAAAFQWWwpUFsEPVBzBD1Qc WwpUFlsKJQAAAAwAAAAHAACAJQAAAAwAAAAAAACAJAAAACQAAAAAAIBBAAAAAAAAAAAAAIBBAAAA AAAAAAACAAAARgAAAKgAAACcAAAARU1GKwhAAAJAAAAANAAAAAIQwNsAAAAAzgAAAAAAAAAK16M+ AgAAAAIAAAACAAAAAgAAAAAAAAACEMDbAAAAAAAAAP8IQAEDSAAAADwAAAACEMDbBQAAAAAAAADG nbJDXA98Q8ad4kNcD3xDxp3iQ/aoJUPGnbJD9qglQ8adskNcD3xDAAEBAYEAAAAVQAEAEAAAAAQA AAAAAAAAJAAAACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAXwAAADgAAAADAAAAOAAA AAAAAAA4AAAAAAAAAAAAAQAFAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAlAAAADAAAAAMAAAAlAAAA DAAAAAUAAIBWAAAAMAAAAGQBAACkAAAAxwEAAP4AAAAFAAAAVBbBD1QcwQ9UHFsKVBZbClQWwQ8l AAAADAAAAAcAAIAlAAAADAAAAAAAAIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIA AAAoAAAADAAAAAMAAABGAAAAiAAAAHwAAABFTUYrKkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADA QsadskP2qCVDNkACgFQAAABIAAAAAAAA/wEAAAABAAAAAwAAAEIAaQB0AABo9j0AzUw+AAptPgDN TD6Aeok+AM1MPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAA AAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAGAAAABxAQAAqgAAAIMBAAC7AAAAAQAAAKqq 2EGqqthBcQEAALkAAAADAAAATAAAAAAAAAAAAAAAAAAAAP//////////VAAAAEIAaQB0AAAACwAA AAQAAAAAAAAAJQAAAAwAAAANAACARgAAAIAAAAB0AAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAA GAAAAAAAwEIAAAAAAAAAAAAAwELGnbJD9qglQzZAAoBAAAAANAAAAAAAAP8BAAAAAQAAAAEAAAAt AAAwoT4AzUw+AACAPwAAAAAAAAAAAACAPwAAAAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAA AAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAVAAAAIMBAACqAAAAhgEAALsAAAABAAAAqqrY Qaqq2EGDAQAAuQAAAAEAAABMAAAAAAAAAAAAAAAAAAAA//////////9QAAAALQAAAAAAAAAlAAAA DAAAAA0AAIBGAAAAsAAAAKQAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAYAAAAAADAQgAAAAAA AAAAAADAQsadskP2qCVDNkACgHAAAABkAAAAAAAA/wEAAAABAAAABgAAAHMAdAByAGUAYQBtAICa vT4AzUw+gEXoPgDNTD4A+/8+AM1MPsAyDj8AzUw+gO0lPwDNTD5AqD0/AM1MPgAAgD8AAAAAAAAA AAAAgD8AAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAMAAAA AQAAAFQAAABwAAAAiQEAAKoAAAC3AQAAuwAAAAEAAACqqthBqqrYQYkBAAC5AAAABgAAAEwAAAAA AAAAAAAAAAAAAAD//////////1gAAABzAHQAcgBlAGEAbQAIAAAABAAAAAUAAAAJAAAACQAAAAAA AAAlAAAADAAAAA0AAIBGAAAA7AAAAOAAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAYAAAAAADA QgAAAAAAAAAAAADAQsadskP2qCVDNkACgKwAAACgAAAAAAAA/wEAAAABAAAADAAAAE0AdQBsAHQA aQBwAGwAZQB4AGUAcgAgAADSxj0Azcw+AJRxPgDNzD6AP6g+AM3MPgA1uz4Azcw+gOrSPgDNzD4A 4OU+AM3MPsCqCj8Azcw+gCUUPwDNzD5A4Cs/AM3MPsA1QT8Azcw+gPBYPwDNzD7AJWc/AM3MPgAA gD8AAAAAAAAAAAAAgD8AAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAA ACUAAAAMAAAAAQAAAFQAAACUAAAAbwEAAL0AAAC9AQAAzgAAAAEAAACqqthBqqrYQW8BAADMAAAA DAAAAEwAAAAAAAAAAAAAAAAAAAD//////////2QAAABNAHUAbAB0AGkAcABsAGUAeABlAHIAIAAN AAAACQAAAAQAAAAEAAAABAAAAAkAAAAEAAAACQAAAAgAAAAJAAAABQAAAAAAAAAlAAAADAAAAA0A AIBGAAAA0AAAAMQAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADA QsadskP2qCVDNkACgJAAAACEAAAAAAAA/wEAAAABAAAACQAAAG8AcgAgAEoAbwBpAG4AdAAgAAAq aD7AmRk/gIqjPsCZGT8A9b8+wJkZP4Cq1z7AmRk/wCoBP8CZGT+A5Rg/wJkZP0BgIj/AmRk/ABs6 P8CZGT/A9UU/wJkZPwAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwA AAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAIQAAAB7AQAA0AAAAK8BAADhAAAAAQAA AKqq2EGqqthBewEAAN8AAAAJAAAATAAAAAAAAAAAAAAAAAAAAP//////////YAAAAG8AcgAgAEoA bwBpAG4AdAAgAAAACQAAAAUAAAAEAAAACAAAAAkAAAAEAAAACQAAAAQAAAAAAAAAJQAAAAwAAAAN AACARgAAAKgAAACcAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAA wELGnbJD9qglQzZAAoBoAAAAXAAAAAAAAP8BAAAAAQAAAAUAAABDAG8AZABlAHIAgMqLPgDNTD+A ask+AM1MPwDg+D4AzUw/wCoUPwDNTD+A5Ss/AM1MPwAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAA EgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAGwAAAB/ AQAA4wAAAKoBAAD0AAAAAQAAAKqq2EGqqthBfwEAAPIAAAAFAAAATAAAAAAAAAAAAAAAAAAAAP// ////////WAAAAEMAbwBkAGUAcgAAAAwAAAAJAAAACQAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAA AJgAAACMAAAARU1GKytAAAAMAAAAAAAAAAhAAAJAAAAANAAAAAIQwNsAAAAAzgAAAAAAAAAK16M+ AgAAAAIAAAACAAAAAgAAAAAAAAACEMDbAAAAAL93Rv8IQAEDLAAAACAAAAACEMDbAgAAAAAAAADA bjhDhSuMQ8BuOEMfhf5CAAEBARVAAQAQAAAABAAAAAAAAAAkAAAAJAAAAAAAgD0AAAAAAAAAAAAA gD0AAAAAAAAAAAIAAABfAAAAOAAAAAMAAAA4AAAAAAAAADgAAAAAAAAAAAABAAUAAAAAAAAARne/ AAAAAAAAAAAAAAAAACUAAAAMAAAAAwAAACUAAAAMAAAABQAAgFcAAAAkAAAAtwAAAH4AAAC6AAAA GgEAAAIAAACHC4YRhwv1ByUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAkAAAAAACAQQAAAAAA AAAAAACAQQAAAAAAAAAAAgAAACgAAAAMAAAAAwAAAEYAAABcAAAAUAAAAEVNRisIQAEDPAAAADAA AAACEMDbBAAAAAAAAADUHDxDw7WLQ8BuOEPhOpFDq8A0Q8O1i0PUHDxDw7WLQwABAYEUQAGAEAAA AAQAAAC/d0b/KAAAAAwAAAACAAAAJAAAACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAA JwAAABgAAAACAAAAAAAAAEZ3vwAAAAAAJQAAAAwAAAACAAAAEwAAAAwAAAABAAAAJQAAAAwAAAAI AACAVgAAACwAAAC0AAAAFwEAAL0AAAAjAQAABAAAAMILdxGHCygSTQt3EcILdxElAAAADAAAAAcA AIATAAAADAAAAAEAAAAlAAAADAAAAAAAAIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAA AAIAAABGAAAAlAAAAIgAAABFTUYrCEAAAkAAAAA0AAAAAhDA2wAAAADOAAAAAAAAAArXoz4CAAAA AgAAAAIAAAACAAAAAAAAAAIQwNsAAAAAv3dG/whAAQM0AAAAKAAAAAIQwNsDAAAAAAAAAMBuaENS uLFCxp3KQ1K4sULGncpDPYobQwABAYEVQAEAEAAAAAQAAAAAAAAAJAAAACQAAAAAAIA9AAAAAAAA AAAAAIA9AAAAAAAAAAACAAAAXwAAADgAAAADAAAAOAAAAAAAAAA4AAAAAAAAAAAAAQAFAAAAAAAA AEZ3vwAAAAAAAAAAAAAAAAAlAAAADAAAAAMAAAAlAAAADAAAAAUAAIBXAAAAKAAAAOcAAABXAAAA lwEAAJ0AAAADAAAAhw6OBVQZjgVUGbkJJQAAAAwAAAAHAACAJQAAAAwAAAAAAACAJAAAACQAAAAA AIBBAAAAAAAAAAAAAIBBAAAAAAAAAAACAAAAKAAAAAwAAAADAAAARgAAAFwAAABQAAAARU1GKwhA AQM8AAAAMAAAAAIQwNsEAAAAAAAAANB0zEO4nhpDxp3KQ/aoJUO8xshDuJ4aQ9B0zEO4nhpDAAEB gRRAAYAQAAAABAAAAL93Rv8kAAAAJAAAAAAAgD0AAAAAAAAAAAAAgD0AAAAAAAAAAAIAAAAlAAAA DAAAAAIAAAATAAAADAAAAAEAAAAlAAAADAAAAAgAAIBWAAAALAAAAJEBAACaAAAAmQEAAKYAAAAE AAAAjxmqCVQZWwoZGaoJjxmqCSUAAAAMAAAABwAAgBMAAAAMAAAAAQAAACUAAAAMAAAAAAAAgCQA AAAkAAAAAACAQQAAAAAAAAAAAACAQQAAAAAAAAAAAgAAAEYAAACUAAAAiAAAAEVNRisIQAACQAAA ADQAAAACEMDbAAAAAM4AAAAAAAAA7FE4QAIAAAACAAAAAgAAAAIAAAAAAAAAAhDA2wAAAAAAAAD/ CEABAzQAAAAoAAAAAhDA2wMAAAAAAAAAk2q3QxRupEPGncpDFG6kQ8adykMpnIZDAAEBgRVAAQAQ AAAABAAAAAAAAAAkAAAAJAAAAAAAgD0AAAAAAAAAAAAAgD0AAAAAAAAAAAIAAABfAAAAOAAAAAMA AAA4AAAAAAAAADgAAAAAAAAAAAABAC4AAAAAAAAAAAAAAAAAAAAAAAAAAAAAACUAAAAMAAAAAwAA ACUAAAAMAAAABQAAgFcAAAAoAAAAbAEAAAoBAACYAQAATAEAAAMAAADuFo4UVBmOFFQZ1BAlAAAA DAAAAAcAAIAlAAAADAAAAAAAAIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIAAAAo AAAADAAAAAMAAABGAAAAXAAAAFAAAABFTUYrCEABAzwAAAAwAAAAAhDA2wQAAAAAAAAADn/HQ9dj h0PGncpDXA98Q3+8zUPXY4dDDn/HQ9djh0MAAQGBFEABgBAAAAAEAAAAAAAA/ygAAAAMAAAAAgAA ACQAAAAkAAAAAACAPQAAAAAAAAAAAACAPQAAAAAAAAAAAgAAACcAAAAYAAAAAgAAAAAAAAAAAAAA AAAAACUAAAAMAAAAAgAAABMAAAAMAAAAAQAAACUAAAAMAAAACAAAgFYAAAAsAAAAjwEAAPwAAACc AQAADwEAAAQAAADwGO0QVBnBD7gZ7RDwGO0QJQAAAAwAAAAHAACAEwAAAAwAAAABAAAAJQAAAAwA AAAAAACAJAAAACQAAAAAAIBBAAAAAAAAAAAAAIBBAAAAAAAAAAACAAAARgAAAJwAAACQAAAARU1G KypAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEKL4CBCpHAjQjZAAoBoAAAAXAAAAAAAAP8BAAAA AQAAAAUAAABQAGkAdABjAGgAgLahPgDNTD6Aodo+AM1MPgCX7T4AzUw+QKYCPwDNTD7A+xc/AM1M PgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwA AAAYAAAAJQAAAAwAAAABAAAAVAAAAGwAAABHAAAALQAAAGgAAAA+AAAAAQAAAKqq2EGqqthBRwAA ADwAAAAFAAAATAAAAAAAAAAAAAAAAAAAAP//////////WAAAAFAAaQB0AGMAaAAAAAsAAAAEAAAA BAAAAAgAAAAAAAAAJQAAAAwAAAANAACARgAAANgAAADMAAAARU1GKytAAAAMAAAAAAAAACpAAAAk AAAAGAAAAAAAwEIAAAAAAAAAAAAAwEKL4CBCpHAjQjZAAoCYAAAAjAAAAAAAAP8BAAAAAQAAAAoA AABQAGEAcgBhAG0AZQB0AGUAcgBzAAAukD0Azcw+AO05PgDNzD4AbIw+AM3MPoDWqD4Azcw+AEzY PgDNzD7AsA8/AM3MPoBrJz8Azcw+QEYzPwDNzD4AAUs/AM3MPkA2WT8Azcw+AACAPwAAAAAAAAAA AACAPwAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAAB AAAAVAAAAIgAAAAvAAAAQAAAAH8AAABRAAAAAQAAAKqq2EGqqthBLwAAAE8AAAAKAAAATAAAAAAA AAAAAAAAAAAAAP//////////YAAAAFAAYQByAGEAbQBlAHQAZQByAHMACwAAAAkAAAAFAAAACQAA AA0AAAAJAAAABAAAAAkAAAAFAAAAAAAAACUAAAAMAAAADQAAgEYAAADQAAAAxAAAAEVNRisrQAAA DAAAAAAAAAAqQAAAJAAAABgAAAAAAMBCAAAAAAAAAAAAAMBCwG44Q1wPHEM2QAKAkAAAAIQAAAAA AAD/AQAAAAEAAAAJAAAAUQB1AGEAbgB0AGkAegBlAGQAAET4PQCaGT4AcYA+AJoZPoDmrz4Amhk+ AFzfPgCaGT7AaAc/AJoZPoBDEz8Amhk+QL4cPwCaGT7AEzI/AJoZPoDOST8Amhk+AACAPwAAAAAA AAAAAACAPwAAAAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAA DAAAAAEAAABUAAAAhAAAAMQAAACbAAAACwEAAKwAAAABAAAAqqrYQaqq2EHEAAAAqgAAAAkAAABM AAAAAAAAAAAAAAAAAAAA//////////9gAAAAUQB1AGEAbgB0AGkAegBlAGQAAAAMAAAACQAAAAkA AAAJAAAABAAAAAQAAAAIAAAACQAAAAAAAAAlAAAADAAAAA0AAIBGAAAAqAAAAJwAAABFTUYrK0AA AAwAAAAAAAAAKkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQsBuOENcDxxDNkACgGgAAABcAAAA AAAA/wEAAAABAAAABQAAAFAAaQB0AGMAaACAtqE+gDOzPoCh2j6AM7M+AJftPoAzsz5ApgI/gDOz PsD7Fz+AM7M+AACAPwAAAAAAAAAAAACAPwAAAAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAA AAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAbAAAANcAAACvAAAA+AAAAMAAAAABAAAAqqrY Qaqq2EHXAAAAvgAAAAUAAABMAAAAAAAAAAAAAAAAAAAA//////////9YAAAAUABpAHQAYwBoAAAA CwAAAAQAAAAEAAAACAAAAAAAAAAlAAAADAAAAA0AAIBGAAAA2AAAAMwAAABFTUYrK0AAAAwAAAAA AAAAKkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQsBuOENcDxxDNkACgJgAAACMAAAAAAAA/wEA AAABAAAACgAAAFAAYQByAGEAbQBlAHQAZQByAHMAAC6QPQDNDD8A7Tk+AM0MPwBsjD4AzQw/gNao PgDNDD8ATNg+AM0MP8CwDz8AzQw/gGsnPwDNDD9ARjM/AM0MPwABSz8AzQw/QDZZPwDNDD8AAIA/ AAAAAAAAAAAAAIA/AAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAl AAAADAAAAAEAAABUAAAAiAAAAL8AAADCAAAADwEAANMAAAABAAAAqqrYQaqq2EG/AAAA0QAAAAoA AABMAAAAAAAAAAAAAAAAAAAA//////////9gAAAAUABhAHIAYQBtAGUAdABlAHIAcwALAAAACQAA AAUAAAAJAAAADQAAAAkAAAAEAAAACQAAAAUAAAAAAAAAJQAAAAwAAAANAACARgAAANgAAADMAAAA RU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwELGnYJDpHAdQDZAAoCY AAAAjAAAAAAAAP8BAAAAAQAAAAoAAABDAG8AbQBwAHIAZQBzAHMAZQBkAAClAj4AzUw+AOV9PgDN TD4AaK4+AM1MPoB99T4AzUw+gHkSPwDNTD7AriA/AM1MPoBpOD8AzUw+AL9NPwDNTD6AFGM/AM1M PkDPej8AzUw+AACAPwAAAAAAAAAAAACAPwAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAA FgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAIgAAAARAQAABwAAAGoBAAAYAAAAAQAAAKqq2EGq qthBEQEAABYAAAAKAAAATAAAAAAAAAAAAAAAAAAAAP//////////YAAAAEMAbwBtAHAAcgBlAHMA cwBlAGQADAAAAAkAAAANAAAACQAAAAUAAAAJAAAACAAAAAgAAAAJAAAAAAAAACUAAAAMAAAADQAA gEYAAACwAAAApAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAAABgAAAAAAMBCAAAAAAAAAAAAAMBC xp2CQ6RwHUA2QAKAcAAAAGQAAAAAAAD/AQAAAAEAAAAGAAAAUABpAHQAYwBoACAAAFjUPgDNzD6A oQY/AM3MPkAcED8Azcw+APcbPwDNzD6ATDE/AM3MPkAHST8Azcw+AACAPwAAAAAAAAAAAACAPwAA AAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAA AHAAAAAtAQAAGgAAAFEBAAArAAAAAQAAAKqq2EGqqthBLQEAACkAAAAGAAAATAAAAAAAAAAAAAAA AAAAAP//////////WAAAAFAAaQB0AGMAaAAgAAsAAAAEAAAABAAAAAgAAAAJAAAAAAAAACUAAAAM AAAADQAAgEYAAADYAAAAzAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAAABgAAAAAAMBCAAAAAAAA AAAAAMBCxp2CQ6RwHUA2QAKAmAAAAIwAAAAAAAD/AQAAAAEAAAAKAAAAUABhAHIAYQBtAGUAdABl AHIAcwAAWy0+wJkZP4CYjz7AmRk/AA6/PsCZGT+AeNs+wJkZPwB3BT/AmRk/wAEpP8CZGT+AvEA/ wJkZP0CXTD/AmRk/AFJkP8CZGT9Ah3I/wJkZPwAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAABIAAAAM AAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAMAAAAAQAAAFQAAACIAAAAFQEAAC0A AABlAQAAPgAAAAEAAACqqthBqqrYQRUBAAA8AAAACgAAAEwAAAAAAAAAAAAAAAAAAAD///////// /2AAAABQAGEAcgBhAG0AZQB0AGUAcgBzAAsAAAAJAAAABQAAAAkAAAANAAAACQAAAAQAAAAJAAAA BQAAAAAAAAAlAAAADAAAAA0AAIBGAAAAlAAAAIgAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAY AAAAAADAQgAAAAAAAAAAAADAQsadgkOkcB1ANkACgFQAAABIAAAAAAAA/wEAAAABAAAAAwAAAEIA aQB0AACbYT4AzUw/gLipPgDNTD8Arrw+AM1MPwAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAA AAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAGAAAAAaAQAA QAAAACwBAABRAAAAAQAAAKqq2EGqqthBGgEAAE8AAAADAAAATAAAAAAAAAAAAAAAAAAAAP////// ////VAAAAEIAaQB0AAAACwAAAAQAAAAAAAAAJQAAAAwAAAANAACARgAAAIAAAAB0AAAARU1GKytA AAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwELGnYJDpHAdQDZAAoBAAAAANAAA AAAAAP8BAAAAAQAAAAEAAAAtAIBj1D4AzUw/AACAPwAAAAAAAAAAAACAPwAAAAAAAAAAAAASAAAA DAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAVAAAAC0BAABA AAAAMAEAAFEAAAABAAAAqqrYQaqq2EEtAQAATwAAAAEAAABMAAAAAAAAAAAAAAAAAAAA//////// //9QAAAALQAAAAAAAAAlAAAADAAAAA0AAIBGAAAAsAAAAKQAAABFTUYrK0AAAAwAAAAAAAAAKkAA ACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQsadgkOkcB1ANkACgHAAAABkAAAAAAAA/wEAAAABAAAA BgAAAHMAdAByAGUAYQBtAADO8D4AzUw/gLwNPwDNTD9Alxk/AM1MP4DMJz8AzUw/QIc/PwDNTD8A Qlc/AM1MPwAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYA AAAMAAAAGAAAACUAAAAMAAAAAQAAAFQAAABwAAAAMgEAAEAAAABgAQAAUQAAAAEAAACqqthBqqrY QTIBAABPAAAABgAAAEwAAAAAAAAAAAAAAAAAAAD//////////1gAAABzAHQAcgBlAGEAbQAIAAAA BAAAAAUAAAAJAAAACQAAAAAAAAAlAAAADAAAAA0AAIBGAAAAQAAAADQAAABFTUYrK0AAAAwAAAAA AAAACkAAgCQAAAAYAAAA9+7o/wEAAAD50PVD9qglQ2xm5kLMzKxCKAAAAAwAAAACAAAAJAAAACQA AAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAJwAAABgAAAACAAAAAAAAAOju9wAAAAAAJQAA AAwAAAACAAAAJQAAAAwAAAAIAACAVgAAADAAAADrAQAApQAAAF8CAAD9AAAABQAAALseWwq7HsEP 7iXBD+4lWwq7HlsKJQAAAAwAAAAHAACAJQAAAAwAAAAAAACAJAAAACQAAAAAAIBBAAAAAAAAAAAA AIBBAAAAAAAAAAACAAAARgAAAKgAAACcAAAARU1GKwhAAAJAAAAANAAAAAIQwNsAAAAAzgAAAAAA AAAK16M+AgAAAAIAAAACAAAAAgAAAAAAAAACEMDbAAAAAAAAAP8IQAEDSAAAADwAAAACEMDbBQAA AAAAAAD50PVDXA98Q0q1F0RcD3xDSrUXRPaoJUP50PVD9qglQ/nQ9UNcD3xDAAEBAYEAAAAVQAEA EAAAAAQAAAAAAAAAJAAAACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAXwAAADgAAAAD AAAAOAAAAAAAAAA4AAAAAAAAAAAAAQAFAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAlAAAADAAAAAMA AAAlAAAADAAAAAUAAIBWAAAAMAAAAOoBAACkAAAAYAIAAP4AAAAFAAAAux7BD+4lwQ/uJVsKux5b CrsewQ8lAAAADAAAAAcAAIAlAAAADAAAAAAAAIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAA AAAAAAIAAAAoAAAADAAAAAMAAABGAAAAiAAAAHwAAABFTUYrKkAAACQAAAAYAAAAAADAQgAAAAAA AAAAAADAQvnQ9UP2qCVDNkACgFQAAABIAAAAAAAA/wEAAAABAAAAAwAAAEIAaQB0AACbYT4AzUw+ gLipPgDNTD4Arrw+AM1MPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAA AAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAGAAAAABAgAAqgAAABMCAAC7AAAA AQAAAKqq2EGqqthBAQIAALkAAAADAAAATAAAAAAAAAAAAAAAAAAAAP//////////VAAAAEIAaQB0 AAAACwAAAAQAAAAAAAAAJQAAAAwAAAANAACARgAAAIAAAAB0AAAARU1GKytAAAAMAAAAAAAAACpA AAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEL50PVD9qglQzZAAoBAAAAANAAAAAAAAP8BAAAAAQAA AAEAAAAtAIBj1D4AzUw+AACAPwAAAAAAAAAAAACAPwAAAAAAAAAAAAASAAAADAAAAAEAAAAYAAAA DAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAVAAAABMCAACqAAAAFgIAALsAAAAB AAAAqqrYQaqq2EETAgAAuQAAAAEAAABMAAAAAAAAAAAAAAAAAAAA//////////9QAAAALQAAAAAA AAAlAAAADAAAAA0AAIBGAAAAsAAAAKQAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAYAAAAAADA QgAAAAAAAAAAAADAQvnQ9UP2qCVDNkACgHAAAABkAAAAAAAA/wEAAAABAAAABgAAAHMAdAByAGUA YQBtAADO8D4AzUw+gLwNPwDNTD5Alxk/AM1MPoDMJz8AzUw+QIc/PwDNTD4AQlc/AM1MPgAAgD8A AAAAAAAAAAAAgD8AAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUA AAAMAAAAAQAAAFQAAABwAAAAGQIAAKoAAABHAgAAuwAAAAEAAACqqthBqqrYQRkCAAC5AAAABgAA AEwAAAAAAAAAAAAAAAAAAAD//////////1gAAABzAHQAcgBlAGEAbQAIAAAABAAAAAUAAAAJAAAA CQAAAAAAAAAlAAAADAAAAA0AAIBGAAAAiAAAAHwAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAY AAAAAADAQgAAAAAAAAAAAADAQvnQ9UP2qCVDNkACgEgAAAA8AAAAAAAA/wEAAAABAAAAAgAAAEQA ZQAAoIA9AM3MPgCQOz4Azcw+AACAPwAAAAAAAAAAAACAPwAAAAAAAAAAEgAAAAwAAAABAAAAGAAA AAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAFgAAADyAQAAvQAAAAUCAADOAAAA AQAAAKqq2EGqqthB8gEAAMwAAAACAAAATAAAAAAAAAAAAAAAAAAAAP//////////UAAAAEQAZQAM AAAAAAAAACUAAAAMAAAADQAAgEYAAACAAAAAdAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAAABgA AAAAAMBCAAAAAAAAAAAAAMBC+dD1Q/aoJUM2QAKAQAAAADQAAAAAAAD/AQAAAAEAAAABAAAALQCA PY0+AM3MPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAA FgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAFQAAAAGAgAAvQAAAAkCAADOAAAAAQAAAKqq2EGq qthBBgIAAMwAAAABAAAATAAAAAAAAAAAAAAAAAAAAP//////////UAAAAC0AAAAAAAAAJQAAAAwA AAANAACARgAAAOQAAADYAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAA AAAAwEL50PVD9qglQzZAAoCkAAAAmAAAAAAAAP8BAAAAAQAAAAsAAABtAHUAbAB0AGkAcABsAGUA eABlAHIAAKipPgDNzD6AvfA+AM3MPoAZED8Azcw+QJQZPwDNzD4AbyU/AM3MPsDpLj8Azcw+gKRG PwDNzD5AH1A/AM3MPgDaZz8Azcw+gC99PwDNzD4gdYo/AM3MPgAAgD8AAAAAAAAAAAAAgD8AAAAA AAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAA AJAAAAALAgAAvQAAAFgCAADOAAAAAQAAAKqq2EGqqthBCwIAAMwAAAALAAAATAAAAAAAAAAAAAAA AAAAAP//////////ZAAAAG0AdQBsAHQAaQBwAGwAZQB4AGUAcgAAAA0AAAAJAAAABAAAAAQAAAAE AAAACQAAAAQAAAAJAAAACAAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAANAAAADEAAAARU1GKytA AAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEL50PVD9qglQzZAAoCQAAAAhAAA AAAAAP8BAAAAAQAAAAkAAABvAHIAIABKAG8AaQBuAHQAIAAASKc+wJkZP4C91j7AmRk/ACjzPsCZ GT/AbgU/wJkZP0DEGj/AmRk/AH8yP8CZGT/A+Ts/wJkZP4C0Uz/AmRk/QI9fP8CZGT8AAIA/AAAA AAAAAAAAAIA/AAAAAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUA AAAMAAAAAQAAAFQAAACEAAAACwIAANAAAAA/AgAA4QAAAAEAAACqqthBqqrYQQsCAADfAAAACQAA AEwAAAAAAAAAAAAAAAAAAAD//////////2AAAABvAHIAIABKAG8AaQBuAHQAIAAAAAkAAAAFAAAA BAAAAAgAAAAJAAAABAAAAAkAAAAEAAAAAAAAACUAAAAMAAAADQAAgEYAAAC8AAAAsAAAAEVNRisr QAAADAAAAAAAAAAqQAAAJAAAABgAAAAAAMBCAAAAAAAAAAAAAMBC+dD1Q/aoJUM2QAKAfAAAAHAA AAAAAAD/AQAAAAEAAAAHAAAARABlAGMAbwBkAGUAcgCA7ZE+AM1MP4CNzz4AzUw/AAP/PgDNTD8A 1xQ/AM1MP8CRLD8AzUw/gExEPwDNTD9AB1w/AM1MPwAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAA EgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAHgAAAAH AgAA4wAAAEMCAAD0AAAAAQAAAKqq2EGqqthBBwIAAPIAAAAHAAAATAAAAAAAAAAAAAAAAAAAAP// ////////XAAAAEQAZQBjAG8AZABlAHIAAAAMAAAACQAAAAgAAAAJAAAACQAAAAkAAAAAAAAAJQAA AAwAAAANAACARgAAALQAAACoAAAARU1GKytAAAAMAAAAAAAAAAhAAAJAAAAANAAAAAIQwNsAAAAA zgAAAAAAAADsUThAAgAAAAIAAAACAAAAAgAAAAAAAAACEMDbAAAAAAAAAP8IQAEDSAAAADwAAAAC EMDbBQAAAAAAAAB96BJESKG3Q33oKkRIobdDfegqROE6kUN96BJE4TqRQ33oEkRIobdDAAEBAYEA AAAVQAEAEAAAAAQAAAAAAAAAJAAAACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAXwAA ADgAAAADAAAAOAAAAAAAAAA4AAAAAAAAAAAAAQAuAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAlAAAA DAAAAAMAAAAlAAAADAAAAAUAAIBWAAAAMAAAAEkCAAAgAQAArwIAAHIBAAAFAAAAuyT1Frsq9Ra7 KigSuyQoErsk9RYlAAAADAAAAAcAAIAlAAAADAAAAAAAAIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAA gEEAAAAAAAAAAAIAAAAoAAAADAAAAAMAAABGAAAAnAAAAJAAAABFTUYrKkAAACQAAAAYAAAAAADA QgAAAAAAAAAAAADAQn3oEkThOpFDNkACgGgAAABcAAAAAAAA/wEAAAABAAAABQAAAEMARQBMAFQA IAAA8JI+AACAPgCQ0D4AAIA+gL0EPwAAgD5AeBw/AACAPkCINj8AAIA+AACAPwAAAAAAAAAAAACA PwAAAAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEA AABUAAAAbAAAAGcCAAArAQAAkQIAADwBAAABAAAAqqrYQaqq2EFnAgAAOgEAAAUAAABMAAAAAAAA AAAAAAAAAAAA//////////9YAAAAQwBFAEwAVAAgAAAADAAAAAsAAAAJAAAACgAAAAAAAAAlAAAA DAAAAA0AAIBGAAAAqAAAAJwAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAYAAAAAADAQgAAAAAA AAAAAADAQn3oEkThOpFDNkACgGgAAABcAAAAAAAA/wEAAAABAAAABQAAAEEAdQBkAGkAbwCA35I+ gGbmPoDKyz6AZuY+AED7PoBm5j7AWhU/gGbmPoDVHj+AZuY+AACAPwAAAAAAAAAAAACAPwAAAAAA AAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAA bAAAAGcCAAA/AQAAjwIAAFABAAABAAAAqqrYQaqq2EFnAgAATgEAAAUAAABMAAAAAAAAAAAAAAAA AAAA//////////9YAAAAQQB1AGQAaQBvAAAACwAAAAkAAAAJAAAABAAAAAAAAAAlAAAADAAAAA0A AIBGAAAAvAAAALAAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADA Qn3oEkThOpFDNkACgHwAAABwAAAAAAAA/wEAAAABAAAABwAAAEQAZQBjAG8AZABlAHIAAHU9PoBm Jj+AWpw+gGYmPwDQyz6AZiY/AHv2PoBmJj9A+BI/gGYmPwCzKj+AZiY/wG1CP4BmJj8AAIA/AAAA AAAAAAAAAIA/AAAAAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUA AAAMAAAAAQAAAFQAAAB4AAAAXQIAAFIBAACZAgAAYwEAAAEAAACqqthBqqrYQV0CAABhAQAABwAA AEwAAAAAAAAAAAAAAAAAAAD//////////1wAAABEAGUAYwBvAGQAZQByAAAADAAAAAkAAAAIAAAA CQAAAAkAAAAJAAAAAAAAACUAAAAMAAAADQAAgEYAAACYAAAAjAAAAEVNRisrQAAADAAAAAAAAAAI QAACQAAAADQAAAACEMDbAAAAAM4AAAAAAAAA7FE4QAIAAAACAAAAAgAAAAIAAAAAAAAAAhDA2wAA AAAAAAD/CEABAywAAAAgAAAAAhDA2wIAAAAAAAAAxp3iQyncUEN/PO1DKdxQQwABAQEVQAEAEAAA AAQAAAAAAAAAJAAAACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAXwAAADgAAAADAAAA OAAAAAAAAAA4AAAAAAAAAAAAAQAuAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAlAAAADAAAAAMAAAAl AAAADAAAAAUAAIBXAAAAJAAAAMIBAADOAAAA3QEAANQAAAACAAAAVBwODagdDg0lAAAADAAAAAcA AIAlAAAADAAAAAAAAIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIAAAAoAAAADAAA AAMAAABGAAAAXAAAAFAAAABFTUYrCEABAzwAAAAwAAAAAhDA2wQAAAAAAAAA0HTsQ7ieSkP50PVD KdxQQ9B07EOaGVdD0HTsQ7ieSkMAAQGBFEABgBAAAAAEAAAAAAAA/ygAAAAMAAAAAgAAACQAAAAk AAAAAACAPQAAAAAAAAAAAACAPQAAAAAAAAAAAgAAACcAAAAYAAAAAgAAAAAAAAAAAAAAAAAAACUA AAAMAAAAAgAAABMAAAAMAAAAAQAAACUAAAAMAAAACAAAgFYAAAAsAAAA2AEAAMoAAADsAQAA2AAA AAQAAACPHaoMux4ODY8dcg2PHaoMJQAAAAwAAAAHAACAEwAAAAwAAAABAAAAJQAAAAwAAAAAAACA JAAAACQAAAAAAIBBAAAAAAAAAAAAAIBBAAAAAAAAAAACAAAARgAAAJQAAACIAAAARU1GKwhAAAJA AAAANAAAAAIQwNsAAAAAzgAAAAAAAADsUThAAgAAAAIAAAACAAAAAgAAAAAAAAACEMDbAAAAAAAA AP8IQAEDNAAAACgAAAACEMDbAwAAAAAAAADjTglEXA98Q+NOCUQUbqRDP54ORBRupEMAAQGBFUAB ABAAAAAEAAAAAAAAACQAAAAkAAAAAACAPQAAAAAAAAAAAACAPQAAAAAAAAAAAgAAAF8AAAA4AAAA AwAAADgAAAAAAAAAOAAAAAAAAAAAAAEALgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAJQAAAAwAAAAD AAAAJQAAAAwAAAAFAACAVwAAACgAAAAiAgAA+QAAAD0CAABMAQAAAwAAAFQiwQ9UIo4UqCOOFCUA AAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAkAAAAAACAQQAAAAAAAAAAAACAQQAAAAAAAAAAAgAA ACgAAAAMAAAAAwAAAEYAAABcAAAAUAAAAEVNRisIQAEDPAAAADAAAAACEMDbBAAAAAAAAABoOg5E XE+hQ33oEkQUbqRDaDoORM2Mp0NoOg5EXE+hQwABAYEUQAGAEAAAAAQAAAAAAAD/JAAAACQAAAAA AIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAJQAAAAwAAAACAAAAEwAAAAwAAAABAAAAJQAAAAwA AAAIAACAVgAAACwAAAA4AgAAQgEAAEwCAABQAQAABAAAAI8jKhS7JI4UjyPyFI8jKhQlAAAADAAA AAcAAIATAAAADAAAAAEAAAAlAAAADAAAAAAAAIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAA AAAAAAIAAABGAAAANAAAACgAAABFTUYrCkAAgCQAAAAYAAAA9+7o/wEAAAAWgjRE4TqRQwAAwEKc mZlCKAAAAAwAAAACAAAAJAAAACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAJwAAABgA AAACAAAAAAAAAOju9wAAAAAAJQAAAAwAAAACAAAAJQAAAAwAAAAIAACAVgAAADAAAADSAgAAIgEA ADMDAABwAQAABQAAACEtKBIhLfUWITP1FiEzKBIhLSgSJQAAAAwAAAAHAACAJQAAAAwAAAAAAACA JAAAACQAAAAAAIBBAAAAAAAAAAAAAIBBAAAAAAAAAAACAAAARgAAAKgAAACcAAAARU1GKwhAAAJA AAAANAAAAAIQwNsAAAAAzgAAAAAAAADsUThAAgAAAAIAAAACAAAAAgAAAAAAAAACEMDbAAAAAAAA AP8IQAEDSAAAADwAAAACEMDbBQAAAAAAAAAWgjRESKG3QxaCTERIobdDFoJMROE6kUMWgjRE4TqR QxaCNERIobdDAAEBAYEAAAAVQAEAEAAAAAQAAAAAAAAAJAAAACQAAAAAAIA9AAAAAAAAAAAAAIA9 AAAAAAAAAAACAAAAXwAAADgAAAADAAAAOAAAAAAAAAA4AAAAAAAAAAAAAQAuAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAlAAAADAAAAAMAAAAlAAAADAAAAAUAAIBWAAAAMAAAAM8CAAAgAQAANQMAAHIB AAAFAAAAIS31FiEz9RYhMygSIS0oEiEt9RYlAAAADAAAAAcAAIAlAAAADAAAAAAAAIAkAAAAJAAA AAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIAAAAoAAAADAAAAAMAAABGAAAAnAAAAJAAAABFTUYr KkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQhaCNEThOpFDNkACgGgAAABcAAAAAAAA/wEAAAAB AAAABQAAAFAAaQB0AGMAaAAAJaE+gDOzPgAQ2j6AM7M+gAXtPoAzsz6AXQI/gDOzPgCzFz+AM7M+ AACAPwAAAAAAAAAAAACAPwAAAAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAA ABgAAAAlAAAADAAAAAEAAABUAAAAbAAAAPACAAA1AQAAEQMAAEYBAAABAAAAqqrYQaqq2EHwAgAA RAEAAAUAAABMAAAAAAAAAAAAAAAAAAAA//////////9YAAAAUABpAHQAYwBoAAAACwAAAAQAAAAE AAAACAAAAAAAAAAlAAAADAAAAA0AAIBGAAAAnAAAAJAAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQA AAAYAAAAAADAQgAAAAAAAAAAAADAQhaCNEThOpFDNkACgFwAAABQAAAAAAAA/wEAAAABAAAABAAA AFAAbwBzAHQAAJ8XPgDNDD+AuoQ+AM0MPwAwtD4AzQw/ANvePgDNDD8AAIA/AAAAAAAAAAAAAIA/ AAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABU AAAAZAAAAOACAABIAQAA/wIAAFkBAAABAAAAqqrYQaqq2EHgAgAAVwEAAAQAAABMAAAAAAAAAAAA AAAAAAAA//////////9UAAAAUABvAHMAdAALAAAACQAAAAgAAAAAAAAAJQAAAAwAAAANAACARgAA AIAAAAB0AAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEIWgjRE 4TqRQzZAAoBAAAAANAAAAAAAAP8BAAAAAQAAAAEAAAAtAICQ9j4AzQw/AACAPwAAAAAAAAAAAACA PwAAAAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEA AABUAAAAVAAAAAADAABIAQAAAwMAAFkBAAABAAAAqqrYQaqq2EEAAwAAVwEAAAEAAABMAAAAAAAA AAAAAAAAAAAA//////////9QAAAALQAAAAAAAAAlAAAADAAAAA0AAIBGAAAAsAAAAKQAAABFTUYr K0AAAAwAAAAAAAAAKkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQhaCNEThOpFDNkACgHAAAABk AAAAAAAA/wEAAAABAAAABgAAAGYAaQBsAHQAZQByAIB9CT8AzQw/QFgVPwDNDD8A0x4/AM0MP8BN KD8AzQw/gCg0PwDNDD9A40s/AM0MPwAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAABIAAAAMAAAAAQAA ABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAMAAAAAQAAAFQAAABwAAAABgMAAEgBAAAjAwAA WQEAAAEAAACqqthBqqrYQQYDAABXAQAABgAAAEwAAAAAAAAAAAAAAAAAAAD//////////1gAAABm AGkAbAB0AGUAcgAEAAAABAAAAAQAAAAEAAAACQAAAAAAAAAlAAAADAAAAA0AAIBGAAAAQAAAADQA AABFTUYrK0AAAAwAAAAAAAAACkAAgCQAAAAYAAAA9+7o/wEAAACwGxpECtdJQgAAwEKamZlCJAAA ACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAJQAAAAwAAAACAAAAJQAAAAwAAAAIAACA VgAAADAAAABoAgAAMgAAAMkCAACAAAAABQAAAIcmKAOHJvUHhyz1B4csKAOHJigDJQAAAAwAAAAH AACAJQAAAAwAAAAAAACAJAAAACQAAAAAAIBBAAAAAAAAAAAAAIBBAAAAAAAAAAACAAAARgAAAKgA AACcAAAARU1GKwhAAAJAAAAANAAAAAIQwNsAAAAAzgAAAAAAAAAK16M+AgAAAAIAAAACAAAAAgAA AAAAAAACEMDbAAAAAAAAAP8IQAEDSAAAADwAAAACEMDbBQAAAAAAAACwGxpEH4X+QrAbMkQfhf5C sBsyRArXSUKwGxpECtdJQrAbGkQfhf5CAAEBAYEAAAAVQAEAEAAAAAQAAAAAAAAAJAAAACQAAAAA AIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAXwAAADgAAAADAAAAOAAAAAAAAAA4AAAAAAAAAAAA AQAFAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAlAAAADAAAAAMAAAAlAAAADAAAAAUAAIBWAAAAMAAA AGcCAAAxAAAAygIAAIEAAAAFAAAAhyb1B4cs9QeHLCgDhyYoA4cm9QclAAAADAAAAAcAAIAlAAAA DAAAAAAAAIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIAAAAoAAAADAAAAAMAAABG AAAAnAAAAJAAAABFTUYrKkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQrAbGkQK10lCNkACgGgA AABcAAAAAAAA/wEAAAABAAAABQAAAFAAaQB0AGMAaAAAJaE+AACAPgAQ2j4AAIA+gAXtPgAAgD6A XQI/AACAPgCzFz8AAIA+AACAPwAAAAAAAAAAAACAPwAAAAAAAAAAAAASAAAADAAAAAEAAAAYAAAA DAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAbAAAAIcCAAA7AAAAqAIAAEwAAAAB AAAAqqrYQaqq2EGHAgAASgAAAAUAAABMAAAAAAAAAAAAAAAAAAAA//////////9YAAAAUABpAHQA YwBoAAAACwAAAAQAAAAEAAAACAAAAAAAAAAlAAAADAAAAA0AAIBGAAAA0AAAAMQAAABFTUYrK0AA AAwAAAAAAAAAKkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQrAbGkQK10lCNkACgJAAAACEAAAA AAAA/wEAAAABAAAACQAAAFAAYQByAGEAbQBlAHQAZQByAAA+4z2AZuY+AHVjPoBm5j4AMKE+gGbm PoCavT6AZuY+ABDtPoBm5j7AEho/gGbmPoDNMT+AZuY+QKg9P4Bm5j4AY1U/gGbmPgAAgD8AAAAA AAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAA AAwAAAABAAAAVAAAAIQAAABzAgAATwAAALwCAABgAAAAAQAAAKqq2EGqqthBcwIAAF4AAAAJAAAA TAAAAAAAAAAAAAAAAAAAAP//////////YAAAAFAAYQByAGEAbQBlAHQAZQByAAAACwAAAAkAAAAF AAAACQAAAA0AAAAJAAAABAAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAALwAAACwAAAARU1GKytA AAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEKwGxpECtdJQjZAAoB8AAAAcAAA AAAAAP8BAAAAAQAAAAcAAABEAGUAYwBvAGQAZQByAAB1PT6AZiY/gFqcPoBmJj8A0Ms+gGYmPwB7 9j6AZiY/QPgSP4BmJj8Asyo/gGYmP8BtQj+AZiY/AACAPwAAAAAAAAAAAACAPwAAAAAAAAAAAAAS AAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAeAAAAHoC AABiAAAAtgIAAHMAAAABAAAAqqrYQaqq2EF6AgAAcQAAAAcAAABMAAAAAAAAAAAAAAAAAAAA//// //////9cAAAARABlAGMAbwBkAGUAcgAAAAwAAAAJAAAACAAAAAkAAAAJAAAACQAAAAAAAAAlAAAA DAAAAA0AAIBGAAAAoAAAAJQAAABFTUYrK0AAAAwAAAAAAAAACEAAAkAAAAA0AAAAAhDA2wAAAADO AAAAAAAAAArXoz4CAAAAAgAAAAIAAAACAAAAAAAAAAIQwNsAAAAAv3dG/whAAQM0AAAAKAAAAAIQ wNsDAAAAAAAAAONOCUT2qCVD404JRFK4sUIClBdEUrixQgABAQEVQAEAEAAAAAQAAAAAAAAAJAAA ACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAXwAAADgAAAADAAAAOAAAAAAAAAA4AAAA AAAAAAAAAQAFAAAAAAAAAEZ3vwAAAAAAAAAAAAAAAAAlAAAADAAAAAMAAAAlAAAADAAAAAUAAIBX AAAAKAAAACQCAABXAAAAYAIAAKcAAAADAAAAVCJbClQijgXlJY4FJQAAAAwAAAAHAACAJQAAAAwA AAAAAACAJAAAACQAAAAAAIBBAAAAAAAAAAAAAIBBAAAAAAAAAAACAAAAKAAAAAwAAAADAAAARgAA AFwAAABQAAAARU1GKwhAAQM8AAAAMAAAAAIQwNsEAAAAAAAAACFZF0QpXKpCsBsaRFK4sUIhWRdE exS5QiFZF0QpXKpCAAEBgRRAAYAQAAAABAAAAL93Rv8oAAAADAAAAAIAAAAkAAAAJAAAAAAAgD0A AAAAAAAAAAAAgD0AAAAAAAAAAAIAAAAnAAAAGAAAAAIAAAAAAAAARne/AAAAAAAlAAAADAAAAAIA AAATAAAADAAAAAEAAAAlAAAADAAAAAgAAIBWAAAALAAAAF0CAABVAAAAaQIAAF0AAAAEAAAA1yVT BYcmjgXXJckF1yVTBSUAAAAMAAAABwAAgBMAAAAMAAAAAQAAACUAAAAMAAAAAAAAgCQAAAAkAAAA AACAQQAAAAAAAAAAAACAQQAAAAAAAAAAAgAAAEYAAACMAAAAgAAAAEVNRisIQAACQAAAADQAAAAC EMDbAAAAAM4AAAAAAAAA7FE4QAIAAAACAAAAAgAAAAIAAAAAAAAAAhDA2wAAAAAAAAD/CEABAywA AAAgAAAAAhDA2wIAAAAAAAAAfegqRBRupEPZNzBEFG6kQwABAYEVQAEAEAAAAAQAAAAAAAAAJAAA ACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAACAAAAXwAAADgAAAADAAAAOAAAAAAAAAA4AAAA AAAAAAAAAQAuAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAlAAAADAAAAAMAAAAlAAAADAAAAAUAAIBX AAAAJAAAAKkCAABGAQAAxAIAAEwBAAACAAAAuyqOFA4sjhQlAAAADAAAAAcAAIAlAAAADAAAAAAA AIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIAAAAoAAAADAAAAAMAAABGAAAAXAAA AFAAAABFTUYrCEABAzwAAAAwAAAAAhDA2wQAAAAAAAAAAtQvRFxPoUMWgjREFG6kQwLUL0TNjKdD AtQvRFxPoUMAAQGBFEABgBAAAAAEAAAAAAAA/ygAAAAMAAAAAgAAACQAAAAkAAAAAACAPQAAAAAA AAAAAACAPQAAAAAAAAAAAgAAACcAAAAYAAAAAgAAAAAAAAAAAAAAAAAAACUAAAAMAAAAAgAAABMA AAAMAAAAAQAAACUAAAAMAAAACAAAgFYAAAAsAAAAvwIAAEIBAADTAgAAUAEAAAQAAAD1KyoUIS2O FPUr8hT1KyoUJQAAAAwAAAAHAACAEwAAAAwAAAABAAAAJQAAAAwAAAAAAACAJAAAACQAAAAAAIBB AAAAAAAAAAAAAIBBAAAAAAAAAAACAAAARgAAAIwAAACAAAAARU1GKwhAAAJAAAAANAAAAAIQwNsA AAAAzgAAAAAAAADsUThAAgAAAAIAAAACAAAAAgAAAAAAAAACEMDbAAAAAAAAAP8IQAEDLAAAACAA AAACEMDbAgAAAAAAAAAWgkxEFG6kQ00SWUQUbqRDAAEBgRVAAQAQAAAABAAAAAAAAAAkAAAAJAAA AAAAgD0AAAAAAAAAAAAAgD0AAAAAAAAAAAIAAABfAAAAOAAAAAMAAAA4AAAAAAAAADgAAAAAAAAA AAABAC4AAAAAAAAAAAAAAAAAAAAAAAAAAAAAACUAAAAMAAAAAwAAACUAAAAMAAAABQAAgFcAAAAk AAAALwMAAEYBAABnAwAATAEAAAIAAAAhM44URTaOFCUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQA AAAkAAAAAACAQQAAAAAAAAAAAACAQQAAAAAAAAAAAgAAACgAAAAMAAAAAwAAAEYAAABcAAAAUAAA AEVNRisIQAEDPAAAADAAAAACEMDbBAAAAAAAAAB2rlhEXE+hQ4pcXUQUbqRDdq5YRM2Mp0N2rlhE XE+hQwABAYEUQAGAEAAAAAQAAAAAAAD/JAAAACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAAAAAC AAAAJQAAAAwAAAACAAAAEwAAAAwAAAABAAAAJQAAAAwAAAAIAACAVgAAACwAAABiAwAAQgEAAHYD AABQAQAABAAAACw2KhRYN44ULDbyFCw2KhQlAAAADAAAAAcAAIATAAAADAAAAAEAAAAlAAAADAAA AAAAAIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEAAAAAAAAAAAIAAABGAAAAzAAAAMAAAABFTUYr KkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQpNq80MUbqRDNkACgJgAAACMAAAAAAAA/wEAAAAB AAAACgAAAEMAbwBtAHAAcgBlAHMAcwBlAGQAAOSePQDNTD4Asko+AM1MPoDOlD4AzUw+AOTbPgDN TD7ArAU/AM1MPgDiEz8AzUw+wJwrPwDNTD5A8kA/AM1MPsBHVj8AzUw+gAJuPwDNTD4AAIA/AAAA AAAAAAAAAIA/AAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAA DAAAAAEAAABUAAAAiAAAAO4BAABNAQAARwIAAF4BAAABAAAAqqrYQaqq2EHuAQAAXAEAAAoAAABM AAAAAAAAAAAAAAAAAAAA//////////9gAAAAQwBvAG0AcAByAGUAcwBzAGUAZAAMAAAACQAAAA0A AAAJAAAABQAAAAkAAAAIAAAACAAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAAKgAAACcAAAARU1G KytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEKTavNDFG6kQzZAAoBoAAAA XAAAAAAAAP8BAAAAAQAAAAUAAABBAHUAZABpAG8AAHmsPgDNzD4AZOU+AM3MPsBsCj8Azcw+gCci PwDNzD5Aois/AM3MPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwA AAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAGwAAAAHAgAAYAEAAC8CAABxAQAAAQAA AKqq2EGqqthBBwIAAG8BAAAFAAAATAAAAAAAAAAAAAAAAAAAAP//////////WAAAAEEAdQBkAGkA bwAAAAsAAAAJAAAACQAAAAQAAAAAAAAAJQAAAAwAAAANAACARgAAAJQAAACIAAAARU1GKytAAAAM AAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEKTavNDFG6kQzZAAoBUAAAASAAAAAAA AP8BAAAAAQAAAAMAAABCAGkAdAAAZy4+wJkZP4AekD7AmRk/ABSjPsCZGT8AAIA/AAAAAAAAAAAA AIA/AAAAAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAMAAAA AQAAAFQAAABgAAAA9wEAAHMBAAAJAgAAhAEAAAEAAACqqthBqqrYQfcBAACCAQAAAwAAAEwAAAAA AAAAAAAAAAAAAAD//////////1QAAABCAGkAdAAAAAsAAAAEAAAAAAAAACUAAAAMAAAADQAAgEYA AACAAAAAdAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAAABgAAAAAAMBCAAAAAAAAAAAAAMBCk2rz QxRupEM2QAKAQAAAADQAAAAAAAD/AQAAAAEAAAABAAAALQCAybo+wJkZPwAAgD8AAAAAAAAAAAAA gD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAAB AAAAVAAAAFQAAAAKAgAAcwEAAA0CAACEAQAAAQAAAKqq2EGqqthBCgIAAIIBAAABAAAATAAAAAAA AAAAAAAAAAAAAP//////////UAAAAC0AAAAAAAAAJQAAAAwAAAANAACARgAAALAAAACkAAAARU1G KytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEKTavNDFG6kQzZAAoBwAAAA ZAAAAAAAAP8BAAAAAQAAAAYAAABzAHQAcgBlAGEAbQAANNc+wJkZP4DvAD/AmRk/QMoMP8CZGT+A /xo/wJkZP0C6Mj/AmRk/AHVKP8CZGT8AAIA/AAAAAAAAAAAAAIA/AAAAAAAAAAASAAAADAAAAAEA AAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAcAAAAA8CAABzAQAAPQIA AIQBAAABAAAAqqrYQaqq2EEPAgAAggEAAAYAAABMAAAAAAAAAAAAAAAAAAAA//////////9YAAAA cwB0AHIAZQBhAG0ACAAAAAQAAAAFAAAACQAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAAKAAAACU AAAARU1GKytAAAAMAAAAAAAAAAhAAAJAAAAANAAAAAIQwNsAAAAAzgAAAAAAAAAK16M+AgAAAAIA AAACAAAAAgAAAAAAAAACEMDbAAAAAL93Rv8IQAEDNAAAACgAAAACEMDbAwAAAAAAAACwGzJEUrix QhaCQERSuLFCFoJARIUrjEMAAQEBFUABABAAAAAEAAAAAAAAACQAAAAkAAAAAACAPQAAAAAAAAAA AACAPQAAAAAAAAAAAgAAAF8AAAA4AAAAAwAAADgAAAAAAAAAOAAAAAAAAAAAAAEABQAAAAAAAABG d78AAAAAAAAAAAAAAAAAJQAAAAwAAAADAAAAJQAAAAwAAAAFAACAVwAAACgAAADHAgAAVwAAAAQD AAAaAQAAAwAAAIcsjgUhMI4FITCGESUAAAAMAAAABwAAgCUAAAAMAAAAAAAAgCQAAAAkAAAAAACA QQAAAAAAAAAAAACAQQAAAAAAAAAAAgAAACgAAAAMAAAAAwAAAEYAAABcAAAAUAAAAEVNRisIQAED PAAAADAAAAACEMDbBAAAAAAAAACbbUFEw7WLQxaCQEThOpFDkZY/RMO1i0ObbUFEw7WLQwABAYEU QAGAEAAAAAQAAAC/d0b/KAAAAAwAAAACAAAAJAAAACQAAAAAAIA9AAAAAAAAAAAAAIA9AAAAAAAA AAACAAAAJwAAABgAAAACAAAAAAAAAEZ3vwAAAAAAJQAAAAwAAAACAAAAEwAAAAwAAAABAAAAJQAA AAwAAAAIAACAVgAAACwAAAD+AgAAFwEAAAYDAAAjAQAABAAAAFwwdxEhMCgS5i93EVwwdxElAAAA DAAAAAcAAIATAAAADAAAAAEAAAAlAAAADAAAAAAAAIAkAAAAJAAAAAAAgEEAAAAAAAAAAAAAgEEA AAAAAAAAAAIAAABGAAAAxAAAALgAAABFTUYrKkAAACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQhaC QEQp3AhDNkACgJAAAACEAAAAAAAA/wEAAAABAAAACQAAAFEAdQBhAG4AdABpAHoAZQBkAABE+D0A mhk+AHGAPgCaGT6A5q8+AJoZPgBc3z4Amhk+wGgHPwCaGT6AQxM/AJoZPkC+HD8Amhk+wBMyPwCa GT6Azkk/AJoZPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAA AAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAIQAAAAOAwAAiAAAAFUDAACZAAAAAQAAAKqq 2EGqqthBDgMAAJcAAAAJAAAATAAAAAAAAAAAAAAAAAAAAP//////////YAAAAFEAdQBhAG4AdABp AHoAZQBkAAAADAAAAAkAAAAJAAAACQAAAAQAAAAEAAAACAAAAAkAAAAAAAAAJQAAAAwAAAANAACA RgAAAKgAAACcAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEIW gkBEKdwIQzZAAoBoAAAAXAAAAAAAAP8BAAAAAQAAAAUAAABQAGkAdABjAGgAgLahPoAzsz6Aodo+ gDOzPgCX7T6AM7M+QKYCP4Azsz7A+xc/gDOzPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAEgAA AAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAGwAAAAgAwAA mwAAAEEDAACsAAAAAQAAAKqq2EGqqthBIAMAAKoAAAAFAAAATAAAAAAAAAAAAAAAAAAAAP////// ////WAAAAFAAaQB0AGMAaAAAAAsAAAAEAAAABAAAAAgAAAAAAAAAJQAAAAwAAAANAACARgAAANgA AADMAAAARU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEIWgkBEKdwI QzZAAoCYAAAAjAAAAAAAAP8BAAAAAQAAAAoAAABQAGEAcgBhAG0AZQB0AGUAcgBzAAAukD0AzQw/ AO05PgDNDD8AbIw+AM0MP4DWqD4AzQw/AEzYPgDNDD/AsA8/AM0MP4BrJz8AzQw/QEYzPwDNDD8A AUs/AM0MP0A2WT8AzQw/AACAPwAAAAAAAAAAAACAPwAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwA AAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwAAAABAAAAVAAAAIgAAAAJAwAArwAAAFkDAADAAAAAAQAA AKqq2EGqqthBCQMAAL4AAAAKAAAATAAAAAAAAAAAAAAAAAAAAP//////////YAAAAFAAYQByAGEA bQBlAHQAZQByAHMACwAAAAkAAAAFAAAACQAAAA0AAAAJAAAABAAAAAkAAAAFAAAAAAAAACUAAAAM AAAADQAAgEYAAADYAAAAzAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAAABgAAAAAAMBCAAAAAAAA AAAAAMBCLQTxQ6RwHUA2QAKAmAAAAIwAAAAAAAD/AQAAAAEAAAAKAAAAQwBvAG0AcAByAGUAcwBz AGUAZAAA5J49AM1MPgCySj4AzUw+gM6UPgDNTD4A5Ns+AM1MPsCsBT8AzUw+AOITPwDNTD7AnCs/ AM1MPkDyQD8AzUw+wEdWPwDNTD6AAm4/AM1MPgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAABIAAAAM AAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAMAAAAAQAAAFQAAACIAAAA6QEAAAcA AABCAgAAGAAAAAEAAACqqthBqqrYQekBAAAWAAAACgAAAEwAAAAAAAAAAAAAAAAAAAD///////// /2AAAABDAG8AbQBwAHIAZQBzAHMAZQBkAAwAAAAJAAAADQAAAAkAAAAFAAAACQAAAAgAAAAIAAAA CQAAAAAAAAAlAAAADAAAAA0AAIBGAAAAsAAAAKQAAABFTUYrK0AAAAwAAAAAAAAAKkAAACQAAAAY AAAAAADAQgAAAAAAAAAAAADAQi0E8UOkcB1ANkACgHAAAABkAAAAAAAA/wEAAAABAAAABgAAAFAA aQB0AGMAaAAgAIC+uj4Azcw+gKnzPgDNzD6ATwM/AM3MPkAqDz8Azcw+wH8kPwDNzD6AOjw/AM3M PgAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAA GAAAACUAAAAMAAAAAQAAAFQAAABwAAAABQIAABoAAAApAgAAKwAAAAEAAACqqthBqqrYQQUCAAAp AAAABgAAAEwAAAAAAAAAAAAAAAAAAAD//////////1gAAABQAGkAdABjAGgAIAALAAAABAAAAAQA AAAIAAAACQAAAAAAAAAlAAAADAAAAA0AAIBGAAAA2AAAAMwAAABFTUYrK0AAAAwAAAAAAAAAKkAA ACQAAAAYAAAAAADAQgAAAAAAAAAAAADAQi0E8UOkcB1ANkACgJgAAACMAAAAAAAA/wEAAAABAAAA CgAAAFAAYQByAGEAbQBlAHQAZQByAHMAAE70PcCZGT8A/Ws+wJkZPwB0pT7AmRk/gN7BPsCZGT8A VPE+wJkZP8A0HD/AmRk/gO8zP8CZGT9Ayj8/wJkZPwCFVz/AmRk/QLplP8CZGT8AAIA/AAAAAAAA AAAAAIA/AAAAAAAAAAASAAAADAAAAAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAA AAEAAABUAAAAiAAAAO0BAAAtAAAAPQIAAD4AAAABAAAAqqrYQaqq2EHtAQAAPAAAAAoAAABMAAAA AAAAAAAAAAAAAAAA//////////9gAAAAUABhAHIAYQBtAGUAdABlAHIAcwALAAAACQAAAAUAAAAJ AAAADQAAAAkAAAAEAAAACQAAAAUAAAAAAAAAJQAAAAwAAAANAACARgAAAJQAAACIAAAARU1GKytA AAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEItBPFDpHAdQDZAAoBUAAAASAAA AAAAAP8BAAAAAQAAAAMAAABCAGkAdAAAZy4+AM1MP4AekD4AzUw/ABSjPgDNTD8AAIA/AAAAAAAA AAAAAIA/AAAAAAAAAAAAABIAAAAMAAAAAQAAABgAAAAMAAAAAAAAABYAAAAMAAAAGAAAACUAAAAM AAAAAQAAAFQAAABgAAAA8gEAAEAAAAAEAgAAUQAAAAEAAACqqthBqqrYQfIBAABPAAAAAwAAAEwA AAAAAAAAAAAAAAAAAAD//////////1QAAABCAGkAdAAAAAsAAAAEAAAAAAAAACUAAAAMAAAADQAA gEYAAACAAAAAdAAAAEVNRisrQAAADAAAAAAAAAAqQAAAJAAAABgAAAAAAMBCAAAAAAAAAAAAAMBC LQTxQ6RwHUA2QAKAQAAAADQAAAAAAAD/AQAAAAEAAAABAAAALQCAybo+AM1MPwAAgD8AAAAAAAAA AAAAgD8AAAAAAAAAAAAAEgAAAAwAAAABAAAAGAAAAAwAAAAAAAAAFgAAAAwAAAAYAAAAJQAAAAwA AAABAAAAVAAAAFQAAAAFAgAAQAAAAAgCAABRAAAAAQAAAKqq2EGqqthBBQIAAE8AAAABAAAATAAA AAAAAAAAAAAAAAAAAP//////////UAAAAC0AAAAAAAAAJQAAAAwAAAANAACARgAAALAAAACkAAAA RU1GKytAAAAMAAAAAAAAACpAAAAkAAAAGAAAAAAAwEIAAAAAAAAAAAAAwEItBPFDpHAdQDZAAoBw AAAAZAAAAAAAAP8BAAAAAQAAAAYAAABzAHQAcgBlAGEAbQAANNc+AM1MP4DvAD8AzUw/QMoMPwDN TD+A/xo/AM1MP0C6Mj8AzUw/AHVKPwDNTD8AAIA/AAAAAAAAAAAAAIA/AAAAAAAAAAASAAAADAAA AAEAAAAYAAAADAAAAAAAAAAWAAAADAAAABgAAAAlAAAADAAAAAEAAABUAAAAcAAAAAoCAABAAAAA OAIAAFEAAAABAAAAqqrYQaqq2EEKAgAATwAAAAYAAABMAAAAAAAAAAAAAAAAAAAA//////////9Y AAAAcwB0AHIAZQBhAG0ACAAAAAQAAAAFAAAACQAAAAkAAAAAAAAAJQAAAAwAAAANAACARgAAABwA AAAQAAAARU1GKytAAAAMAAAAAAAAAEwAAABkAAAAEwAAAAgAAAB1AwAAhQEAABMAAAAIAAAAYwMA AH4BAAApAKoAAAAAAAAAAAAAAIA/AAAAAAAAAAAAAIA/AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAiAAAADAAAAP////9GAAAAHAAAABAAAABFTUYrAkAAAAwAAAAAAAAADgAAABQAAAAA AAAAEAAAABQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAEMAbwBtAHAATwBiAGoAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEgACAf///////////////wAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA4AAABzAAAAAAAAAAMATwBiAGoASQBuAGYAbwAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIBDwAAABMA AAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEAAAAAYAAAAAAAAAVgBp AHMAaQBvAEQAbwBjAHUAbQBlAG4AdAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAABwAAgH///////////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAE AQAA3mkAAAAAAABWAGkAcwBpAG8ASQBuAGYAbwByAG0AYQB0AGkAbwBuAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAIgACABIAAAAUAAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAABEAAAAcAAAAAAAAAAAAAwANAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD+/wAABAACAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA/v8AAAUBAgAAAAAAAAAAAAAA AAAAAAAAAgAAAALVzdWcLhsQk5cIACss+a5EAAAABdXN1ZwuGxCTlwgAKyz5roABAAA8AQAACwAA AAEAAABgAAAAAgAAAGgAAAAOAAAAdAAAAA8AAACAAAAAFwAAAKAAAAALAAAAqAAAABAAAACwAAAA EwAAALgAAAAWAAAAwAAAAA0AAADIAAAADAAAAAQBAAACAAAA5AQAAB4AAAAEAAAAAAAAAB4AAAAE AAAAAAAAAB4AAAAYAAAAQnJvYWRjb20gQ29ycG9yYXRpb24AAAAAAwAAAAAADAALAAAAAAAAAAsA AAAAAAAACwAAAAAAAAALAAAAAAAAAB4QAAADAAAACAAAAFBhZ2UtMQBODAAAAFJlY3RhbmdsZQBF ThQAAABEeW5hbWljIGNvbm5lY3RvcgBlbgwQAAAEAAAAHgAAAAgAAABQYWdlcwAAAAMAAAABAAAA HgAAAAgAAABNYXN0ZXJzAAMAAAACAAAAVmlzaW8gKFRNKSBEcmF3aW5nDQoAAAAAAAALAN5pAAAA hAEAFAAAABxPGAJ1aAAAaQEAAFIAAAAAAAAAAAAAANDeEgAAAAAAAAAAAAAAAAAAAAAAAAAAAMje EgAAAAAAAAAAAH6FdTIAAAAAAEjABAAAAADcAAAAAAAAAAAAAADtaOrxAxnp8v///4MA//by//P7 8AcB+PEAiYDq8RYEgBYAIAAZAID/AMDAwADm5ub/AM3NzQCzs7PvAJqamiEAgABm/2ZmAE1NTQAz PzMzABoaGtzyGAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAGAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAVRrr8P/y8B3r8Bzo86D69AcP6/D28eb1Suf0BDrr8ETr 8AEAVDUB6POA8vFJD1sP8vHg+zEB5vVo6uf0EOvwWOvwAgBUhRjr8EDf/JQBlAHm9QGumQIAAAMu BAXr8Aaq6/AH6/AI6/AJ6/AKquvwC+vwDOvwDevwDqrr8A/n9JPj+C/IAlXnAQEBwgHnAQIAYlcB AP4dEgEjFgIjFl8DYgAA/urxhcAE+urxmRATMHoUrkeP4XqEP74DowbHAQIMTBOUAQIAXBYmFUIQ MRdyfxIDowYmEwRiAiUUVQWVFgaVFgeVFgi5ESZ/Egkhqhbq8YbEBOrxAQPDABMRbwZNEwwkZBMg KQ1Q4PvwPxoflBw7F7MUSCQVwhJ/EgbrF38SB4ssUQiLLOoYHRIKgSYLIyZKohMM1SwNNSaiEw7r FAvwP0QSh8gE8BUTECMvOC8/TRM4NOA/IFUw8vBD70ELD7oKfhTVLAHVLAUC1SwD1SdxL4MlUDaU JIhZNtoVlhUJlRbKKOsboAJ9COAQE2s/wQMdFJYVlBkAlhVsKZYVwh/UG/0/yiIdElULlRYMlRYN lRYOlRZVD5UWEJUWEZUWEpUWJRNBFaTQBOrxMwYj4vkN/W8CCQQ+U0AjQjrjAV8CAGEA/l9SAWVV VQJlVQNlVQRlVQVlVQUGZVUHZVCjT7VO2k/sRE7rAQIQYK8RX1IRZVVVEmVVE2VVFGVVFWVVVRZl VRdlVRikVRmVFi8aYgkE4lMb6VJ/EiUcBTuE1ATq8RE1T00VAwCj2AQlBRMQPlGkaXE7UuNSAOlS 41IByGcCyGcJA+lS6vGn3ARxaD5RwGrS6vG34ATq8fcQEyAgGKMGL3+jBggIQX9ef0YngSAlFoNw ojOpF6MdfxIEOpZ8BakcBmIIl0bXcFSlI5Z8CZZ8CpZ8C6kX0urxveQE6vGxdmnwPw+IE4gTPU9z PB0XkjNU6xTjNQJ4jAOLLASLLEEFiyyKL6obrFLq8b7oBArq8VEQFAJtZVmPIBu6EEJxXQGCVUCU kVHq8b/sBALq8fp2b3afTBVfX3Ffg18AlV+nX7VPx0/JX+tP/U8PWSodEhSVFhVBFcDwBGCSWCxY wwGjBMA/ha3Qlq5Z2KiujKhzVc6h5cymAHif7q+Gn5ifqp+8n86f4J9A8p8EoSQVEK8ipH8SELum SqITEaq8EpemohMTyrypFKmmohMV6rwWqrwXJqq8GCDNpaITGSrHQGSqlhUblRYclRYdlRYeMkEV yPQEGz/QP43Pn8YASltEIcMAFxCVZcMBgjrRtQCXNdG1pzXRtbc10bWnhQwwkLKKDDCPRCQVByMR DLIIqhK1CRK1ChK1C1G1ELojFhEjFhJiB0MTkgL4CB7LAGifsN/C39Tf5t9o+N8K7+X2mscAAw3X AZbLAQEBLOYD1wHHAQFBAd8BMOLeAjDjSuYD3wEgROPjAU7jcuZx5wPnAU7jipDmAZDmA+sBTuOu 5gGiruYD7wFO48zmAczmAyjzAU7j6uYB6uYD9wFO44oI9gEI9gP7AU7jJvYBIib2A5QBTuNE9gN1 YzcBmlj2AVj2AxJL5nb2AaJ29gNGgU7jlPYBlPYDCRRL5rL2AbL2sfJE47LxIgEjFUvm5Pbj9wPk 8UTjpOTxASMWLeYWBgAWBgMQFgFE4xYBASMXS+ZIBkcHAQNIAUTjSAFCA58BTuN6BkkDhAFE4xlL 5pgGA5gBnJIHASUAAGl6NH0xHILWwlSEASxFdTmIZvdklPK1aFQsV9sHECcAMGbOolXF3K4FNb8w 9K4RCdsGiGZJJARbT21Pf0OAJd+RRZUjIxZI31rfFLMMKhK1DRK1DhK1DxKwpLOoRRaztLssM1km NFkmNZKBJjY8FnMTOUZh+mFqUI8FzQ/fD4MrlbVoOAwXLDg/xSYhM9gi80QXRRYBAT1UOxfcrxU/ dz+JP5s/EK041yavdBQsBRQnNd7gVUEI6VJevxJMWSAF0gxZN4SGt/pha0vmdC+GL0JKiAC1aHVp mOOAcHPEnWEuMZIBBUTp8lRjQfhjBnB2TwZwIKYBiE0NS5IBS3Vo92RYEQEoAUdO8bRpWBELSVtF nEUFk7VoLv2ExsB5dyOXI5VRAkXWLLH6YYWFaC01g/8wehSuR+F6hIi4xHB2kgECYoiDEvphhkKr SEf9hLFtsWPFL1ANSxS/wXtch/ZIsIfKsl8LaM4HauA/II2ABnDvQQAcPyU/Nz9JP1s73DyXEnQS AQXsNWohyA8KTxxPpCpGURC1L8cv2S/oRv71L0t/XX8AFJqxElNFbiVjT6N/h0+ZTwjXbLpFpgEk vGdcldlFpgEAy3lbRf1PD18hXzNfRV9XXwBpX3tfjV+fXwd/w1/VX1ZCAPg1kn8wnzaSLJVImc9/ sk0AkgEuYL5lXuGiTa5JYRUsj6A+j1CPYozoUWEVnMKDMjccx3EOoKw/Da8fpjBhEyVvWJ9JbwIA H6ZmZApvrAJvrANvp1NFWBFfTwAtn7J/xH+nT7lGNuFFEcpEQIQBymk24Tbh6U8UEwWun6DAn9Kf 5J9tjcGEMtlI/6EAGcWaimmkijsBcBYCA2IXr4/Bj9OP5Y/3j95fAP61Cq8coCzPBLM3r1KvW6/r AgIULAOJPARiAgqJ06QTGDYPwwS5OXNjIOIWxvCFxlJi7ZIBYQEi7zK/VLiuwZ/QB7jPXp8iPNMA Q9A91u2SAkPUxA8owW/Tb/oOcjxHBL+zHRQCNUPAKx89H08fYRvvsrYUFALl7zLJVLg8UmC9oCe5 ADbiN7U7sl3Rl+WMkqTlPNIIseV3cr7lAKcxy+X/ctjlAQD+4eXlEfLy5S/y/+VN8gAM9WHyGfV/ 8ib1nfIz9bvyAED17fJN9SkCWvVRAmf1gwLYI7aYAYH1ABqL+AAbBpj4ACKl9K7RVkI7sbWvAFpg 0K/cr+6vAL/rf/1/D44EinVbvS9stnOweL+KuogUqOqQ/rVhFXEPyBgmwQEPGUl5wCTPNs9IzHoC eFXxeBIBEfAAeoSRgP76AwNikABidACjYn0JFG2FDAU4foR7foaEGkZ3vwABxr94S/N4EBcQdQBi ex4YAKLftN/G39jeWMDs19lF/tElVdgSxQnvG+7pGMU17wBH5VBIXdG9/8//TyY7sV0tCP7/sk1X sSzHR0jyPLhXsQBXtRwFinUMJVsPcb9/D5EGBK6IapVNrg/AD9IP5ArWMUkDluKnMQFXsikZM36E f08b6LSBTmtEH4jRvJc2415sbc/kPyukAqBEMZ3PkxkCJkxtBemQdQYh8GADRdYB3JEfZU9VZAEV gpFKwnMfhR+XH6kfFAUKwx8nbqW83x/xHwMvapYAUuP9RoF2jZE0L0YvP59RnwB8LxcPKQ87D00P 3i9xDwI/9JUPzhN8D8Mb6O73HwAct8njPD9OP+ENyvsCABYU4QkZZBJqAENiwQkfGx8tH4eDHWCi CKY/XRYCEwBtH6VPt0/JTwC1H8cf2R8RXyNfDy8hLIPTAGJfdF+GX5hfql+8X85f4F/A8l8EbxZv KG86b0xoHvJ/9OwAH9Dc7mBv+HJvhG/zAjIAYkAAe2Liom5xAGLSt28EyW8xhS02HF+j+W/KfSZ/ ADh/Sn9cf25/gH+Sf6R/tn8Ai5+tofED5H/2fwiPGo8sjwA+j1CPYo90j4aPmI+qj7yP/kxoILDF 4wAh7cPx+OKP9I8Gn5ZlcgArYr6ia5GxYuQ5n0ufAF2fb5+Bn5OfpZ+3n8mf258A7Z//nxGvI681 ryIrejFZrwBrr32vj6+hr7Ovxa/Xr+mvAPuvDb8fvzG/Q7o4MldiWr8AbL9+v/QBAh9jE55tvL/O vwDgv/K/BM8WzyjPOs9Mz17PAHDPgs+Uz6bPuM9A09DP4s8A9M8G3xjfKt88307fYN9y3/CE35bf qN9DuiLd4s0fACPv8eTQ3+Lf9N/6GJjLomsuAGJMAANi3SfvOe9L713vb++B7wCT76Xvt+/J79vv 7e//7xH/BCP/IisNRP9W/2j/ev+M/wCe/7D/wv/U/+b/+P8KDxwPPMmP24Akxc2pRw9ZDyhrDxif qGUx+iKwng+wDwDCD9QP5g/4DwofHB8uH0AfQFIfZB92H4gfmh/Qew67HwDNH98f8R8DLxUvJy85 L0sv4F0vby+BL5MvR20l2t8/yAAm7O/ivi/QL3jiL+9CDjA/AGLHAD4PRQBi2xU/Jz85P0s/AF0/ bz+BP5M/pT+3P8k/2z8Q7T//PxFP0HsPMk9ET1ZPAGhPek+MT55PsE/CT9RP5k/4+E8KX7fdJ+Lp 8wCHKK3BNF9GX1hfZXOS5wBiZJsMp7JzAGIBvIxfnl+wX8Jf1F/mX/hfAApvHG8ub0BvUm9kb3Zv iG8CR6sQqW+7b81v32/xbwN/ABV/J385f0t/XX9vf4F/kYY9cc6TKe/x57JzrX8ov3/Rf96RMHFS 3gOE5ZcACo8cjy6PQI9Sj2SPdo+IjwCaj6yPvo/Qj+KP9I8Gn8XEAREgnzKfRJ9Wn2ifep+Mn4Ce n7Cfwp/Un+af+J+33Sr/fJzNACvg6PPgI681r0evScJ9oGsAYj2b7n5qAGLcA4+BrwCTr6Wvt6/J r9uv7a//rxG/QCO/Nb9Hv1m/a781+xKMvwCev7C/wr/Uv+a/+L8KzxzP4C7PQM9Sz2TPpS0s2+M/ 8gAtqL/hj8+hzzyzz+J1cQBi2Xdb9MAPdQBiuebP+M8K3xzfAC7fQN9S32Tfdt+I35rfrN8Qvt/Q 3+LfrBsTA+8V7yfvADnvS+9d72/vge+T76Xvt+/4ye/b76UtG+ju9wAHLq3Cq38X/yn/xMZu8AHh SP7/f2v/ff+P/6H/s/8Axf/X/+n/+/8NDx8PMQ9DDwRVD2cPFHoPjA+eD7APwg8A1A/mD/gPCh8c Hy4fQB9SH/5kHS/z9OwAMOjD7N19H48foR+uMi0A32JAAGLid1szAB9iRABi19Qf5h9DNT04ASwx 4OjZFC+QIf5+VD0AYjsAYtOgYEg4L0ovXC9uLm2AKRiALjSUL6YvuC/KL59GfSJiAfBgSKRhV6P2 Lwg/Gj8sPwA+P1A/Yj90P4Y/mD+qP7w/PM4/aRgy6u3gDVP1PwgHTxlPPGExMEJH/GdbUk88ZE92 TjP09/CLT51IHkRASQBi5a9PwU/TTwDlT/dPCV8bXy1fP19RX81vAheVFn5fkF+iX7Rfxl/YXwDq X/xfDm8gbzJvRG9Wb2QdezTzFnAi3eLNgW9Ik2+lb8iCOTZ67XQyMEIBy9hv6m/8bwojfWMaf+t/ AD5/UH9if3R/hn+Yf6p/vH8Qzn/gf8CfhbMXC48djy+PAEGPU49lj3ePiY+bj62Pv4/40Y/jj2Qd Nfr6+gAPNuns3w6fIJ8yn1WyIoGj69HLvGCoQNhln3efAImfm5+tn36v0Z/jn/WfB68AGa8rrz2v T69hr3Ovha90wQCbr62vv6/Rr+Ov9a8Hvxm/QCu/Pb9Pv2G/c7+OI3HT0w8aRne/obOcv66/wL8E tBRZ8HvytOrX+b8Lz3ZOAB/jNM+fRmPrXc9vz4HPk8+AfC+OL8nP28/tz9YvcQUZACXfN99J31vf bd9/35Hfo9/Atd/H39nf69/93w/tN0gheCLvNO9G760efGrvfO8Eju93TR0RL0HPKfHZ7+vvAP3v D/+fz7HPRf9X/2n/+c8kC9/pJBpuFKP8BLP/PCQBCM30TilzG3EZ+v8MDx4PAGES7vU9D08BYRFX D2kPj/X0CxE1BTN4RE8b6LQngU5rJh+YRhCc8hg26nqyh+v4nHJkHMdxhmMwrD9iP3Q2NQO2BOAz PyCRMKcA70F1GeQWqHr4OzN0Nv47MgHEPAICxDwDxDdWH2gfeh+MH3kkCi7/J8M1vMEf0x/lHy1G EjsyB2FGerJqLggOTyBPithIldhIe0pHrQ+2BjAFMypS84JHg0Z8AKcDekYQaUqqPHFfOvJrIgTE T9ZPiqFciKVYB0pHOPQGgRsAHi8wLwVpdQETbWIvdC+GLwCYL6ovvC/OL+Av8i8EPxY4AHvzJz85 P0s/XT8vf4E/kz8AXl+3P8k/2z/tP/8/mF+qXwA1T0dPWU9rT31Pj0+hT7NPANR/5n/pT/tPFl8f XzFfQ18AVV8vn3lfi1+Zj6uPwV/TXwBdkupf/F/Ollxh3J0rbz1vAE9vYW9zb4Vvl2+pb7tvzW8A 32/xbwN/FX8nf/mvS39dfwAon4F/k3+lf7d/yX9in3SfAP9/EY8jjzWPR49Zj2uPfY8Anr+wv7OP xY/gj+mP+48NnwAfn/nPQ59Vn2PPdc+Ln52bFRLp8hXj+AHc8AAAGPxsAVVG6/D/8vAD6/BE6PNm +vQAA/HyDg///+rxUB8N4Pv28eb1aOf0BOvwvSjr8AEAVBjr8BBQ3/xZAVkB5vUB6/ACQgi1kuP4 hIkCVUHo8whtQKsNwD+9Bb9J6POj8D/PBnwJ4/hQ4PvwqT/mD+H6AfXxIOnyt3SFBOrxT6UDEBBA 5vXMSB9aFgQEWh93HwHDkunyyIkE6vFIpQMIFNBxP6wfvhYBG/A/ATYUVQchFg/oEqRCCF6lAzri +f4gAgkE/TAGBxf2iAEEMYkBG2ATYv8AbwB4ACwAcv8AZQBjAHQAYb8AbgBnAGxJICw7AGJPIHMA aUsgFSAS6vFlhQSIARxeB+ENcSUq4vmD4/gKpQMFohTo86FpiQR1L4cv3/yU4/hYfAUnfAcQJwAw VSkytcUIHiHQBL8w3f4HAHwGNxWdEnX96fIg8/AAAO/p8v+Y/1DwAAAH4PAEDxYPKA86DwAAD/r/ /4yIBI////B7jABXAov/BfBfD4BxD4MPlQ+nD7kPyw/VA4wHAOL//arr8EQAcgBhAP9nACAAdABo AHVl+fBz/fBhAHD/8ldvAG778G/5+GTz8qd3AGkRAPjxcPXyZfsALuvwIABjAHVUAwAUAW0pAHr/ 8GT58JdiAHkfBmf38CoFYVQRAFgBc//wbP/wY/vwFWkPAiAFAm4hAHoBOgEDAAAAAAAAbWLj+Oi/ 6PPgP/T0/gME4L8b6LSBTm8b6L8cFQLgPxwDa+g/FAPg+/QAwDQLDfD79PC/lUjr8P/y8AXr9Ovw RCLr8AQCBPDzEQ0a7/QjCULk9wHr8PbxPQXq8Wjn9DUG6/AwPgJUGOvwaAFQ4vldAV0BOQkC6/AD AgRqRQWb4/j1jQJVQejzU/g/tw/ABPC/Bei/BXfgP1E2Cc8DFo0B2wR1kABgIOUFBf6m/AIFdZQA BhlCkQECTPbxQQEBBPbxBQEHRAKAlAFAE5ABQBNBAUAT6vEHVQAoEgMuFgU4FgFCFoCDElASgxJa EoMSZBJNAoSySggdsQM1DQAMjQECv4BMAABA/urxo5SNBOrx1rED/fLw9xH+9PwUNQuJjQEBYD9D /wBvAHAAeQBy7wBpAGdUAHQAIL8AKABjACkuIDKuYAAwADMuIE0mIGMuJCBvAHMeIGYsIh0hqXIg IFkhYSwgaR4gbvsALi4gIABBAGyqdCAgJChzeCJlTCBl6iQgdoogZGwgAAD+3S+NAmASViYgcwBf XwBTAGJiIC4yIP9oAG0AIQAjAH80ADUANwA4vyAEmyDq8aCRBOrxaAHAHMcaOuvwpAYGAAAt3i8D I5sJBAMn8D85CQkCv6JCCCYNP0wGNwoQ5/RsBD4EXQdVaA96D4wP6PMVIVjm9ZQBwBEBB+rxikoI xVSxCgByIDoGeTECACoCEmJDAP4RQgEYEhxGKQNEAvbxCngyA4MSO0JCTQKL5RjIIbQPDE4BHk8A 3xJVQdUlYU/UBHpPIUmIQQBRRQQ1rEsERsNNHU3jSUUFqTL4TzsFD7kyckEBCtcAwP5dUgFjUwHA Ut8Scbk09vEsYgcU3/woN0WwPwI3meP4Vq1K4gYo3BDnXDEx/AQiAdQQ/AKAF1MHGIxPyFNNBdZb xwbgBT/4Vb/vVgFpx038AiZDAAcYxlXoFdZf1ATvXkQwA28AFW8mXcpR1SWwawRGWmfIZyBubxhX im+caQQ1XLFvVmoA9l/ibwkbaX/r8CgS5/R5Mf5lAD3b1kr8GjD/SIaqHf6BmggD+2cCCJQxZQQq jwI36vEVyeP4EbkyVEEBvBG8EdLVIrTSKF1TVAcUJkACt3EBZDLV2OvwSOvwhOvwPQD3LQAC6fLB y4Qf9Q/r8P8MAAAAnfT/blqfVeSjAAD9Af30AAAiSo9Mr03w0CXi+Qbn9Eb+6/BcIjICqxMA/wBB AQAA1gAj/uvw/KkQAuwUAOsASj8CCuvwREgOvwI2FQAAemEG5K+QAAKwbwAEZwEAfSvr8LTFNAK0 bwD5Mz8C8vGkHjIC5wZvABcDUAGtAZUe6/AC6/AE8/Tp8h8EAATy8QBVRuvw//LwBOvwROjzZvr0 AALx8g4P///q8VAfDeD79vHm9Wjn9APr8L0k6/ABAFQY6/AMkN/8WQFZAeb1AQYC6POS2uP4lQYC VT/o8whAtqcNwD+5Bb9J6PPwUT/LBkUJ4/hQ4PvwpwR0Ch+BBfdDACAAEQYBLQO5BhP+6vHIhQTq 8SVIoQrQwQVaH9CnBAMV1/A/AUMUB4EF/w9KjxKkiQh8oQPi+f4gAscJBP0wBgMXiAEET/4GARtg ImwAaQD/bgBlACwAYwC5b+oQ6xFjAHTuHm8rAHLuEHLyEHX8EO0RX2IAYQBz6BBjuxES6vFvhQSI ARxeB90NNiUK4vmo4/hmoQPGEcYRMgNtAYgAAQHGFQH8WQCdHQYCYAlDAi68EAsmBgEBYLsRoSIK py+5IkLq8WWJBDovTC/f/IPj+CUKoQMFSRTo82lWBPIvlAQ/3/yU4/hYrBdFBxCvJwAwVaYyxQQe IRbMBL8w3f4HRQZEFYwC66QB6vEg8/AAAJb+7/KN/16Af////79///4Df//8+gsA+AsA8AAAB/T/ A3/39Ad/9/e1nyAAHyAEgAAtA/99fDUMcBf3/2BJAt4WAf//YB9WAD//3/99g/8BYAEBwf//AfiD eAF/i/f/AfdhAI+K/wb/8zgzM/9/iv//gwADA/+Pif//B/MP//8D/3/vif8Fh44AB4eH/wGPg/8H fw/53pABhf8Bf6AAjwCfmZAP/4+qAJ8C/5n5XQCpBv8AjgC3B/D+wgcDj//wgzMBz4ODM4K5AhMB hgCTBAPDAHUBhcsBkAD/gp8AAakAvgAMEQIfFB+P3/8zkzM/LBbwCfeZAD8eFvD/n/D+RxGHeAZw ///w5XiTAAZeENwAiv8Cf3/wgwCL/wQmEO+PjP8EBAB/jP8Ng50AtP/9wOvwRAByAGEA/2cAIABv AG4AdXT78CD/8GgAZfnwXXD18mUALAMGbvnwcWTz9gYFHgFkAHMDAr4CAWIAbAB1CQJ4pwAZIDwB /PEgOwBoUvXwcAkAPAEo8/BlIwB3IABpNwJpAGP18FV0ZQRj+/JuCQBj//BdafvyKQAu6/BsdQIN ZZ8CAACtYuP44L/09D/8/L//G+i0gU4b4L8FHBUDPxwFFAX0/wYLlUjr8P/y8AXr9OvwRIrr8ATr 8ALp8vLxEQ0HCO/0Iwnk9wHr8PbxPQXq8dVo5/QJ6/A8PgJUGILr8CTf/F0BXQE5CQkBA6gCBPbx IwEG6/AI5/SbeuP4RkEAAgBVQOjzq/g/ww3w1Aa/wwTgbssF4L9QNgoCHAYCf3UKAABgdQwQEP8D 5GICAAaAQe8AAED+CBIBdQsqEBENFRwXBgECExIOEkUEOhgDLxIqElAUGpEBGQL28UEBAQT28RQQ QABtGwYBBHUJAGAg6AVVBWgYA3QWBX4WAYgSUwV1lACSHhUGAQb5BloeEw4GAQdh5hkI8xWK6vGd SgiOBgLCBdwGAD1AKS3wPwEPbhJEAiR4EksnAm4cVS0BdBZ/JxEDoxyJKOrxnAYGngC9Cv7uCDJy HMdxHMe/4T9An/RJ4SDPtT/XJNHfJb8/+QpAUwAWCROPG/4KMgHDHKv+KkETTBAQgUEBLd4gEGIF AHLq8QMHB8AFgQkBXQBsGYISqBJqghIECjIDPDJ1HxAQk4AuIBEKMgSFMhU4EDSkFoISBQoyBXWT EBQ5AurxoJEE6vFoAcAATAZICJcBL03p8qMCBOrx1pIaI/3y8BJB/hdENQsv/gYCYBJWAGkAc/8A XwBTAGIAYe8ALgBjVABtACH/ACMANAA1ADh7ADBXQAAA/oklEv9gP0MAbwBwAN15TDBpAGdUAHQA VyAAKEdAKXhAMldCVTN4QE05QGNMMG87QCdvAGZ2QmdBcmpAo0FVYXZAaWhAbkVAIHhAV0EAbL5A IG5Ic8JCVWU7QGVMMHbUQGRFQMRcQOrxpEIIRjC/AQgCeAkBCQMaQQIACQQeR7vwPzkGCCMEA0Kn NKEE6vETDEcBADAQM0FJAVpC6vGEnQTq8SvdNKDe/XonghJlOOrxv6UIJgBnX0sHOAmTFL5APwOU AWwBBwEAVWgBghHi+ZQB1DXgiAnq8V9B+FnAAAEBD4rq8YpKCDQaKiAQ4BUBCQCmMQ40YkMApTIk NGhiAurxi7MoQmvcCFtrVCBsagS0YerxaQYEQQEIEWUEA0WEOQm9XZTj+FUBalTjZxA3JwAy2SJx vB5W5vVRId8FWkHp8ukhRQnjbRAmcE8yhFfq8ULROKYxZQD/MpBVEDaXpk//laP7tkyTQLyBZwIE BQFoAWUEv3/ha8kq4/gRBgJUQQEd9HStIuG00TgxYfBwkxQmQPwARFG9UtXY6/BI6/CE6/C4APct AALp8mLThB/1H+vw/wwAAACvOP9C+FFf1+oAAPcBAB3+8wCEh3JfQjYp4HHi+Qbn9P1G6/CsHDIC rBn/AAB1AQAA1gD9I+vwBKsQAiEbdwAAN08ARgAK6/D/LGMEAlgcAACRhz8CZgHq8d9vAOrxRLsA K/PxxjQCgAEi/D8C8vHkMTICAR0PAACUA1ABrQEAAAQAAAAdAAAAAAAAAAAABAAAAKV06/A07/Tq 8QL79AH/paxldZ7TIkD/exSuR+F6EEDo4/j+9erxC+vw5HMCvwIyIQAACuvwQT4tAgR1AgI8Nwbq 8QEB3PQAAB0AAAAfAAAAAAAAAAAABAAAAKV06/A07/Tq8QL79AH/paxldZ7TIkD/exSuR+F6EEDo 4/j+9erxC+vw9HUCvwKMIQAACuvwQT4tAhRrAgKWNwbq8QEB3PQUAAAAQgAAAAAAAAAAAAAAAAAA AAMAAAAAAAAAQAAAACQiMgJGIQAARgAAAFIAAAAAAAAAAAAAAAAAAAAAAAAAQAAAAMxbNAKgIQAA RgAAAFIAAAAAAAVI6/A83P8EDxYPKA4FSOvwPNz/BA8WDygOFAAAAEIAAAAAAAAAAAAAAAAAAAAD AAAAAAAAAEUAAAA0JTICPCIAAA0AAABSAAAAAAAAAAAAAAAAAAAAAAAAAEUAAAAEXDQCSSIAAA0A AABSAAAAAACVquvwKOvwdtz/4foC6uvwBef0HuvwtCEyvwL+GAAAm+vwU/sAyevwbLgqApmvGQAA E+vwQiEC1P8ZMgKVIAAAnfovAj/r8FxNGALm7yEAAFLr8FAARH7r8MQkMgJWImIFABIFgwEUAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAU06/Ao3P8EDxYMVRXr8P/y8Dzr8GDo8/z69PHyAACHGQIwBuD78L/c/+D78vEJA0wP UwAA9vHm9Ubn9ATr8J1E6/ABAFR5AerxA4BCBpANBQekB+D7dQHm9WjugQYAACTr8AIAVIUY6/AM 3/zYAdgB5vUB1N0C6POS4/jQhQJVQf+lrGV1ntMiQP9BexSuR+F6EFtAQejzwD84Fb84FM3wQBXw PwcRXx8AUPITDD9hH+H6Af8DAC8EPABMhQEN2AELEVcBARuxFiexFg2xFpURsRY2sRY3sRbq8ZuU BBTq8ZEgE0Dm9Q0t8PE/Hy0UKHMZAc8DELiFAQsSBxEBBBbdAQRtdQsQYCDm9QX+UiI0+wJcIgVi IgV12ABsKYQ7EggYkCATBRIHETEkwJQnJcUm0M8u2OEuxShzBVUHMeUFNpEfJz8DEwsVEwBvWSXr 8BzdB8QJWCYq3/yo4/hmIBP98vCXMcyxA/QRAQGXMQMS/At3AAAd3QJgCUNIMH1uwzBlAGMAdEgw dXLr8P61MAACAfrxRf64Mgq+P9Ay6vFlCBQVH+vwlN0HfN/8CUEJQaj/DdgBdQEF6/AG6/AHKuvw COvwCevwCuvw1TEU5wHOEQ7r8A/r8FIh2BGVEuvwE+vwFOv0YiEXFOvw4wEZ6/Aa6/C6EVUxCrgx Huf0g+P4XEEjEAcRAUjl9rRBXBXARNgBy0sLFQDARHUBy0v7BcBETEHLS8QFAMBEUEHLS0hBvEhc QctLTEEAvEhYQctLUEG8SJBBy0tUQQC8SLoRy0tYQbxIVTHLS1xBILxIbSHLS9UxvEghyE7nBSLA RCXITs4RvEgmyE5sQSC8SMQRy0twQbxIKMhOUiEivEgpyE7YEbxIKshOfEECvEgryE6AQbxICxHL S4RBALxIzhHLS+7xvEjYEctLYiEgvEh8QctLkEG8SDHITuMBArxINchOmEG8SOIRy0ucQSC8SOwR y0u6EbxIOMhOVTEivEg5yE64MbxIOshOrEWqwEQ7yErJ+AiKBxACQwBQYiF1AXUBBxERrEEkrEEL EREoRdgBEW0hbSFidQER3AHcAVhBES7r8IwLkc4RES/r8BiR2BERiBIBEgF8QRFTYVNhkEERiHRh dGG6ERHEEcQRVTERiLZhtmFtIRHXYddhMmERiPhh+GFTYREZcRlxdGERMSzr8I2RxBERLevwmpEy GXERvnGnlREy6/C0kcbfcREz6/DBkeIRETQI6/DOkewREd9x33FjgellAO6fAK8SrySvNq/n9MtB BxEAS1V9BHUBn1NcFfLxSEGQDwBFDVSl6vF+VdAFSEGNkeAEgEx1biZaWDRP6PP6Eeb1Rf4HJMIS ySzDcQr/QEC8HoXrUbhn5j9AhmMmIJiZMrF16Sa14C+12T9R46nKp2OE+xhTryXkqwBCiHkiuhHO ggKItoey2BEBgQzEEc6DprbDEqCyI2Kk0AgYmpFutlwT/goCCQSk57N+GwH0guvwhvgIPbw0dLEB HLfJ45ZASYN5wKIQsAIVNhU2cx4Bal1ThcEIIyATMLggskueP3WCATEmCxEC8xSgaURcFQuR4ADm 9UMNsEz+q8AgAEEAdQBkqwBpSDAKDbBuyTBvCsLAZc8wCsTAshRSP2Q/yOD7DUHm9Vhrt8QHECcb ADIHMlXFFT4yYR4CahIG6bEFCcQGAYDj+EFx1QS/r9Gv46/1qpnj+AASAQohOrYOL3wX77fM1VwV KWTd2i622Scmv+zWB+kCnEEDr7evuFriwxKqvczVAAsV3N4xthctIgAJ7/sFI+Vg7dY07evXTu+I vwEDdOMCIv8B+wLQ0cQFnu816/zfAuX2Z8uI4wHgBLH/BNuzgRwv1dgRAwBUBxG4MbgxgPsCtmFd VWEB4ABcHy/V+LYVNPA/h6///wcRbDhJRExBJZECAFXjAa79kEkFtN9CRzPRiewOcLWKEPsYhEEK Ly7WiwgYugUotgDLDxtV5Q4xtdMH8sW6DiEfAF1Vug9DrqYhWK+eVKYlea8Ai6+dr6+vj9+h33MP 968Jun+UvJfJFGwaHb8AL79Bv1O/Zb93vqCxjLO9JiC8IqCzAIHOg9smAcmSv78A0b/jv/W/B88Z zyvPPc9Pz0Bhz3PPhc+Xz6nPu81E1MAAz8/hz7f/Bd8X3ynfO99N3wBf33Hfg9/8Hw4vIC/L393f AO/fhf8T7xIfN+8L/w+y2y8AcuWVX4zvYP+w78Lv1O88HwD4731fWuO9LzT1V29O/9NfAHL/NV+W /6j/uv/M/97/8P8AAgTLVUV/Iw9kectRRQ9XDwBpD3sPjQ+fD7EPbx/VD+cPAPkPFY8dHy8fQR9T H2+Pdx+AvkWRHBJjvoHLUa4f8YsQAC5K5UXLVewV+CHOkSBEHEEg1UmgEfgi0Rx+cQigQUKA+Qow L7lEVUDzSiSnswzHbCBUJjOakvNXP0CaZyLJmJXjoZXTuT99OTdAFx1BAnXfDAAAYHVskGAEW4BB KFD+FR1BA6KW1N2TaEGdTViYnIVfxO//ZU/Kj5gTvyvv8x999o+XAQMpws2ScGiRvoGgEDunAQCd gPggQHoz35E1ogOYO69NrgORAwK+gTdCAlCHor2CkaNoQaeieBMPszX9PgINHUJgAZVh95IkkyHc FTWgL2dCJIOyuXWBAYyilrO+gfMrZJFoQVdGRCMIAsQwNbAYNQIhP5z2Z+OSC3RO1JC+gXqFc9+W DGeyaJFws2hBnNRllD8pNK2VkraUMnL/HMdxHMfhP0BXn/RJvrDPs7XRvLUBv7ybEmOVMWyVTTSo NpFy4rY5DcMx3pBZ4gIBYlsAACfDeoMoUIBcMID2kytxCQUCBBJlWc8uNfANv/GP//++gXpxtXX5 oYCKdNOQ1ErVc6d/xX9FIgEIjCPjfyBuuTKgCo+duaCQAAjb0G/ib/RvBn8Yf4QSvXWAxo+ZFN+P 0t8DnxWfJ5M4AB1HOjRjRmSRTJ9en8tVdJ//VUCxWMWMuRb+YyCRwvUoXI/ySqGYzIni7LOXiOTc vJ2y45IC6ZzMkgN1xKBgIXX4INuV+J9FI61v5VQmNbVv5cR54cL5MqfMkQhElCH3Ua8DIf92r7gi RdULAQOHogFz92hB/VDKVgEm4q85Tzcw5darr72vz68A4a/zqZwRW2MiBnLyF78pvwA7v02/X7qc EXC+zJGFvz8pBqy1QNCc5LS/xr/Yv+q6/qE1TxvotIFOax8/GkZ3vwvPooU/zwFUJ01iz3TPxdGg 0Y7PoM8Ass/Ez9bP6M9GIgjVooYQ3wQALs8D7DrfTN+k/3Dfgt8AlNl+caDRqN/K397SN0HWKQBr YcQbJ0VQVSGVy1E6MSBE0EgFuDtrL1rpkb8ESd5/y2QKtiFAQHjiP8LxP0BgZnIxl5ApoOVVcTPW eDXZ+gauQW8BzwMWHUEEdctQvWAa5OA/Bf6jMgVBdb6ArTmHo35xN0IEbfMA1TblQvjxUFUic6P6 wlKuEQ8CAoBM2uV6ccJRgn8QlH/DH7h/2x8A7h+igZ+PRMDyozIAqTnfQLgzAb85QMlBGy2cRXE2 rU+rN/C2NQDSTzaM8056NQxfHl9xV2CNAJxOW1+/T39fi42cTwVfxV+gKV9CgE2BlfUzQ1Qi5U+/ AHUAdABwX2IK+wBBX2BkAGkAb/pN4FNxYGcAbgBhAnlwCnFgvIN/ATiRWiRbHZSRoq8clENIWNZ3 WxcQLycAMlXmYsV2BiLlESF3FTNBBuLpbgRoAHcGAJwxson88a0vvy/RL+Mv9S/gBz8ZPys/PT9P PZK8l//JFGwRQEAA1/+jcD0K1z9AlYSzUm4waHI1/HPeAAWD1gCTP6U/tz/JOfzx2Tlghu4/AABP Ek8kTzZPSE9aT2xPfk8AkE/+XxBvIm8vX+pO/HY4nwBtX1yfOl+AnAWFl5+pn95fAJlfMZTmn0qf Cq/oXymfkp+gUq+2n0JqiqFTaENzYG2+Y2ByAGUAc/SgZXRvYGprQnFgdAAt9KAFdPCiYeyghG+W b6hvum8AzG/eb/BvAn8UfyZ7tgH88QA8fPzztgXF3wLLMO/VhZN5AKDRSBFaJOXXNMdbGemJVu3+ vgUavyvzH33+QP3oQ2EAoTIkgDCfw+Ax1QfZP1Feye8SfYpm5JGOGWDgMvGFAQL/xv7COgKTDBtL Jh3WAR3ROgMlREsmO9YBO9E6C5X1LeLJAB+TawNpD3sDGuQOEUJVGhWNuADynj+KgS4aWCEBqB+V Q2EflSpQa1AMsmNyMcAK9dDxoS0AZnFgBWxhYGXwoB+zlfEpvzu/AE2/X79xv4O/lb+nv7m/4cMC 94MsVycAIbIZ+LH5heWPAqXjmc6IvoHCAbHGi64p9JB1EYTXCfUflWQa+qXG2UYPJfC/KfZE+XVx AybXACbYl/Iw0iHdCfUjJWD7t8YEgDAz/b+L/zvfr/UYD8n3gEJVYPU0pnH9DycGD//PAQEDsfOT DxkjDfFbFRn+cfwAOf/QhkovXC9uL4Avki89wqL5hTwZE+y1gR/4j9TwDT8Cz///z7GygdmFvo8A 0I/y7/SPBpyl4xuZ2eGHrwDUFnKZKyWfxjwvQlVWLqjFAEQnh3UrLpIv/RUrL7F6PHEBCjV/R39Z f2t/fX+Pf6F/+LN/xX/Xf0BgyrRU+7+KAVD2KFyPwv31pcBaNBLwzasCrsBwU1XCM3fwyzMdjy+P AEGPU4r6IWSJJkZ5j+LPnY8Ar4/FH9cf6R/7HwmfG5/dLwCbr62vup91nsI2/k/4nyJfAMWfRlzL NV1fb19prySv90QArF8QX9Bfc6/vT1hfGG98X7rNqi7bq0kAbu6gdVQP4P2rUwSwZ7BgYXdAAB+/ KO8670zvXu9w74LvlO8Epu/Buw28RAI/gHSIMfS/gLZ7ScczMb9FiDVdNVjRNAAZF+FheDkiwVjS F8yfQXQx8M+x+iGYPeHEQMfvyn/8R58TQEBevDC/KFz/P0BgcvWaIr4iyV2FboBog7nkNl9EtRdp QnVoQGB1+iBgVwSAQXBBFRkRA2eG1KKDZEGd4kjqhsT8It/+L3vSElSIkrzfl8kUbBRUhwEAjSlp QnArtEQiwACXATsAgFjQQHozcEH6ghEDAJ8SnxkScE1xHsRSl4I/nwFSn2SeIEODcfxiAlAgQoJy qJNkQaegSBOGwxn9MDKgc2ABhEE4YkRBMCLFj9FUTJtmAwODcRATghamARamo5K7B13RdDVXjobD CAICa0CIM4TRAjiB315HqIILdE6+UE1xC3pzpIYMiKIqkZGjZEGpnOm4hcVwaYWge4Qy/3Icx3Ec x+E/r0Cf9EnfoM/UpdFC3aW/gYul06HRMYU9rtzEj0LC2Q3P0aOAlvICAbdiAABIs3qDV2CAAeNw RT7B5UVzFINxdr+K2Rvwv7Z///+DcXhBs0UBJIVHaEF4OdNDpU/DT4fCAQHbT9cuzaW/A0ovrjBy 9QApywwfHh8wH0IfVB+KclPRAId/EDRT0aR/88/If9p/7H9A/n8QjyKPlz6SoIQixk6C7xdAQMZZ gI/C8Tc/QIBMy+m/nNTKprPZv4KPlI9A/s/SA411g3BgdU/QoIW9jQr0MzCFIfrWghZAQC72uzEo +OWFGEBAvIAhsglhI5SzRTKRQeZonwKAQO9S702TKpFH1JLnf+8BQJLvjZ+fkqXBqJnk7APk5gUD IEIBBPe+n9Cf4p/0nwAGraXBZsN29hPysQN29j2vAE+vYa9zr4WloqeXqIgxpq9suKnrBjgzLgHj v9WvAOev+a8Lvx2/L79Bv1O/bBsApcHWD4m/+Q2lwa+1u0HAAQCIRAbXYnO7Qexpr0nuvwDPEBLP VWsyz30dqKfUlRd5LQGQTMW4G2LPdM+Gz5jPqs8CyMISTfTJzFDzUCETNerPAIMnUd+bRVk5Fxdv MiYfOB/4ShmcObylF6uYMdfLAiAu50CUYcZRQJh0p9MeNeAnNdk/Ud8pUNxCQ0pFhFdOYDFCNium mIA2fzLk4gw38haeNgFMnjHk4gs5K6a8NgO8MQiYM571PFUtYznlQ8nzx/8QX0PBVGwB2rUjeQxQ QYwKov8RAeVFlGHlRSyha1CuwGB0AGMj0Ap2QG+qgtB06TBmwGBstmBlAwBy0mQ8Ud1v728BfxN/ ACV/N39Jf1t/bX9Fgz9jaESA/RF8EbZJeSFBZeYtL1WZAM8oHFGrETI2hB6sVNPxBUcEjFXEBWSd WiY2ZYAUVaxWBMdZogEDpzenOBpisTKiPUCMVZhl41s4Ni8wtl2/Dm8GvD8BAzRjm28c85BR2rUA 41WEFvRdq1eJb4A/rWUWfwDHZy9VnF70XLxf0SbxHwMvABUvJy85L78CFNVstG0lBI81+O4FP4Mv //9QIQ8RAKshqyEcUR8Uon09hXRf2CMAFNVWH2g1cB+CHydUmx/pWAC/j8odID7Hhy9Vro4VnxTV AK6PtUrJwc/RVC9P9M/RcS8Ar5+VL6cvtbmtQsa/2L/iNf76L1VAtVjFjLn9FiYwFa5H4XoUQwBA C59bf0M/VTpTYz8y9zJC/+LaJQECqaaoohCYMxtBk3PHpgEbQbYz2j+A7D/+PxBPXElstG9PgUFh TpZAYQBtlECTRUWGQhVpW7BhkkBvlk+oT7pPAMxP3k/wTwJfFF8mXzhfSl8AXF9uX4Bf0Cdebq1f d38YZQCpr61lus/HZ8QFnFvxXwNvAM5aVm9ob3pvEWzHrzJlfN8ATGcLpZxV4W/zbyTfS39dfwCY z4F/k39R77d/yX/bf+19AJqVse8Xj9DpmpE5j0uPXY8Ab4+Bj5OPpY9jn8mP24/tj4CB/xGfI581 n0ef2/9rnxsATsSFnGYDKgHtoaKfXQvKJyA7IVHF7aWzJeShNEnniLEgl7nQkeSixZyrIQgMwXzg +Qocr2E1QCd5L5NXKdgEFThwCtWaZaJxyQQVFRAPE7k/UBL7bpqRAgJ12TBgddgAb2AEgEG1EP4V SeFRAw4WSRMLwZ0uuOrTFb/flfmPPgMVOL6/Esksw3EGFTcBCwApUBJwOyFB8dCQpxd3AQCA5KBA ejNLEYahEgBw1AHMlNAXvR8BwNAf4h85EqcfCy7KIQMChCoBo7ICRSIpAk8jC8Gn6tWYE93z/U8A AQANBkniYAH9sWMSVjHQleOxhI+sCsIupKIqAUDjvSYBQL0mSiI+d+mh0AELwVfd80cIAgI7EO2j ELECDb+c/adPEgt0Tngw1AF6hXNLFgwvMobhODMLwZzk0QQrqVgQFVKjuT8y/3Icx3Ecx+E/r0Cf 9EmGMM97NdFChDW/KBtBAyJB2AU9L0zET+JDSQ1QQUoQGWICAbdiAADvM3qD6vCAgeNgFZfhMbwq AR1PFrnwDb9dD///KgHm4SH17JNg+OL3nwmv7aU1/1UBeKMMT//i/rk/QnN6/2k1D9AA0Es8707v YO9y74TvhJKIsQCBn8+UpAVPD7Sfxp/Yn+qf8HJPhE8gr8JCHL8r878fff4/QK9Coa6BCUevWa9r r32r7AQoXmBEpKKW1gKW1pNiwaMl1caSsmYDsmHBoyYKVtBmAQDQYd+v8a8DvxW/J785v0u/rl24 UQB1cbBua7J6AGG0fL+Ov6C/sr/Ev9a/6L8A+r8Mzx7PMM9CzxT/Zs863wCKzyDvrsmyb8zFw4/m zz/bAAbfGN9x2NBvjtU+n6jX9mUAAe9R32PfdYb2VT/Vwt/U3wBr35VvzcQAr++H+d8L76KPAC/v Dl+PrzJfRF+J75vspA0Auu/M741RkXHm7/jvCv8c/wCST0D/tk8QD9pPiP+a/76/AL7/0P/i//T/ GM8YDy9xLg8AsdQvcUsPmc9vD4EPkw+lD+C3D8kPCW45MBzC5Yv4/7/sSfM/QHwUKa4sYJQQgBOV nD5lQ9OEFRBM09koHzofGSB10gONdYIgYHXtYEcVZB0K/MOgHcG1WMWMuRaqOmBILWFHghAVI2L2 DClnnxQAcHXSKCTn1g8vAQHm3/jfHSORcSclOeYl7wEDOO80L0YiL3FPKYrsk2ICrHIDRSIZoq7j ZS93L4kvAJsvrS2eZv+XGa/iL/QvBj8AGD8qN3XROD6RcU0/XzkSg1+gvEAsM9Tx43s/6I0/nz+x OyPANE8b6P+0gU5rPxpGdxG/0z9LxQdPVC90ZQ8zTwCFCi9xVkXAYdShM7SsxxKDAMBhnHlauW6/ p08+z1DK2E8QGc8+ZCEX5r1880tKFXmvAIuvna+vr8Gvb1L9cWxfulQA/XSLz55fsF/CX9Rf5l/I D/wKbzUT+yL+L3vS/xBAQF6PwvUos1z/RGU6YNjMsiHsBERosSPcVm9ob3pvjGVYlpECWJYHMqxj J96GJjYDICYxrGPh4YETRDYB4eHob0D6bwx/Hn8wf0hxRnhCUnLVLSjQdIBwZWJyCgDVTXJwbHhy cCKwZQBdeH5yIABv7jJK8jBdaXZyIABD8jBkfn8AkH+if7R/xn/Yf+p//H8OjwAgjzKPRI9Wj2iP eo+Mj56PArCPACY/1YVPX++PAZ7DJQAanz1fRD9Qlcpfap8SXrEmAJafLZ8JP9aERW97V7uVPqUA TMaeXeSfr5k2r7wsK19prwB/H5Yfox+1H8Ov1a/5FXp/AAG/mXmJQSO/Nb9Hv1m/2w8Afb//D03P Ix/Fv9e/So+tKUAHzySOi488zy6P4RQlaM+AesfzgYjPJZ+sz77P0MM4gBJ3Xw0N0fXPB996L+gm QK0bztL+P54nmlbVNCrU8vvRlbnaletl3YjyoQLSlobfmN8bhuo0FBr6wpXEoiCPwvE/QAEcwpU6 1d7UN+PtxHSnTu8YYON5rybuAHAR0RTlx6YCs68Bxq9z7xfy84Gu6Ri3AKBBMmBpNhIxAv48TRhA OJYAyO/a7+zv/u8Q+fOBCXPHtgAnsjTzx7ZD/1X/Z/95/4v1UBHRl/+p/4csluaVLe+UANv/7f// /xEPIw81BoywQQAAAjFIDxNy84EN33UPMN3zgQCbBUcnwXJWL2gvajX+f/QPQs6HoOaVGR9HjbfW Kt2UAHcfEX9KvzV/R3/RGBIx94+AcYQSMRSfiu84n0qff5M0AAbnnqXaRS8hgZ+Tn34tWMD+NhLE Esksw3EK7UBjpkDIsiL8P0DFlEUTvzP0tyA989m/bPacGqICdV3QYHXBwMAppRqvLK++wU9wNxGS 3/+V+Y8+A0BAr15noBSuCUCVJ0q38YNHA2ulqKMrJNf33K8BgNf/6f9vo4lB9KUqBhYPAyApDwG/ NBERMjhCAiKyRGIAfwgHMhy5mwdvv4G/k7+lvwC3vRI2RGdeb+y//r8QzyLPQDTHU8dIyIlBV8+K Kc6yIg30pVWwPIbPmM+qz7zPQM7P4MbTUOzA14X3z1QSMQBTLyDfdi0SMUbVXVFgcb90AJ3nnkNd UShJ5nmF35ffyo8A3IrI3/klGmI58Qo/co0sNgGcPvTz3wXvF+8p7zvvTegAaVFc73GEaVF57/8/ ne+v78Bcn26f5e/376SfF/gvVq8xY64FxPDDU6WIAFDzO/arRdmQT/T1nWb/eP/0iv+c/cUj8g5A QLx/HoXrUbjmP733HlSh9Sj4P2ykUQOipSBRVj8PKANVX7SvcCyy9aQEo1aPXwGiX2QPdgLqMX8J AvRcA/RWIbQUZzK/8Q9WvwBov8AP0g/kD/YPD2YRE8NmACNiJRPDZjQfRh9YH2ofNsUgT3GIH5of rBytRqoNpMwfqN4f8B8CKj0QJLgvUp6xP4EyJC4IUAAa6qJi2wAA9XN6g46ggOMECVU9L1TqMSOP Zi9GjeoxAIwvni+wL8Iv1C/mL/gvNT8g0t2tRtqPRz2tRpi3RGg/AHo/Rm+eP7A/TehogeVh1T8A 9z880kaRypnvcYmLh4U5pQAwRTZhR0FOlPNlJic0Z2WZ+odJkcqExjAbuyX18UzwrbbyC0BAYP9N PCsaCfA/QEbzY+A/ZKR0oPNj0KV23YBDkc8DFkuRBHV6NmBgR0TgPwX+l6LDBXVt0KGpdQMNhQEE AJUDyaYgQplh+JXAM0Rq+JIeXdECAoBMCVVG0UaRAHNhkGFshFGdc2FzZY+N5WEMIEWjj1UAt4/D Ec2PQ5Iil6IAnakeQKyjAbOpvbEg5I2QtWWmob+fp/Cqpca/ABCc575upQDPEs9lxw+Vh4UAkL5P z7O/c88PlSBFkL/5vyi5zx3PfpAOM2gqS5N7hX9QAGkAdABjL0DdClHQYQByX9BtAFVlVdBlYdBz ekBpkZRARpENAU6Uh4lllVNtlGu4+Vj0p4eHECcAMlVi1NLFIXZ7hXpU8L+QYQJSQukZdBNwBZaQ oXuFpZEA+vGhn7OfxZ/Xn+mf+58Nr/gfrzGvQ63i8mSE1rVAsvYBY6hAM/Th4wR1qPPj04evma+r r72p+vEAzalO9uKv9K8Gvxi/Kr88vwBOv2C/cr+Ev/LPBN8W3yPPAN6/8L8wD2vPJs9mD+7qhQ8A lw/Sz43P04TUDzgP+A/cz6AXD4APQB+kDzbaPkTbUesAdV/QblXQaQB6AmfQZFvSVN9m33jfit+c 3wCu38Df0t/k3/bfCO/FNUPRACbvOO9K71zvbu+A75LvpO/0tu9DrbwmUgpAQEm+4+FHDUBANPTi 8w8/QMDMjTFZoIMz+eACjDPcDP8e/zD/valD0c2pAOc2Z/95/4v/nf+v/8H/0/8A5f/3/wkPdx+J H5sfqA9jDgCDNr9P5g/jT7MPB1yMNR5fADBfVx8SH9OEbV/RT5FfYR+AsE8ZX9lfPV822lQhR9hD qwBv/hBwBiBlCCBzUuQeIPQfBiRC7BItCCABdHdi/RELLx0vLy9BL1MvoGUvdy+JL5svFOsrfpAH AMAvcnFsc9QlAEHaL+YjZksQhXNiRx41DpEkS5cpQV2Z6K+DZk9ZPUULZPoi/v8ve9IWQEBejx/C 9Shc//HliTnz5iKeNVH9ebiCBEpxEk8WMhVg1KINP5UCmoaZgoGRIwEMgZFCk7iGAYGRsoITCzU/ ltaGAdaB0INcQaWkRjgtfYkWM/6wcgmBBBeThVMbZ8A0N9EWNV6oyRuqb7xhRAAgLf4QdXQ4QN8R cDhAZQB4BCJVCnFgcikwSnFgadwSFSBskmNxYGSElMBhSTEAyG/ab+xv/m8QfyJ/NH9GfyhYf3yD fkMs3ncUxZwAQSiARWxPWkOZiUhkC2Tz5iASbiEUKZEfl/whA8GHwYgA5aLLgryNnqUWNUVBsqeJ Ng3cFFXwvyEW2amepUk1CK6ujDbIrb/Zr9aP/aWtvwAXt/x1rqW7XEC3FFebv5qPEwED/6MozwHS IqKhq1UQKbubOMevVL1nBrZOQ+F05LfP1pvJiUiygQMAVFC9cRFBEUOvgCh9SDzJE7SLdRbf+BSk 8D+hf/8B/25xOUFgRUVPV0+Hr3tPAI1MbNOiSW6hDm88NvlJwNUAOobR3/x1696MNdnXq1XA3iAn 75LVwN/bmsMhMW9/gXcAj+GPf8Hvs3/Ff9d/6X/7f/wNjx+Gx+/K/Eef3xtAQK1HezEJQNlAPJEJ wpiZZfHpP0pa9OBi9dlcjwNLU32PEhYyQpWCx7YCx7a68rKDCTkGttn2Adnx7o8AnxKfBCSfNp84 yRvrH/0Za5OhnwCzn73P15/pn/ufDa8frzGvAEOvVa9nr3mvf9+drim1bfZAv6+Jz3kXFjWupVv8 2UG/ABXP4vfi+MGzDy8Xt0k1ix4AZPaKv7cf4r8TZeMdDM+jvwDD+cG8xPhWz08v5hmrH5jPAKrP vM/Oz+DP8s8E2Kzlvz8AKd/eOY/hS99d32/fgd+T3wCl37ffde/b39Qcek8O7Vv+AC/vaU7QT2Tv c0+U5MexO0FAl+yMI8exWWG072tbECJKEDtFWWXy5fbxNP/nlgG4CZCT4fby1+wfQQgaEa+ACvgu /6VgF1NgiWSW4c84GEBAVfBT8AZAl0BYZhVh5mL4FGPWanT3UCBLF0wBAnUJ8HtgdeZQYASAQW0g bf5GYgN1QsBgdfbwFFVlF0GdxxgK3uAkilPyX0cDQEDFSvIZUPc3AQApTAEBcEfRRSHck+C3ZwEA gPbwQHoZM1lhsWIDcI/hjCHAYSDhZs1vwmjBb9NpAuBvG34SfVEDxPKxAgJVct+CX3iEuvKxAgNV cumCf3MXQaeqHEgTFlP90OABJ8ICBwBgAQBBc2JwgeLl8QEEofwWQi62+9+M7y/5895ROhdBVxZT CAICSWBFI9IaAQIfDw8HFUwBC3RpTpKAR9F6+lBA/lmCIQxfgo/haIMXQZzfVD357hVDoLxAzuPn PzL/chzHcRzH4T+vQJ/0SbaAz6uF0VK0hb82a4wjheNYIxZT/zBPG+i0gU5rfz8aRne/AAFmJgUC J8ACA5elMcoVCNTtcWwznygJ8L9rX///7XEA9DFFIV7BpTEERH1XFUNewRChCStJP08XUgGK811P G1+IknSIT9+YsGX08pcO7VAAFWUVq0o/XD9uP4A/kj+W4gE2OV9LV62hWV/fr31fj1/AoV+zX8Vf 118w/jQTVUD9YQNiHUBAvB6FV+tRuBpgYOQVmmXyEcmHtZiwkrO5Nm9Ib2uF0QORtm5vF0Hq2pSW vN+XyRRsHH64LFafMWOuBR5+t69kA2NwEjR4K8YffwFwjgUAEnFUxkDPXshdzyV7K89FfwBXca2h X3nOx5V/p3+5f8t/EN19raFSo0DWAUDWzcIPjwAhjzOPRY9XikUhaI6mwX2PEj35UJO1ZPO5q4+9 j8+PauGLPfCEuIGynj8LQvgEnlhgCCICAWIAADZy43qDbSCA4+O1HZ8BVK2hoO9Gn8PtraFslfrn AAZCCf8b/0Uls5/FnwKv9tYE8p+/RUDkFVP7Tq9gr3KvCISvlq+0ojeur0tX6/HOrwAdD/KvBL8W vyi/Or9Mv16//tqUIKN/B5y2IfNAQErjGmAQAd+8A5p4GmCStscDLGCkv7a/6Mi/2r/HGJjalBir mOcx1wK7AIC2KZtmf91gaiJAQLQbxQEMsWSmwYHFaRaSz3MYch8ATnrr8V95uhfoz/rPDN9RVYwm 3+PjAwPr8ZDzLCYBBOvxz/EB8OJs337T0hOK34Cc31uGzNduiOvx0N/i2R4A2gX33wnvG+8t7z/v Ue9j7wB174fvm/Hr8YI/s++lPevxANnv6+/97w//If8z/0X/V/8AIEnHBn7/kP+DT7T/xv/Y/wMA OJSU8/8VD6WCOgH6SUBOMeg7l5WO5U4FZfEge0egUiUGl0VHlUnuWZHalJP8AMKSYADXo3A9Cu/X P0CVFqLxP0ChaBVllVPoIJ5T1gQ2gG43Ac8DFkAxBHVl8HlgPFGAkOA/Bf7HUoMFdSeg0VlVc81B X3IEAHVz+VY+AhbRjuXe8xQqjuIe0jECAoBMIRVskToBANwR+RF8lIFN3BHcFZ+dl5EMPgX+P1UA Ek/gkd2fNwIix1IAzVnt8NxTAeNZ7WEgP03AZZVW0W/PV/DaVfZvACCsF36eVTB/Qn+Vdz9Fl5UA wG5/f+Nvo38/RT4FwG8pf4jpf01/gJAODHx+QNo1Q/8AbwBtAHAAcu8AZQBzjYBlAGSumABBAHWT gGmDgAq7AEKdgHQALY2AdAqJgmGFgAqdgCWjOgFlAaB+RJeZgQWZSY+RlJtoWHwkZ5eXECcAMlUa konFgCbaNSGbNfkRjpLpQHgkciB9BsBR2jXVQTns/wDAp2iRDA+anzAPQg9UD2YP4HgPig+cDrzQ H4L7Iv73L3vSF8BI4XoUf65HAUBAMDNFod/jP0DLzE6hAMCaQ6TTS6Xwv+oP/A9Abf52ogN18RBg dZXA5AgV570Kt5Aggcjvyrf8R599sK1HPKEJ90BAFz4SIEBAxL/1KFyPwvEhxQEDcDF+yOimfh/y qPGvmM8AZfGuxT6yJr8pwz6/qB9okVS+GYu8A4u2A8QSAau3ANQf5h/4Hx7fMN1okXJDHcYAurJY 0x3GZ995320vfy+RJcAdsbvfry/f3ESmm5Pnv9DhL/MvBT8XOiMlNE8b/+i0gU5rPxpGI3e/OD8j hWw/VGiRY98AlT+G3WiRuzXRgU3RyzStlwAEQ9GB0YmCaZZvDE8eTyiLAD5PxM0a7+Y1NO5OolKg aE8gek+MT55PsE/g+DrOT+BPAPJPBF8WXyhfOl9MX15fcF9/QOxvTNRoA7sA/UU8ogNAQGBNPH8r GgnwP0BARaUKpfPgrPXTt1/JX9tf7VkA3eH9WQkGEm8AzzZvSG9abwBsb35vkG+ib7RvD+80j0aP AFN/Dn6l9uEPkX8FH15/KRwArvVAH1IfAo+9fy5Ejx/zD4CzHwyP0g87H/sfXx9mij66dItRmYBh AG6ngGkrAHqRhFClgmPKkKQhqWGJgLSBZaeAZYmAcwC3j8mP24/tj/+PEZ8jnzWfBEefWZw73u/w 7wL/FP8m/+A4/0r/XP9u/4D/QF+Jr2SW4TiJUEk8og3XQECcllXATqLsPwg9ROPAR0Pcx//Z/+v/ 7VkAdjH9WaJGIg80D0YPWA9qDwB8D44PoA+yD8QPMi9EL1YvAGMfDn4+RnpfoR+eX24fwlwAR0XZ X+tfEi/NH890KG+MXwBMbxwva1/UX5Rv+F9mig8xCHePiYymJyCvL8EkpI+2jwDWL+gv+i8MPx4/ MD9CP1Q/AFiZAQAAAAIAAAACAAAAAAAAABFVYunyAuvwBOvwHevwVR7r8B/r8CDr8CTr8FUl6/Am 6/An6/Ao6/BVKevwKuvwK+vwLOvwVS3r8C7r8C/r8DDr8FUx6/Ay6/Az6/A06/ABNdzwAAAAAAAA BAAAAKV06/A07/Tq8Rjr8AL+6/ABpaxldZ7T/yJAexSuR+F6QxBA4/gCAQIB6vEL6/D/5MQ0AmJO AAD1UuvwQy0CdGsCAlW0NwAK6/BB6fIB3PQUAAAAFgAAAAAAAAAAAAAAAAAAAAEAAAAAAAAAQAAA AJS3FgK+TgAATwAAAFIAAAAAAAVI6/A83P8EDxYPKA4UAAAAFgAAAAAAAAAAAAAAAAAAAAEAAAAA AAAARQAAAKyAFgI/TwAADQAAAFIAAAAAABQAAABcAAAAAAAAAAAAAAABAAAABAAAAAAAAAAVAAAA 5IAaAmEjAADwKgAA0wDJAAAArJIMAlFOAAARAAAAQAA/AAAABIAaAg1PAAAuAAAAUABEAAAAXLcW AkxPAAAuAAAAUAAAAAAAAAAAAFUx6/D/8vAD6/Ao6PMVHOvwDNz/APbx9vHm9VUB6/AC5/RPIghE JgJhVUEB6PPy8VUP//9OBZBqCeD7KQHm9WgiCAEBAhcAVBjr8AR9D4sK6vFVvOP4KxYCVenyCa8N rnwOHwYI6fLJiQgeQiYCUS0BvwoZBS0CSCoEAD0PKABYD2kT9vF3HbEPkA+oog+0D+H6m+P4RRYC URD6DwgvGi8sLwD9ZuvwZQEAAEEA/3IAaQBhAGwA3yAAVQBu+fBjAN9vAGQAZf/wTQBpU9z/3/z/ NgPpPy8F9wE/YDQBAgsGBAUCUgEEJAH9ZuvwBQEAAFMA/3kAbQBiAG8AoWzc/xIPJA82D4Dq8QX/ BQECAQcGAgUBBzcB/Wbr8AUBAABXAH9pAG4AZwBk9/Shc9z/GA8qD+L5gOrxBQA3Cv1m6/BFAQAA QQB/cgBpAGEAbNz/3BAPIg8Ah3rp8oAI3uf0/wEAQEQA/wIvCwYEAlIBBCQB/Wbr8KUBAABTAHdp AG318HUAbtz/tBIP3f4D6fIOCOb1AfsABOnyAgEGAAMBAVMBJAH9ZuvwpQEAAFAA/00AaQBuAGcA RUz58FXc/xYP4foD6fJ3LggW5/QBABDp8gMCAjYDVQP9ZuvwpQEAAE0A/1MAIABQAEcA/28AdABo AGkA8WPc/xoP5fa/AgCg3/v8x2gQ5/SfAP8CQAAA198CC/8GAAcCBQgCBAAkAf1m6/ClAQAARAB/ bwB0AHUAbdz//BAPIg8ArwIAsPvvfNdpMOf0nwAI/0AAANffAgsGBwAAAVMBJAH9ZuvwRQEAAFMA 93kAbO3wYQBlAPFu3P8UD9/8hwYABN7i+Z8AACDq8QEK/wUCBQMGAwMDACQB7Wbr8EUB8PAAcwD/ dAByAGEAbgD/ZwBlAGwAbwBVIPXwZAMAc/fwYdz/tuH6h3oVAoAI5/T//wEBIAAAKCAC/wsGBAMF BAQCAQQkAf1m6/BFAQAAVgD/cgBpAG4AZAAxYdz/Eg/d/gMA8/Dj+A44BQEBBk0AUgEkAf1m6/BF AQAAUwD/aAByAHUAdAAxadz/Eg8kDwAEJg/r8AcCAAU8CP1m6/BFAQAATQBfYQBuAGf38Gzc/yQS DyQPgCUP6PMEPgj9ZuvwRQEAAFQAf3UAbgBnAGHc/0gQDyIP6/BAJg/p8gQ+CP1m6/BHAQAAUwDf ZQBuAGT58HkAsWHc/xQP3/zfewECgP0I5/T/AQEgAAB/KCACCwYEAlIBAQQkAf1m6/BFAQAAUgAd YffwdgBp3P8QDyIPEuvwAiYPNQIFPAj9ZuvwRwEAAEQAf2gAZQBuAHXc/3wQDyIPAN97AGHr8PuA COf0/wEBIAD/ACggAgsGBAICUgEEJAH9ZuvwRQEAAEwAH2EAdABo9/D//xEPziMPAAAQJg/r8AIA AQQ8CP1m6/BFAQAARwDfYQB1AHT38G0AkWnc/xQP3f4gJg/r8AIDAAU8CP1m6/BFAQAAQwD/bwBy AGQAaQD/YQAgAE4AZQBxd9z/Gg/l9gMAAPPw+uT3ATgEAgsDBAICUgEEJAH9ZuvwRwEAAE0A/1MA IABGAGEAH3IAcwBp3P8WD+H6e997/fAAAIAI5/T//wEBIAAAKCBfAgsGBAJSAQQkAf1m6/ClAQAA RwB/dQBsAGkAbdz//BAPIg8ArwIAsPvvfNdpMOf0nwAI/0AAANffAgsGBwAAAVMBJAH9ZuvwRQEA AFQA/2kAbQBlAHMA1yAATvvwd//wUgCdb/nwYQBu3P/d/ofteunygAjn9P8BAP1ARAD/AgIGAwUf BAUCAwQkARgAAAAQAgAAAAAAAAAAAAAAAAAAAAAAABgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA 1wAAANQ+FQKNUAAARQAAAEIA1wAAAFw+FQLSUAAALgAAAEIA1wAAAOQ9FQIAUQAAJQAAAEIA1wAA AGw9FQIlUQAANQAAAEIA1wAAAPQ8FQJaUQAAMgAAAEIA1wAAAHw8FQKMUQAAMwAAAEIA1wAAAAQ8 FQK/UQAARwAAAEIA1wAAAIw7FQIGUgAAOgAAAEIA1wAAABQ7FQJAUgAANwAAAEIA1wAAAJw6FQJ3 UgAATAAAAEIA1wAAACQ6FQLDUgAALQAAAEIA1wAAAKw5FQLwUgAAJwAAAEIA1wAAADQ5FQIXUwAA IgAAAEIA1wAAALw4FQI5UwAAIwAAAEIA1wAAAEQ4FQJcUwAAOQAAAEIA1wAAAMw3FQKVUwAAIgAA AEIA1wAAAFQ3FQK3UwAAOQAAAEIA1wAAANw2FQLwUwAAJgAAAEIA1wAAAGQ2FQIWVAAAJwAAAEIA 1wAAAOw1FQI9VAAAOgAAAEIA1wAAAHQ1FQJ3VAAAPgAAAEIA1wAAAMwKFQK1VAAAOgAAAEIA1wAA AEQLFQLvVAAARwAAAEIAAAAAAAEAAABHAHUAaQBkAGUAAAD1EuvwAevwVABoAH9lAEQAbwBj3PD1 EuvwAevwUABhAH9nAGUALQAx3PD1IuvwAevwRwBlAH9zAHQAdQBy9/DfIABGAG//8G0ADWH78AAA 9Rjr8ALr8FIAZQD/YwB0AGEAbgA3ZwBs9/AAAAAAAAD1FuvwAevwVgBpAP1z9/BvACAAOQABMNzw 9Rjr8AHr8EMAbwB9bvnwZQBjAHT38AFy3PD1FuvwAevwVgBpAH1z9/BvACAAMAEAAwAA9Rbr8AHr 8FYAaQD9c/fwbwAgADAAATHc8PUW6/AB6/BWAGkA/XP38G8AIAAwAAEy3PD1FuvwAevwVgBpAP1z 9/BvACAAMAABM9zw9Rbr8AHr8FYAaQD9c/fwbwAgADEAATDc8PUW6/AB6/BWAGkAfXP38G8AIAAx AQADAAD1FuvwAevwVgBpAP1z9/BvACAAMQABMtzw9Rbr8AHr8FYAaQD9c/fwbwAgADEAATPc8PUW 6/AB6/BWAGkA/XP38G8AIAAyAAEw3PD1FuvwAevwVgBpAP1z9/BvACAAMgABMdzw9Rbr8AHr8FYA aQB9c/fwbwAgADIBAAMAAPUW6/AB6/BWAGkA/XP38G8AIAAyAAEz3PD1FuvwAevwVgBpAP1z9/Bv ACAANQABMNzw9Rbr8AHr8FYAaQD9c/fwbwAgADUAATHc8PUW6/AB6/BWAGkA/XP38G8AIAA1AAEy 3PD1FuvwAevwVgBpAP1z9/BvACAANQABM9zw9Rbr8AHr8FYAaQD9c/fwbwAgADcAATDc8PUW6/AB 6/BWAGkA/XP38G8AIAA4AAEw3PABAAAAQgBhAHMAaQBjAAAA9R7r8AHr8EIAYQD/cwBpAGMAIAD3 UwBo9/BkAG8AAXfc8AAAAAD1GuvwA+vwdgBpAH9zAFYAZQBy+fAfaQBvAG7c8PUc6/AB6/BSAGUA /2MAdABhAG4Ad2cAbPfwLgAy3PD1KOvwAuvwRAB5AP9uAGEAbQBpANdjACABAG/58G4AFWUBAHQH AHLc8PUs6/AB6/BEAHkA/24AYQBtAGkA12MAIAEAb/nwbgD1ZQEAdAcAcgAuAAE03PAAAAAA9R7r 8AHr8EIAbAD/bwBjAGsAIAB9TvnwcgBtAGH38AMAAAAAAAD1HOvwAevwUgBlAP9jAHQAYQBuAHdn AGz38C4AOdzw9R7r8AHr8FIAZQD/YwB0AGEAbgD3ZwBs9/AuADIAATPc8PUu6/AB6/BEAHkA/24A YQBtAGkA12MAIAEAb/nwbgDVZQEAdAcAcu3wMgABN9zw9R7r8AHr8FIAZQD/YwB0AGEAbgD3ZwBs 9/AuADIAATjc8PUu6/AB6/BEAHkA/24AYQBtAGkA12MAIAEAb/nwbgDVZQEAdAcAcu3wMwABMtzw 9R7r8AHr8FIAZQD/YwB0AGEAbgB3ZwBs9/AuADMJAAMAAPUu6/AB6/BEAHkA/24AYQBtAGkA12MA IAEAb/nwbgDVZQEAdAcAcu3wMwABN9zw9S7r8AHr8EQAeQD/bgBhAG0AaQDXYwAgAQBv+fBuANVl AQB0BwBy7fAzAAE43PD1LuvwAevwRAB5AP9uAGEAbQBpANdjACABAG/58G4A1WUBAHQHAHLt8DMA ATnc8PUe6/AB6/BSAGUA/2MAdABhAG4A92cAbPfwLgA0AAEz3PD1LuvwAevwRAB5AP9uAGEAbQBp ANdjACABAG/58G4A1WUBAHQHAHLt8DEAATPc8PUu6/AB6/BEAHkA/24AYQBtAGkA12MAIAEAb/nw bgDVZQEAdAcAcu3wMQABN9zw9R7r8AHr8FIAZQD/YwB0AGEAbgD3ZwBs9/AuADEAATjc8PUe6/AB 6/BSAGUA/2MAdABhAG4A92cAbPfwLgA0AAE53PD1LuvwAevwRAB5AP9uAGEAbQBpANdjACABAG/5 8G4A1WUBAHQHAHLt8DUAATPc8PUu6/AB6/BEAHkA/24AYQBtAGkA12MAIAEAb/nwbgDVZQEAdAcA cu3wNQABNNzw9S7r8AHr8EQAeQD/bgBhAG0AaQDXYwAgAQBv+fBuAFVlAQB0BwBy7fA1GQADAAD1 LuvwAevwRAB5AP9uAGEAbQBpANdjACABAG/58G4A1WUBAHQHAHLt8DUAATfc8AAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAGAAAACAIAAAFAAAAAAAAAAAAAABQ AAAAUAAAAAAAAAAzAAAADJMMAgpXAAAQAAAARQAzAAAA3JIMAhpXAAAVAAAARwAzAAAAhJMMAi9X AAAVAAAARwAzAAAAfLcqAkRXAAAlAAAARwAzAAAA5LAUAmlXAAAcAAAARwAzAAAAjJMAAoVXAAAE AAAARQAzAAAAxLEUAolXAAAZAAAARwAzAAAAxJsUAqJXAAAaAAAARwAzAAAAZLsUArxXAAAZAAAA RwAzAAAARJ0UAtVXAAAZAAAARwAzAAAARKkUAu5XAAAZAAAARwAzAAAAZKwUAgdYAAAZAAAARwAz AAAApK0UAiBYAAAZAAAARwAzAAAAxK0UAjlYAAAZAAAARwAzAAAAZK0UAlJYAAAZAAAARwAzAAAA JK0UAmtYAAAZAAAARwAzAAAARK0UAoRYAAAZAAAARwAzAAAARKwUAp1YAAAZAAAARwAzAAAAJK8U ArZYAAAZAAAARwAzAAAAZK8UAs9YAAAZAAAARwAzAAAAxKwUAuhYAAAZAAAARwAzAAAARK4UAgFZ AAAZAAAARwAzAAAAhK8UAhpZAAAZAAAARwAzAAAAxLAUAjNZAAAZAAAARwAzAAAAZLAUAkxZAAAZ AAAARwAzAAAAhLAUAmVZAAAZAAAARwAzAAAAVJMMAn5ZAAAQAAAARQAzAAAA7LoqAo5ZAAAiAAAA RwAzAAAAnJMAArBZAAAEAAAARQAzAAAApLAUArRZAAAdAAAARwAzAAAABLEUAtFZAAAfAAAARwAz AAAAtIcVAvBZAAApAAAARwAzAAAARIUVAhlaAAAuAAAARwAzAAAAlJMAAkdaAAAEAAAARQAzAAAA xLoqAktaAAAiAAAARwAzAAAAFJAAAm1aAAAEAAAARQAzAAAAxLIUAnFaAAAfAAAARwAzAAAAnLoq ApBaAAAiAAAARwAzAAAAhJ8WArJaAAAvAAAARwAzAAAAJLoqAuFaAAAiAAAARwAzAAAAvJ8WAgNb AAAvAAAARwAzAAAA5LMqAjJbAAAiAAAARwAzAAAAlL4WAlRbAAAvAAAARwAzAAAAzL4WAoNbAAAv AAAARwAzAAAAPL8WArJbAAAvAAAARwAzAAAA/LkqAuFbAAAiAAAARwAzAAAABL8WAgNcAAAvAAAA RwAzAAAAdL8WAjJcAAAvAAAARwAzAAAA1LkqAmFcAAAiAAAARwAzAAAArLkqAoNcAAAiAAAARwAz AAAArL8WAqVcAAAvAAAARwAzAAAABKMWAtRcAAAvAAAARwAzAAAAPKMWAgNdAAAvAAAARwAzAAAA rKMWAjJdAAAvAAAARwAzAAAAJJAAAmFdAAAEAAAARQAzAAAA1I8fAmVdAAAEAAAARQAzAAAAnI8f AmldAAAEAAAARQAzAAAAxIsfAm1dAAAEAAAARQAzAAAAvI8fAnFdAAAEAAAARQAzAAAAxI8fAnVd AAAEAAAARQAzAAAABJQfAnldAAAEAAAARQAzAAAA/JMfAn1dAAAEAAAARQAzAAAAdJQfAoFdAAAE AAAARQAzAAAAzIsfAoVdAAAEAAAARQAzAAAAPJQfAoldAAAEAAAARQAzAAAA1JUAAo1dAAAEAAAA RQAzAAAA5JMfApFdAAAEAAAARQAzAAAA5JUAApVdAAAEAAAARQAzAAAANJAfApldAAAEAAAARQAz AAAAZJQfAp1dAAAEAAAARQAzAAAAFJAfAqFdAAAEAAAARQAzAAAA7I8fAqVdAAAEAAAARQAzAAAA HJAfAqldAAAEAAAARQAzAAAA3JUAAq1dAAAEAAAARQAzAAAAlI8fArFdAAAEAAAARQAzAAAAfI8f ArVdAAAEAAAARQAzAAAAhI8fArldAAAEAAAARQAzAAAAdI8fAr1dAAAEAAAARQAzAAAA1I4fAsFd AAAEAAAARQAzAAAAbI8fAsVdAAAEAAAARQAAAAAAAQAAAAIAAAADAAAABAAAAAUAAAAGAAAABwAA AAgAAAAJAAAACgAAAAsAAAAMAAAADQAAAA4AAAAPAAAAEAAAABEAAAASAAAAEwAAABQAAAAVAAAA FgAAABcAAAAYAAAAGQAAABoAAAAbAAAAHAAAAB0AAAAeAAAAHwAAACAAAAAhAAAAIgAAACMAAAAk AAAAJQAAACYAAAAnAAAAKAAAACkAAAAqAAAAKwAAACwAAAAtAAAALgAAAC8AAAAwAAAAMQAAADIA AAAzAAAANAAAADUAAAA2AAAANwAAADgAAAA5AAAAOgAAADsAAAA8AAAAPQAAAD4AAAA/AAAAQAAA AEEAAABCAAAAQwAAAEQAAABFAAAARgAAAEcAAABIAAAASQAAAEoAAABLAAAATAAAAE0AAABOAAAA TwAAAAAAAABVYuX2BuvwB+vwCOvwVQnr8Arr8Avr8Azr8FUN6/AO6/AP6/AQ6/BVEevwEuvwE+vw FOvwVRXr8Bbr8Bfr8Bjr8FUZ6/Aa6/Ab6/Ai3PAAAAAAAAAEAAAApXTr8DTv9OrxGOvwAv7r8AGl rGV1ntP/IkB7FK5H4XpDEEDj+AIBAgHq8Qvr8P8U0TQCzWQAAPVO6/BDLQL0awICXxtlAAAK6/BB 6fIBAdz0AAABAAAAAwAAAAAAJXTr8DTv9OrxAtz/3f70/vH+9Qvr8GRzAgJfdmUAAArr8EEtAp+k kwACgDcA/vFBAunyARAEFAAAAEoBAAAAAAAAAAAAAAAAAAAPAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAEAAAABkoBYCJWUAAFEAAABSAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAEAAAACcoBYCgmUAADYAAABSAAAAAAAFSOvwPNz/BA8WDygOBUjr8Dzc/wQPFg8o DhQAAABKAQAAAAAAAAAAAAAAAAAADwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAABFAAAADKEWAuZmAAANAAAAUgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABFAAAA 1KAWAvNmAAANAAAAUgAAAAAAVRLn9AH18FD58FTc8hgAAAAaAAAAAQAAAAAAAAAAAAAAAQAAAAEA AAAAAAAALwAAACSTDAIuaAAADQAAAEMAAAAAAAAAAAB78ALr8AEAAOTz8E7g+wEAA+vwCgUE6/B9 CevwdwEBACLc/+A0D0YPWA/5/uT3hxkC8TCGAWgP6PMorjwg/5k4w+NAKOtn4zqiqwKnBqcOQBz/ nQT3/CC7Ku/x6fIWvusIFMMWAoDr8FXu6/BCABfr8FRPGFMC1eMB6/BQCxL8ERBR+RUWLx9BHxrr 8MQREP0d8/B+EgAA0gD9POvwNJsMApsTdwAACuvwRgBD+wGvdQICpW8QBuvwRPsAHevwHFYYAqy6 IgBm6/BSACTr8MwffhgCEiOMECsf6fI9IRMBfxgCMqUQGBNVKevwJKEQVqUQC5cS9Sfr8JShEH5P AAD1aBkSMevwdIAWAtXq7RCj6/DSrh8A2L7r8GyXFgI2BhDQrvPwUAAy6/DcHyDJtV3r8AcOEAA/ 6/Ckvh8guGUAAConIkS+hRGhFgIAZ0gl/+br8CygViDn9GAALr7r8PSfFgI78BA2AKkWAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAUAUwB1AG0AbQBhAHIAeQBJAG4AZgBvAHIAbQBh AHQAaQBvAG4AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAoAAIB/////xUAAAD/////AAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAOgEAAPRQAAAAAAAABQBEAG8AYwB1AG0AZQBu AHQAUwB1AG0AbQBhAHIAeQBJAG4AZgBvAHIAbQBhAHQAaQBvAG4AAAAAAAAAAAAAADgAAgD///// //////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAAAAAIAAAAAAABf ADEAMwA0ADkAMAAzADUAMAA3ADQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAGAABAf//////////GAAAAALOAgAAAAAAwAAAAAAAAEYAAAAA0E27vGtyywHQTbu8a3LL AQAAAAAAAAAAAAAAAAEATwBsAGUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAKAAIB////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAGgAAABQAAAAAAAAA/v8AAAUBAgAAAAAAAAAAAAAAAAAAAAAAAQAAAOCF n/L5T2gQq5EIACsns9kwAAAAxFAAAAsAAAABAAAAYAAAAAIAAABoAAAAAwAAAHQAAAAEAAAAgAAA AAUAAACkAAAABgAAALAAAAAHAAAAvAAAAAgAAADIAAAAEgAAAOwAAAANAAAABAEAABEAAAAQAQAA AgAAAOQEAAAeAAAABAAAAAAAAAAeAAAABAAAAAAAAAAeAAAAHAAAAEp1aW4tSHdleSAoUmF5bW9u ZCkgQ2hlbgAAAAAeAAAABAAAAAAAAAAeAAAABAAAAAAAAAAeAAAABAAAAAAAAAAeAAAAHAAAAEp1 aW4tSHdleSAoUmF5bW9uZCkgQ2hlbgAAAAAeAAAAEAAAAE1pY3Jvc29mdCBWaXNpbwBAAAAAwORX mlFyywFHAAAArE8AAP////8OAAAAAQAAAIwAAAAAAAAAAAAAAHoAAAA1AAAAAAAAAAAAAADoDAAA mwUAACBFTUYAAAEApE8AAAMAAAABAAAADwAAAGwAAAAAAAAAgAcAALAEAAAIAgAARQEAAAAAAAAA AAAAAAAAAEDvBwCI9QQAVgBJAFMASQBPAAAARAByAGEAdwBpAG4AZwAAAAAAAABMAAAABE8AAAAA AAAAAAAAegAAADUAAAAAAAAAAAAAAHsAAAA2AAAAIADMAAAAAAAAAAAAAACAPwAAAAAAAAAAAACA PwAAAAAAAAAA////AAAAAABkAAAAKAAAAIwAAAB4TgAAKAAAAHsAAAA2AAAAAQAYAAAAAAB4TgAA AAAAAAAAAAAAAAAAAAAAAP////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /////////////////////////////////////////////////////wAAAAAAAP////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /////////////////wAAAAAAAP////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /8zMzP////Pz89nZ2f////Pz87+/v7+/v9nZ2dnZ2f///////////////////////8zMzP////Pz 89nZ2f////Pz87+/v7+/v9nZ2dnZ2f////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /////////////////////////////////////////////////////////wAAAAAAAP////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /////////////////////////////////////////9nZ2fPz8////+bm5v///7+/v7+/v9nZ2b+/ v9nZ2f///////////////////////9nZ2fPz8////+bm5v///7+/v7+/v/Pz87+/v9nZ2f////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /////////////////////wAAAAAAAP////////////////////////////////////////////// /////////////////////////2BfXwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA ABAQEO/v7////////////2BgYAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACAg IP////////////////////////////////////////////////////////////////////////// /////////////////////////////////////5+fnwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAgICNfX1////////////39/fwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAgICNfX1////////////////////////////////////////////wAAAAAAAP////// /////////////////////////////////////////////////////////////////z8/Prmyrvfu 6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6B8fHt/f3////////////0BAQL+/v/////// /////////////////////////////////////wAAAP///////////8zMzNnZ2aOjo5SUlMzMzLGx sf///////////////////////////////////////8zMzNnZ2aOjo5SUlMzMzLGxsf////////// /39/f39/f////////////////////////////////////////////0BAQL+/v////////////2Bf X5qUkffu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6D49PL+/v/////////////////// /////////////////////////wAAAAAAAP////////////////////////////////////////// /////////////////////////////z8/Prmyrvfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu 6Pfu6B8fHt/f3////////////0BAQL+/v/////////////Pz8/Pz8/////////Pz8/////////// /wAAAP////////////Pz8////////9nZ2f////////////////////////////////////////// //////Pz8////////9nZ2f///////////////////39/f39/f/////////Pz8/////Pz8/Pz8/// /+bm5v///////////0BAQL+/v////////////2BfX5qUkffu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu 6Pfu6Pfu6Pfu6D49PL+/v////////////////////////////////////////////wAAAAAAAP// /////////////////////////////////////////////////////////////////////z8/Prmy rvfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6B8fHt/f3////////////0BAQL+/v/// /////6Ojo7+/v7+/v8zMzLGxsb+/v9nZ2f///////wAAAL+/v+bm5szMzNnZ2aOjo8zMzL+/v/Pz 8+bm5ubm5r+/v7Gxsf///////////////7+/v+bm5szMzNnZ2aOjo8zMzL+/v/Pz8+bm5ubm5r+/ v7GxsX9/f39/f////+bm5pSUlMzMzMzMzL+/v6Ojo4WFhbGxsf///////0BAQL+/v/////////// /2BfX5qUkffu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6D49PL+/v/////////////// /////////////////////////////wAAAAAAAP////////////////////////////////////// /////////////////////////+fn5////z8/Prmyrvfu6Ovj3dLLxZ6YlPfu6N/X0evj3Z6YlNLL xffu6Pfu6B8fHt/f3////+fn5////0BAQL+/v////+bm5v////////////////Pz8/////////// /////wAAAL+/v+bm5tnZ2b+/v8zMzMzMzL+/v+bm5ubm5v///7+/v6Ojo////////////////7+/ v+bm5tnZ2b+/v8zMzNnZ2b+/v+bm5ubm5v///62traOjo39/f39/f////+bm5v////////////// //Pz8////////////////0BAQL+/v////+fn5////2BfX5qUkevj3evj3Z6YlLmyruvj3ffu6Lmy rp6YlLmyrvfu6Pfu6D49PL+/v////////////////////+fn5////////////////////wAAAAAA AP////////f391BQUEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQIdfQ0hDQEBAQBgYGCgoKDU0 NLmyruvj3aulod/X0d/X0ffu6N/X0ffu6N/X0d/X0ffu6Pfu6A8PDjg4OEBAQB4eHigoKDQ0NL+/ v/////////Pz89nZ2ebm5szMzLGxsf///////////////wAAAEBAQEBAQEBAQEBAQEhISO/v7/// /////////////////////////////////////////////////////////////////5eXl0BAQEBA QCAgIBYWFk9PT39/f/////////Pz8/Pz88zMzLGxsczMzPPz8////////////yQkJDAwMEBAQB4e Hh8fH0JBQZqUkdLLxca+uuvj3ffu6Pfu6Ovj3bmyruvj3evj3ffu6Pfu6CMiIjAwMEBAQEBAQEBA QEBAQEBAQB4eHigoKM/Pz////////////wAAAAAAAP///////////9fX17+/v7+/v7+/v7+/v7+/ v7+/v7+/v7+/v7+/v7+AVb+2sL+/v6enp+/v7z8/Prmyrvfu6Pfu6Pfu6Pfu6Pfu6N/X0ffu6Pfu 6Pfu6Pfu6Pfu6BwcG6enp7+/v62tre/v70BAQL+/v////////7GxsdnZ2ebm5rGxsczMzP////// /////////wAAAL+/v7+/v7+/v7+/vxgYGN/f3/////////////////////////////////////// /////////////////////////////2BgYHh4eL+/v6ioqN/f339/f39/f////////7Gxsebm5tnZ 2aOjo9nZ2fPz8////////////zw8PI+Pj7+/v6ioqOfn52BfX5qUkffu6Pfu6Pfu6Pfu6Pfu6N/X 0ffu6Pfu6Pfu6Pfu6Pfu6Do5OI+Pj7+/v7+/v7+/v7+/v7+/v62tre/v7////////////////wAA AAAAAP///////////////////////7+/v7+/v4WFhb+/v9nZ2bGxsf///8eIXffu6P////////// /z8/Prmyrvfu6Pfu6NLLxca+uuvj3ca+uquloffu6Pfu6Pfu6Pfu6B8fHt/f3////////////0BA QL+/v////////////////////////////////////////////wAAAP///////////////yAgIN/f 3////////////////////////////////////////////////////////////////////2BgYJ+f n////////////39/f39/f////////////////////////////////////////////0BAQL+/v/// /////////2BfX5qUkffu6Pfu6NLLxca+uuvj3dLLxY+Kh/fu6Pfu6Pfu6Pfu6D49PL+/v////+bm 5ubm5v///////////////////////////////wAAAAAAAP///////////////////////8zMzNnZ 2fPz8+bm5v///+bm5v///8eIXffu6P///////////z8/Prmyrvfu6Pfu6Ovj3ffu6Pfu6Pfu6Pfu 6Pfu6Pfu6Pfu6Pfu6B8fHt/f3////////////0BAQL+/v/////Pz88zMzPPz8/Pz8+bm5rGxsf// /////////////wAAAP///////////////yAgIN/f3/////////////////////////////////// /////////////////////////////////2BgYJ+fn////////////39/f39/f////////7+/v/// /+bm5ubm5tnZ2f///////////////0BAQL+/v////////////2BfX5qUkffu6Pfu6Ovj3ffu6Pfu 6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6D49PL+/v6Ojo7+/v7GxsbGxsbGxsdnZ2f////////////////// /wAAAAAAAP///////////////////////////////+bm5v///////9nZ2f///8eIXffu6P////// /////z8/Prmyrvfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6B8fHt/f3/////////// /0BAQL+/v////////8zMzNnZ2dnZ2f///6Ojo/Pz8////////////wAAAP///////////////yAg IN/f3////////////////////////////////////////////////////////////////////2Bg YJ+fn////////////39/f39/f////////8zMzPPz87+/v////7+/v9nZ2f///////////0BAQL+/ v////////////2BfX5qUkffu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6D49PL+/v7+/ v/////////////////Pz8////////////////////wAAAAAAAP///////////////////////+bm 5r+/v7+/v6Ojo/Pz87Gxsf///8eIXffu6P///////////09PTy8uLT49PD49PD49PD49PD49PEI/ PD49PD49PD49PD49PD49PBAPD+fn5////////////0hISDAwMEBAQEBAQEBAQEBAQEBAQEBAQEBA QEBAQEBAQEBAQEBAQAgICP///////////////yAgIN/f3/////////////////////////////// /////////////////////////////////////2BgYJ+fn////////////4eHhyAgIEBAQEBAQEBA QEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQBAQEMfHx////////////3BvbycmJT49PD49PD49PD49 PD49PEI/PD49PD49PD49PD49PD49PBAPD8/Pz/Pz89nZ2ebm5szMzMzMzPPz8/////////////// /////wAAAAAAAP///////////////////////////////////////////////////8eIXffu6P// /////////+/v77+/v7+/v7+/v7+/v7+/v7+2sL+XfL+/v7+/v7+/v7+/v7+/v9fX1/////////// /////+fn57+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v9fX1/////////////// /yAgIN/f3/////////////////////////////////////////////////////////////////// /2BgYJ+fn/////////////f397+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v8fH x/////////////////f397+/v7+/v7+/v7+/v7+/v7+/v7+SdL+/v7+/v7+/v7+/v7+/v8/Pz/// /7GxsdnZ2ebm5rGxsdnZ2fPz8////////////////////wAAAAAAAP////////////////////// /////9nZ2aOjo8zMzLGxsfPz8////8eIXffu6P///////////////////////////////////+XG sdOigf////////////////////////////////////////////////////////////////////// /////////////////////////////////////////yAgIN/f3/////////////////////////// /////////////////////////////////////////2BgYJ+fn/////////////////////////// //////////////////////////////////////////////////////////////////////////// //////Xr48WEWP////////////////////////////////////Pz8/////////////////////// /////////wAAAAAAAP///////////////////////////8zMzMzMzNnZ2dnZ2f///////8eIXffu 6P///////////////////////////////////+fMudeqjP////////////////////////////// //////////////////////////////////////////////////////////////////////////// /+/v7xISEp+fn/////////////////////////////////////////////////////////////// /////2BgYJ+fn/////////////////////////////////////////////////////////////// //////////////////////////////////////////////fu6MeIXf////////////////////// /9nZ2bGxsaOjo9nZ2ZSUlKOjo////////////////////////wAAAAAAAP////////////////// /////////////////////////////////8eIXffu6P////////////////////////////////// /+fMudeqjP////////////////////////////////////////////////////////////////// /////////////////////////////////////////////zg4ON/f3/////////////////////// /////////////////////////////////////////////2BgYJ+fn/////////////////////// //////////////////////////////////////////////////////////////////////////// //////////fu6MeIXf///////////////////////////+bm5v////////////////////////// /////////////wAAAAAAAP///////////////////////////////////////////////////8eI Xffu6f///////////////////////////////////+fMudeqjP////////////////////////// /////////////////////////////////////////////////////////+fn50dHRz8/Pj8/Pj8/ Pj8/Pj8/PjExMD8/Pj8/Pj8/Pj8/Pj8/Pj8/Pr+/v////////////9/f30dHRz8/Pj8/Pj8/Pj8/ Pj8/Pj8/PhgYFykpKT8/Pj8/Pj8/Pj8/Pj8/Pj8/PmBfX/////////////////////////////// //////////////////////////////////////////////////fu6MeIXf////////////////// /////////////////////////////////////////////////////wAAAAAAAP////////////// /////////////7+/v7+/v7+/v7+/v7+/v7+/v6qRgb+/v7+/v7+/v7+/v7+/v7+/v+/v7/////// /////+fMudeqjP////////////////////////////////////////////////////////////// /////////////////////7+/vy8uLbmyrrmyrrmyrrmyrrmyrrmyrrmyrrmyrrmyrrmyrrmyrl1a V39/f////////////5+fn0ZEQrmyrrmyrrmyrrmyrrmyrrmyrqihnrmyrrmyrrmyrrmyrrmyrrmy rrmyrgAAAP////////////////////////////////////////////////////////////////// //////////////fu6MeIXf////////////////////////////////////////////////////// /////////////////wAAAAAAAP///////////////////////7+/vxAPDz49PD49PD49PD49PD49 PD49PD49PD49PD49PD49PD49PB8eHn9/f////////////+fMudeqjP////////////////////// /////////////////////////////////////////////////////////////7+/vz49PPfu6Pfu 6NLLxdLLxauloaulobmyrtLLxffu6Pfu6Pfu6Ht3dH9/f////////////5+fn11aWPfu6Pfu6NLL xd/X0bmyrrmyrquloca+ut/X0dLLxffu6Pfu6Pfu6Pfu6AAAAP////////////////////////// //////////////////////////////////////////////////////fu6MeIXf////////////// /////////////////////////////////////////////////////////wAAAAAAAP////////// /////////////7+/vz49PPfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Ht3dH9/f/// /////////+fMudeqjP////////////////////////////////////////////////////////// /////////////////////////7+/vz49PPfu6Pfu6Lmyrt/X0ca+up6YlKuloauloffu6Pfu6Pfu 6Ht3dH9/f////////////5+fn11aWPfu6Pfu6Ovj3d/X0bmyrtLLxca+utLLxdLLxca+uuvj3ffu 6Pfu6Pfu6AAAAP////////////////////////////////////////////////////////////// //////////////////fu6MeIXf////////////////////////////////////////////////// /////////////////////wAAAAAAAP///////////////////////7+/vz49POvj3bmyrsa+uuvj 3dLLxdLLxbmyruvj3auloffu6Pfu6Ht3dH9/f////////////+fMudeqjP////////////////// /////////////////////////////////////////////////////////////////7+/vz49PPfu 6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Ht3dH9/f////////////5+fn11aWPfu6Pfu 6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6AAAAP////////////////////// //////////////////////////////////////////////////////////fu6MeIXf////////// /////////////////////////////////////////////////////////////wAAAAAAAP////// /////////////////7+/vz49PPfu6N/X0d/X0ffu6LmyrtLLxevj3ffu6Ma+utLLxffu6Ht3dH9/ f////////////+fMudeqjP////////////////////////////////////////////////////// /////////////////////////////7+/vz49PPfu6Ma+uqulod/X0ca+uo+Kh9LLxauloffu6Pfu 6Pfu6Ht3dH9/f////+fn5/f395+fn11aWPfu6Pfu6N/X0Y+Kh6uloca+uqulodLLxY+Kh/fu6Pfu 6Pfu6Pfu6Pfu6AAAAP////////////////////////////////////////////////////////// //////////////////////fu6MeIXf////////////////////////////////////////////// /////////////////////////wAAAAAAAP///////////////////////7+/vz49PPfu6Pfu6Pfu 6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Ht3dH9/f////////////+fMudeqjP////////////// /////////////////////////////////////////////////////////////////////7+/vz49 PPfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6GRhXgAAAAAAAAAAAAYGBlRUVF1aWPfu 6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6AAAAP////////////////// //////////////////////////////////////////////////////////////fu6MeIXf////// /////////////////////////////////////////////////////////////////wAAAAAAAP// /////////////////////7+/vz49PMa+ut/X0aulod/X0auloY+Kh4F8ed/X0YF8eZ6YlPfu6Ht3 dH9/f////////////+fMuc2ihczMzLGxsbGxsaOjo5SUlLGxsYWFhbGxsbGxsebm5ubm5v////// /////////////////////////////////7+/vz49PPfu6Pfu6NLLxffu6Lmyruvj3d/X0evj3evj 3evj3ffu6Ht3dH9/f////+fn59/f35+fn11aWPfu6N/X0ffu6Ovj3ffu6Ovj3ffu6Kuloffu6N/X 0evj3evj3evj3ffu6AAAAP////////////////////////////////////////////////////// //////////////////////////fu6MeIXf////////////////////////////////////////// /////////////////////////////wAAAAAAAP///////////////////////7+/vz49PMa+uvfu 6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Ht3dH9/f////////////+fMuc2ihczMzP////// /////////////////////////////////////////////////////////////////////////7+/ vz49PLmyro+Kh7myrvfu6KuloauloYF8ebmyro+Kh6uloevj3Xt3dH9/f////////////5+fn11a WN/X0YF8edLLxZ6YlNLLxbmyrvfu6J6YlKuloYF8ebmyro+Kh6uloevj3QAAAP////////////// //////////////////////////////////////////////////////////////////fu6MeIXf// /////////////////////////////////////////////////////////////////////wAAAAAA AP///////////////////////7+/vz49PPfu6Pfu6Pfu6Pfu6Pfu6Ma+utLLxevj3ffu6Pfu6Pfu 6Ht3dH9/f////////////+fMudeqjP///////////////////7+/v9nZ2fPz8/////////////// /////////////////////////////////////7+/vz49POvj3evj3ffu6Pfu6Pfu6Ovj3ffu6Pfu 6Pfu6Pfu6Pfu6Ht3dH9/f////////////5+fn11aWOvj3ffu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Ovj 3ffu6Pfu6Pfu6Pfu6Pfu6AAAAP////////////////////////////////////////////////// //////////////////////////////fu6MeIXf///7+/v9nZ2b+/v9nZ2b+/v7+/v/Pz87+/v/Pz 8+bm5v///////////////////////////wAAAAAAAP///////////////////////7+/vz49PPfu 6Pfu6Pfu6J6YlPfu6NLLxZ6YlOvj3ffu6Pfu6Pfu6Ht3dH9/f////////////+fMudeqjP////// /////6Ojo////9nZ2aOjo/Pz8/////////////////////////////////////////////////// /7+/vz49PNLLxca+uuvj3d/X0dLLxdLLxZ6YlKulobmyrp6YlPfu6Ht3dH9/f////////////5+f n11aWPfu6Ovj3auloevj3d/X0dLLxbmyroF8ebmyrrmyrp6YlPfu6Pfu6Pfu6AAAAP////////// //////////////////////////////////////////////////////////////////////fu6MeI XaOjo/Pz87GxsfPz87+/v7+/v7Gxsf///7+/v7+/v////////////////////////////////wAA AAAAAP///////////////////////7+/vy8uLbmyrrmyrrmyrrmyrrmyrrmwq7myrrmyrrmyrrmy rrmyrl1aV39/f////////////+fMudeqjP////////////////////////////////////////// /////////////////////////////////////////7+/vy8uLZ2XlJ2XlLmyrrmyrrmyrqehnaeh nbmyrqehnbCqprmyrl1aV39/f////////////5+fn0ZEQrmyrrCqpouFgrmyrrmyrrmyrp6Wkaeh nbmyrqehnbCqprmyrrmyrrmyrgAAAP////////////////////////////////////////////// //////////////////////////////////fu6MeIXf////////////////////////////////// /////////////////////////////////////wAAAAAAAP///////////////////////9/f3z8/ Pz8/Pj8/Pj8/Pj8/Pj8/PqdtRUtEPz8/Pj8/Pj8/Pj8/Pj8/Pq+vr////////////+fMudeqjL+/ v6Ojo6Ojo7GxsaOjo9nZ2dnZ2bGxsbGxsZSUlP////////////////////////////////////// /////9/f3z8/Pz8/Pj8/Pj8/Pj8/Pj8/PlNIPz8/Pj8/Pj8/Pj8/Pj8/Pj8/Pq+vr/////////// /9fX1z8/Pz8/Pj8/Pj8/Pj8/Pj8/Pj8/PothQ2dQQT8/Pj8/Pj8/Pj8/Pj8/Pj8/PlBPT/////// //////////////////////////////////////////////////////////////////////////fu 6MeIXf///////////7+/v////7GxsaOjo9nZ2f////////////////////////////////////// /wAAAAAAAP///////////////////////////////////////////////////8eIXffu6P////// /////////////////////////////+fMudeqjObm5vPz8////////////////////////////9nZ 2f///////////////////////////////////////////////////////////////////////8+Z dPny7f///////////////////////////////////////////////////////////////////9eq jOfMuf////////////////////////////////////////////////////////////////////// //////////////////////////////////////fu6MeIXf///////////7+/v////////9nZ2f// /////////////////////////////////////////wAAAAAAAP////////////////////////// /////////////////////////8eIXffu6P///////////////////////////////////+fMudeq jP////////////////////////////////////////////////////////////////////////// /////////////////////////////////////8eIXffu6P////////////////////////////// /////////////////////////////////////9eqjOfMuf////////////////////////////// //////////////////////////////////////////////////////////////////////////// //fu6MeIXf///////+bm5vPz8/Pz8/////Pz8/Pz8+bm5v////////////////////////////// /////wAAAAAAAP///////////////////////////////////////////////////8eIXffu6P// /////////////////////////////////+fMudeqjP////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /8eIXffu6P////////////////////////////////////////////////////////////////// /9eqjOfMuf////////////////////////////////////////////////////////////////// //////////////////////////////////////////fu6MeIXb+/v7+/v7+/v/Pz86Ojo+bm5v// /9nZ2ZSUlKOjo////////////////////////////////wAAAAAAAP////////////////////// /////////////////////////////8eIXffu6P///////////////////////////////////+fM udeqjP////////////////////////////////////////////////////////////////////// /////////////////////////////////////////8eIXffu6P////////////////////////// /////////////////////////////////////////9eqjOfMuf////////////////////////// //////////////////////////////////////////////////////////////////////////// //////fu6MeIXf////////////////////////////////////Pz8/////////////////////// /////////wAAAAAAAP///////////////////////////////////////////////////8eIXffu 6P///////////2BfXwAAAAAAAAAAAAAAAAAAACoaD1o4IQAAAAAAAAAAAAAAAAAAABAQEO/v7/// //////////////////////////////////////////////////////////////////////////// /////8eIXffu6P////////////////////////////////////////////////////////////// /////9eqjOfMuf///////////////////////////3BwcAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAgICOfn5/////////////////////fu6MeIXf////////////////////// /////////////////////////////////////////////////wAAAAAAAP////////////////// /////////////////////////////////8eIXffu6P///////////z8/Prmyrvfu6Pfu6Pfu6Pfu 6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6B8fHt/f3/////////////////////////////////////// /////////////////////////////////////////////8eIXffu6P////////////////////// /////////////////////////////////////////////9eqjOfMuf////////////////////// /////2BfX5qUkffu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6B8fHt/f3/////////// //////////fu6MeIXf////////////////////////////////////////////////////////// /////////////wAAAAAAAP///////////////////////////////////////////////////8eI Xffu6P///////////z8/Prmyrvfu6Pfu6Ovj3d/X0ffu6Pfu6NLLxevj3evj3ffu6Pfu6B8fHt/f 3/////////////////////////////////////////////////////////////////////////// /////////8eIXffu6P////////////////////////////////////////////////////////// /////////9eqjOfMuf///////////////////////////2BfX5qUkffu6Pfu6Ovj3evj3ffu6Ovj 3ffu6N/X0ffu6Pfu6Pfu6B8fHt/f3/////////////////////fu6MeIXf////////////////// /////////////////////////////////////////////////////wAAAAAAAP////////////// /////////////////////////////////////8eIXffu6P///////////z8/Prmyrrmyrsa+urmy rrmyrp6YlPfu6NLLxY+Kh6uloevj3ffu6B8fHt/f3/////////////////////////////////// /////////////////////////////////////////////////8eIXffu6P////////////////// /////////////////////////////////////////////////9eqjOfMuf////////////////// /////////2BfX5qUkffu6N/X0Y+Kh7myrsa+urmyrquloY+Kh6uloffu6Pfu6B8fHt/f3/////// //////////////fu6MeIXf////////////////////////////////////////////////////// /////////////////wAAAAAAAP////////////////////////////////////////////////// /8eIXffu6P///////////z8/Prmyruvj3ffu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6B8f Ht/f3/////////////////////////////////////////////////////////////////////// /////////////8eIXffu6P////////////////////////////////////////////////////// /////////////9eqjOfMuf///////////////////////////2BfX5qUkffu6Ovj3ffu6Pfu6Pfu 6Pfu6Ovj3ffu6Pfu6Pfu6Pfu6B8fHt/f3/////////////////////fu6MeIXf////////////// /////////////////////////////////////////////////////////wAAAAAAAP////////// /////////////////////////////////////////8+Zdb93Rr93Rr93RsqPZ1dJP7myruvj3aul obmyrquloaulobmyroF8eevj3Z6YlPfu6Pfu6INWN793Rr93Rr93Rr93Rr93Rr93Rr93Rr93Rr93 Rr93Rr93Rr93Rr93Rr93Rr93Rr93Rr93Rr93Rr93Rr93Rr93Rr93RsOAUvv28/////////////// /////////////////////////////////////////////////////9+7or93Rr93Rr93Rr93Rr93 Rr93Rr93RsiJX3hlWZqUkd/X0aulobmyrqulobmyrp6YlIF8eevj3Z6YlNLLxffu6INWN793Rr93 Rr93Rr93Rr93Rr93Rr93Rs+Zdf////////////////////////////////////////////////// /////////////////////wAAAAAAAP////////////////////////////////////////////// /////////////////////////z8/PrCqpquloffu6NLLxffu6N/X0evj3d/X0ffu6N/X0evj3ffu 6B8fHt/f3/////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /////////////////////////////////////////////////////2BfX5qUkauloffu6NLLxffu 6N/X0evj3d/X0ffu6N/X0d/X0ffu6B8fHt/f3/////////////////////////////////////// /////////////////////////////////////////////////////////////wAAAAAAAP////// //////////////////Pz8+bm5qOjo9nZ2bGxsbGxsb+/v7+/v6Ojo7+/v8zMzPPz8z8/Prmyrvfu 6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6B8fHt/f3////////////////////7Gxsf// /////7+/v7+/v6Ojo9nZ2bGxsaOjo/////////////////////////////////////////////// //////////////////////////////Pz87Gxsf///////7+/v7+/v4WFhb+/v7+/v6Ojo/////// /////////////////2BfX5qUkffu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6Pfu6B8fHt/f 3/////////////////////////////////////////////////////////////////////////// /////////////////////////wAAAAAAAP///////////////////////9nZ2czMzPPz8/Pz8/Pz 8/Pz8/////Pz8/Pz8/Pz8/Pz8////z8/Prmyrvfu6Pfu6NLLxd/X0d/X0bmyrquloffu6Pfu6Pfu 6Pfu6B8fHt/f3////////////////////7+/v/////////////Pz8/Pz8/////Pz8/Pz8/////// //////////////////////////////////////////////////////////////////////Pz87+/ v////////////9nZ2ebm5v////Pz8/Pz8////////////////////////2BfX5qUkffu6Pfu6NLL xd/X0d/X0bmyrquloffu6Pfu6Pfu6Pfu6B8fHt/f3/////////////////////////////////// /////////////////////////////////////////////////////////////////wAAAAAAAP// /////////////////////////////////////////////+bm5v///////////////////z8/Prmy rvfu6Pfu6NLLxd/X0ffu6Ovj3ffu6Pfu6Pfu6Pfu6Pfu6B8fHt/f3/////////////////////Pz 8/Pz8/Pz8/Pz8/Pz8/Pz8////+bm5vPz8+bm5v////////////////////////////////////// //////////////////////////////////////Pz8/Pz8/Pz8/Pz8/Pz8+bm5v///+bm5v////Pz 8////////////////////2BfX5qUkffu6Pfu6NLLxd/X0ffu6Ovj3ffu6Pfu6Pfu6Pfu6Pfu6B8f Ht/f3/////////////////////////////////////////////////////////////////////// /////////////////////////////wAAAAAAAP////////////////////////////////////Pz 87GxsfPz89nZ2ZSUlP///////////////09PTy8uLT49PD49PD49PD49PD49PD49PD49PD49PD49 PD49PD49PBAPD+fn5////////////////7+/v/Pz87GxsfPz87GxsaOjo5SUlObm5oWFhaOjo8zM zP///////////////////////////////////////////////////////////////////////7Gx sfPz87GxsfPz86Ojo6Ojo4WFhf///4WFhbGxsebm5v///////////////////3BvbycmJT49PD49 PD49PD49PD49PD49PD49PD49PD49PD49PD49PBAPD+fn5/////////////////////////////// /////////////////////////////////////////////////////////////////////wAAAAAA AP////////////////////////////////////////Pz8////////////////////////////+/v 77+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v9fX1////////////////////+bm 5v////////////////////////////////////////////////////////////////////////// /////////////////////////////////////+bm5v////////////////////////////////// //////////////////////////f397+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/v7+/ v9fX1/////////////////////////////////////////////////////////////////////// /////////////////////////////////wAAAAAAAP////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////Pz8/////Pz87+/v7+/v/////// //////////////////////////////////////////////////////////////////////////// /////////////+bm5v///8zMzL+/v7+/v/////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /////////////////////////////////////////////////////////////////////////wAA AAAAAP////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /////////////9nZ2czMzP///+bm5ubm5v////////////////////////////////////////// /////////////////////////////////////////////////////7+/v/Pz8////7+/v/Pz8/// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /////////////////////////////////////wAAAAAAAP////////////////////////////// //////////////////////////////////////////////////////////////////////////// /////////////////////////////////////////////////////////+bm5v////////////// //////////////////////////////////////////////////////////////////////////// //////////////////Pz8/////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /wAAAAAAAP////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /9nZ2dnZ2ZSUlKOjo5SUlIWFhaOjo4WFhb+/v7+/v4WFhczMzLGxsf////////////////////// /////////////////////////////////////////7+/v+bm5pSUlJSUlLGxsZSUlLGxsYWFhb+/ v7+/v3R0dLGxsdnZ2f////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /////////////////////////////////////////wAAAAAAAP////////////////////////// //////////////////////////////////////////////////////////////////////////// /////////////////////////////////////////////8zMzP////////////////////////// /////////////9nZ2f////////////////////////////////////////////////////////// /////+bm5ubm5v////////////////////////////////////////Pz8/////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /////wAAAAAAAP////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /////////////////////////////////////////////wAAAAAAAA4AAAAUAAAAAAAAABAAAAAU AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIAAAAAEAAAAAAAAACgA AAABAAAAYAAAAAIAAABoAAAAAwAAAHQAAAACAAAAAgAAAA4AAABfUElEX0xJTktCQVNFAAMAAAAV AAAAX1ZQSURfQUxURVJOQVRFTkFNRVMAAAIAAADkBAAAQQAAAAIAAAAAAAAAHgAAAAQAAAAAAAAA AQAAAggAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAEA/v8DCgAA/////wLOAgAAAAAAwAAAAAAAAEYXAAAATWljcm9zb2Z0IEVxdWF0aW9u IDMuMAAMAAAARFMgRXF1YXRpb24ACwAAAEVxdWF0aW9uLjMA9DmycQAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADAAQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABwAAAACAAvDUQAAAAAAAAC4HRkARPYYAAAAAAAD AQEDCgoBEoNkAAKCKAASg24AAoIpAAKGPQASg3MAAoIoABKDbgACgikAAoYSIhKDYgACgSAAEoNz AAKCKAASg24AAoYSIhKDcAACgikAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQBDAG8AbQBwAE8AYgBq AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgEXAAAA GQAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAbAAAAZgAAAAAAAAAD AE8AYgBqAEkAbgBmAG8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAEgACAf////8aAAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AB0AAAAGAAAAAAAAAEUAcQB1AGEAdABpAG8AbgAgAE4AYQB0AGkAdgBlAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAgAAIA////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAHgAAAG0AAAAAAAAAXwAxADMANAA5ADAAMwA0ADgANAA0AAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABgAAQElAAAASAAAAB0AAAACzgIAAAAA AMAAAAAAAABGAAAAANBNu7xrcssB0E27vGtyywEAAAAAAAAAAAAAAAABAE8AbABlAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACgACAf////// /////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACAAAAAUAAAAAAAAAAEA QwBvAG0AcABPAGIAagAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAASAAIBHAAAAB4AAAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA IQAAAGYAAAAAAAAAAwBPAGIAagBJAG4AZgBvAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAABIAAgH/////HwAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAjAAAABgAAAAAAAABFAHEAdQBhAHQAaQBvAG4AIABOAGEAdABpAHYAZQAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIAACAP///////////////wAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACQAAAChAAAAAAAAAAEAAAIIAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAP7/AwoAAP// //8CzgIAAAAAAMAAAAAAAABGFwAAAE1pY3Jvc29mdCBFcXVhdGlvbiAzLjAADAAAAERTIEVxdWF0 aW9uAAsAAABFcXVhdGlvbi4zAPQ5snEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAwAEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAcAAAAAgALw4UAAAAAAAAAeG8YAFRsGAAAAAAAAwEBAwoKARKDSAADDwEACwES g3AAEoNyABKDZQAAEQAKAoIoABKDegACgikAAoY9AAMOAAABEoNEAAKCKAASg3oAAoIpAAABEoNT AAKCKAASg3oAAoIpAAAAAoY9AAKIMQAChhIiEoNiAAKBIAASg3oAAw8AAAsRAQKGEiISg3AAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAAACCAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAXwAxADMANAA5ADAAMwAwADMA MAA5AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABgAAQH///////// /yIAAAACzgIAAAAAAMAAAAAAAABGAAAAANBNu7xrcssB0E27vGtyywEAAAAAAAAAAAAAAAABAE8A bABlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAACgACAf///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACcA AAAUAAAAAAAAAAEAQwBvAG0AcABPAGIAagAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAASAAIBIQAAACMAAAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAKAAAAGYAAAAAAAAAAwBPAGIAagBJAG4AZgBvAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgH/////JAAAAP////8AAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAqAAAABgAAAAAAAAABAP7/AwoAAP////8CzgIAAAAA AMAAAAAAAABGFwAAAE1pY3Jvc29mdCBFcXVhdGlvbiAzLjAADAAAAERTIEVxdWF0aW9uAAsAAABF cXVhdGlvbi4zAPQ5snEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAwAE AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAcAAAAAgALwx0AAAAAAAAACBEYAByKGgAAAAAAAwEBAwoKAQKIMAAChmQiEoNiAAKGPAACiDEA AAAAAAAAAAAAAQAAAggAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAEA/v8DCgAA/////wLOAgAAAAAAwAAAAAAAAEYXAAAATWljcm9zb2Z0 IEVxdWF0aW9uIDMuMAAMAAAARFMgRXF1YXRpb24ACwAAAEVxdWF0aW9uLjMA9DmycQAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADAAQAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEUAcQB1AGEAdABpAG8AbgAgAE4A YQB0AGkAdgBlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgAAIA//////////// ////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAKwAAADkAAAAAAAAAXwAxADMA NAA5ADAAMwA0ADIANwA0AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA ABgAAQEgAAAAKgAAACcAAAACzgIAAAAAAMAAAAAAAABGAAAAANBNu7xrcssB0E27vGtyywEAAAAA AAAAAAAAAAABAE8AbABlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAACgACAf///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAACwAAAAUAAAAAAAAAAEAQwBvAG0AcABPAGIAagAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIBJgAAACgAAAD/////AAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAALQAAAGYAAAAAAAAAAwBPAGIAagBJAG4AZgBvAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgH/////KQAAAP// //8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAvAAAABgAAAAAAAABFAHEAdQBh AHQAaQBvAG4AIABOAGEAdABpAHYAZQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA IAACAP///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADAAAABH AAAAAAAAAF8AMQAzADQAOQAwADMANAAyADkAOAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAYAAEB/////zkAAAAsAAAAAs4CAAAAAADAAAAAAAAARgAAAADQTbu8a3LL AdBNu7xrcssBAAAAAAAAAAAAAAAAAQBPAGwAZQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAoAAgH///////////////8AAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAyAAAAFAAAAAAAAAAcAAAAAgALwysAAAAAAAAAiBAYAJSp GgAAAAAAAwEBAwoKAQKCKAASg0YAAw8BAAsBEoNzAAARAAoCgi8AEoNwAAKCKQAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEAAAIIAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAB AP7/AwoAAP////8CzgIAAAAAAMAAAAAAAABGFwAAAE1pY3Jvc29mdCBFcXVhdGlvbiAzLjAADAAA AERTIEVxdWF0aW9uAAsAAABFcXVhdGlvbi4zAPQ5snEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAwAEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAcAAAAAgALwxoAAAAAAAAAoAsZAIxyGAAAAAAAAwEBAwoKARKD RgADDwEACwESg3MAABEAAAAAAAAAAAAAAAAAAQAAAggAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEAQwBvAG0AcABPAGIAagAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIBKwAAAC0AAAD///// AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAMwAAAGYAAAAAAAAAAwBPAGIAagBJ AG4AZgBvAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIA AgH/////LgAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA1AAAABgAA AAAAAABFAHEAdQBhAHQAaQBvAG4AIABOAGEAdABpAHYAZQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAIAACAP///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAADYAAAA2AAAAAAAAAF8AMQAzADQAOQAyADcAOQA0ADkAMQAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAYAAEBDQAAADQAAAAxAAAAAs4CAAAAAADAAAAAAAAA RgAAAADQTbu8a3LLAdBNu7xrcssBAAAAAAAAAAAAAAAAAQBPAGwAZQAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAoAAgH///////////////8A AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA3AAAAFAAAAAAAAAABAEMAbwBtAHAA TwBiAGoAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEgAC ATAAAAAyAAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADgAAABmAAAA AAAAAAMATwBiAGoASQBuAGYAbwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAASAAIB/////zMAAAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAOgAAAAYAAAAAAAAARQBxAHUAYQB0AGkAbwBuACAATgBhAHQAaQB2AGUAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACAAAgD///////////////8AAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAA7AAAAOQAAAAAAAAABAP7/AwoAAP////8CzgIAAAAAAMAAAAAA AABGFwAAAE1pY3Jvc29mdCBFcXVhdGlvbiAzLjAADAAAAERTIEVxdWF0aW9uAAsAAABFcXVhdGlv bi4zAPQ5snEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAwAEAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAcAAAA AgDowR0AAAAAAAAAcIYaAPyaGgAAAAAAAwEBAwoKATKDZAAGCAAAAoIoABKDbgACgikAAAAAAAAA AAAAAQAAAggAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAEA/v8DCgAA/////wLOAgAAAAAAwAAAAAAAAEYXAAAATWljcm9zb2Z0IEVxdWF0 aW9uIDMuMAAMAAAARFMgRXF1YXRpb24ACwAAAEVxdWF0aW9uLjMA9DmycQAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADAAQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAF8AMQAzADQAOQAyADcAOQA0ADkANgAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAYAAEB//////////82AAAAAs4C AAAAAADAAAAAAAAARgAAAADQTbu8a3LLAdBNu7xrcssBAAAAAAAAAAAAAAAAAQBPAGwAZQAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAoAAgH/ //////////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA8AAAAFAAAAAAA AAABAEMAbwBtAHAATwBiAGoAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAEgACATUAAAA3AAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAD0AAABmAAAAAAAAAAMATwBiAGoASQBuAGYAbwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIB/////zgAAAD/////AAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAPwAAAAYAAAAAAAAARQBxAHUAYQB0AGkAbwBuACAATgBhAHQAaQB2 AGUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACAAAgD///////////////8AAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAAAAAOQAAAAAAAABfADEAMwA0ADkAMAAz ADQAOAAxADEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAGAABAP// ////////OwAAAALOAgAAAAAAwAAAAAAAAEYAAAAA0E27vGtyywHQTbu8a3LLAQAAAAAAAAAAAAAA AAEATwBsAGUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAKAAIB////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAQQAAABQAAAAAAAAAAQBDAG8AbQBwAE8AYgBqAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgE6AAAAPAAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAABCAAAAZgAAAAAAAAAcAAAAAgDowR0AAAAAAAAA6OgYAAR4GAAAAAAA AwEBAwoKATKDcwAGCAAAAoIoABKDbgACgikAAAAAAAAAAAAAAQAAAggAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEA/v8DCgAA/////wLO AgAAAAAAwAAAAAAAAEYXAAAATWljcm9zb2Z0IEVxdWF0aW9uIDMuMAAMAAAARFMgRXF1YXRpb24A CwAAAEVxdWF0aW9uLjMA9DmycQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAADAAQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAABwAAAACAAvDXQAAAAAAAAAI2xcAfIAaAAAAAAADAQEDCgoBMoNzAAYIAAACgigAEoNu AAKCKQAChj0AMoNkAAYIAAACgigAEoNuAAKCKQAChisAEoNiAAKBIAAyg3MABggAAAKCKAASg24A AoYSIhKDcAACgikAAAAAAAAAAAAAAQAAAggAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAMATwBiAGoASQBuAGYAbwAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIB/////z0AAAD/////AAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAARAAAAAYAAAAAAAAARQBxAHUAYQB0AGkAbwBu ACAATgBhAHQAaQB2AGUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACAAAgD///// //////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABFAAAAeQAAAAAAAABf ADEAMwA0ADkAMAA1ADEAOAA3ADMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAGAABAVcAAAD/////QAAAAALOAgAAAAAAwAAAAAAAAEYAAAAA0E27vGtyywHQTbu8a3LL AQAAAAAAAAAAAAAAAAEATwBsAGUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAKAAIB////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAARwAAABQAAAAAAAAAAQBDAG8AbQBwAE8AYgBqAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgE/AAAAQQAAAP////8AAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAAAZgAAAAAAAAADAE8AYgBqAEkAbgBmAG8A AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEgACAf////9C AAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEoAAAAGAAAAAAAAAEUA cQB1AGEAdABpAG8AbgAgAE4AYQB0AGkAdgBlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAgAAIA////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA SwAAALoAAAAAAAAAXwAxADMANAA5ADIANwA5ADYAMgAwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAABgAAQEvAAAAYQAAAEUAAAACzgIAAAAAAMAAAAAAAABGAAAAANBN u7xrcssB0E27vGtyywEAAAAAAAAAAAAAAAABAP7/AwoAAP////8CzgIAAAAAAMAAAAAAAABGFwAA AE1pY3Jvc29mdCBFcXVhdGlvbiAzLjAADAAAAERTIEVxdWF0aW9uAAsAAABFcXVhdGlvbi4zAPQ5 snEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAwAEAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAcAAAAAgALw54A AAAAAAAA0NoYABR6GAAAAAAAAwEBAwoKARKDSAADDwEACwESg3AAEoNvABKDcwASg3QAABEACgKC KAASg3oAAoIpAAKGPQADDgAAATKDUwAGCAAAAoIoABKDegACgikAAAEyg0QABggAAAKCKAASg3oA AoIpAAAAAoY9AAMOAAABAogxAAABAogxAAKGEiISg2IAAoEgABKDegADDwAACxEBAoYSIhKDcAAA AAAAAAAAAAAAAAABAAACCAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAQD+/wMKAAD/////As4CAAAAAADAAAAAAAAARhcAAABNaWNyb3Nv ZnQgRXF1YXRpb24gMy4wAAwAAABEUyBFcQEATwBsAGUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAKAAIB////////////////AAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAATgAAABQAAAAAAAAAAQBDAG8AbQBwAE8AYgBqAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgFEAAAARgAA AP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABPAAAAZgAAAAAAAAADAE8A YgBqAEkAbgBmAG8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAEgACAf////9HAAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAFEA AAAGAAAAAAAAAEUAcQB1AGEAdABpAG8AbgAgAE4AYQB0AGkAdgBlAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAgAAIA////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAUgAAAEcAAAAAAAAAdWF0aW9uAAsAAABFcXVhdGlvbi4zAPQ5snEAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAwAEAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAcAAAAAgDowSsAAAAAAAAA8H0a APyaGgAAAAAAAwEBAwoKAQKCKAASg0YAAw8BAAsBEoNzAAARAAoCgi8AEoNwAAKCKQAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEAAAII AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAABAP7/AwoAAP////8CzgIAAAAAAMAAAAAAAABGFwAAAE1pY3Jvc29mdCBFcXVhdGlvbiAzLjAA DAAAAERTIEVxdWF0aW9uAAsAAABFcXVhdGlvbi4zAPQ5snEAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAwAEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAABfADEAMwA0ADkAMAA1ADEAMwAxADAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAGAABAWsAAABSAAAASgAAAALOAgAAAAAAwAAA AAAAAEYAAAAA0E27vGtyywHQTbu8a3LLAQAAAAAAAAAAAAAAAAEATwBsAGUAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAKAAIB//////////// ////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAVAAAABQAAAAAAAAAAQBDAG8A bQBwAE8AYgBqAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA ABIAAgFJAAAASwAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABVAAAA ZgAAAAAAAAADAE8AYgBqAEkAbgBmAG8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAEgACAf////9MAAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAFcAAAAGAAAAAAAAAEUAcQB1AGEAdABpAG8AbgAgAE4AYQB0AGkAdgBlAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgAAIA////////////////AAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAWAAAAIEAAAAAAAAAXwAxADMANAA5ADAANQAxADMANAAw AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABgAAQH//////////08A AAACzgIAAAAAAMAAAAAAAABGAAAAANBNu7xrcssB0E27vGtyywEAAAAAAAAAAAAAAAABAE8AbABl AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA CgACAf///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAFsAAAAU AAAAAAAAAAEAQwBvAG0AcABPAGIAagAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAASAAIBTgAAAFAAAAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAXAAAAGYAAAAAAAAAHAAAAAIAC8NlAAAAAAAAAEANGADMdhgAAAAAAAMBAQMKCgES g0gAAw8BAAsBEoNwABKDcgASg2UAABEACgKCKAASg3oAAoIpAAKGPQADDgAAAQKIMQAAAQKIMQAC hisAEoNhAAKBIAASg3oAAw8AAAsRAQKGEiISg3AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQAAAggAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEA/v8DCgAA /////wLOAgAAAAAAwAAAAAAAAEYXAAAATWljcm9zb2Z0IEVxdWF0aW9uIDMuMAAMAAAARFMgRXF1 YXRpb24ACwAAAEVxdWF0aW9uLjMA9DmycQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAADAAQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAABwAAAACAAvDXAAAAAAAAAA4nBoAzNkYAAAAAAADAQEDCgoBEoNIAAMPAQAL ARKDcAASg28AEoNzABKDdAAAEQADAE8AYgBqAEkAbgBmAG8AAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEgACAf////9RAAAA/////wAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAF4AAAAGAAAAAAAAAEUAcQB1AGEAdABpAG8AbgAgAE4AYQB0 AGkAdgBlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgAAIA//////////////// AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAXwAAAHgAAAAAAAAAXwAxADMANAA5 ADAANQAxADQAMgAwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABgA AQBNAAAAPgAAAFQAAAACzgIAAAAAAMAAAAAAAABGAAAAANBNu7xrcssB0E27vGtyywEAAAAAAAAA AAAAAAABAE8AbABlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAACgACAf///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAGEAAAAUAAAAAAAAAAoCgigAEoN6AAKCKQAChj0AAogxAAKGKwASg2EAAoEgABKDegAD DwAACxEBAoYSIhKDcAAAAAAAAAAAAAAAAAABAAACCAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQD+/wMKAAD/////As4CAAAAAADAAAAA AAAARhcAAABNaWNyb3NvZnQgRXF1YXRpb24gMy4wAAwAAABEUyBFcXVhdGlvbgALAAAARXF1YXRp b24uMwD0ObJxAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAMABAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHAAA AAIAC8OHAAAAAAAAACCVGgB0qRoAAAAAAAMBAQMKCgESg0gAAw8BAAsBEoNwABKDcgASg2UAABEA CgKCKAASg3oAAoIpAAKGPQADDgAAAQKIMQAChhIiEoNiAAKBIAASg3oAAw8AAAsRAQKGEiISg3AA AAAACgECiDEAAoYrABKDYQACgSAAEoN6AAMPAAALEQEChhIiEoNwAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAQBDAG8AbQBwAE8AYgBqAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgFTAAAAVQAAAP////8AAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAABiAAAAZgAAAAAAAAADAE8AYgBqAEkAbgBmAG8AAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEgACAf////9WAAAA/////wAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAGQAAAAGAAAAAAAAAEUAcQB1AGEAdABp AG8AbgAgAE4AYQB0AGkAdgBlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgAAIA ////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAZQAAAKMAAAAA AAAAXwAxADMANAA5ADAANQAxADUAMAAyAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAABgAAQD//////////1kAAAACzgIAAAAAAMAAAAAAAABGAAAAANBNu7xrcssB0E27 vGtyywEAAAAAAAAAAAAAAAABAE8AbABlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACgACAf///////////////wAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAGgAAAAUAAAAAAAAAAEAQwBvAG0AcABPAGIAagAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIBWAAAAFoAAAD/////AAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAaQAAAGYAAAAAAAAAAwBPAGIAagBJAG4A ZgBvAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgH/ ////WwAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABrAAAABgAAAAAA AABFAHEAdQBhAHQAaQBvAG4AIABOAGEAdABpAHYAZQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAIAACAP///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAGwAAACnAAAAAAAAAAEAAAIIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAP7/AwoAAP////8CzgIAAAAAAMAAAAAAAABGFwAAAE1p Y3Jvc29mdCBFcXVhdGlvbiAzLjAADAAAAERTIEVxdWF0aW9uAAsAAABFcXVhdGlvbi4zAPQ5snEA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAwAEAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAcAAAAAgALw4sAAAAA AAAAuAgZAJR5GAAAAAAAAwEBAwoKARKDSAADDwEACwESg3AAEoNvABKDcwASg3QAABEACgKCKAAS g3oAAoIpAAKGPQADDgAAAQKIMQAChisAEoNhAAKBIAASg3oAAw8AAAsRAQKGEiISg3AAAAAACgEC iDEAAoYSIhKDYgACgSAAEoN6AAMPAAALEQEChhIiEoNwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAABAAACCAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAXwAxADMANAA5ADIANwA5ADkAMAAwAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAABgAAQH//////////14AAAACzgIAAAAAAMAAAAAAAABGAAAA ANBNu7xrcssB0E27vGtyywEAAAAAAAAAAAAAAAABAE8AbABlAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACgACAf///////////////wAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAG8AAAAUAAAAAAAAAAEAQwBvAG0AcABPAGIA agAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIBXQAA AF8AAAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAcAAAAGYAAAAAAAAA AwBPAGIAagBJAG4AZgBvAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAABIAAgH/////YAAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAByAAAABgAAAAAAAAD9////hAEAAIMBAACFAQAAhwEAAIgBAADoAQAAiQEAAIoBAACMAQAAiwEA AI0BAACPAQAAjgEAAJABAACRAQAAkgEAAJQBAACTAQAAlQEAAJYBAACXAQAAmAEAAJkBAACbAQAA mgEAAJwBAACdAQAAngEAAKABAACfAQAAoQEAANsBAADaAQAAowEAAKQBAAClAQAApgEAAKcBAACo AQAAqQEAAKoBAACrAQAArAEAAK0BAACuAQAArwEAALABAACxAQAAsgEAALMBAAC0AQAAtQEAALYB AAC3AQAAuAEAALkBAAC6AQAAuwEAALwBAAC9AQAAvgEAAL8BAADAAQAAwQEAAMIBAADDAQAAxAEA AMUBAADGAQAAxwEAAMgBAADJAQAAygEAAMsBAADMAQAAzQEAAM4BAADPAQAA0AEAANEBAADSAQAA 0wEAANQBAADVAQAA1gEAANcBAADYAQAA2QEAAOoBAADcAQAA3QEAAN4BAADgAQAA3wEAAOEBAADj AQAA4gEAAOQBAADmAQAA5QEAAOcBAADpAQAA9QEAAP7////2AQAA6wEAAOwBAADtAQAA7gEAAO8B AADwAQAA8QEAAPIBAADzAQAA9AEAAP7////+////DQIAAP3////5AQAA+gEAAPsBAAD8AQAA/QEA AP4BAAD/AQAAAAIAAAEA/v8DCgAA/////wLOAgAAAAAAwAAAAAAAAEYXAAAATWljcm9zb2Z0IEVx dWF0aW9uIDMuMAAMAAAARFMgRXF1YXRpb24ACwAAAEVxdWF0aW9uLjMA9DmycQAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADAAQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABwAAAACAOjBWAAAAAAAAABYhhoA/Joa AAAAAAADAQEDCgoBEoNIAAMPAQALARKDcAASg3IAEoNlAAARAAoCgigAEoN6AAKCKQAChj0AAogx AAKGEiISg2IAAoEgABKDegADDwAACxEBAoYSIhKDcAAAAAAAAAAAAAAAAAAAAAAAAQAAAggAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEA /v8DCgAA/////wLOAgAAAAAAwAAAAAAAAEYXAAAATWljcm9zb2Z0IEVxdWF0aW9uIDMuMAAMAAAA RFMgRXF1YXRpb24ACwAAAEVxdWF0aW9uLjMA9DmycQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAARQBxAHUAYQB0AGkAbwBuACAATgBhAHQAaQB2AGUAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAACAAAgD///////////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAABzAAAAdAAAAAAAAABfADEAMwA0ADkAMgA3ADkAOQAwADUAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAGAABAVwAAAB/AAAAYwAAAALOAgAAAAAA wAAAAAAAAEYAAAAA0E27vGtyywHQTbu8a3LLAQAAAAAAAAAAAAAAAAEATwBsAGUAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAKAAIB//////// ////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAdQAAABQAAAAAAAAAAQBD AG8AbQBwAE8AYgBqAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAABIAAgFiAAAAZAAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAB2 AAAAZgAAAAAAAAADAE8AYgBqAEkAbgBmAG8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAEgACAf////9lAAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAHgAAAAGAAAAAAAAAEUAcQB1AGEAdABpAG8AbgAgAE4AYQB0AGkAdgBlAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgAAIA////////////////AAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAeQAAAIUAAAAAAAAAXwAxADMANAA5ADIAOAAxADQA NwA4AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABgAAQD///////// /2gAAAACzgIAAAAAAMAAAAAAAABGAAAAANBNu7xrcssB0E27vGtyywEAAAAAAAAAAAAAAAABAE8A bABlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAACgACAf///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHwA AAAUAAAAAAAAAAAAAwAEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAcAAAAAgDowWkAAAAAAAAAOOcYAFx5GAAAAAAAAwEBAwoKARKDSAAD DwEACwESg3AAEoNvABKDcwASg3QAABEACgKCKAASg3oAAoIpAAKGPQADDgAAAQKIMQAAAQKIMQAC hhIiEoNiAAKBIAASg3oAAw8AAAsRAQKGEiISg3AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAAACCAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQD+/wMKAAD///// As4CAAAAAADAAAAAAAAARhcAAABNaWNyb3NvZnQgRXF1YXRpb24gMy4wAAwAAABEUyBFcXVhdGlv bgALAAAARXF1YXRpb24uMwD0ObJxAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAMABAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAQBDAG8AbQBwAE8AYgBqAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAABIAAgFnAAAAaQAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAB9AAAAZgAAAAAAAAADAE8AYgBqAEkAbgBmAG8AAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEgACAf////9qAAAA/////wAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAH8AAAAGAAAAAAAAAEUAcQB1AGEAdABpAG8AbgAgAE4A YQB0AGkAdgBlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgAAIA//////////// ////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgAAAAE0AAAAAAAAAXwAxADMA NAA5ADAAMwA1ADEANQA2AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA ABgAAQAWAAAAhAAAAG0AAAACzgIAAAAAAMAAAAAAAABGAAAAANBNu7xrcssB0E27vGtyywEAAAAA AAAAAAAAAACBAAAA/v////7///+EAAAA/v////7///+HAAAAiAAAAIkAAAD+/////v///4wAAAD+ /////v////7////+////kQAAAP7////+////lAAAAP7////+////lwAAAP7////+////mgAAAP7/ ///+////nQAAAP7////+////oAAAAKEAAACiAAAAowAAAP7////+////pgAAAP7////+////qQAA AKoAAACrAAAA/v////7///+uAAAA/v////7///+xAAAAsgAAALMAAAD+/////v///7YAAAD+//// /v////7////+////uwAAAP7////+////vgAAAL8AAAD+/////v///8IAAAD+/////v///8UAAAD+ /////v///8gAAAD+/////v///8sAAAD+/////v///84AAAD+/////v///9EAAAD+/////v///9QA AAD+/////v///9cAAAD+/////v///9oAAAD+/////v///90AAAD+/////v///+AAAAD+/////v// /+MAAAD+/////v///+YAAAD+/////v///+kAAAD+/////v///+wAAAD+/////v///+8AAAD+//// /v////IAAAD+/////v////7////+////9wAAAP7////+/////v////7////8AAAA/v////7///// AAAAAAEAABwAAAACAOjBMQAAAAAAAABQhhoA/JoaAAAAAAADAQEDCgoBMoNkAAYIAAACgigAEoNu AAKCKQAChj0AEoNkAAKCKAASg24AAoIpAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAQAAAggAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEA/v8DCgAA/////wLOAgAAAAAAwAAAAAAAAEYX AAAATWljcm9zb2Z0IEVxdWF0aW9uIDMuMAAMAAAARFMgRXF1YXRpb24ACwAAAEVxdWF0aW9uLjMA 9DmycQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADAAQAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABwAAAACAAvD uQAAAAAAAADA9BgAJGkYAAAAAAADAQEDCgoBMoNzAAYIAAACgigAEoNuAAKCKQAChj0AEoNkAAKC KAASg24AAoIpAAKGKwASg2IAAoEgADKDcwAGCAAAAoIoABKDbgAChhIiEoNwAAKCKQAChj0AEoNz AAKCKAASAQBPAGwAZQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAoAAgH///////////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAACCAAAAFAAAAAAAAAABAEMAbwBtAHAATwBiAGoAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEgACAWwAAABuAAAA/////wAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAIMAAABmAAAAAAAAAAMATwBiAGoASQBuAGYAbwAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIB/////28AAAD///// AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAhQAAAAYAAAAAAAAARQBxAHUAYQB0 AGkAbwBuACAATgBhAHQAaQB2AGUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACAA AgD///////////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACGAAAA1QAA AAAAAACDbgACgikAAoYSIhKDYgACgSAAEoNzAAKCKAASg24AAoYSIhKDcAACgikAAoYrABKDYgAC gSAAMoNzAAYIAAACgigAEoNuAAKGEiISg3AAAoIpAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAEAAAIIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAP7/AwoAAP////8CzgIAAAAAAMAAAAAAAABGFwAA AE1pY3Jvc29mdCBFcXVhdGlvbiAzLjAADAAAAERTIEVxdWF0aW9uAAsAAABFcXVhdGlvbi4zAPQ5 snEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAwAEAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAcAAAAAgDowSEA AAAAAAAAyHoaAPR6GgAAAAAAAwEBAwoKAQKCewASg3MAAoIoABKDbgACgikAAoJ9AAAAAAAAAQAA AggAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAF8AMQAzADQAOQAyADgAMQA1ADMAOQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAYAAEBmAAAALYAAAByAAAAAs4CAAAAAADAAAAAAAAARgAAAADQTbu8a3LLAdBN u7xrcssBAAAAAAAAAAAAAAAAAQBPAGwAZQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAoAAgH///////////////8AAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAACKAAAAFAAAAAAAAAABAEMAbwBtAHAATwBiAGoAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEgACAXEAAABzAAAA/////wAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIsAAABmAAAAAAAAAAMATwBiAGoASQBu AGYAbwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIB /////3QAAAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAjQAAAAYAAAAA AAAARQBxAHUAYQB0AGkAbwBuACAATgBhAHQAaQB2AGUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAACAAAgD///////////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAACOAAAAPQAAAAAAAABfADEAMwA0ADkAMgA4ADEANQA0ADEAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAGAABAf//////////dwAAAALOAgAAAAAAwAAAAAAAAEYA AAAA0E27vGtyywHQTbu8a3LLAQAAAAAAAAAAAAAAAAEATwBsAGUAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAKAAIB////////////////AAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAjwAAABQAAAAAAAAAAQBDAG8AbQBwAE8A YgBqAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgF2 AAAAeAAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACQAAAAZgAAAAAA AAABAP7/AwoAAP////8CzgIAAAAAAMAAAAAAAABGFwAAAE1pY3Jvc29mdCBFcXVhdGlvbiAzLjAA DAAAAERTIEVxdWF0aW9uAAsAAABFcXVhdGlvbi4zAPQ5snEAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAwAEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAcAAAAAgDowSUAAAAAAAAAoA0ZADzpGAAAAAAAAwEBAwoK AQKCewAyg3MABggAAAKCKAASg24AAoIpAAKCfQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEAAAIIAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAP7/AwoAAP////8C zgIAAAAAAMAAAAAAAABGFwAAAE1pY3Jvc29mdCBFcXVhdGlvbiAzLjAADAAAAERTIEVxdWF0aW9u AAsAAABFcXVhdGlvbi4zAPQ5snEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAMATwBiAGoASQBuAGYAbwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAASAAIB/////3kAAAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAkgAAAAYAAAAAAAAARQBxAHUAYQB0AGkAbwBuACAATgBhAHQAaQB2AGUAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAACAAAgD///////////////8AAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAACTAAAAQQAAAAAAAABfADEAMwA0ADkAMgA4ADEANQA0ADUAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAGAABAHUAAACOAAAAfAAAAALOAgAA AAAAwAAAAAAAAEYAAAAA0E27vGtyywHQTbu8a3LLAQAAAAAAAAAAAAAAAAEATwBsAGUAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAKAAIB//// ////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAlQAAABQAAAAAAAAA AQBDAG8AbQBwAE8AYgBqAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAABIAAgF7AAAAfQAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AACWAAAAZgAAAAAAAAADAE8AYgBqAEkAbgBmAG8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAEgACAf////9+AAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAJgAAAAGAAAAAAAAAEUAcQB1AGEAdABpAG8AbgAgAE4AYQB0AGkAdgBl AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgAAIA////////////////AAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAmQAAAE0AAAAAAAAAXwAxADMANAA5ADIAOAAw ADEAOAAwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABgAAQH///// /////4EAAAACzgIAAAAAAMAAAAAAAABGAAAAANBNu7xrcssB0E27vGtyywEAAAAAAAAAAAAAAAAA AAMABAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAHAAAAAIA6MExAAAAAAAAAIjaFwD8mhoAAAAAAAMBAQMKCgEyg3MABggAAAKCKAASg24A AoIpAAKGPQASg3MAAoIoABKDbgACgikAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAABAAACCAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQD+/wMKAAD/////As4CAAAAAADAAAAAAAAARhcA AABNaWNyb3NvZnQgRXF1YXRpb24gMy4wAAwAAABEUyBFcXVhdGlvbgALAAAARXF1YXRpb24uMwD0 ObJxAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAMABAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHAAAAAIAC8Pz AAAAAAAAAMiTGgB0qRoAAAAAAAMBAQMKCgESg2MAAoY9AAMOAAABAx0BAAESg3MAAoIoABKDbgEA TwBsAGUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAKAAIB////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA mwAAABQAAAAAAAAAAQBDAG8AbQBwAE8AYgBqAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAABIAAgGAAAAAggAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAACcAAAAZgAAAAAAAAADAE8AYgBqAEkAbgBmAG8AAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEgACAf////+DAAAA/////wAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAJ4AAAAGAAAAAAAAAEUAcQB1AGEAdABpAG8AbgAg AE4AYQB0AGkAdgBlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgAAIA//////// ////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAnwAAAA8BAAAAAAAAAAKC KQASg3MAAoIoABKDbgAChhIiEoNwAAKCKQAACwESg24AAoY9AAKIMQAAARKDTAAADQKGESIAAAoB Aw0AAAEDHQEAARKDcwADDwAACxEBAogyAAAACgKCKAASg24AAoIpAAMdAQABEoNzAAMPAAALEQEC iDIAAAAKAoIoABKDbgAChhIiEoNwAAKCKQAACwESg24AAoY9AAKIMQAAARKDTAAADQKGESIAAAsB EoNuAAKGPQACiDEAAAESg0wAAA0ChhEiAAALEQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEAAAIIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAP7/AwoAAP////8CzgIAAAAAAMAAAAAA AABGFwAAAE1pY3Jvc29mdCBFcXVhdGlvbiAzLjAADAAAAERTIEVxdWF0aW9uAAsAAABFcXVhdGlv bi4zAPQ5snEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAwAEAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABfADEA MwA0ADkAMAA0ADIAMAA0ADUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAGAABAf////+JAAAAhgAAAALOAgAAAAAAwAAAAAAAAEYAAAAA0E27vGtyywHQTbu8a3LLAQAA AAAAAAAAAAAAAAEATwBsAGUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAKAAIB////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAApAAAABQAAAAAAAAAAQBDAG8AbQBwAE8AYgBqAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgGFAAAAhwAAAP////8AAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAClAAAAZgAAAAAAAAADAE8AYgBqAEkAbgBmAG8AAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEgACAf////+IAAAA /////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAKcAAAAGAAAAAAAAAEUAcQB1 AGEAdABpAG8AbgAgAE4AYQB0AGkAdgBlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAgAAIA////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAqAAA AM8AAAAAAAAAXwAxADMANAA5ADAANAAyADUAMAAwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAABgAAQD//////////4sAAAACzgIAAAAAAMAAAAAAAABGAAAAANBNu7xr cssB0E27vGtyywEAAAAAAAAAAAAAAAABAE8AbABlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACgACAf///////////////wAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAKwAAAAUAAAAAAAAAAEAQwBvAG0AcABPAGIAagAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIBigAAAIwAAAD/ ////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAArQAAAGYAAAAAAAAAHAAAAAIA C8OzAAAAAAAAABCVGgB0qRoAAAAAAAMBAQMKCgESg2MAAoZIIgKEsgMChj0AAw4AAAEDHQEAARKD cwACgigAEoNuAAKCKQASg3MAAoIoABKDbgAChhIiEoNwAAKCKQAACwESg24AAoY9AAKIMQAAARKD TAAADQKGESIAAAoBAx0BAAESg3MAAw8AAAsRAQKIMgAAAAoCgigAEoNuAAKGEiISg3AAAoIpAAAL ARKDbgAChj0AAogxAAABEoNMAAANAoYRIgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEAAAIIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAP7/AwoAAP////8CzgIAAAAAAMAAAAAAAABG FwAAAE1pY3Jvc29mdCBFcXVhdGlvbiAzLjAADAAAAERTIEVxdWF0aW9uAAsAAABFcXVhdGlvbi4z APQ5snEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAwAEAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADAE8AYgBq AEkAbgBmAG8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA EgACAf////+NAAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAK8AAAAG AAAAAAAAAEUAcQB1AGEAdABpAG8AbgAgAE4AYQB0AGkAdgBlAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAgAAIA////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAsAAAANMAAAAAAAAAXwAxADMANAA5ADIAOAAxADYANwA4AAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABgAAQH//////////5AAAAACzgIAAAAAAMAAAAAA AABGAAAAANBNu7xrcssB0E27vGtyywEAAAAAAAAAAAAAAAABAE8AbABlAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACgACAf////////////// /wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAALQAAAAUAAAAAAAAABwAAAACAAvD twAAAAAAAACoEBgAdKkaAAAAAAADAQEDCgoBEoNiAAKGPQADAgEAAQUBAQEDAwAAARKDYgADDwEA CwESgm0AEoJhABKCeAAAEQAACgECgWkAAoFmAAABEoNjAAKGPgACiDEAAAESg2IAAw8BAAsBEoJt ABKCYQASgngAABEAChKDYwAAAQKBaQACgWYAAAESg1QAAoZkIhKDYwAChjwAAogxAAABAogwAAAB AoFpAAKBZgAAARKDYwAChjwAEoNUAAAAAAKWewAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAABAAACCAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQD+/wMKAAD/////As4CAAAAAADAAAAAAAAARhcA AABNaWNyb3NvZnQgRXF1YXRpb24gMy4wAAwAAABEUyBFcXVhdGlvbgALAAAARXF1YXRpb24uMwD0 ObJxAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAMABAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQBDAG8AbQBw AE8AYgBqAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIA AgGPAAAAkQAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAC1AAAAZgAA AAAAAAADAE8AYgBqAEkAbgBmAG8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAEgACAf////+SAAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAALcAAAAGAAAAAAAAAEUAcQB1AGEAdABpAG8AbgAgAE4AYQB0AGkAdgBlAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgAAIA////////////////AAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAuAAAAD4AAAAAAAAAXwAxADMANAA5ADIAOAAwADQANwA1AAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABgAAQBDAAAAcAAAAJUAAAAC zgIAAAAAAMAAAAAAAABGAAAAANBNu7xrcssB0E27vGtyywEAAAAAAAAAAAAAAAAcAAAAAgDowSIA AAAAAAAA8H0aAPyaGgAAAAAAAwEBAwoKARKDYgADDwEACwESgm0AEoJhABKCeAAAEQAAAAAAAQAA AggAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAEA/v8DCgAA/////wLOAgAAAAAAwAAAAAAAAEYXAAAATWljcm9zb2Z0IEVxdWF0aW9uIDMu MAAMAAAARFMgRXF1YXRpb24ACwAAAEVxdWF0aW9uLjMA9DmycQAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAADAAQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABwAAAACAOjBaQAAAAAAAABwhhoA/JoaAAAAAAADAQED CgoBEoNIAAMPAQALARKDcAASg28AEoNzABKDdAAAEQAKAoIoABKDegACgikAAoY9AAMOAAABAogx AAABAogxAAKGEiISg2IAAoEgABKDegADDwAACxEBAoYSIhKDcAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEATwBsAGUAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAKAAIB ////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAuQAAABQAAAAA AAAAAQBDAG8AbQBwAE8AYgBqAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAABIAAgGUAAAAlgAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAC6AAAAZgAAAAAAAAADAE8AYgBqAEkAbgBmAG8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEgACAf////+XAAAA/////wAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAALwAAAAGAAAAAAAAAEUAcQB1AGEAdABpAG8AbgAgAE4AYQB0AGkA dgBlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgAAIA////////////////AAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAvQAAAIUAAAAAAAAAXwAxADMANAA5ADIA OAAwADQAOAA4AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABgAAQGn AAAArAAAAJoAAAACzgIAAAAAAMAAAAAAAABGAAAAANBNu7xrcssB0E27vGtyywEAAAAAAAAAAAAA AAABAE8AbABlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAACgACAf///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAMAAAAAUAAAAAAAAAAEAQwBvAG0AcABPAGIAagAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIBmQAAAJsAAAD/////AAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAwQAAAGYAAAAAAAAAAwBPAGIAagBJAG4AZgBvAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgH/////nAAAAP////8AAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADDAAAABgAAAAAAAAABAAACCAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQD+/wMK AAD/////As4CAAAAAADAAAAAAAAARhcAAABNaWNyb3NvZnQgRXF1YXRpb24gMy4wAAwAAABEUyBF cXVhdGlvbgALAAAARXF1YXRpb24uMwD0ObJxAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAMABAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAHAAAAAIA6MEmAAAAAAAAANBaGQCENhkAAAAAAAMBAQMKCgESg2IAAoEg ABKDegADDwAACxEBAoYSIhKDcAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAAACCAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQD+/wMKAAD/////As4CAAAAAADA AAAAAAAARhcAAABNaWNyb3NvZnQgRXF1YXRpb24gMy4wAAwAAABEUyBFcUUAcQB1AGEAdABpAG8A bgAgAE4AYQB0AGkAdgBlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgAAIA//// ////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAxAAAAEIAAAAAAAAA XwAxADMANAA5ADIAOAAwADQANwA4AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAABgAAQD//////////58AAAACzgIAAAAAAMAAAAAAAABGAAAAANBNu7xrcssB0E27vGty ywEAAAAAAAAAAAAAAAABAE8AbABlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAACgACAf///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAMYAAAAUAAAAAAAAAAEAQwBvAG0AcABPAGIAagAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIBngAAAKAAAAD/////AAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAxwAAAGYAAAAAAAAAdWF0aW9uAAsAAABFcXVh dGlvbi4zAPQ5snEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAwAEAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAc AAAAAgDowTMAAAAAAAAAiOYYAJR5GAAAAAAAAwEBAwoKARKDSAADDwEACwESg3AAEoNvABKDcwAS g3QAABEACgKCKAASg3oAAoIpAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAEAAAIIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAABAP7/AwoAAP////8CzgIAAAAAAMAAAAAAAABGFwAAAE1pY3Jv c29mdCBFcXVhdGlvbiAzLjAADAAAAERTIEVxdWF0aW9uAAsAAABFcXVhdGlvbi4zAPQ5snEAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAwAEAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADAE8AYgBqAEkAbgBmAG8A AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEgACAf////+h AAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAMkAAAAGAAAAAAAAAEUA cQB1AGEAdABpAG8AbgAgAE4AYQB0AGkAdgBlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAgAAIA////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA ygAAAE8AAAAAAAAAXwAxADMANAA5ADIAOAAwADQAOAA0AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAABgAAQD//////////6QAAAACzgIAAAAAAMAAAAAAAABGAAAAANBN u7xrcssB0E27vGtyywEAAAAAAAAAAAAAAAABAE8AbABlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACgACAf///////////////wAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAMwAAAAUAAAAAAAAAAEAQwBvAG0AcABPAGIAagAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIBowAAAKUA AAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAzQAAAGYAAAAAAAAAAwBP AGIAagBJAG4AZgBvAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAABIAAgH/////pgAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADP AAAABgAAAAAAAABFAHEAdQBhAHQAaQBvAG4AIABOAGEAdABpAHYAZQAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAIAACAP///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAANAAAAB0AAAAAAAAAF8AMQAzADQAOQAyADgAMAA0ADgAMQAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAYAAEBnQAAAKIAAACpAAAAAs4CAAAAAADA AAAAAAAARgAAAADQTbu8a3LLAdBNu7xrcssBAAAAAAAAAAAAAAAAHAAAAAIA6MFYAAAAAAAAALA1 GQCMEhkAAAAAAAMBAQMKCgESg0gAAw8BAAsBEoNwABKDcgASg2UAABEACgKCKAASg3oAAoIpAAKG PQACiDEAAoYSIhKDYgACgSAAEoN6AAMPAAALEQEChhIiEoNwAAAAAAAAAAAAAAAAAAAAAAABAAAC CAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAQD+/wMKAAD/////As4CAAAAAADAAAAAAAAARhcAAABNaWNyb3NvZnQgRXF1YXRpb24gMy4w AAwAAABEUyBFcXVhdGlvbgALAAAARXF1YXRpb24uMwD0ObJxAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAMABAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHAAAAAIA6MEvAAAAAAAAALgRGQBc5xgAAAAAAAMBAQMK CgESg0gAAw8BAAsBEoNwABKDcgASg2UAABEACgKCKAASg3oAAoIpAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAE8AbABlAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACgACAf////////// /////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAANIAAAAUAAAAAAAAAAEAQwBv AG0AcABPAGIAagAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAASAAIBqAAAAKoAAAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA0wAA AGYAAAAAAAAAAwBPAGIAagBJAG4AZgBvAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAABIAAgH/////qwAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAADVAAAABgAAAAAAAABFAHEAdQBhAHQAaQBvAG4AIABOAGEAdABpAHYAZQAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIAACAP///////////////wAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAANYAAABLAAAAAAAAAIghxaCoCYKGAIoaXVYhGRJEpLFG GyXWWGOJFS1q5I+yCstIUdBSiDVd0aKvbuOaUj/KVlutoN66vlopFbbahVfd1lZ0tVV0UVq7trLf 58w9k+HmTnK5JIDvkcdnzjlzZs55zu/8mTPnTvo4jjMbFID+4Lq+jnNzHw4S7r1LHGd0meMUn3H2 VMfp4zxxj+Ns4Jxse0KCPRKvIzCW5ThnhNIr2v2gv7N8V7bDBZyxoBhwuTF93D7OcI7zQVZ+/DUl a0xA524H24DOHe9mmfP8+3qVR7vZTh5xcke5OcHxka7jFBIWA2TF+TYXjblOlstxOdB9IOcl/Q/3 3rkTuLJ/rmWF2/M43B1z+/SdxsHERHgOLHfHlcuqp93/ePUdVz4e8O9Oux3/smrLinf69OlzE+d/ gfLmw34ZHKeEY+Xza0BlUflHJuA43iQOE67jWHmkSpwJQOeL5Swr9Hg/qNP/RxFi72OPKYt3AeHj FccFi8GcBFTOHfm7hu4YKpQW7MgXdg21aalyzx5zqhfK+2T8CecFx8p7KbB5UB1vAnGgOi5z/fIX 4Se0ylDi2Goac7P6jSdM9zJ19MOyCnNK4n/K82TnPOcs/l/D8aluljcLbvDjd1M0o5vOq3bO4Jw6 p9hxPvwOV/EqnJYthr2nsivldxcM93n5qYYJg60LHxcQaMul/OtYbijQscLOA81kQHn4cPfu3VDg WhSI6+9kk/M5zhXOpaZd+qF7+//NxsSWJO6Rf+PmbF1+w9zN2QK3MO6QS1453R4bQyTUm7c5ezvn bNte4EzN+vXp/pm0E/zNFfLn+0ExyvVWgfNFwhY8vzl7Gunk993tCyfM4Rr4Jzu3LywaN6Xu05IB 7vsfFC0o5oRPCM+H6UaMe4N85nO0mIDt5hrDFgzLH7aAWxj3+tsFzhj8iw/Dzq/fnP2Hx14+/YVh JScP/+aD5hLDr9ucnZ849x3SN13nl8bapspeB2aDMjxjQB2YCPjnrF27bbKguNOAbKMGhB1ZNK6e wA67ObpYdXsIGABygZy970DklI23gzVANj7G7bCNMo5lx0Ugyt67st/JznynCUu5xvF++50K7+tv V3gbf2HYcT81HP/KFypN+K+PMkyG9sF+bbmkWVjPSgLOA7VAcWE9v0RAd3p2lKNDS+mZMNNOem4g bh3IVM+FpG0GuKA/0Nh3hnO1MxM1m5zLaIFOWRUaHotedxv25r/oc/HHfvgvDkXTYyvjN4827LmT DfesxgvIVwkaSt8GMBPwL9BYcZcC5b8GhJ212QYCk8oW2GyMOGltofS2nq39riZsBUjWuyfG1bFc txioL8+C5Zpe/8Hkwqvbqlu/taNKrPEzeexUuxFQu9txUtedYM7tYKnY3TipdFYLbNibgV/3LABT 0fwGuAbWeXV9VhbO77+y0J5PfXj2mGgvlN/J+BPOC451jVKQB3SsfkP1Vguku+YgEzkeD6RVNiz3 +uw/VjWcc391/OpYtbSS/5crHjSscLS7qdDt4+VzbiLN7lBeIrU7jvOLgPISdrZMCh8ClF8kCHSK UdYL8I8F80EVkbfCtbDu39C/rHBZbGVhQ/+doGwPvVJdjyRpaSe9Vuhk4OvlZLkclwMzT4BfAnK5 W79fofyHoXB73jPXTzVzPpWhOBFudXg0/zljk+IbWxcafffFNjWApbLNHMLlpl7/jH+Q+L/VX+ns MecGttlI+KegCUxHc+W7pN/KwpJ+Gwfb8/fFNnXfdG3ztsc2VDneompp9W/z52CTvzB+G55sm9I9 3AZGci+B4qVtp1YvW1aVP5Vdhe30q5yzE3wdfAaywds52waX9Csd3NBv22Ad2+tJu1TXI0nadtrG yauB7LTMjZ4DDF1SM6Vh8MJqzQGaOb8J4IIxqx+ec2hl8xjN9H/nuk3oxCz20ysrDH/UZthr3+D7 j3rPcPzFgYxqnBcxGyjgurZdFyWOoT3mslYPTMzMBZrhiRhHDgFXwkNhxdm5gOIKCVCea0DY2XGq nsBweZQHOw/oanxq57w1IHl8SlfbWaRtALhA2xw8oXnJsXmM+W9XOH9dZNib+7Dvf/c/ff9vPvT9 EYr6Y4bKI02KgI7lhgIdKyxZ09mEVaJbEYkuh0fDSm81VdxxBCivNSDswpqGyhGM+1ZXpbH3HYhB TMW/DshGM9WzmbSpbPXLzFUvZ45lnmwad04ys6t+30C5Yyu9BcsNOzse9/0z/8vn2Yd0OavK1FYn oZ2eA2bByc8BitNzQHe2Gi6PngGygbVTsVyyttsI2wT2RlutL2hsVz+wmrQrAC6wVT13VDNvvYp+ YA5z17n8d5lqYPu9k5xf1lQ4P5peIfZeajXsbH/NcPyhnEoT//shhuNrxhh2d7q+/8wLDHtZVxvu yf5iOXm+Do3PBpvBVwH/AttWnOa7KluUbTcSl6rceYRLf+zZsNqH/LEEouqmnfg1INO6mUXaBoAL 6kb3DrW/f3gI5TdVOE1vGY7PzkXZTRXecSMMu/+sGiC+x/uRVWjZCB4H1wL+BVor7hqgvEZpXb9H OU5Kqx9pJc1ykKmedaSdBnCBnjRNp4p+ZB5d4vEV8buGVXpHtRh2Cr/vc49q18j9FqPNXHA7WAr4 F2inuCVA+YrSTuVI5Nnolo1fLrlv8AijXBnr5ZK2HOACvZTXcY734pO9oI3K9R1uIG2+C5K1UZy0 4V+kNrUmf/wPl6zHOsLaQKb200zaJoAL9FCfPheV1U9eSo/plG/1regDp1LW5G7MM+ztPtz3P3mS YWf+dJ8jrCvTcehOxJF+/xck66c46dfdOBQuT4zyZQO1ZfVzPdnXnfXxx5NvfGWJGYdmce1Ufd1Z jO+J2ejOBjPKeMffYthp+7HvH6a+j9HoTPV9Nb3Q151M3nLR7Xz4CJh/QXtV3DACpE9Ue60nLlSO tMb1dtKsAXtjq+noGRo7Frroxtgw+i7D3r0/9/2FnxiOL9Vo3htjx4mU60g0uxA+HuZfoKfixhLQ nZ6hcsTseqlsU7YqNwrkgX2Zg4b1bOZaqdp+eM7mLBvnt/2HV/m9Y9s/+/5hv/L5hUSfsKSoV9r+ CeRxPNp9BZ4M8y/QVXGVBHTX9sPlSXcO6nLtcpBsq7mu/wxbRNyfd7xj+kytK+ldlsP7mDV33VVV SpzqKQuMTMDZi+fzd3b8mVQd9a3r2LqPcaEL8I8Hy8EIyv8u3AzLvpbFSguWxXYNXTZI0HFpgU2b yZrGVK7ZDmy7jblOlotf2ti1oJc4ltuX94TSUO8B41fPCliaWm1tvNY5b+JevbGedLwpRef/Wf00 f7HHaB3Uwy2EF6P/H+E5QT3sWQeqE5s2k3qwNqU8qE5UHy1A9lnmRq+PWP26ekdSa+aKM51ZFO/+ Cu/kRyuc/FcNeznv+3xxVqUJf/QwnyNG+Myf4R+kLKeg3c9hrRvzL2jnipuS0LWG47ALP8N3lGNq MIfM5uRYIoHV3/afrYQvB8ltvMxNT8860k4DuGDupPqpNTMn57uTGHPyKr0f32nYKX7M5x7VrpH7 3Q2moc+/whfC/Au0U9wMApSvKO1UjkSejW7YtnHJenmEzgOZ6uWSthzgAr2UV+be332zF7RRuR4F 0uYncLI2ipM2ykOUNrXEjUuYULIe64hrA5nq0UzaJoAL9OiH5+u0x9kobd6wMcTJiuKvHmLYO/8L ht0tRxl2fljmx19W5fsjrCvTufcT5Ef6/RpO1k9x0k95jtKvnrhwebLxhxE193Y5rxwka5vrdh5/ LyuZVvWn9lHVf9v9TNXdjL8jSSfQRU8ylHQcHk/Dx7FQn34SaR4Ad4CPQA4o6XtRbknf0sElAwQd X5RrbSKTPl39eDtYA1TOrsbW7t7HcInIPTjSR+8WLowVVFuWXlY3G9+bY6v0S+Wsfn2JtMecG4yt 4wlvA63gY6DrJNeB/DZtJvVQyjXzgPKgOlF9tADVSZkbPRZY/dIfWymYGVvFGlsNM7aacMZWwxEt OPOxtZayPAKuBXGg/s6uj9dy/DSQrl214Y6x9fy0xtYtXO85kNx+09WzhbRLAS7oG9VX1LIiMces Slxme8iKN1kl573D/JmshsPjmcNon8fdT/n+fOYw+OMFh7NKTvgvTzTstk8y7Mw83+ce1V15V+fz LLgFvAzCuivu34HKFKV7I3Epymv0VzrVmXhggrMTDAVtiTjzvmILYb1RH2eYlaIrWCW6lPq4hrpx /s9h6M3biieX+nzZI4adI17w/Zpbas/IHfnozt6Qs0sNe6Xlht03v2iYjNMerAsfFxCo9io9ixLH UJfviVQfLvg1WALeBuH6cPFvB93VR+fyXmHqI0Za6S+kUx+rOW8FSG4fXY0B6T5fjeW6xUDPZ1mw nJ6r9H5k/EOLDKuvPxieoWaQt5FA9WHm/VSK3S+ybNDKwtcG98x+Edu3N3CfWuCPt6n3i7wyfGG1 9oVoP6U0k1/7RWw42nXaL1LKNWWT0lvlEbBgNfOE6zjWOccB2a6tn8RJQbtV+BCga8pO7fiGnZmx UXX8V1BF5GBQC2R7zw4uK9yRv7Lw2cE7Qc/uHSnn+iqntMt1O8/D9KyZu/WsKq2DaB6mc/dFk71Z BzmJe70LPKB+kXWPvjxz92cdBOi4tK/VMJM5wlSu2Q7WAN92nCyXY2midgalvXeGUyPnatLw6Js/ q1J7tSxNrbY2/mCcqy2jYCeCP4CbgF8Pe9aB6mRf6sHalG3Pqo8WoDopc6Pnala/g32u9iBlOR38 HNQAmnUwV1PcdCBdFRd2qddBji62653qG2KJBFb/fZkbhPVs4bpL/Wt/budqd5D/c8DroBGEdVfc V0F3c4MDOVdLpz7OZQfUpbx7msf6wg3+6kLuIH8udtmt/tzs1Sd8f+07huO/93eWeE+x7sdczZ1Q Yjj+7VMNe9O+1EtztX9C76+BV8FcEK4PxV0LuquPFOU145HSWahd2LahMLnk9nEfYa1AfcwY1x/T 8Hb5m5nuntft/slyrqM+TWNIX10Up72Aej4Xn/fEv1YNvu+vhntrzqb+RM7uC/R9HTooX1YTzjX7 KJXnRjADeEB9k84r6ce6SL+VOW/1X5lj0/TnFHvMKd5I/idwONlQ0rHmPrp+HtA1NfY2gFrgj72p 5205L/5P1eiGtWaslG7y31j4U8MKT563Hai9lIsox/ngVnAxyAZb+385hz2/OVv7vwi+vId2Q4iX FmoDe6ujtCsH0lPa5bqd523P3/Ojatmb5m2bNj203+Ztl1GgB8jX8/Df4BxQN/Si3LqhPxtcN1zQ 8edjjU0aah/q8/f8PmBparW18QfjvK0J/dvQ/gX4s6Ae9qwD1Ym1vZ5oz8u5z0LQ3bzN6qd5Wx3n TwO4YJ6h/iGxb2jWn/ydfLlX8S6BvRM33GqYXqYH1xEauZ/2hq2Fvwmvh6Fgjqa4n+JXvmpA2Nk5 msrRW3udwnq53EdtHxfopbyO4/Xx/b2gjcolTaTNLXCyNoqTNspDlDa1xJE/46y9DaQK1Y+tA20g eRwuc6Pn/WE9mknbBHCBHv3whPcGadXQWM8Z7/gKfbrbcPykQbxfwareH2VYq4dd7cwp4Lq2zy5K HENdrlE1E39rQr/HU+inOOmnPEfpV09cuDwx/NlAfeuABKBgHLHavkfYGyBTbR8i7X0AF2ir8p/N nFPvsq5gx9Ns5pwLzLyz2HF+9TytlZ04rdeiL7s7K9sNO6es9/1/+JXvfyOxC3RtYhfoqEJ031Th vuHvAo2fMtn4vQ/rDcdHXW44qtVnUi+tlOMatH8Kfgp+CYaCNq+4TfhV3hoQdrbNy+6itKAOTN2o nixUVzoWqw7FqkO55HaxhbDnQKZ110LapQAX1J3up9XcYp44r6H29NRQzO/9Zvm7fm5L7CPd9Yzv X+vvEIyX9TG7gLy3/N1AbuMIf1fQEacZju86o8tdQpm941Deb6QOtsCPwL+DoaB+FPdb/CpTVP00 EpeivIHuqgtbDzrWtcRyyfXRTtgakGl9zCJtA8AF9aG61z5u/3eX3sZv0TpqKr0rXzTsvKH9bTWV 8TK1DsLfMrvUGfd6cuybTR4Wo+Wb8Fr4Yziss+J2Eqa8RulcT1xHOfw2Iy0tOOyk52rCVoBkPWOu k+USXg72dg3OPn+NJW1xIr2eO+TiV0+ovnnMCvYeTTBrcAfTc9dI8vc4+EfwJDgczZXvZYfccuiy Q3p2ba2B69YC6R7128pXhk+s/rcff8topf1aP4yXG78NP1ieuR6jHIehlfrwY+Bs+NlDrkez0oKz YR3bdqz57RDi8wCnBvbIYVrPrpqrtILlwNfOyXI5LgeZ2qnSlibSa24ppzViPec+f8/GgCfddsRB 9XtL5fm/wAxwJ2LeAyv/vNvJqRsm/HRAWHd7zClpaa3rq550Temejs1umbep+ol7D+fZbKN57t3w /M+N34anstmRXFsgW5MMJR2rDdq86ziV/cRIfAFxY4F+J3E3fBf8L3A2WDVs7YC6YRflDoN1bK+3 P+0x3XeMqezR/gZLuuq9mXj0mlt6zR6PR7NUzuomm7DHOWgvG5S9zEZz7X1fDfSuUeftGLqyEHsE r+TbNNLdHnNKr9njoMUvVG+duczopTWExRt+afw2PNkeycs+/waYaxhny5eOza5Cq7+Sag08GMhm Pyx6P79u2M8GPzHs/Xwd2+v1hM2u4fotQH1omRv9jPetN4qr2pqPqUrv3Y73E//Z17v/Vc2ZKpw7 2YcDu//CPhw4/in7cBTe43OnMynLODALcPk95k6KmwhyQA0IO/sMUU9gxz4c/zsgqgMhlkhg9R9I 9tUfbgObgDQc4/p9JV6njOORcJE8Sd8RCuu5mtgV5pyOuWgu/s7vOubyw4cW/5nu/b9HUZ7tZq8w 7P1H4tmuZpvxx9cNNs9o3jeOM+xOPdmw0+eLvv/UC31/RA0UcP88gAmmvf9jOeeeAFywGJwLlN7u g1LcOUBli9K/kbhU5dY7Ns1fVXditA+eD2w4QUFfZOtmC2HPgb2pG81JGwYPN78vaiHtUoDbrbII ul/ye48mtk9/Msm5e1iFcy37z2Dv7UcMOw2/NRyf1b/ShFfzK1fi3af5phMcH3KSH/7SWYZ7tk0o 7yr/98AhZP4hOFwnilsLVKau6iRFeY19KJ1FNsdCLBEGdaqPdsLWgEzrYxZpGwAuqA/ZREebVd+j ffTqe8x+evoesfoesfoeEx5h+Zk9H88mD1vBI+ATEAdhnRX3NDgY+p6wfa8mTysALtCz674nr9Ix fQ9s+h5235u+B7/pe9hHbfoewk3fk1fp9z3Em74Hv+l78EfUQKZ9zzMU4lmgcfNlOKy/4v4d7M++ Zxv32wT2xtafv2dJdVvzka7G2ai6SfVbc+eBLDPCOm8t8nnWw4a9Rbt8/ocC+ntG3oeLDXtXTDDs PHO27y+cadjdfIMf3sN1cyiV0R8troSL4HDdKG4oYd3VTapy55FO/Y8dD+z4ECNMUJzcKKBz7biw juM2kGndNJO2CeCCdtMPT0c/VOw49a+gPeuu9/dDU9ZTh+cbdv58hGGv5kQ/vD89v8byHtS8mbyM Q9fD4XPhMXBYc8WNJkx5rgFhl3ouVNxpT3KUttu4WG/Yfap5gXNR4reTqxO/nRyUWC1d8JpZNXWu z/FXRfuMNuxtcA27d7KSx++snfOafP+yOb6/B+tgOTp8AZ3HwXoPmzwfVdxE4rqz+1TlznQ+1Mr9 lK9M7b6OtNMALrB7Pdsl3rV971L1NpXeEz8w7Dy8wecIVTMbaxu5n9bkXHgsfC4MBfNMxZ2DX/mK sm2Vo/feD3b04S73KQe4QC/lld9jzVTvUBM5Cmamjcqltu3C4+FzYd3PzsEVJ20UFqVNLXGyWbn9 1W+G36c5L33oW9EGvg+FQu6dtFNZ1SY9y+BfVmnYqZrhc4R1ZTKPaKbMJyGOCzfC58LSyuqnOOnX Xb8ZLk+M87OBHZui+k2Xc2QryW0z1+28n+QvW3aZdbXZX397v+4n+SPlf4A8XpLVsZ9k/VEX5a4/ 6meDd4wSdPz52E8iDbUWNP3Hh5nvm4m1n8Rqa+P1buImytwbe/llE6mcbXfqw+wx55q1zfGE/YF6 aIMbqYfPYF0nuQ7kt2mZe3n2mFMzWmfTuLEQyD7L3Oi1DatfGvtJ/PGiIjFeVCTGi8iZUGZ9YiN5 7oNOa+Eh8Hr5gW3Tivsp/gM1XoT1csmH+gBc6vGih7Wp40bSRNoMTaGN4qSN9Npf40VYj2bu2wRw gR7qe8P9a2K8qPD88aIiMV5UJMaLisR4UZEYLyIVLOC6eUBlLUocQ93uJxmW0O/MFPopTvr1xngR bo9j3PTWHcPaTiNfEwEu0JbsOlPYP+LcxzsYPZO0nM5Y3LPPJvXcYyciPwXncMOX4HB7VNwmwpSX rmxO+VR9ydm+zT7feYTNA8ljaZmbXr/FaVHtcNK3u1Qksz5K7bAfBZYmA1JoojhpIp260iRq3rad dNtApnq0kXY1wAW2omeQq1LuPXLaNf/nyfcpzeTgYf6Tr9uqmRzWdI12eG2qiL97vGGvbaph9+ha P/7Vy/3wExYZLp5/mx8eMdPLTPMW8n9IQvP6FJorTpqrnFGaNxEXoUGwJpHNORYDEsfiGBDLJdvv FsKeA5nWVwtplwJcUF+6V4r9N9Me9PcXPfUf/hNz6UeG41f7T87uc/4XyJzF/hfI4n853Twpe499 2X9yXnVJl0/OmdWN8i79t8CV8O/gcB+huN8SpjJF1U0jcSnKa9LkEKc6sXWgY11LLJdcHxsIWwcy rY+FpG0GuKA+tB6mPmxmx56j92+nvfCk8+3Nhp2+ZmdkpbPV33MUv30s7YD4mimG3f4NhpnQ8Rxp Xfi4gMA8IO2KEsdQl2PaAuL7o++bcDX8MRzWXnE7dQ6I0r6BuD3L5udDGlvNdSzIJev9GGHtIFnv mOtkuYSXg+R9Hum+V59G2omJ9LIDuUXvFEzRt5EWvVMWsPbaKNxyV39DZSTXEKiJSYaSjpHMuQRM AJpniuUsS+Hj/aBO/7faRD0LnMXFi6mgR+A5sMq0fuSuoevLhNKC9SOFA/NtJKtfV++NO/6Oxf75 +ydWT9l0HZgNLke7MQTcDk+EFWefDxR3GgHSNcre64kLlcO0ObtGp/U9OXtfO0dpJ2wNSLbxMjd6 jpKOnpMZkw/03+OQZpVotgquTaHnwfL3OMJ6LqQumgGuiz7amT/J/x3fV9b4v+Pbor91wrcWbvyT YfeTAv/3eo+W+b/ju6qqF36/t4BMXoy256HtXfDMJI0VdylhvdlHryYPK0Cy/fZEHz2W6xYD9fEU xTj97mf067XViye+b1jrMr2xJrO3/fAMcqe+/1n01rPMf8KvwMp33fBBA9uPHDTQtn3qw7PHRGe0 DtNAwlog3aP2jG6Z9x777aabtSxpJv/dw2YYVvjBst9pG1ptpCzvwdvgbLDqyLV5648aNHDVkXPA 2jyrl7QbQnwe4PSgP+UwLR2ncmIrWA587Zwsl+NykOlcQmlLE+k1Psu9+aUBZj1RrD16lfcOMizN DxZ7VZ5HYKB/gSfBNQiq/NeN4DdpI1YWth+z/78JMuu0QVP0TRD1ydJNfn0TxIanstmR5FnABCLn XdZ+1B5T2U+MxBcQpz5nAidVoUUNrHErm7BVx5QVrh+5snDVMTvB/v3ex9Al5Wb+6ezn757+ivKP oPwnYxTNcA46lIwrLSgZt2toycmCjj8f3z2Vhpq76++YWNac3mpr47GvXlvn39sxRfY4HmymHorR /xTqYU5QD3vWgerE2nhPjC9ruG8LUB9Z5kbPRa1+Xc3tO/ZDHLjvnn6Ihqeg3aFomPzdU8Xt3XdP /fUg9QsCfYdxVn87tw9rOMZNb202rGcDV601V+6Yi+p+tawVaO1bXwAzXz7lC2BiZ/xjPr/2G58j VgQKuEYeQI601wOaOPcNdNI3KnfDyd+oVNwM4pS/GhB2di9JHYE27/a5KEo7j3MpYad5Zbq26JK2 HOCCebzKyzv3lj91qU5ma1UqmwZP6dMXTtZHcdJHeYjSp5a4cUAu2ZbWEdYGkufZ6erRTNomgAv0 0PuI8DdDvUUDWElib169/+1T52X/26f6cpyxM74cZ+L5cpyxtx60r2by0i+h3zEp9FOc9FOeo/Sr TypPNv4wotaZXM4rB8na5rqd37mvO+LwKeFvb40kneCkOfew9Rojge3fZ9F2TuIKL8IenAOWHcu3 tI7lm05jBR1/Pr6rJX30Pa11R5wUsPSyutn43hxnpV8qZ7XHlIL2xblBPWi94kTiNsI3wX497FkH qhN7nf05zlr90h9nKdgB+AbqMrQ7He3aYbVT9Xd2DU1x0xO6dtWGQ/OFTvsNSR7UnR1ntxD2HEhu v131jWE9W0i7FOCCvlF9Rec9d85B+w3UuWirPUkb4EY4rLvivkqYyhSlu9KkKK/RX+nUFsRoHrwj kV/OtgdbHxsIWwcyrY+FpG0GuKA+aGvJ7xIOmjW4G9D3a+TvJ/BcOKy94q4lTPmP0r6BuH15T3If 6VtBst5drcHty3eztKahPVJi/S42q/Uww+rTe2NtQ7Ynt7ffzRqJ9jNI1wg/CKvfrxvBt45G8N2s Ufv/u1klFx42Rd/HUt8j3eTXd7NsOPrt8b3TA/XdLK0bn49WV8AXw9lg6yi+m3Us380axXezOLZt Hrve5/W4FVzfA7Lfrmw23Xd7Y7lWMdB6HkUwTu/0tJ70//5u/EG3DjeSHP6AjOq59DlY63DK9/qy lYWDxu7/NbisV04wa3B6FpVm8msNzoYn2ylZPSC/k30GkbRO9zJcC8tOd40pKywZt7Jw15idoGfX 6faHnWp9SpqfXXDhQWmnJ9CJyk71N2CtnZacvLLwjFP3v51e93C9sdPbGi42WskvO7XhqexUbU2g x5lkKOlY7c72bToeAvIARQ7C7bOb+pmvoYNs8DrY2uDYU8sK6/9uZeHYU3eCnrXBTdwzDtRXdjW/ vePKj8y3AfS8sJrzZbu4YD6ViyfVb6kO9r/TugOtzwZnUjnJf6dVcQ1AZatRaUPOrkE1Epaq3HmE h+e3mnPIT10He1U4DGzAznXbCVsDkude6dbNLNI2AFxQN7p3x/t754D9ndb+aNyInifA18L8C57l FHcNAcprlNb1e5YjeJaQruqr5Wxbs3q2ErYcZKpnHWmnAVygJ03T/s5nv/2d1j+jzVwwAJ2WwvwL tFPcEqB8RWmncvTWb5PCfYPLfcoBLtBLee3Nv9OaiybSZmAKbRQnbZSHKG1qTf74Hy7ZftYR1gYy tZ9m0jYBXKCH1hz32Gt+gP9O6+CEfnpnm2xbipN+3a2ThssTo3zZQG1ZbVOQS9b2PsJaQbK2Xc3Z 9+U5U2uIek4ausQNWN8LO9ieM19Acz1nZtOg9btDtetlY1fmlIwXbjnU6qhnJXvMKWntVyjlxDyg a04FDaAWqA6i9n1876rJU/StMOmmd5y3/1Ol8dvw5HnRgXrO/AzdDpOtUjj7rbCG8dcfWjK+tODt E64/VMdWr554znwO3Z5IaFfmdvVO050y+vXY/zff/hiEvssp9/nwXbD6Vrsmq7g78avNR/W3jcQd yG9/yI5tfcwiL2oDuKB/zsHTsWZ84L79UYyW3yMvk+GH4LDOiltLmPIapXM9cR3lSO8d8zbSbALJ fXKZm559rybtCoAL9NT8ufNa8NzEd4d4z3yQfvtjABo/Qt4vhONwWH/FPU1Yd88Gqcpt31Wr7tRO mKsa1pgpv8LlRoE8YOeymdTNuOz/zuzbH9vv9b/K/aPpvN2uqfBeajXsbH/NcPyhHN7Z1lQ4v+fL Q/oGyBrz5dIKd6fr+8+8wLCXdbV/Xg++212OJu9SGfThThV9fhEcrhvFDSWsu7rpyee2ddyvDexN uwnXTTNpmwAuaDeac3W032LHe/pJtOeXTz/4yOdz+qIt3wAp87/LFV880vi90yYaZkqQ8hccBVxX diXNpJ2O5aSZjhVmbU/n1IFmoN/VHQ6PhsfAYc0VN5ow5bkGhJ19Vq4nMFwetQPZvBADsn05e+/e tvtUbdP50seTzDc8/vYN/5dL8x4w7P3mZ76/7r8Nx8uH+79Y+uBk/xdLGyv9XyzdPMP3P3mZ7+/B OliONu8j+jhYc/UKOFwHiptIWHd2n6rcmfZJrdxP+crU7mVb0wAusHu6VvusyvO1+W3SUS2GncLv +xyhamZ7bRq53060c+FsdD0XDuuqOL0LVb6ibFvl6K3n63A/4XKfcoAL9FJe/edraRXV6jPTRuVS 23bhWAptFCdtlIcobWqJk83KJbftdYS1gUztp5m0TQAX6KE+KPw86vB8bazoA0djUaW7kTmHfvG2 +3Df/yRfy5Ny86f7HGFdBVw3D6isRYljqNt+U2sQLudpH/C5sNLb+bLipF93/Wa4PDHOzwY5QH1m VL85nLh8kKyt6kmaDQG61nwM+2YyZf0z8f8j/tFmhskJxqUeS2xsMo8gQFqp758Kh/OS5XaMOYOI KwRy+UDH/wt7BcQdigIAAEQAZAAAAAAAAAAKAAAAAAAAAAAAAAAAACQJQAHoA+gDAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAPAATwYgAAALIECvAIAAAAAwQAAAAKAABjAAvwJAAAAH8A gACAAARBAwAAAD8BAAAGAL8BDAAfAP8BAAAIAD8DEAAQADMAIvESAAAAvwEAAGAAvwMAAIABPwQA AAcAAAAQ8AQAAAACAACAMgAH8NQBAAADBGMw7eS9TyTkHzoAc4Xv7Jb/ALABAAABAAAA6DQAAAAA EgBgIRvwqAEAAGMw7eS9TyTkHzoAc4Xv7JbaAgAAAAAAAAAAAACgDgAAAAIAAJirFgAgGQMAdgEA AAD+eAGNUj1PAkEQfbuAwIGIVGIsTmIMFhhjYWFi4g9QY4QYS0DPaMIBCSTkOgpsrCz8AXYWxsLe r4pf4Y/QECsTcWbkFi/GxE1mb9/Mm903c6MQB0KuAjTmwCtCFtf+SSsKiU/p4XAop2U1M/IlDC+p oa9SGfIvTkxhFsQdAmnCz3Tqk5VTQJ8uS4w4SWxX2iclr+kAu/L2h+5+UpDWBT+KsEa6dOo6LXvH 6dh7DbdSx8HdoFMhszffO9fZ416YeAVix+i7apXVyySnLnk+zlmM8wbbcca2wWcTQf5aJMhv6WB8 QfqR97hXP/Uy6/96FVj3KyX5OqdFd9PouhHddYO7ortlcC3GL1YNfhLdY/5lmONj/q3oHsfXwfGj X3WQU3MdRc+tNmrAYdZ6XHnL/Nn37zqUqeMhwfcWjK5SNIj3Q4w3PIt2nhRLJoV9GM1AGlFB9zJb Sutc0Wu1HRc9/t+gjHmcSwJvW4OM9LJA9fk9DUn+F/ggeq/1AgAARABkAAAAAAAAAAoAAAAAAAAA AAAAAAAA7AmUAugD6AMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA8ABPBiAAAAsgQK 8AgAAAAEBAAAAAoAAGMAC/AkAAAAfwCAAIAABEEEAAAAPwEAAAYAvwEMAB8A/wEAAAgAPwMQABAA MwAi8RIAAAC/AQAAYAC/AwAAgAE/BAAABwAAABDwBAAAAAMAAIAyAAfwPwIAAAMELdYV5v6StcgX RefUK3yzz/8AGwIAAAEAAAByNwAAAAASAGAhG/ATAgAALdYV5v6StcgXRefUK3yzz9gDAAAAAAAA AAAAAOAPAAAgBAAAAJwYAOBkBgDhAQAAAP54AZVTPUgjURD+3myMZpMzIYpEsdi7wvMKRYOtoKCH hV5hLGyEjbDH3WkuSxRCrKwsTGNrpZYiCIKIxRWCNlZ22tpdY6GICIrmZt5uVpPDwgezzDf/73s7 ChHAuFL8QSfkNLBEqKqRYpe2KapUKlrrVSnfFg3iYmSFLuNJtneF4+gAx1aABOMj1g5YNuPAuQFE /ZgYJrKLP6ZKrgNcoomtD1xNMuSEWHq4tUzTSqDRsGgtrGUjMucjWVxfzpqeMERITP3MOQvWN6do TeZz2d+Y3rstZlmsobvi39T3Q6+m0r3S5r7qisvd3BIzgLR5RBuKx4NbcByvw/IzQz7v7UB6+mue S26VNm06/SCdlgpVvB0TPOvjshFueu0vGythwRnf76rPNX5XzWn/SFBv3Xidb9NXCB4r1DPFRhKm MqXcbH4euE+ZO303yTeZEsauFQWMrTRL3R6/7gs/9XUv2s2Bt+v+z89W1Ktb5eckIngwuF9bg4fr 34WD8J6Xr72PTTP6Xaygz7Epffp9XDZ2Ne9fAvyrUfzdPnbVWY3fVeUav03DIYmv5tv0RF6+yVbZ E1Pvidigt++REmjU6I/eLEX0KVNaWHRyOJQ/GJzxEas6QQLGb5P+prxwKv8w8A9G7bB2PAIAAEQA ZAAAAAAAAAAKAAAAAAAAAAAAAAAAAEgDFwHoA+gDAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAPAATwYgAAALIECvAIAAAABQQAAAAKAABjAAvwJAAAAH8AgACAAARBBQAAAD8BAAAGAL8B DAAfAP8BAAAIAD8DEAAQADMAIvESAAAAvwEAAGAAvwMAAIABPwQAAAcAAAAQ8AQAAAAEAACAMgAH 8IYBAAADBEEZm5PGCS1jhbQc1lJxh2r/AGIBAAABAAAAZzoAAAAAEgBgIRvwWgEAAEEZm5PGCS1j hbQc1lJxh2oKAgAAAAAAAAAAAABABQAAwAEAAKgiCABItQIAKAEAAAD+eAFjZOBkYGBmZWRgYGKQ YQABViDmZIKxmBiBUmAxRqb///+DWXqMElAxbrg6HqYDjA6sQkBxNTZ+BikGoNr/DAwCQP4BIOsY VGsb0DBuqBoeBt/EkoyQyoJUBgYDsN2/mRr+ASWBYALIUgYWJgaBkMzc1GIFv9RyhaD83MQ8hoiN n8sTgVjB4Wu5v1iaHwtQnS5QNQeQNuJKYMxhAWk1rITxTRhAfINKkN/QzGcCmR9cmZuUn8PAECDG teXCByEc5jMygOz5gGRPLjPIXBu4PfbgcFqMYQ9IFSn+gNjDCPdPDBPIhKRKLiAJClkucMiCxBig YSbAwA7m7QHHBSMTk1JwZXFJai6DHyh8GIA6FBm6wBpAxJtPQmC/6ALdC/MTM1g/AL1JX/tDAgAA RABkAAAAAAAAAAoAAAAAAAAAAAAAAAAADANoAegD6AMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAA8ABPBiAAAAsgQK8AgAAAAGBAAAAAoAAGMAC/AkAAAAfwCAAIAABEEGAAAAPwEAAAYA vwEMAB8A/wEAAAgAPwMQABAAMwAi8RIAAAC/AQAAYAC/AwAAgAE/BAAABwAAABDwBAAAAAUAAIAy AAfwjQEAAAMEi6k9ZljwH3jEYebHSLTrv/8AaQEAAAEAAACjPAAAAAASAGAhG/BhAQAAi6k9Zljw H3jEYebHSLTrvxoCAAAAAAAAAAAAAOAEAABAAgAAmI0HAPh8AwAvAQAAAP54AY2RP0vDYBDGn/dN /NO0YigIIg6pg+hQBfEDdHLTQSs4miGiQ2whhZCts1M/giC4uzt09lN0EgoOVXESjM+dSXDswSV3 995zv7wXgxrgrBjAYhNiC/SaLSNreKQ1Y/M812jPrBe1etXXsB07cZusby+uYgPszQGf+ZjRI/3e BV45rF70NHAcDq67WT8CLpX9bYc/PKSNBArXwu/exFESnERpcNqLw1tcPH2mIT3ofKXva1dTDkWb 3ct8H3hD03JFupuVec9Kvl/lh5B8J5O7/udJdX6egXBnFJWcc0cm9CvOi3KOCk7AXYiNdJvzc/hN VM2o+uNMzINOSDKPddm0p5sWNood+ljS7Fn/jbF26yxLBlGMqewLVLRwpwJ5vH00ldHm3PJOjup/ Aa77aX4SAgAARABkAAAAAAAAAAoAAAAAAAAAAAAAAAAAFwFoAegD6AMAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAA8ABPBiAAAAsgQK8AgAAAAHBAAAAAoAAGMAC/AkAAAAfwCAAIAABEEH AAAAPwEAAAYAvwEMAB8A/wEAAAgAPwMQABAAMwAi8RIAAAC/AQAAYAC/AwAAgAE/BAAABwAAABDw BAAAAAYAAIAyAAfwXAEAAAME9EJH4BoX0WzAWd6smoqX8P8AOAEAAAEAAADmPgAAAAASAGAhG/Aw AQAA9EJH4BoX0WzAWd6smoqX8JIBAAAAAAAAAAAAAMABAABAAgAASLUCAPh8AwD+AAAAAP54AWNk 4GRgYD7JwMDAxCADJBkYWIGYkwnGYmJkhIoxMv3//x8sq8coARXjhqvjYXJgOsAoBBRXY+NnkGIA qv3PwCAA5B8AspYBcQPQoGdAzA1Vw8Pgm1iSEVJZkMrAkAC2+zeTAlAPCEwA28rCxCAQkpmbWqzg l1quEJSfm5jHELHxc3kiECs4fC03E0v7ygJUrQs0lQNIG3E9YFQE6yyuBPnlN1PDPyAFBKSax8gA MvcD3NwGRjcGIIfBrZILSIJ8yAX2IUiMAep2AQZ2MG8POEwYmZiUgiuLS1JzGb6C3MkA1KHI0AXR AFTn81kIbAfI7RC7GBmYwfoBEPpMYicCAABEAGQAAAAAAAAACgAAAAAAAAAAAAAAAAAcAnwB6APo AwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADwAE8GIAAACyBArwCAAAAAgEAAAACgAA YwAL8CQAAAB/AIAAgAAEQQgAAAA/AQAABgC/AQwAHwD/AQAACAA/AxAAEAAzACLxEgAAAL8BAABg AL8DAACAAT8EAAAHAAAAEPAEAAAABwAAgDIAB/BxAQAAAwSNxaoFcqyrp9kFVBsSo0ij/wBNAQAA AQAAAPhAAAAAABIAYCEb8EUBAACNxaoFcqyrp9kFVBsSo0ijwgEAAAAAAAAAAAAAYAMAAGACAADA OgUAMK4DABMBAAAA/ngBjZC/SgNBEMa/nYvRXAIesVGxOC1ECy0srfIAUURTWObQC1pcFBI4thFr q7yK2FqkTso8gW8gBCsh5zebTUjpwuzOn2/47YxBBQi+AAj2eANrtIosPDHG54wUReGqp2bb56pL XU3a0g7qzB+WN7ELagsgYjyk90GLA2CL+qrX1HCR9B9a9jkFrhz7V15nLPIMFIqSIGo9Zmkvvkzz +PopS7q4fZ/mCS1u/OTlnc6kRN0J1Rt8z8KhGYi2HttF3HD/P/LxGHfQ+ovVWVd5mv0/z0C53yvc keN0l9xzx7m3IXW6kdBtRCnws0ZYd9Gn26ERObixvX6aYaJzgR37eJs3UNec1h1TZ52zDbhQnj8y qVG6JgIAAEQAZAAAAAAAAAAKAAAAAAAAAAAAAAAAAPMBQAHoA+gDAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAPAATwYgAAALIECvAIAAAACQQAAAAKAABjAAvwJAAAAH8AgACAAARBCQAA AD8BAAAGAL8BDAAfAP8BAAAIAD8DEAAQADMAIvESAAAAvwEAAGAAvwMAAIABPwQAAAcAAAAQ8AQA AAAIAACAMgAH8HABAAADBOmY4mUUUqGWwmFLUF4mquP/AEwBAAABAAAAH0MAAAAAEgBgIRvwRAEA AOmY4mUUUqGWwmFLUF4mquPCAQAAAAAAAAAAAAAgAwAAAAIAAOjWBAAgGQMAEgEAAAD+eAFjZOBk YGB+yMDAwMQgAyQZGFiBmJMJxmJiZISKMTL9//8fLKvHKAEV44ar42FiYFJgFgKKq7HxM0gxANX+ Z2AQAPIPAFnHgPgBUO0xoGHcUDU8DL6JJRkhlQWpDAwBYLt/MzX8A0oCwQSQpQwsTAwCIZm5qcUK fqnlCkH5uYl5DBEbP5cnArGCw9dyNsk0ERagOl2gag4gbcSVwJjBBNKqWQnjS4DdrwHln2KwYgDJ 11WC/IpsH0iUePsYGUD2fkCydxHYnjy4va5ge4oruYDqQCHCBQ4RkC0MUL8KMLCDeXvAYcjIxKQU XFlckprLIALyFwNQhyJDF0QDUJ3PZyGwnSC/QuxmZGAG6wcAYKVNmqQCAABEAGQAAAAAAAAACgAA AAAAAAAAAAAAAAB0CXwB6APoAwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADwAE8GIA AACyBArwCAAAAAoEAAAACgAAYwAL8CQAAAB/AIAAgAAEQQoAAAA/AQAABgC/AQwAHwD/AQAACAA/ AxAAEAAzACLxEgAAAL8BAABgAL8DAACAAT8EAAAHAAAAEPAEAAAACQAAgDIAB/DuAQAAAwSRWKb1 twrfTLZEygCfW1Hf/wDKAQAAAgAAAEVFAAAAABIAYCEb8MIBAACRWKb1twrfTLZEygCfW1HfCgMA AAAAAAAAAAAAIA8AAGACAADgcRcAMK4DAJABAAAA/ngBY2TgZGBgbmVkYGBikGEAAVYg5mSCsZgY gVJgMUam////g1l6jBJQMW64Oh6mBCYFfiGguBobP4MUA1DtfwYGASD/AJC1BYgf8DEwiADVc0PV 8DD4JpZkhFQWpDIwBIDt/s3U8A8oCQQTQJYysDAxCIRk5qYWK/illisE5ecm5jFEbPxcngjECg5f y/dIpvWxANXpAlVzAGkjrgOMWXwgrZqVMP4GLhBfA8rXYrzECeLXweUrwHwFOP86G6r+DlZk/acY VrGg6k9nQlUvAQ4vhH1WDBD1oLBF9h9IlHj/MTKA/PkBqAnmr/m8IBMK4O625Abx8+D8u2B/FcP5 RzhA8klwviDYnwj1TWB/pcDlF4L9gZB3BfujuBLdH0BDmUD+CK7MTcrPYWBYzcX12+CjEM54gviD Ee4Pbx6Qu3Th9i5jB/G14XwvZhDftpILSIJSFhc4ZYHEGKBpRoCBHczbA06LjExMSsGVxSWpuQx9 oPTBANShyNAF1gAifD4LgcNSF+g/WJgyg/UDAINniZ4NAwAARABkAAAAAAAAAAoAAAAAAAAAAAAA AAAAYwrQAugD6AMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA8ABPBiAAAAsgQK8AgA AAALBAAAAAoAAGMAC/AkAAAAfwCAAIAABEELAAAAPwEAAAYAvwEMAB8A/wEAAAgAPwMQABAAMwAi 8RIAAAC/AQAAYAC/AwAAgAE/BAAABwAAABDwBAAAAAoAAIAyAAfwVwIAAAMEtvebCVHOLyF2Wv8P lXtI7f8AMwIAAAEAAADpRwAAAAASAGAhG/ArAgAAtvebCVHOLyF2Wv8PlXtI7RwEAAAAAAAAAAAA AKAQAACABAAAuMQZAPD5BgD5AQAAAP54AZVTv0sbcRR/3/eN0VxSPRIFqRnODmKHFM1YaMkgItJ2 aDI4GiHS0p4XSCCkgzg4OXXs0OL9CenmICLOAR26ZulfEAldKvV87903Zy7FwS98j/d57/N+31dB CkBPI30gD3wm6KYIhxIqZSSFQRCI9YWaN7p0xMvgQcK3s6RfSs7AUyBuAGATPifpjO42gdcJgLTh ZOBttfmh0q7XAPowRdq/FI1IcogIBUrN1cxiCb8nWcqRdJIa6jbTQ13Z5tpv0KGcfL5K1QkEu/LR rTWcd7WW895zq3uw9XPQqtJ1Sn9adn43CPMoyV+0HP1jhvutt2kqULRAHytm1L1GM8xwcEuQzmMz IHCcPtXFnRatri5Nc6Yv3hA7TxjvGNzT+1Oj9p6+SDJeM/aO+hSzd9SV2MvG7mNfj/r7uA6MN7zx SZESeVLltrvjfaZpz1m5levsg5MK+8BoYv+kj4KJez+f8bjHeevw4bj/z2c+w/UWon4WLMavIuxO hHh8L0SCx2w+3k9X/5I9OCZPV79Mc55VgztqT+xD3NNLKbY/N/ae/j3JeNngAd7KXvYj/1yM31GX Mf45XMf4Pn5LjMb3sSJ7XfYsysJvzZK3xhyQF3yDNkwKOpXXqRCflduNZs2FgP94II9FOBIHJrwZ ZOXfLKj7HWjxvwNP+bkSQQIAAEQAZAAAAAAAAAAKAAAAAAAAAAAAAAAAAAwDaAHoA+gDAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAPAATwYgAAALIECvAIAAAADAQAAAAKAABjAAvwJAAA AH8AgACAAARBDAAAAD8BAAAGAL8BDAAfAP8BAAAIAD8DEAAQADMAIvESAAAAvwEAAGAAvwMAAIAB PwQAAAcAAAAQ8AQAAAALAACAMgAH8IsBAAADBN2C6KQIg6FDNqBQsd8gQqT/AGcBAAABAAAA9koA AAAAEgBgIRvwXwEAAN2C6KQIg6FDNqBQsd8gQqQaAgAAAAAAAAAAAADgBAAAQAIAAJiNBwD4fAMA LQEAAAD+eAGNkT9Lw2AQxp/3TfzTtGLopOKQOogOVRA/QKdO1UErOJohokO0kELI1tmpH8HJxcnd wdlPUTp0LhQnwfjcmQTHHlxyd+8998t7MagBzoYBLHYhtkKv2TKyhkdaMzbPc42OzFZRq1d9Ddux E7fJ+v7qJnbA3hzwmX8weqE/u8CMw+pFTwNn4fCunw0i4EbZ33b0w0PaWKBwLfz+fRwlwXmUBheP cfiA67dFGtKDzlc6276dcija7F7n+8QbmZYr0sOszFMr+XGVn0Lyg0zu+p8n1eV5BsKdU1RyrhyZ MKg4n8rpFpyAuxAb6zaX5/CbqJpT9ceZmFedkGQe67JpTzctbBQ79LGm2bv+G2Pt3mWWDKMYU9kX qGjhSQXy6C2aymhzbnknR/W/gO9osMECAABEAGQAAAAAAAAACgAAAAAAAAAAAAAAAABsB2wC6APo AwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADwAE8GIAAACyBArwCAAAAA0EAAAACgAA YwAL8CQAAAB/AIAAgAAEQQ0AAAA/AQAABgC/AQwAHwD/AQAACAA/AxAAEAAzACLxEgAAAL8BAABg AL8DAACAAT8EAAAHAAAAEPAEAAAADAAAgDIAB/ALAgAAAwSxoY2PhkoMqXaNuWV4Zbwj/wDnAQAA AQAAADdNAAAAABIAYCEb8N8BAACxoY2PhkoMqXaNuWV4ZbwjWAMAAAAAAAAAAAAA4AsAAOADAABY aBIACAEGAK0BAAAA/ngBlVI9SwNBEJ2dy+claIgWGkVPC1EkomlVsBCxMBYmhY2QCCcKnjmSQDgr 61Q2KliIP8DGxsoiP8LKQgKSziISbBLwnNlcLhq1yMLuzZudnXlvbgQEAZQ7QQeMAy8v7SB2LBR0 JX0CbduW1oIYcXwhNy6MVaUaipJ/xjcIMaBYGyBCuELWA+3bEMClAkAfGROGZLZ4mLZMnWIgQN4m ZeMXvDy041Sa2Qwj4J6PrSGyrkLMs4Ua5ed1Lhl6ECLpI0MvaNt6SdvJGdkT2L1vlLK0tbWP0lPs oNbOKWSthPqCryprMy3qACTUCt4IogdmXtfbFc4+CdLqtwJK9nXixaoSalkpB7nSab6D9QDjrIMz eKF8v8/gBjDezPcqJSey0pRl7OeOAZ7H1Mbie/Rfpay4LtBVvC4Vx528XX29ed8G1NH/8/7Wt+xn vvOunkkv49W/6kA/f+on/7IyIfuouX1M+rjOkoNNcS372sEZXPHw/ZzLq4mMZ/MqnTynqpxT9oGc /hZGwC/Ro5xsgTidsgpF3YAaTxDQiykoywccsNWIOpPa7QnPEMAXcnKh4pkCAABEAGQAAAAAAAAA CgAAAAAAAAAAAAAAAABsB5AB6APoAwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADwAE 8GIAAACyBArwCAAAAA4EAAAACgAAYwAL8CQAAAB/AIAAgAAEQQ4AAAA/AQAABwC/AQwAHwD/AQAA CAA/AxAAEAAzACLxEgAAAL8BAABgAL8DAACAAT8EAAAHAAAAEPAEAAAADQAAgDIAB/DjAQAAAwQQ zpgp4VnyfxNFWJJEfHe3/wC/AQAAAQAAAPhPAAAAABIAYCEb8LcBAAAQzpgp4VnyfxNFWJJEfHe3 HAMAAAAAAAAAAAAA4AsAAIACAABYaBIAaN8DAIUBAAAA/ngBlVI9SwNBEH27+b4LGFKISIozhSgS URsrwUosVNSksD3DSQTPiyQSzipgZyFiJYJwf8HKRtC/4D+wt1GiVcBzZpNsiJoiC/vxZmfmzdsd gRQQuRSARA48YjRTsneSgq6UTcgwDNVpXkx0bab2S8umfDWzZJ+Oj2ES5BsCGcLPdHqgGZjAMvnT pnzS2LTrlZJfdYCq4m5Li2J4XCnWqESmdOg6NWvLaVi7nmsfY+++1bBpWqtfjZ3cQTlK3gUqMUn7 kvGJvMn1Vn1SRRjyTrBH1avVWV1bNr9pozEqgwDneddMgXhKMdOZ12EOxG2Ssa3xW2Twfg2M171O HX2lZJSstOi7+94RUBk3XhY+skOVduoQWrFvcN5CN29f3++8NzkjPTzvX30nCc47p/Wcxhiv/MeD UX5qsP5AFNU7WprnPM48ixpfRxnParyt3nXGM8jKvWaoXmMfdLsog4RCj6o7hZT5ol+rOy7K3DGg iClcqABeNlpZ9bcF6rneH0dU/A9AZ5aB4gIAAEQAZAAAAAAAAAAKAAAAAAAAAAAAAAAAAFgHlALo A+gDAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAPAATwYgAAALIECvAIAAAADwQAAAAK AABjAAvwJAAAAH8AgACAAARBDwAAAD8BAAAGAL8BDAAfAP8BAAAIAD8DEAAQADMAIvESAAAAvwEA AGAAvwMAAIABPwQAAAcAAAAQ8AQAAAAOAACAMgAH8CwCAAADBPFjC+5+z5ve3flYHOnlGpb/AAgC AAABAAAAkVIAAAAAEgBgIRvwAAIAAPFjC+5+z5ve3flYHOnlGpa4AwAAAAAAAAAAAADACwAAIAQA ACA3EgDgZAYAzgEAAAD+eAGVUz0sBEEUfvN2787tSlzOT+SiWFcIkSUu0ZEgIRoKpxDdXrJC4rgc yWVVapVWZTvK1St0misUQqNTaBQXoiGs9+bmljsUJpnZ9733Zr43384TkATQ7gQt0Ac8YjST2LBQ UEj6BIZhKK0R0at8ZpTXjpZ+YabJPxDvgAxQbgiQInxB1gnNfRPgUgOgj8xphwVnd33ZK7mUA23k faXTeAcPnaZN1FxNF07hbJytTrLWTK7zDS06n8ehrFBHSC1vFN0da9GtWEvbRWcLVs6eKw5Na+ql cp1Zu6+fKSRXzrC0wOC7lTzmzhm6OFWYFCEM2rGgcqFUdt064/4HQRr/ZUR5mxrVWWeqavkkM++V G9huY+woHIiZpnggsjJeUHEfj7Tv+32cA8bz5VZlyImsTN4rFrY3Sclu4/aqlv5TGVaoJjBSKCUV sRWvLkJZl614vvRo5VntM0b/5vmpR2eC6x+O7p+RuMHrY0+M45O/8cJ//nzz/arag9TVUrxVbTzO PGNRHTdN8UBMNMV9nNY5f0jl+/iOjAfLBq3cB4bsA/aB7K43TEFConPZOQIxm/d2dt0i3PMLBdrR DwdyAyc8PqVVJ3xpxm8S4BN6v63f4AIAAEQAZAAAAAAAAAAKAAAAAAAAAAAAAAAAAKgHlALoA+gD AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAPAATwYgAAALIECvAIAAAAEAQAAAAKAABj AAvwJAAAAH8AgACAAARBEAAAAD8BAAAGAL8BDAAfAP8BAAAIAD8DEAAQADMAIvESAAAAvwEAAGAA vwMAAIABPwQAAAcAAAAQ8AQAAAAPAACAMgAH8CoCAAADBCnyZTWXpMePYPuNPyvRw7X/AAYCAAAB AAAAc1UAAAAAEgBgIRvw/gEAACnyZTWXpMePYPuNPyvRw7W4AwAAAAAAAAAAAABADAAAIAQAAGj9 EgDgZAYAzAEAAAD+eAGVUztIA0EQnZ07E3MJGKKFBpHTQhSJxLQiBAuxUBG1sAsRIgqeF0gknJW1 lZ/KQohY2J5YKRbphNQ2giDWNopoYcA4s9l8JUUG9pg3n30zszcCfADas6APDAJLFx0fVjUU5JI2 geVyWWqTol/Z/LW4AJp6PBAi+6inB8JAsWWAIOECaVeMAgAPGoBfxQRgMZndWnPSKYqBbrL+0G2c waLTiRA1V9OHcbzxsNZL2ref6yyhSfezHMkKdYTg2raVyphLqZy5YlvJXVi//swl6Zjxr1x0YDNR uVNIrpgxpMX83FvaYe6YoYtZhWkihEE7F5yRtjPZCuPBL0GSThlRdvNOdVaYitqTj5n37Sq+7Wa8 obArXpr8rriT/qTy5/FNa8zP4xwwnrdbJ0NG5MmsOtaGvQPwahiF6Eeo7WS433eBtQmdGnxvRPHq 4kzh1nm08jyGjen2PP/nceFt5HHFscQTtX73utg/o/qrvwMZoZOXb+6vqFlyzqbiKWo3HuaZUtgV mSa/Ky6b/Hk80Tl+vFbnsnyXMdsgK++BIfeAY0BuVwmD4JXoXm6OQBxZdTLZlAUJ/kOBMobhUCZw wMJnSG1CfWa0QiR/BwGod5UCAABEAGQAAAAAAAAACgAAAAAAAAAAAAAAAAAwB5AB6APoAwAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADwAE8GIAAACyBArwCAAAABEEAAAACgAAYwAL8CQA AAB/AIAAgAAEQREAAAA/AQAABgC/AQwAHwD/AQAACAA/AxAAEAAzACLxEgAAAL8BAABgAL8DAACA AT8EAAAHAAAAEPAEAAAAEAAAgDIAB/DfAQAAAwTtxoixJQAFEibFIg8Q0E+7/wC7AQAAAQAAAFNY AAAAABIAYCEb8LMBAADtxoixJQAFEibFIg8Q0E+7HAMAAAAAAAAAAAAAgAsAAIACAABI0xEAaN8D AIEBAAAA/ngBlVK7TgJBFD0zPHchkVAYJRarhZFijdhYEamMhVgIhY0FxDWauEAWErJW1BbGT6Dh A6xsNLH1Q+glaEXieu8AQ1ApmGQe586dOffMHAEDCD0IQGID3CLUDTldSUFbKiZkEARqtSvWJrGE zkvKjuwk0hTfjq4gA8oNgBThN1o9Uy8kgAPKp0nlJFGstK7LfsMBGop7JC06w+1RsYYlUuUb12la p07bOqu7lRrOn4btCnWr8NV+Xb3qhynbphLjNO+bn3g3ud6GT6oIQ/ZEiLHnOKxuJDvfNFFblkGA mT40U1dcGMx0542ZuyIfZ1zVuBea3z8C42NvXMdMKQUlKy35brV+C1yum7W9QXqh0nEdQiu2lGJ7 cu9M3+97Uxkzuvjev/qyMa7X1npyEcb5/3iwzE/N198VA/VuluY5jDJPTuNimHFWY0O9645nUpS9 ZiqvcQ4mLkohptCLcqeQcqvkN1uOiz47BnRiE/fqAA8nw7T6W5s8N/1j9gzwA61cmVnAAgAARABk AAAAAAAAAAoAAAAAAAAAAAAAAAAAqAdsAugD6AMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAA8ABPBiAAAAsgQK8AgAAAASBAAAAAoAAGMAC/AkAAAAfwCAAIAABEESAAAAPwEAAAYAvwEM AB8A/wEAAAgAPwMQABAAMwAi8RIAAAC/AQAAYAC/AwAAgAE/BAAABwAAABDwBAAAABEAAIAyAAfw CgIAAAME0FjfYXWQ3JbVE3xXwcrnU/8A5gEAAAEAAADoWgAAAAASAGAhG/DeAQAA0FjfYXWQ3JbV E3xXwcrnU1gDAAAAAAAAAAAAAEAMAADgAwAAaP0SAAgBBgCsAQAAAP54AZVSv0vDUBC+d+nPJGCp DioO0UF0qGhHQegkDupgO3RMC5EKtim0UOLkKJ2EDjoq6CC4CCIODp38Cxwc3VwrrQ5WWu9e01ar HXrwwn337n133+UEBAGUa0EfmAE2L50gdj0UdCVjAtvttvSWxKQb03p5Or4oMT1M8XnfGEwD5bYB QoSr5N0z0gFOFQDNzdFhK1XMJJy8RTkQoOgnsfELNg+dCJXmbiYQ8NHH3jh5Hxr32USD+NmOZYce hFBiL2sVjG2rZOzY2VQOkjf1UoqOEXsvPU/tNjqcQtaKqq+4qrG2vEMTgKhaxUvBGXm7UOxUOGwR JBu1Asrua9QXq4qqFeUpyJUO7C6+DTBOu9jEL+XnvYnrwHjDHlRKQWSlcSebtvcByrp6tfwWHqqU 9dQE9hSfqMwbcXn7+gZ5z3U1OZz3r74zf4e3o8/EIy/jtf/qwCh/6nf/FSUj52j05njn4zorLs6J lpxrF5t44eH7xd6cTTnnBVulKO+pKveUc0BufxND4JfoQW62QJyLO4WilYUGbxDQi1koywecsFkP u5vanwmtONk3C8Kj0V8CAABEAGQAAAAAAAAACgAAAAAAAAAAAAAAAADDBHwB6APoAwAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADwAE8GIAAACyBArwCAAAABMEAAAACgAAYwAL8CQAAAB/ AIAAgAAEQRMAAAA/AQAABgC/AQwAHwD/AQAACAA/AxAAEAAzACLxEgAAAL8BAABgAL8DAACAAT8E AAAHAAAAEPAEAAAAEgAAgDIAB/CpAQAAAwSKE99ePvYPJF6uCkdebPgK/wCFAQAAAQAAAKhdAAAA ABIAYCEb8H0BAACKE99ePvYPJF6uCkdebPgKWgIAAAAAAAAAAAAAoAcAAGACAADY0AsAMK4DAEsB AAAA/ngBjZG/S8NQEMe/7yVtbVowFAeRDNFBdKiIoyB0cVMrtoNjo40omFZoIGQR506dnP0DXARH h879Q5zV4iQY784kKiJ4cC/3M/d57xTKgFFXgIYDlgJpWWeWVpSSmNJJkoi1pubTWCWvq+qOvinV KL5cnMUCqDYBbPLHZN2TdkrAHNVX0poqdr3wtB1f+MC+zH7TV++UJBnxUJgadvss8Afunh+5B/3A 6+Hwbhp5pG7jNXpxTpom1TH9DH03rLF6LHLrapz51wX2V3J/pH/mG3K/LD/BMTh/GfNbfOfh6P95 FJjriZoyDle4ejnHrcl/7Ob+RDi+8pvC0f3FQU2aOVpxcNQ/pxmOtb3+XPvzXT45VM4RGjx3K7bo 5E1ZsimOId2BDVoUyYPsVmm91IoHoR+gye8N6ljEUBr42JnW5K514s/ubEj/B22kb+cTAwAARABk AAAAAAAAAAoAAAAAAAAAAAAAAAAAQBVAAegD6AMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAA8ABPBiAAAAsgQK8AgAAAAUBAAAAAoAAGMAC/AkAAAAfwCAAIAABEEUAAAAPwEAAAYAvwEM AB8A/wEAAAgAPwMQABAAMwAi8RIAAAC/AQAAYAC/AwAAgAE/BAAABwAAABDwBAAAABMAAIAyAAfw XQIAAAMEKMVb3SFrclxFEnAD2lKRuf8AOQIAAAEAAAAHYAAAAAASAGAhG/AxAgAAKMVb3SFrclxF EnAD2lKRuXoEAAAAAAAAAAAAAAAiAAAAAgAA+LQ0ACAZAwD/AQAAAP54AY1Tz0sbQRR+M4k/sokm praa+GvjQSw1Ij0WhILxIP5ANAcRhCSYYKnRQISwHsTiSb3UQ+nVk0dByEFExJMH/xNvLSqIgvG9 l81MRhBcmN355vu+N7vfzhPgA/CMSgAJ3UBXAw4f4upMCuHOhKxUKswOi053za90AYkF+sO4PtAY hCigtgIQQnyJsysclzGAKyzmdzUBmE5vrCSdQhZglvd+ktvPSOL1mzYFr4RQ8kc+W7RnsiV7bj2f XoOFk9tSGof9/b50GsmNeVEXR3UzPr9aKTEeI+tnp4b3ewgPuvgajrsJbyk+ydhWuBwx/Y8f6/0p 8dBOWOuXwqa+tc3UT7aa/IFVz19D2UdYv88iY13/rNH0/2yo96dETpp8B/8v/b3foFqf/m19vrT6 /nwFUM7/0FTLddOmCgWVW3sv4TWFTznXosKHXcRnFL7pNP07n0z/Aueu/QnOXfv/cs56v9UQ+bV+ u8WsH/Wb9S84Z60/aiZe17/jnHX9X17il9X7H3LOmk9wzkXndc5okpTzvJPPrK8CHESswsj/8Jvn uJqzUDkP9NG+cbXvTpTwF4X/dJj80AcTDwcJjyr9UMDkd5sI63oTnqrewlXqZIs7mdbA7dEQNDE6 594XUvbPO8WNbB7GqB8BHTHYYwPdpm7DfHbimFftDHnY/wJkhq6cJgIAAEQAZAAAAAAAAAAKAAAA AAAAAAAAAAAAAJQCQAHoA+gDAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAPAATwYgAA ALIECvAIAAAAFQQAAAAKAABjAAvwJAAAAH8AgACAAARBFQAAAD8BAAAGAL8BDAAfAP8BAAAIAD8D EAAQADMAIvESAAAAvwEAAGAAvwMAAIABPwQAAAcAAAAQ8AQAAAAUAACAMgAH8HABAAADBOvkv91l kh6iPL4XTPZhQOn/AEwBAAABAAAAGmMAAAAAEgBgIRvwRAEAAOvkv91lkh6iPL4XTPZhQOnCAQAA AAAAAAAAAAAgBAAAAAIAAOBkBgAgGQMAEgEAAAD+eAFjZOBkYGB+yMDAwMQgAyQZGFiBmJMJxmJi ZISKMTL9//8fLKvHKAEV44ar42FiYFJgEQKKq7HxM0gxANX+Z2AQAPIPAFnHgPgBMwPDMaBh3FA1 PAy+iSUZIZUFqQwMAWC7fzM1/ANKAsEEkKUMLEwMAiGZuanFCn6p5QpB+bmJeQwRGz+XJwKxgsPX chXZtGYWoDpdoGoOIG3ElcC4hwnkJM1aGD8H7H4NOF+MAWR0dS3Ir8j2gUSJt4+RAWTvByR7v4Ht yYPbcwpsT3EtF1AdKES4wCECsoUB6lcBBnYwbw84DBmZmJSCK4tLUnMZmkH+YgDqUGTogmgAqvP5 LAS2E+RXiN2MDMAABQIAB7xQxC4CAABEAGQAAAAAAAAACgAAAAAAAAAAAAAAAACoAkAB6APoAwAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADwAE8GIAAACyBArwCAAAABYEAAAACgAAYwAL 8CQAAAB/AIAAgAAEQRYAAAA/AQAABgC/AQwAHwD/AQAACAA/AxAAEAAzACLxEgAAAL8BAABgAL8D AACAAT8EAAAHAAAAEPAEAAAAFQAAgDIAB/B4AQAAAwRJagNBDtX+7b5eUZO3iiKq/wBUAQAAAQAA AEBlAAAAABIAYCEb8EwBAABJagNBDtX+7b5eUZO3iiKq0gEAAAAAAAAAAAAAQAQAAAACAAAYlgYA IBkDABoBAAAA/ngBY2TgZGBgfsnAwMDEIAMkGRhYgZiTCcZiYmSEijEy/f//HyyrxygBFeOGq+Nh YmByYBECiqux8TNIMQDV/mdgEADyDwBZx0A8FgaGY0DDuKFqeBh8E0syQioLUhkYAsB2/2Zq+AeU BIIJIEsZWJgYBEIyc1OLFfxSyxWC8nMT8xgiNn4uTwRiBYev5Rqyae5AQxl0gao5gLQRVwLjJyaQ kzRrYfxFYPdrQPmnGI4wgIyug8uLgfnVtSC/I9sPUkW8/YwgzzF8QHKHDhPIhDy4PefB9hTXcgFF QSHEBQ4hkBoGqN8FGNjBvD3gMGVkYlIKriwuSc1lcAf5kwGoQ5GhC6IBqM7nsxDYTpDfIXYzMjCD 9QMAyaxRn2ACAABEAGQAAAAAAAAACgAAAAAAAAAAAAAAAACIBEAB6APoAwAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAADwAE8GIAAACyBArwCAAAABcEAAAACgAAYwAL8CQAAAB/AIAAgAAE QRcAAAA/AQAABgC/AQwAHwD/AQAACAA/AxAAEAAzACLxEgAAAL8BAABgAL8DAACAAT8EAAAHAAAA EPAEAAAAFgAAgDIAB/CqAQAAAwQG3IEX3JU7RZ4wfL0Up8Uk/wCGAQAAAQAAAG5nAAAAABIAYCEb 8H4BAAAG3IEX3JU7RZ4wfL0Up8UkWgIAAAAAAAAAAAAAQAcAAAACAAAwPQsAIBkDAEwBAAAA/ngB Y2TgZGBg1mVkYGBikGEAAVYg5mSCsZgYgVJgMUam////g1l6jBJQMW64Oh4mBiYHdiGguBobP4MU A1DtfwYGASD/AJB1DMRjZ2A4BjSMG6qGh8E3sSQjpLIglYEhAGz3b6aGf0BJIJgAspSBhYlBICQz N7VYwS+1XCEoPzcxjyFi4+fyRCBWcPhariOb9pgFqA7keg4gbcSVwFjNBtKqWQnja7OC+BpwfjoT qrw42H8w+VMMVgwg+bpKUFgguwckSrx7GBlA7voA1ARzx1awO/Lg7uhkAZlYDOcvBLsDIe8Cdkcx hjuAmphA7giuzE3Kz2Fg4JLl8jP4KIQzXCDuYIS7w5sZZK9tJReQBMUUFzimQGIM0DgQAEUUEOwB xy0jE5NScGVxSWouw2NQeANtZFBk6AJrABE+n4XAftUFuh/mZ2awfgBaLWoUnQMAAEQAZAAAAAAA AAAKAAAAAAAAAAAAAAAAAGMKeAXoA+gDAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP AATwYgAAALIECvAIAAAAGAQAAAAKAABjAAvwJAAAAH8AgACAAARBGAAAAD8BAAAGAL8BDAAfAP8B AAAIAD8DEAAQADMAIvESAAAAvwEAAGAAvwMAAIABPwQAAAcAAAAQ8AQAAAAXAACAMgAH8OcCAAAD BGjqsCumuv2cqSmnGttdIYj/AMMCAAABAAAAzmkAAAAAEgBgIRvwuwIAAGjqsCumuv2cqSmnGttd IYjmBQAAAAAAAAAAAACgEAAAwAgAALjEGQAIkA0AiQIAAAD+eAGlVM9rE1EQnp3dbDY/tEvaQyse Vg+lPSSYIN4K9aSHtAdbTD3VGGMVTFo2lbAHMaB4EA/+AQp6FkQvelAwFw/iQW9twSABBQs9KEEQ i8Z5s+/tJo05FB9smO998+abmfcyGsQA9A4CGHAYxIrQFyPsW6hp0tKw2+0ym9HG5V4i8Eti03po p2h/0hyBQ0C+XQCbcJOs1/SdJ3DDAkhInyTMFdcvL3prZQAHiYBfFI2ceBn0myZpkc0Y/jSfoLBG 8aP5KfB1aEcs31eTvp/Jw/ddtib1/jjLVkb3uYvGXbbGUFmjZGVsX88xbssYjjFvi77sYvM3i8E9 0RA0EOwFr3Jh9SpAaSJeOPY9BUvPOvUifc7sj/rm+KVtXxtB1JaLn4g+0sXRL57CT61e/A43DB/7 eg71T6xBva2J+Mvhejr345umdPPWKY47I3Xz1tmY0FG4rV+P+NjXbfxh2X/oUp1Lw3VR6uqy3pbZ OCDipqWuiyuJXtwwTvPbmvH21iu8RH8Xr1TKNWe+XHfOrFaK1SH93VtvIZLlevNBvW+571WJC5FZ rj/kN/keFL8DK9wPxbf1LseryjzD/uwvz8H+fDgoIqzJvFqmmxRY5dEyb8UFrgV8KdrPX+M8Fe/i Cz6v4rl4js+reC6mue7Qf6MvnouPzV69hnEcBC4N3A9t/sf95K3n3M9scD9vOC+Fs+ZJfic5yWfN HOelcFvf4bqzA/exv7wG7+P+iKh3Ouj3V85jKsDT/E5CfovzUryLN/m9K97F91xXyD/oO+/iHPd/ youTqpiacZ6aIgfgWbyLNkQZveI5qyEeXfBq6+UKbIv5AnTiCNzhAzr55TspOQvD/4TYB/gLJuIK x1ADAABEAGQAAAAAAAAACgAAAAAAAAAAAAAAAAAQCSgF6APoAwAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAADwAE8GIAAACyBArwCAAAABkEAAAACgAAYwAL8CQAAAB/AIAAgAAEQRkAAAA/ AQAABgC/AQwAHwD/AQAACAA/AxAAEAAzACLxEgAAAL8BAABgAL8DAACAAT8EAAAHAAAAEPAEAAAA GAAAgDIAB/CaAgAAAwRGb+fqTK0iXygl1VL4GJm4/wB2AgAAAQAAAGttAAAAABIAYCEb8G4CAABG b+fqTK0iXygl1VL4GJm4PgUAAAAAAAAAAAAAgA4AAEAIAABgehYAwMkMADwCAAAA/ngBpVS/ixNR EJ43u9lNNjGGmOI8TtgTkbM40WCjcJA/4LTwUtgmEjnBJAeJhg0Ii0UKK9FCEMF/QAsrBS1OEGvR TguvsUghnJxR8PDWmdm32TObFIcP3ubNmx/fN99uRkEGwHiC9IAF4JWinSE7PKFS+qQwCALxnlZz +i47jsthJe3ni3R/0joM80CxAUCB7E06vaBdyQP8sgGyOiYHF2vd9aq30QDYUmm6/U3VOIOXSXuZ oJlNCV1z2+TTETqdyDPPXdz8Qz+07jFBNBEKa16z3r4BUJ93bp75XoQrz3d6NdpuZdR7X7pWCmsq YKyy88X6luLUr15of0Rb22F9l/jzSta/sOAszq6Pwn2bdAvrjuwHFuOsaJyh8UxwVrwQx98TmCk4 1Met2ThK46DGGViPHcZZ1jh9vJPdb/vmG2M/D988L+/2leYR98tRrGf1erPRcS81eu7ldrPWmqHn ZL+Qui39rmoeI3tO7Ja2f8BD6T/yD42y2K2EHgfjkdRjS/rf0LgD61GGK0Y8BtZLm+3OWK93Obaj +D76omcU38eq5Mfxe5If+z9In5HfN88B17ua0Jcu/0PfkR0IzlnN+1OqnWacsraHxlHtn/J9HQA3 qedT0efUWM+C6LE01u/+IeYR+fv4c8L/VnjG/q7otzTtvf/zf/6cdY7N/h9Mfn++uY7Mo+459ORp 5Mg04juQGbeLBaBBROu1zC+FeHzN63QbTSjxnADKWIS7ksCP1Z2inkexJobk/wUA//hY3gIAAEQA ZAAAAAAAAAAKAAAAAAAAAAAAAAAAAIcJYAToA+gDAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAPAATwYgAAALIECvAIAAAAGgQAAAAKAABjAAvwJAAAAH8AgACAAARBGgAAAD8BAAAGAL8B DAAfAP8BAAAIAD8DEAAQADMAIvESAAAAvwEAAGAAvwMAAIABPwQAAAcAAAAQ8AQAAAAZAACAMgAH 8CgCAAADBJySlkM8sO8OyJZa9o22e7X/AAQCAAABAAAAu3AAAAAAEgBgIRvw/AEAAJySlkM8sO8O yJZa9o22e7W2AwAAAAAAAAAAAABADwAAAAcAABijFwBY2QoAygEAAAD+eAGVUztLA0EQnt29XMwl mhAsVFKcKewMam0kChaCD9AUNhannCh4SXxgvE5SiFiJv+BAsLe38H/Y3A/wiVgpxpndvQsnWLhw 3H4z3+zMN8wwyACIBwbAoQR0UvhleHTjDF3Sxni325W3ChvStmzMy3FI1/JFtI+ZeRgB5HYBCojv 8XZDKA9wYwJkNScHS87RTt1vuQCXjHJ/8tNv/OG5pKTc4FBY873N5h5WULKGJ96KsH773nbws2sf 7Y3SdsdA4jiy+/A/ZV2lGpxCX3yFz02FnzXmIukvGwo/af8dU/hRY9sUOXpvWuNQ7Pcn8UqW8LX2 B2xO8mdi/rzsX9X/rY+iSF9913MP7WW3ba82Pafxhz4GpPMVg5Qu2xyUddTjOg1Zx1ac90DW0cNf GcoY8UNxnCLc85cF4c1YRzXxXsC8hD8UFVD837rwkX/oQi7yX7FHka6zNI3U7naEFwzKM6FxKFoD hCdjnOQHbFb2JfIHLPKrOm2cSTpyvv5RZ6//uC04ZzXjQQi8ec6JqyzAx4zIYqGHtsCSW0D1gtyt T16AtER3cm8Y5+U1//DI9aBDswwYMQoXKgB5i+9F2R+a76hPlAPgB5zdo4odAgAARABkAAAAAAAA AAoAAAAAAAAAAAAAAAAAuAFoAegD6AMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA8A BPBiAAAAsgQK8AgAAAAbBAAAAAoAAGMAC/AkAAAAfwCAAIAABEEbAAAAPwEAAAYAvwEMAB8A/wEA AAgAPwMQABAAMwAi8RIAAAC/AQAAYAC/AwAAgAE/BAAABwAAABDwBAAAABoAAIAyAAfwZwEAAAME 24UOQ7u3GMjBFyZzfio2nv8AQwEAAAEAAACZcwAAAAASAGAhG/A7AQAA24UOQ7u3GMjBFyZzfio2 npQBAAAAAAAAAAAAAMACAABAAgAAQEMEAPh8AwAJAQAAAP54AY1QPU8CQRB9M+AHB4kXKjWYHBR2 UPAL+AFooRSWLskaLE4NkJx0V1PxK+ztLaj9FVR0JCbEisTz7XJQM8nMzs6+mXlvBSWg8A1AUWME jugl3WUqktdEsyzzry05z2vlPa6iHZ1rlfXr4zNcgtgMCHmfM/ugp8QuOaycYyq4MeNBb/JmgUe/ e6MRe5zN3FIUFWHvObaj6NYm0d1rbF7w8LlODD3q/CYXV0/DInFNoqkC7WAhKxSYxebdOjUbTf94 0GZex+ETBW7yDyef8mwHqbTgSPVNwOg0Bl6jJ5qzD3HCN+DL/4qoNu4no7GNMXRMwY46ptsGorrr qt/h2G93iecO/AOIW0uPwAIAAEQAZAAAAAAAAAAKAAAAAAAAAAAAAAAAAKgHbALoA+gDAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAPAATwYgAAALIECvAIAAAAHAQAAAAKAABjAAvwJAAA AH8AgACAAARBHAAAAD8BAAAGAL8BDAAfAP8BAAAIAD8DEAAQADMAIvESAAAAvwEAAGAAvwMAAIAB PwQAAAcAAAAQ8AQAAAAbAACAMgAH8AoCAAADBLsCD+cJ/COf0MeYhodlqh3/AOYBAAABAAAAtnUA AAAAEgBgIRvw3gEAALsCD+cJ/COf0MeYhodlqh1YAwAAAAAAAAAAAABADAAA4AMAAGj9EgAIAQYA rAEAAAD+eAGVUr9Lw1AQvnfpzyRgqYJWHKKD6FDRbgpCF8VBHWwHx7QQqWCbQgslTo7iJHTQ0UE3 F8WKg0Mn/wL/ADfXStHBSuvda9pqtUMPXrjv3r3v7rucgCCAci3oA1PA5qUTxI6Hgq5kTGCr1ZLe gphwY1o3T8cXJa6HKT7rG4FJoNwWQIhwlbwHRjrAuQKguTk6bKWKmaSTtygHAhT9JDZ+weahE6XS 3M0YAj752Bsl70PjPhtoED/bqezQgxBK7metgrFtlYwdO5vKwe5NvZSiY8TfS8uRvUqbU8haMfUV VzTWlndoAhBTq3glOCNvF4rtCkdNgmTDVkDZfY36YlUxtaw8B7nSod3BdwHGaReb+KX8vDdxHRhv 2P1KKYisNOFk0/YBwFpEHV98Cw9UynpqAruKz1Tmjbq8PX39vLcRNTOY96++C3+bt63PxGMv49X/ 6sAwf+p3/2UlI+dodOd47+M6Sy7OiaacawebeOnh+/nunE055zlbpSjvqSr3lHNAbn8DQ+CX6FFu tkCcSTiFopWFCm8Q0ItpOJEPOGGzHnY3tTcTWnGyb+roncJQAgAARABkAAAAAAAAAAoAAAAAAAAA AAAAAAAAMAJAAegD6AMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA8ABPBiAAAAsgQK 8AgAAAAdBAAAAAoAAGMAC/AkAAAAfwCAAIAABEEdAAAAPwEAAAYAvwEMAB8A/wEAAAgAPwMQABAA MwAi8RIAAAC/AQAAYAC/AwAAgAE/BAAABwAAABDwBAAAABwAAIAyAAfwmgEAAAMExkPA+zUlNVwG 1mBSQzK0CP8AdgEAAAEAAAB2eAAAAAASAGAhG/BuAQAAxkPA+zUlNVwG1mBSQzK0CFICAAAAAAAA AAAAAIADAAAAAgAA+GsFACAZAwA8AQAAAP54AZVRvUoDQRD+di7+5BLxvEqDxWkhWCSIT5AXMIVJ YXuBE4WcCSRwrFVKsfIRUtraGBALnyRInkBQq4DnzORyFkHQgWFnZ2e+79sZgyLgHBqAsAuxFfYi LSIy/KQ5Q2maalQz21mulNeVCTR0fM4frG6iAq5NAY/vLxyN2esOMGawUlZTxkk4uGjZXgQcKfeM Au4Ru1PWAsFrXcZRP2hESXDajcMrnD28JyF7UP9MNirnkwJXVxl1nc9j9wNPJHp7Vv4yo+EXH2z/ xTMQ3Lccd2QaqujaznlGpgbhaWc8P7o5SaK7aeN2twPc77iPnVf/V91zHpPr31L91SX9jIu/z2NZ /1T1BtZlHNmQqxuSPyCbvYc1vT3rTg3RftP2B1GMicwZ3LGH23kD191MfZ2RzH7xB14w2zfmyHTb OwIAAEQAZAAAAAAAAAAKAAAAAAAAAAAAAAAAAFsDfAHoA+gDAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAPAATwYgAAALIECvAIAAAAHgQAAAAKAABjAAvwJAAAAH8AgACAAARBHgAAAD8B AAAGAL8BDAAfAP8BAAAIAD8DEAAQADMAIvESAAAAvwEAAGAAvwMAAIABPwQAAAcAAAAQ8AQAAAAd AACAMgAH8IUBAAADBPk6YuT6YSgEOqe2fUeluvH/AGEBAAABAAAAxnoAAAAAEgBgIRvwWQEAAPk6 YuT6YSgEOqe2fUeluvEMAgAAAAAAAAAAAABgBQAAYAIAAOBTCAAwrgMAJwEAAAD+eAGNUbFOw0AM fXdJA00rEXUCxBAYEAxl4AtYQAzAAB2YqmQIgiE0UiNFYerM1I8AdnYGPoVPqFQxgQjPVlJ1rCXn bN97frHPoA04ngEsdiDWordtE1nDK60ZW1WVRkdms651FriujWzU6rG+721gG8RWQMD8k9EbPWRj j/hOjeniMs7vB2WWAJlq/9jJHy9pUxGFaxEMHtJkHF4lRXg9SuNH3L7Pi5gennwXp1t3Q5e4PtHr PI/9iXlxhXpYNnnkSH5QymzL/aW6en8D0Zkt6fxq36eFzhmk43mtE3J2salub3Ud/hNZM7L4Kpzn y7waqWSjce7zlN36ultRQ721AGuafehrGGv3bspxnqQYyoZAxi6elSCfi3lPVfpUaaZylP8PVSZh tpUCAABEAGQAAAAAAAAACgAAAAAAAAAAAAAAAAAwB5AB6APoAwAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAADwAE8GIAAACyBArwCAAAAB8EAAAACgAAYwAL8CQAAAB/AIAAgAAEQR8AAAA/ AQAABgC/AQwAHwD/AQAACAA/AxAAEAAzACLxEgAAAL8BAABgAL8DAACAAT8EAAAHAAAAEPAEAAAA HgAAgDIAB/DfAQAAAwQSSEJeUoVDSdAsmmDVoOZs/wC7AQAAAQAAAAF9AAAAABIAYCEb8LMBAAAS SEJeUoVDSdAsmmDVoOZsHAMAAAAAAAAAAAAAgAsAAIACAABI0xEAaN8DAIEBAAAA/ngBlVK7TgJB FD0zPHchkVCJsVgtjBRrxMaKSGUsxEIobDSBZIkmrpCFhKwVtYXxEwgJH2BlY2Fr47dAwIrE9d4B huCjYJJ5nDt35twzcwQMIPQoAIlNcItQN+R8JQVtqZiQQRCo1Z5Yn8USOi8pO7KTSFN8J7qGDVBu AKQIv9HqhXohARxSPk0qJ4lipXVd9hsO0FDcE2nRGW5PijUskSrfuE7TOnPa1nndrdzh4nnUrlC3 Cp/tQabWC1O2TSXGaT4wx3g3ud6GT6oIQ/ZFiLHnOKxuIjtfNFFblUGAmQaaqSsuDWa696bMXZGP M65q3A8t7x+D8Yk3rWOhlIKSlZZ8t1q/BWoZ82N/mP5X6bQOoRVbSrE9u3eh7+e944x59f+9v/Vl Y1yvrfXkIozzf/FglZ9arr8rhurdLM1zFGWenMbFMOOsxoZ6113PpCh7zVRe4xzMXJRCTKFX5U4h 5XbJb7YcFz12DOjEFh7UAR5OR2n1tzZ5bv7H7BngG00mmmU7AgAARABkAAAAAAAAAAoAAAAAAAAA AAAAAAAAIAN8AegD6AMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA8ABPBiAAAAsgQK 8AgAAAAgBAAAAAoAAGMAC/AkAAAAfwCAAIAABEEgAAAAPwEAAAYAvwEMAB8A/wEAAAgAPwMQABAA MwAi8RIAAAC/AQAAYAC/AwAAgAE/BAAABwAAABDwBAAAAB8AAIBfADEAMwA0ADkAMgA4ADAANAA5 ADYAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAGAABAbEAAABmAAAA rgAAAALOAgAAAAAAwAAAAAAAAEYAAAAA0E27vGtyywHQTbu8a3LLAQAAAAAAAAAAAAAAAAEATwBs AGUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAKAAIB////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA2AAA ABQAAAAAAAAAAQBDAG8AbQBwAE8AYgBqAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAABIAAgGtAAAArwAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAADZAAAAZgAAAAAAAAADAE8AYgBqAEkAbgBmAG8AAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEgACAf////+wAAAA/////wAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAANsAAAAGAAAAAAAAAAEAAAIIAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAP7/AwoAAP////8C zgIAAAAAAMAAAAAAAABGFwAAAE1pY3Jvc29mdCBFcXVhdGlvbiAzLjAADAAAAERTIEVxdWF0aW9u AAsAAABFcXVhdGlvbi4zAPQ5snEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAwAEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAcAAAAAgDowSYAAAAAAAAAsBYZAIQ2GQAAAAAAAwEBAwoKARKDYgACgSAAEoN6AAMP AAALEQEChhIiEoNwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEAAAIIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAP7/AwoAAP////8CzgIAAAAAAMAAAAAAAABG FwAAAE1pY3Jvc29mdCBFcXVhdGlvbiAzLjAADAAAAERTIEVxRQBxAHUAYQB0AGkAbwBuACAATgBh AHQAaQB2AGUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACAAAgD///////////// //8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADcAAAAQgAAAAAAAABfADEAMwA0 ADkAMgA4ADAANAA5ADIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA GAABAP//////////swAAAALOAgAAAAAAwAAAAAAAAEYAAAAA0E27vGtyywHQTbu8a3LLAQAAAAAA AAAAAAAAAAEATwBsAGUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAKAAIB////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAA3gAAABQAAAAAAAAAAQBDAG8AbQBwAE8AYgBqAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgGyAAAAtAAAAP////8AAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAADfAAAAZgAAAAAAAAB1YXRpb24ACwAAAEVxdWF0aW9uLjMA 9DmycQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADAAQAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABwAAAACAOjB LwAAAAAAAAAwGBkAhDYZAAAAAAADAQEDCgoBEoNIAAMPAQALARKDcAASg3IAEoNlAAARAAoCgigA EoN6AAKCKQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAQAAAggAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAEA/v8DCgAA/////wLOAgAAAAAAwAAAAAAAAEYXAAAATWljcm9zb2Z0IEVx dWF0aW9uIDMuMAAMAAAARFMgRXF1YXRpb24ACwAAAEVxdWF0aW9uLjMA9DmycQAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADAAQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAMATwBiAGoASQBuAGYAbwAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIB/////7UAAAD///// AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA4QAAAAYAAAAAAAAARQBxAHUAYQB0 AGkAbwBuACAATgBhAHQAaQB2AGUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACAA AgD///////////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADiAAAASwAA AAAAAABfADEAMwA0ADkAMgA4ADIANgA0ADIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAGAABAXoAAADFAAAAuAAAAALOAgAAAAAAwAAAAAAAAEYAAAAA0E27vGtyywHQ Tbu8a3LLAQAAAAAAAAAAAAAAAAEATwBsAGUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAKAAIB////////////////AAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAA5AAAABQAAAAAAAAAAQBDAG8AbQBwAE8AYgBqAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgG3AAAAuQAAAP////8A AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADlAAAAZgAAAAAAAAADAE8AYgBqAEkA bgBmAG8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEgAC Af////+6AAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAOcAAAAGAAAA AAAAAEUAcQB1AGEAdABpAG8AbgAgAE4AYQB0AGkAdgBlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAgAAIA////////////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAA6AAAAEMAAAAAAAAAXwAxADMANAA5ADIAOAAyADYANQA2AAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABgAAQH//////////70AAAACzgIAAAAAAMAAAAAAAABG AAAAANBNu7xrcssB0E27vGtyywEAAAAAAAAAAAAAAAAcAAAAAgDowScAAAAAAAAAuA8YAPyaGgAA AAAAAwEBAwoKARKDdwADDwEACwESg2kAABEACgKCKAASg24AAoIpAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEAAAIIAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAP7/ AwoAAP////8CzgIAAAAAAMAAAAAAAABGFwAAAE1pY3Jvc29mdCBFcXVhdGlvbiAzLjAADAAAAERT IEVxdWF0aW9uAAsAAABFcXVhdGlvbi4zAPQ5snEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAwAEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAcAAAAAgDowScAAAAAAAAA8H0aAPyaGgAAAAAAAwEBAwoKARKDdwAD DwEACwESg28AABEACgKCKAASg24AAoIpAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEATwBsAGUAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAKAAIB////////////////AAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA6gAAABQAAAAAAAAAAQBDAG8AbQBwAE8A YgBqAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgG8 AAAAvgAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADrAAAAZgAAAAAA AAADAE8AYgBqAEkAbgBmAG8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAEgACAf////+/AAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAO0AAAAGAAAAAAAAAEUAcQB1AGEAdABpAG8AbgAgAE4AYQB0AGkAdgBlAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgAAIA////////////////AAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAA7gAAAEMAAAAAAAAAXwAxADMANAA5ADIAOAAyADkANwA5AAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABgAAQH/////ygAAAMIAAAACzgIA AAAAAMAAAAAAAABGAAAAANBNu7xrcssB0E27vGtyywEAAAAAAAAAAAAAAAABAE8AbABlAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACgACAf// /////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAPAAAAAUAAAAAAAA AAEAQwBvAG0AcABPAGIAagAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAASAAIBwQAAAMMAAAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAA8QAAAGYAAAAAAAAAAwBPAGIAagBJAG4AZgBvAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgH/////xAAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAADzAAAABgAAAAAAAAABAAACCAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQD+/wMKAAD/////As4CAAAAAADA AAAAAAAARhcAAABNaWNyb3NvZnQgRXF1YXRpb24gMy4wAAwAAABEUyBFcXVhdGlvbgALAAAARXF1 YXRpb24uMwD0ObJxAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAMABAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA HAAAAAIA6MEaAAAAAAAAAPB9GgD8mhoAAAAAAAMBAQMKCgESg3AAAw8BAAsBAogwAAARAAAAAAAA AAAAAAAAAAEAAAIIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAABAP7/AwoAAP////8CzgIAAAAAAMAAAAAAAABGFwAAAE1pY3Jvc29mdCBF cXVhdGlvbiAzLjAADAAAAERTIEVxdWF0aW9uAAsAAABFcXVhdGlvbi4zAPQ5snEAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEUAcQB1AGEAdABpAG8AbgAgAE4AYQB0AGkAdgBl AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgAAIA////////////////AAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA9AAAADYAAAAAAAAAXwAxADMANAA5ADIAOAAy ADkANgA3AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABgAAQC7AAAA wAAAAMcAAAACzgIAAAAAAMAAAAAAAABGAAAAANBNu7xrcssB0E27vGtyywEAAAAAAAAAAAAAAAAB AE8AbABlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAACgACAf///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA APUAAAAUAAAAAAAAAAEAQwBvAG0AcABPAGIAagAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAASAAIBxgAAAMgAAAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAA9gAAAGYAAAAAAAAAAwBPAGIAagBJAG4AZgBvAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgH/////yQAAAP////8AAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD4AAAABgAAAAAAAABFAHEAdQBhAHQAaQBvAG4A IABOAGEAdABpAHYAZQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIAACAP////// /////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAPkAAAA2AAAAAAAAAF8A MQAzADQAOQAyADgAMwA2ADIAOAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAYAAEA///////////MAAAAAs4CAAAAAADAAAAAAAAARgAAAADQTbu8a3LLAdBNu7xrcssB AAAAAAAAAAAAAAAAAQBPAGwAZQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAoAAgH///////////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAD6AAAAFAAAAAAAAAAAAAMABAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHAAAAAIA6MEaAAAAAAAAADgEGQBMeBgA AAAAAAMBAQMKCgESg2IAAw8BAAsBAogwAAARAAAAAAAAAAAAAAAAAAEAAAIIAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAP7/AwoAAP// //8CzgIAAAAAAMAAAAAAAABGFwAAAE1pY3Jvc29mdCBFcXVhdGlvbiAzLjAADAAAAERTIEVxdWF0 aW9uAAsAAABFcXVhdGlvbi4zAPQ5snEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAwAEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAcAAAAAgDoweUAAAAAAAAAyA8YAPyaGgAAAAAAAwEBAwoKATKDcwAGCAAAAoIo ABKDbgACgikAAoY9ADKDZAAGCAAAAoIoABKDbgACgikAAoYrABKDdwADDwEACwESg28AABEACgKC KAASg24AAoIpAAKBIAASg2IAAw8BAAsBAgEAQwBvAG0AcABPAGIAagAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIBywAAAM0AAAD/////AAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA+wAAAGYAAAAAAAAAAwBPAGIAagBJAG4AZgBvAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgH/////zgAA AP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD9AAAABgAAAAAAAABFAHEA dQBhAHQAaQBvAG4AIABOAGEAdABpAHYAZQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAIAACAP///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP4A AAABAQAAAAAAADEAVABhAGIAbABlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAOAAIAAQAAAAMAAAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAA+AEAANApAAAAAAAAAQEAAAIBAAD+////BAEAAAUBAAAGAQAABwEAAAgBAAAJ AQAACgEAAP7///8MAQAADQEAAA4BAAAPAQAA/v///xEBAAD+//////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////+IMAAAEQAKAoEgADKDcwAGCAAAAoIoABKDbgAChhIiEoNw AAMPAQALAQKIMAAAEQAKAoIpAAKGKwASg3cAAw8BAAsBEoNpAAARAAoCgigAEoNuAAKCKQACgSAA EoNiAAKBIAAyg3MABggAAAKCKAASg24AAoYSIhKDcAACgikAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD+/wAABQECAAAAAAAA AAAAAAAAAAAAAAABAAAA4IWf8vlPaBCrkQgAKyez2TAAAACUAQAAEQAAAAEAAACQAAAAAgAAAJgA AAADAAAApAAAAAQAAACwAAAABQAAANQAAAAGAAAA4AAAAAcAAADsAAAACAAAAAABAAAJAAAAJAEA ABIAAAAwAQAACgAAAFABAAAMAAAAXAEAAA0AAABoAQAADgAAAHQBAAAPAAAAfAEAABAAAACEAQAA EwAAAIwBAAACAAAA5AQAAB4AAAAEAAAAAAAAAB4AAAAEAAAAAAAAAB4AAAAcAAAASnVpbi1Id2V5 IChSYXltb25kKSBDaGVuAAAAAB4AAAAEAAAAAAAAAB4AAAAEAAAAAAAAAB4AAAAMAAAATm9ybWFs LmRvdG0AHgAAABwAAABKdWluLUh3ZTIAB/CFAQAAAwSKFrKp7LQEuPentXWvWm/K/wBhAQAAAQAA AJZ/AAAAABIAYCEb8FkBAACKFrKp7LQEuPentXWvWm/KDAIAAAAAAAAAAAAAAAUAAGACAAA4wAcA MK4DACcBAAAA/ngBjZGxTsNADIb/u6SBpJWoOgFiCAwIhlKJJ2BCDJQBOjCS4RAMgYhWisLUmakv UcTKztBH4REqVUxIhN9WUnWsJedsn+0v9hmEgBcYwGIPIg1qaGvLGl5pzNiyLNU6MdtVrLnMa9k7 i0aH8cNgC7tgbgm06c9ovVNnPhAwv1nltNBPRg+DInNApuxfO/7jJWUiUPgW7cFj6obxlcvj6+c0 ecLt5yJPqPHZT/6xc99jU3SZvcnzNBqbvi+lx0Xth574R4XMttpfouv3NxDOfIUz1b6vS845pONF xYk5u8hEt7c+h//Eqjmr+Cqc59tMjUcre3Eu4im7jXS3QkO1tTY21PvS1zDWHtwUw5FL0ZMNgRX7 eNMC+VwuOkrpklJPJQzgHyBsYXZOAgAARABkAAAAAAAAAAoAAAAAAAAAAAAAAAAAMAJAAegD6AMA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA8ABPBiAAAAsgQK8AgAAAAhBAAAAAoAAGMA C/AkAAAAfwCAAIAABEEhAAAAPwEAAAYAvwEMAB8A/wEAAAgAPwMQABAAMwAi8RIAAAC/AQAAYAC/ AwAAgAE/BAAABwAAABDwBAAAACAAAIAyAAfwmAEAAAMEcmrAdZvo259076CM/guwp/8AdAEAAAEA AADRgQAAAAASAGAhG/BsAQAAcmrAdZvo259076CM/guwp1ICAAAAAAAAAAAAAIADAAAAAgAA+GsF ACAZAwA6AQAAAP54AZVRwUrDQBB9O6nWpAVjThIsrAqChxbxC/oD9mB78JpCRKGxhRbCeupRPPkJ /QVPBfHgp0i/QFBPBePsJI2HIujAsLOzM++9nVFwAedYAYQ9WNtgd2kVkeInySnKskyiltotcrWy rk6gqRNw/mhzGyG4NgN8vr9wNGdvO8CcwWpFTR1n0eSqZ0YxcCLcS9LcY+1BWCsEv3edxGPdiVN9 PkyiG1w8vqcRu25/pm542ahwdZNRt/g89T7wRFbvyNi/LGn6xQfbf/EULO5biTtTHVF0a3KemWrB 8vQLnh/dnCSru2uS/nAAVEPvcPAa/Ko751Gl/h3R31zTz7j4+zzW9S9ErzYe49gNebIh+wcUs/dR lduz7FQRHXTNeBInaNg5gzv2cZ83cN3dIpAZ2dmv/sALZvsGRUhvDjsCAABEAGQAAAAAAAAACgAA AAAAAAAAAAAAAAAgA3wB6APoAwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADwAE8GIA AACyBArwCAAAACIEAAAACgAAYwAL8CQAAAB/AIAAgAAEQSIAAAA/AQAABgC/AQwAHwD/AQAACAA/ AxAAEAAzACLxEgAAAL8BAABgAL8DAACAAT8EAAAHAAAAEPAEAAAAIQAAgDIAB/CFAQAAAwT4O5MG X/koFg54sF330DMs/wBhAQAAAQAAAB+EAAAAABIAYCEb8FkBAAD4O5MGX/koFg54sF330DMsDAIA AAAAAAAAAAAAAAUAAGACAAA4wAcAMK4DACcBAAAA/ngBjZGxTsNADIb/c9JC0iKiTqhiCAwIhjDw BEyIpQzQgZEMh2AIRLRSFKaOiKkvUXZ2hj4K4gkqISYkwu9TUnWsJedsn+0v9hkEgNc2gGAXKi1q II0lhlcuZqSqKmcdm5061lnmdeVG0OoxftDeRh/MrYCI/pzWG3XuA23md+qcLgbp+G5Y5hbIHftX Jn+8pEwVCl8QDe8zO4ovbBFfPmbpA67fv4uUGp/+FFv924RNkTB7k+dJODEDX0uPysYPPPUPS51t tb9G1+9voJzFCmfm+j4vOWfQjuc1J+bsKlO3vfU5/CdWLVjFV+E8n2ZmPFr5k7UhT91t6HarNNRb i7DhvA/3GkZk/6ocjW2GRDcEVuzh1RXo5+Wr5ygJKc1UygD+AVaLXy83AgAARABkAAAAAAAAAAoA AAAAAAAAAAAAAAAAWAJoAegD6AMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA8ABPBi AAAAsgQK8AgAAAAjBAAAAAoAAGMAC/AkAAAAfwCAAIAABEEjAAAAPwEAAAYAvwEMAB8A/wEAAAgA PwMQABAAMwAi8RIAAAC/AQAAYAC/AwAAgAE/BAAABwAAABDwBAAAACIAAIAyAAfwgQEAAAMEjSC1 JzhuYHSlieil6GjJ4f8AXQEAAAEAAABahgAAAAASAGAhG/BVAQAAjSC1JzhuYHSlieil6GjJ4QoC AAAAAAAAAAAAAMADAABAAgAA0M8FAPh8AwAjAQAAAP54AY1RIU/DYBB933VjrFtCmSIEkoIgIIZA oooHBEwgqShhooNkS5q6atR+wSQehUHsp6DQSxYUCeXd0S2Tu+Tau/vevXfffQ5NwKs7QLAHtTq9 KYtIHI+s5qQsS4tO3U5Vay1xbYlk6nVYP9rYwi6ILYGA+ZTRK73wgC+StSpMG1fx6LGXPyfAvWn/ SPHLQ9pYRVETBL1+mgzD6yQLb57SeIC7t3kW08PoO9vef5jUiOsSvcn/mV+4hqetJ/kif7f5j3O9 2yq/otbnd1Cd2YrOhSjDYKkTQfOs0gl5d7Wxqa+vw5nYNWPX//yf7twY+rnPum7Wt82qFqqdBWhY 9mFv4UQOb/PhKEkx0f2AHQd4sQb9XM47ptEl7+JOfBjaH8xPYWU4AgAARABkAAAAAAAAAAoAAAAA AAAAAAAAAAAAfwJoAegD6AMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA8ABPBiAAAA sgQK8AgAAAAkBAAAAAoAAGMAC/AkAAAAfwCAAIAABEEkAAAAPwEAAAYAvwEMAB8A/wEAAAgAPwMQ ABAAMwAi8RIAAAC/AQAAYAC/AwAAgAE/BAAABwAAABDwBAAAACMAAIAyAAfwggEAAAME6N32x9et 0XmWn272NCiWE/8AXgEAAAEAAACRiAAAAAASAGAhG/BWAQAA6N32x9et0XmWn272NCiWEwoCAAAA AAAAAAAAAAAEAABAAgAAQDIGAPh8AwAkAQAAAP54AY1RsU7DQAx956RA00pEnRAiUmBAMJSBpQtD fgAG6MBIhiAY0oJaKbqtM1O/gp2doZ/CBEuHCsSERHg+kqpjLTmxfc/v+XwGTcBrGECwB7UGvSl1 JIZHrmakLEsXnZidqtZa4tqSCPwO64cb29gFsSUQMp8xeqbPPOCdZK0K08Z5Or7r24cMuHHaPzL5 5SFtqqLwBWH/Ps9G8UVWxJfDPB3g+uWrSOlx8l30otvIJ65L9Bb/p8HE9DxtPbZ1/uHmP7J6t1V+ Ra3Pb6A6ixWdR1GGwVIngeZFpRPz7mpTp76+Dmdi14Jd//O/mTPHMLQB67rZwG1WtVDtLMSmy17d WxiRgys7Gmc5It0P2LGPJ9egn/lnx2l0yVvfiQ9D+wN3/WAwFwIAAEQAZAAAAAAAAAAKAAAAAAAA AAAAAAAAACwBaAHoA+gDAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAPAATwYgAAALIE CvAIAAAAJQQAAAAKAABjAAvwJAAAAH8AgACAAARBJQAAAD8BAAAGAL8BDAAfAP8BAAAIAD8DEAAQ ADMAIvESAAAAvwEAAGAAvwMAAIABPwQAAAcAAAAQ8AQAAAAkAACAMgAH8GEBAAADBE6IEFg8s21L mA0UPJoC55X/AD0BAAABAAAAyYoAAAAAEgBgIRvwNQEAAE6IEFg8s21LmA0UPJoC55WSAQAAAAAA AAAAAADgAQAAQAIAAOjnAgD4fAMAAwEAAAD+eAFjZOBkYGA+ycDAwMQgAyQZGFiBmJMJxmJiZISK MTL9//8fLKvHKAEV44ar42FyYHrAKAQUV2PjZ5BiAKr9z8AgAOQfALKWAfECoEHPgJgbqoaHwTex JCOksiCVgSEBbPdvJgWgHhCYALKUgYWJQSAkMze1WMEvtVwhKD83MY8hYuPn8kQgVnD4Wq4jm+bD AlSnC1TNAaSNuB4wmoPda1AJ8stvpoZ/QAoIJoBFiTePkQFk7ge4uQ2MWQwgJxVUcgFJkA+5wD4E OxPqdgEGdqAcA8MecJgwMjEpBVcWl6TmMviA3MkA1KHI0AXRAFTl81kIbAfI7RC7GBmYwfoBOpZK FRcCAABEAGQAAAAAAAAACgAAAAAAAAAAAAAAAAAEAWgB6APoAwAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAADwAE8GIAAACyBArwCAAAACYEAAAACgAAYwAL8CQAAAB/AIAAgAAEQSYAAAA/ AQAABgC/AQwAHwD/AQAACAA/AxAAEAAzACLxEgAAAL8BAABgAL8DAACAAT8EAAAHAAAAEPAEAAAA JQAAgDIAB/BhAQAAAwSsx7021M76acmncGt1QVKg/wA9AQAAAQAAAOCMAAAAABIAYCEb8DUBAACs x7021M76acmncGt1QVKgkgEAAAAAAAAAAAAAoAEAAEACAAAQhAIA+HwDAAMBAAAA/ngBY2TgZGBg PsnAwMDEIAMkGRhYgZiTCcZiYmSEijEy/f//HyyrxygBFeOGq+NhcmBawCgEFFdj42eQYgCq/c/A IADkHwCylgFxAtCgZ0DMDVXDw+CbWJIRUlmQysCQALb7N5MCUA8ITABZysDCxCAQkpmbWqzgl1qu EJSfm5jHELHxc3kiECs4fC0P4k1bxgJUpwtUzQGkjbgeML5kAGk1qAT55TdTwz8gBQQTwL4g3jxG BpC5H+DmNjDqgc1NquQCioN8yAX2IdiZULcLMLCDrGLYAw4TRiYmpeDK4pLUXIZlIHcyAHUoMnRB NABV+XwWAtsBcjvELkYGZrB+AMDVS6ZCAwAARABkAAAAAAAAAAoAAAAAAAAAAAAAAAAAiBOQAegD 6AMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA8ABPBiAAAAsgQK8AgAAAAnBAAAAAoA AGMAC/AkAAAAfwCAAIAABEEnAAAAPwEAAAYAvwEMAB8A/wEAAAgAPwMQABAAMwAi8RIAAAC/AQAA YAC/AwAAgAE/BAAABwAAABDwBAAAACYAAIAyAAfwjAIAAAMEsNafCss3gGYdaoCEk3HqSv8AaAIA AAEAAAD3jgAAAAASAGAhG/BgAgAAsNafCss3gGYdaoCEk3HqShoFAAAAAAAAAAAAAEAfAACAAgAA uHEwAGjfAwAuAgAAAP54AZWUz0sbQRTH30xiNJu0xq0/EpvWbQ9FIxERT6LoxZv2oB48GjX+gCZR IoTtoXjwFBD6Bwiihx6KF8Gjh3gSvLR4U/A/aEEsejDQ9b23m5lspaALu5nPvDffl33f2REQBghs SwAJSaCrAe8wsjuSQngjIR3H4WifiHtzEZUXlZtyrMvE+Q+hZugEzHUAYsgVHB0RdQEMYn7Ey4nC ZGZjZcZeywLMce2q3PyLQby+UlEISojNrOayRetjtmRNFXKZPMwe/ill8LbGbktOcikRxLw0Zjfh 74BREdtvaWmPXeOLTuJuj1OimiD+ouJ7zJbi3TjFNaeZtd5Ve71eReRfEeu42VwfT4nBl8S63uUL Yq0/HPXzWYRY6302iGv/vyIuQ/54uaE+fgYHQWJdb0X68+PsZ00vJYbAzSfvq9JCz+h6bv8FkA/X KOX23ZKnJun2e3215Di/d7/t1tE+U9ZzfHbrCOX3yRtSWFP+LbwmzisOsf9Fxb/Zz3nFyx3+/KU2 4pKK37f49XfYX60/yv5q/TL7qfWD7KfOP2/y67eynzq+xf4tqvr77JeOj7NfRa+P2i9SfXofH/s1 x/t61atryR9hUiw88gsnJdWZtnPzhU/YqaRx9/Pa/O93+a9f60nSTav3G2ol7lX8PeaPf2v0xycC xCO2gU86aQw+aWgOvDMkBo1Mx3w2CSnfT9vFjWwOEnReAK54B2VeQI9fNybv3TT2ubaHA7z+Acm4 4tekAgAARABkAAAAAAAAAAoAAAAAAAAAAAAAAAAAdAl8AegD6AMAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAA8ABPBiAAAAsgQK8AgAAAAoBAAAAAoAAGMAC/AkAAAAfwCAAIAABEEKAAAA PwEAAAYAvwEMAB8A/wEAAAgAPwMQABAAMwAi8RIAAAC/AQAAYAC/AwAAgAE/BAAABwAAABDwBAAA ACcAAIAyAAfw7gEAAAMEkVim9bcK30y2RMoAn1tR3/8AygEAAAIAAAA5kgAAAAASAGAhG/DCAQAA kVim9bcK30y2RMoAn1tR3woDAAAAAAAAAAAAACAPAABgAgAA4HEXADCuAwCQAQAAAP54AWNk4GRg YG5lZGBgYpBhAAFWIOZkgrGYGIFSYDFGpv///4NZeowSUDFuuDoepgQmBX4hoLgaGz+DFANQ7X8G BgEg/wCQtQWIH/AxMIgA1XND1fAw+CaWZIRUFqQyMASA7f7N1PAPKAkEE0CWMrAwMQiEZOamFiv4 pZYrBOXnJuYxRGz8XJ4IxAoOX8v3SKb1sQDV6QJVcwBpI64DjFl8IK2alTD+Bi4QXwPK12K8xAni 18HlK8B8BTj/Ohuq/g5WZP2nGFaxoOpPZ0JVLwEOL4R9VgwQ9aCwRfYfSJR4/zEygPz5AagJ5q/5 vCATCuDutuQG8fPg/LtgfxXD+Uc4QPJJcL4g2J8I9U1gf6XA5ReC/YGQdwX7o7gS3R9AQ5lA/giu zE3Kz2FgWM3F9dvgoxDOeIL4gxHuD28ekLt04fYuYwfxteF8L2YQ37aSC0iCUhYXOGWBxBigaUaA gR3M2wNOi4xMTErBlcUlqbkMfaD0wQDUocjQBdYAInw+C4HDUhfoP1iYMoP1AwCDZ4meAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAFAFMAdQBtAG0AYQByAHkASQBuAGYAbwByAG0AYQB0AGkAbwBuAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAKAACAQIAAADRAAAA/////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAMBAADEAQAAAAAAAAUARABvAGMAdQBtAGUAbgB0AFMAdQBtAG0AYQByAHkA SQBuAGYAbwByAG0AYQB0AGkAbwBuAAAAAAAAAAAAAAA4AAIB////////////////AAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACwEAACwBAAAAAAAAAQBDAG8AbQBwAE8AYgBqAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAgD///////// //////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQAQAAeQAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAP///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAHkgKFJheW1vbmQpIENoZW4AAAAAHgAAAAQAAAAxMQAAHgAAABgAAABNaWNyb3Nv ZnQgT2ZmaWNlIFdvcmQAAABAAAAAACyYziEAAABAAAAAAFSPuEJyywFAAAAAAHIOp2tyywEDAAAA BwAAAAMAAACICgAAAwAAAAk8AAADAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP7/AAAFAQIAAAAAAAAAAAAAAAAAAAAAAAEA AAAC1c3VnC4bEJOXCAArLPmuMAAAAPwAAAAMAAAAAQAAAGgAAAAPAAAAcAAAAAUAAACQAAAABgAA AJgAAAARAAAAoAAAABcAAACoAAAACwAAALAAAAAQAAAAuAAAABMAAADAAAAAFgAAAMgAAAANAAAA 0AAAAAwAAADdAAAAAgAAAOQEAAAeAAAAGAAAAEJyb2FkY29tIENvcnBvcmF0aW9uAAAAAAMAAACA AAAAAwAAACQAAAADAAAAbUYAAAMAAAAAAAwACwAAAAAAAAALAAAAAAAAAAsAAAAAAAAACwAAAAAA AAAeEAAAAQAAAAEAAAAADBAAAAIAAAAeAAAABgAAAFRpdGxlAAMAAAABAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAQIAAAICAAADAgAABAIAAAUCAAAGAgAABwIAAAgCAAAJAgAACgIAAAsCAAAMAgAA /v////7///////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////9qBBMAEgABAAsBDwAHAAQABAAEAAAABAAIAAAAmAAAAJ4AAACeAAAAngAAAJ4AAACe AAAAngAAAJ4AAACeAAAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAAdgIAAHYC AAB2AgAAdgIAAHYCAAB2AgAAdgIAAHYCAAB2AgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAAPgIA ADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAA NgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAAKgAAAA2 BgAANgYAABYAAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAALgAAAA2BgAANgYAADYG AAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAABoAQAASAEAADYGAAA2BgAANgYA ADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAA NgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2 BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYG AAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYA ADYGAAA2BgAANgYAADYGAAA2BgAAsAMAADYGAAAyBgAAGAAAAMADAADQAwAA4AMAAPADAAAABAAA EAQAACAEAAAwBAAAQAQAAFAEAABgBAAAcAQAAIAEAACQBAAAwAMAANADAADgAwAA8AMAAAAEAAAQ BAAAMgYAACgCAADYAQAA6AEAACAEAAAwBAAAQAQAAFAEAABgBAAAcAQAAIAEAACQBAAAwAMAANAD AADgAwAA8AMAAAAEAAAQBAAAIAQAADAEAABABAAAUAQAAGAEAABwBAAAgAQAAJAEAADAAwAA0AMA AOADAADwAwAAAAQAABAEAAAgBAAAMAQAAEAEAABQBAAAYAQAAHAEAACABAAAkAQAAMADAADQAwAA 4AMAAPADAAAABAAAEAQAACAEAAAwBAAAQAQAAFAEAABgBAAAcAQAAIAEAACQBAAAwAMAANADAADg AwAA8AMAAAAEAAAQBAAAIAQAADAEAABABAAAUAQAAGAEAABwBAAAgAQAAJAEAADAAwAA0AMAAOAD AADwAwAAAAQAABAEAAAgBAAAMAQAAEAEAABQBAAAYAQAAHAEAACABAAAkAQAADgBAABYAQAA+AEA AAgCAAAYAgAAVgIAAH4CAAAgAAAAT0oEAFBKBABRSgQAX0gBBG1ICQRuSAkEc0gJBHRICQQAAAAA SgAAYPH/AgBKAAwQAADXRNgAAAAGAE4AbwByAG0AYQBsAAAADAAAABJkFAEBABSkyAAYAENKFgBf SAEEYUoWAG1ICQRzSAkEdEgJBGgAAUABAAIAaAAMFA8AvV8GAJAACQBIAGUAYQBkAGkAbgBnACAA MQAAABMAAQAFJAEGJAETpOABFKQAAEAmAAAnADUIgUIqB0NKHABPSgUAUEoAAFFKBQBcCIFeSgAA YUocAHBoNl+RAABoAAJAAQACAGgADBwQAL1fBgCQAAkASABlAGEAZABpAG4AZwAgADIAAAATAAIA BSQBBiQBE6TIABSkAABAJgEAJwA1CIFCKgdDShoAT0oFAFBKAABRSgUAXAiBXkoAAGFKGgBwaE+B vQAAAAAAAAAAAAAAAAAAAABEAEEg8v+hAEQADAwAAAAAAAAQABYARABlAGYAYQB1AGwAdAAgAFAA YQByAGEAZwByAGEAcABoACAARgBvAG4AdAAAAAAAUgBpAPP/swBSAAwdAAAAAAAAMAYMAFQAYQBi AGwAZQAgAE4AbwByAG0AYQBsAAAAHAAX9gMAADTWBgABCgNsADTWBgABBQMAAGH2AwAAAgALAAAA KABrIPT/wQAoAAANAAAAAAAAMAYHAE4AbwAgAEwAaQBzAHQAAAACAAwAAAAAAFwA/g+iAPEAXAAM AAEAvV8GAJAADgBIAGUAYQBkAGkAbgBnACAAMQAgAEMAaABhAHIAAAAnADUIAUIqB0NKHABPSgUA UEoAAFFKBQBcCAFeSgAAYUocAHBoNl+RAABcAP4PogABAVwADAACAL1fBgCQAA4ASABlAGEAZABp AG4AZwAgADIAIABDAGgAYQByAAAAJwA1CAFCKgdDShoAT0oFAFBKAABRSgUAXAgBXkoAAGFKGgBw aE+BvQAAbAD+TwEAEgFsAAwAAABXZScAAAANAFMAcABlAGMAUABhAHIAYQAgADEALQA5ADkAAAAj ABEAAyQDCiYAC0YCAA3GBwGgBQHQAgASZGgBAQAUpAAAYSQDABQAQ0oYAE9KAABQSgAAUUoAAGFK GAB0AP5PAQAiAXQADAAAAFB0fQAAAAoAcwBwAGUAYwBwAGEAcgBhADAAMQAAADIAEgADJAMKJgAL RgMADcYWAQAABtACoAVwCEALEA7gEAAAAAAAABJkaAEBABSkAABhJAMUAENKGABPSgAAUEoAAFFK AABhShQAUEsDBBQABgAIAAAAIQCCirwT+gAAABwCAAATAAAAW0NvbnRlbnRfVHlwZXNdLnhtbKyR y2rDMBBF94X+g9C22HK6KKXYzqJJd30s0g8Y5LEtao+ENAnJ33fsuFC6CC10IxBizpl7Va6P46AO GJPzVOlVXmiFZH3jqKv0++4pu9cqMVADgyes9AmTXtfXV+XuFDApmaZU6Z45PBiTbI8jpNwHJHlp fRyB5Ro7E8B+QIfmtijujPXESJzxxNB1+SoLRNegeoPILzCKx7Cg8Pv5DCSAmAtYq8czYVqi0hDC 4CywRDAHan7oM9+2zmLj7X4UaT6DF9jNBDO/XGD1P+ov5wZb2A+stkfp4lx/xCH9LdtSay6Tc/7U u5AuGC6Xt7Rh5r+tPwEAAP//AwBQSwMEFAAGAAgAAAAhAKXWp+fAAAAANgEAAAsAAABfcmVscy8u cmVsc4SPz2rDMAyH74W9g9F9UdLDGCV2L6WQQy+jfQDhKH9oIhvbG+vbT8cGCrsIhKTv96k9/q6L +eGU5yAWmqoGw+JDP8to4XY9v3+CyYWkpyUIW3hwhqN727VfvFDRozzNMRulSLYwlRIPiNlPvFKu QmTRyRDSSkXbNGIkf6eRcV/XH5ieGeA2TNP1FlLXN2Cuj6jJ/7PDMMyeT8F/ryzlRQRuN5RMaeRi oagv41O9kKhlqtQe0LW4+db9AQAA//8DAFBLAwQUAAYACAAAACEAa3mWFoMAAACKAAAAHAAAAHRo ZW1lL3RoZW1lL3RoZW1lTWFuYWdlci54bWwMzE0KwyAQQOF9oXeQ2TdjuyhFYrLLrrv2AEOcGkHH oNKf29fl44M3zt8U1ZtLDVksnAcNimXNLoi38Hwspxuo2kgcxSxs4ccV5ul4GMm0jRPfSchzUX0j 1ZCFrbXdINa1K9Uh7yzdXrkkaj2LR1fo0/cp4kXrKyYKAjj9AQAA//8DAFBLAwQUAAYACAAAACEA lrWt4pYGAABQGwAAFgAAAHRoZW1lL3RoZW1lL3RoZW1lMS54bWzsWU9v2zYUvw/YdyB0b2MndhoH dYrYsZstTRvEboceaYmW2FCiQNJJfRva44ABw7phhxXYbYdhW4EW2KX7NNk6bB3Qr7BHUpLFWF6S NtiKrT4kEvnj+/8eH6mr1+7HDB0SISlP2l79cs1DJPF5QJOw7d0e9i+teUgqnASY8YS0vSmR3rWN 99+7itdVRGKCYH0i13Hbi5RK15eWpA/DWF7mKUlgbsxFjBW8inApEPgI6MZsablWW12KMU08lOAY yN4aj6lP0FCT9DZy4j0Gr4mSesBnYqBJE2eFwQYHdY2QU9llAh1i1vaAT8CPhuS+8hDDUsFE26uZ n7e0cXUJr2eLmFqwtrSub37ZumxBcLBseIpwVDCt9xutK1sFfQNgah7X6/W6vXpBzwCw74OmVpYy zUZ/rd7JaZZA9nGedrfWrDVcfIn+ypzMrU6n02xlsliiBmQfG3P4tdpqY3PZwRuQxTfn8I3OZre7 6uANyOJX5/D9K63Vhos3oIjR5GAOrR3a72fUC8iYs+1K+BrA12oZfIaCaCiiS7MY80QtirUY3+Oi DwANZFjRBKlpSsbYhyju4ngkKNYM8DrBpRk75Mu5Ic0LSV/QVLW9D1MMGTGj9+r596+eP0XHD54d P/jp+OHD4wc/WkLOqm2chOVVL7/97M/HH6M/nn7z8tEX1XhZxv/6wye//Px5NRDSZybOiy+f/Pbs yYuvPv39u0cV8E2BR2X4kMZEopvkCO3zGBQzVnElJyNxvhXDCNPyis0klDjBmksF/Z6KHPTNKWaZ dxw5OsS14B0B5aMKeH1yzxF4EImJohWcd6LYAe5yzjpcVFphR/MqmXk4ScJq5mJSxu1jfFjFu4sT x7+9SQp1Mw9LR/FuRBwx9xhOFA5JQhTSc/yAkArt7lLq2HWX+oJLPlboLkUdTCtNMqQjJ5pmi7Zp DH6ZVukM/nZss3sHdTir0nqLHLpIyArMKoQfEuaY8TqeKBxXkRzimJUNfgOrqErIwVT4ZVxPKvB0 SBhHvYBIWbXmlgB9S07fwVCxKt2+y6axixSKHlTRvIE5LyO3+EE3wnFahR3QJCpjP5AHEKIY7XFV Bd/lbobod/ADTha6+w4ljrtPrwa3aeiINAsQPTMR2pdQqp0KHNPk78oxo1CPbQxcXDmGAvji68cV kfW2FuJN2JOqMmH7RPldhDtZdLtcBPTtr7lbeJLsEQjz+Y3nXcl9V3K9/3zJXZTPZy20s9oKZVf3 DbYpNi1yvLBDHlPGBmrKyA1pmmQJ+0TQh0G9zpwOSXFiSiN4zOq6gwsFNmuQ4OojqqJBhFNosOue JhLKjHQoUcolHOzMcCVtjYcmXdljYVMfGGw9kFjt8sAOr+jh/FxQkDG7TWgOnzmjFU3grMxWrmRE Qe3XYVbXQp2ZW92IZkqdw61QGXw4rxoMFtaEBgRB2wJWXoXzuWYNBxPMSKDtbvfe3C3GCxfpIhnh gGQ+0nrP+6hunJTHirkJgNip8JE+5J1itRK3lib7BtzO4qQyu8YCdrn33sRLeQTPvKTz9kQ6sqSc nCxBR22v1VxuesjHadsbw5kWHuMUvC51z4dZCBdDvhI27E9NZpPlM2+2csXcJKjDNYW1+5zCTh1I hVRbWEY2NMxUFgIs0Zys/MtNMOtFKWAj/TWkWFmDYPjXpAA7uq4l4zHxVdnZpRFtO/ualVI+UUQM ouAIjdhE7GNwvw5V0CegEq4mTEXQL3CPpq1tptzinCVd+fbK4Ow4ZmmEs3KrUzTPZAs3eVzIYN5K 4oFulbIb5c6vikn5C1KlHMb/M1X0fgI3BSuB9oAP17gCI52vbY8LFXGoQmlE/b6AxsHUDogWuIuF aQgquEw2/wU51P9tzlkaJq3hwKf2aYgEhf1IRYKQPShLJvpOIVbP9i5LkmWETESVxJWpFXtEDgkb 6hq4qvd2D0UQ6qaaZGXA4E7Gn/ueZdAo1E1OOd+cGlLsvTYH/unOxyYzKOXWYdPQ5PYvRKzYVe16 szzfe8uK6IlZm9XIswKYlbaCVpb2rynCObdaW7HmNF5u5sKBF+c1hsGiIUrhvgfpP7D/UeEz+2VC b6hDvg+1FcGHBk0Mwgai+pJtPJAukHZwBI2THbTBpElZ02atk7ZavllfcKdb8D1hbC3ZWfx9TmMX zZnLzsnFizR2ZmHH1nZsoanBsydTFIbG+UHGOMZ80ip/deKje+DoLbjfnzAlTTDBNyWBofUcmDyA 5LcczdKNvwAAAP//AwBQSwMEFAAGAAgAAAAhAA3RkJ+2AAAAGwEAACcAAAB0aGVtZS90aGVtZS9f cmVscy90aGVtZU1hbmFnZXIueG1sLnJlbHOEj00KwjAUhPeCdwhvb9O6EJEm3YjQrdQDhOQ1DTY/ JFHs7Q2uLAguh2G+mWm7l53JE2My3jFoqhoIOumVcZrBbbjsjkBSFk6J2TtksGCCjm837RVnkUso TSYkUiguMZhyDidKk5zQilT5gK44o49W5CKjpkHIu9BI93V9oPGbAXzFJL1iEHvVABmWUJr/s/04 GolnLx8WXf5RQXPZhQUoosbM4CObqkwEylu6usTfAAAA//8DAFBLAQItABQABgAIAAAAIQCCirwT +gAAABwCAAATAAAAAAAAAAAAAAAAAAAAAABbQ29udGVudF9UeXBlc10ueG1sUEsBAi0AFAAGAAgA AAAhAKXWp+fAAAAANgEAAAsAAAAAAAAAAAAAAAAAKwEAAF9yZWxzLy5yZWxzUEsBAi0AFAAGAAgA AAAhAGt5lhaDAAAAigAAABwAAAAAAAAAAAAAAAAAFAIAAHRoZW1lL3RoZW1lL3RoZW1lTWFuYWdl ci54bWxQSwECLQAUAAYACAAAACEAlrWt4pYGAABQGwAAFgAAAAAAAAAAAAAAAADRAgAAdGhlbWUv dGhlbWUvdGhlbWUxLnhtbFBLAQItABQABgAIAAAAIQAN0ZCftgAAABsBAAAnAAAAAAAAAAAAAAAA AJsJAAB0aGVtZS90aGVtZS9fcmVscy90aGVtZU1hbmFnZXIueG1sLnJlbHNQSwUGAAAAAAUABQBd AQAAlgoAAAAAPD94bWwgdmVyc2lvbj0iMS4wIiBlbmNvZGluZz0iVVRGLTgiIHN0YW5kYWxvbmU9 InllcyI/Pg0KPGE6Y2xyTWFwIHhtbG5zOmE9Imh0dHA6Ly9zY2hlbWFzLm9wZW54bWxmb3JtYXRz Lm9yZy9kcmF3aW5nbWwvMjAwNi9tYWluIiBiZzE9Imx0MSIgdHgxPSJkazEiIGJnMj0ibHQyIiB0 eDI9ImRrMiIgYWNjZW50MT0iYWNjZW50MSIgYWNjZW50Mj0iYWNjZW50MiIgYWNjZW50Mz0iYWNj ZW50MyIgYWNjZW50ND0iYWNjZW50NCIgYWNjZW50NT0iYWNjZW50NSIgYWNjZW50Nj0iYWNjZW50 NiIgaGxpbms9ImhsaW5rIiBmb2xIbGluaz0iZm9sSGxpbmsiLz4AAAAAkUYAAAoAAIIAAAAA//// /wAIAAAsDQAAzBIAACIZAAD3IwAA5SUAAMcmAABFKAAAvCoAAPQrAADvLAAAci8AAK4yAABROAAA Fj8AAGtAAACSRQAAckYAALpHAADTSAAAgU4AAJFPAAAoAAAAKgAAACsAAAAtAAAALgAAAC8AAAAw AAAAMQAAADMAAAA0AAAANQAAADYAAAA3AAAAOAAAADkAAAA6AAAAOwAAADwAAAA9AAAAPwAAAEAA AAAACAAAnhgAAGQpAADsRwAAkU8AACkAAAAsAAAAMgAAAD4AAADMCgAA5goAAOgKAACeEAAAuBAA ALoQAADfGwAA8xsAAPUbAAAuHAAAQhwAAEQcAABxHAAAhRwAAIccAACvHQAAwx0AAMUdAADRHQAA 5R0AAOcdAACGHgAAmh4AAJweAACyHgAAxh4AAMgeAAAiHwAANh8AADgfAABuHwAAgh8AAIQfAAAo IAAAPCAAAD4gAAB+IAAAkiAAAJQgAADUIAAA6CAAAOogAABkIQAAeCEAAHohAACAIQAAlCEAAJYh AAD3IgAACyMAAA0jAAATIwAAJyMAACkjAADgIwAA9CMAAPYjAAA7JAAATyQAAFEkAADbJAAA7yQA APEkAAD3JAAACyUAAA0lAADEJQAA2CUAANolAACUKQAAqCkAAKopAABBKgAAVSoAAFcqAACaKgAA rioAALAqAADRKgAA5SoAAOcqAADDNQAA1zUAANk1AAABNgAAFTYAABc2AABHNgAAWzYAAF02AAB5 NgAAjTYAAI82AADMNgAA4DYAAOI2AABXNwAAazcAAG03AAChNwAAtTcAALc3AADVPAAA6TwAAOs8 AADxPAAABT0AAAc9AAByPQAAhj0AAIg9AACOPQAAoj0AAKQ9AADuPgAAAj8AAAQ/AAC/PwAA0z8A ANU/AACRRgAAEzqU/5WMEzqU/5WMEzqU/5WEEzqU/5WEEzqU/5WEEzqU/5WEEzqU/5WEEzqU/5WA EzqU/5WAEzqU/5WEEzqU/5WEEzqU/5WAEzqU/5WEEzqU/5WEEzqU/5WEEzqU/5WEEzqU/5WAEzqU /5WAEzqU/5WAEzqU/5WEEzqU/5WAEzqU/5WAEzqU/5WAEzqU/5WEEzqU/5WEEzqU/5WEEzqU/5WA EzqU/5WEEzqU/5WEEzqU/5WEEzqU/5WEEzqU/5WEEzqU/5WEEzqU/5WEEzqU/5WEEzqU/5WAEzqU /5WAEzqU/5WAEzqU/5WAEzqU/5WEDwAA8DgAAAAAAAbwGAAAAAIIAAACAAAAAQAAAAEAAAABAAAA AgAAAEAAHvEQAAAA//8AAAAA/wCAgIAA9wAAEAAPAALwkgAAABAACPAIAAAAAQAAAAEEAAAPAAPw MAAAAA8ABPAoAAAAAQAJ8BAAAAAAAAAAAAAAAAAAAAAAAAAAAgAK8AgAAAAABAAABQAAAA8ABPBC AAAAEgAK8AgAAAABBAAAAA4AAFMAC/AeAAAAvwEAABAAywEAAAAA/wEAAAgABAMJAAAAPwMBAAEA AAAR8AQAAAABAAAAAAAAAA8jAAASIwAA8zIAAPUyAAAMMwAADjMAAGA/AABiPwAAcz8AAHU/AADm QAAA6UAAAJNGAAAHAB0ABwAcAAcAHAAHABwABwAcAAcAHAAHAAAAAACJAAAAiQAAAKEAAADqAAAA qwYAAKsGAAC7BgAAvAYAAL8GAADABgAA0gYAAP4GAABsCQAAbQkAAG4JAABuCQAAlQkAAJYJAACG HgAAnR4AALIeAADJHgAAIR8AACEfAAAiHwAAIh8AACggAAA/IAAAmCEAAFgiAAD3IgAADiMAABMj AAAqIwAA4CMAAPcjAADbJAAA8iQAAPckAAAOJQAAxCUAANslAADRKgAA6CoAAA40AAC5NwAA8TwA AAg9AAByPQAAiT0AAI49AAClPQAAoz4AAKQ+AACoPgAAqT4AAOw+AAAFPwAACD8AAAk/AAAKPwAA Cz8AAA0/AAAOPwAAWz8AAKU/AACmPwAA2D8AAOlAAADpQAAArEMAAK1DAACuQwAAuUMAALpDAACQ RgAAk0YAAAMABAADAAQAAwAEAAMABAADAAQAAwAEAAMABAADAAQAAwAEAAMABAADAAQAAwAEAAMA BAADAAQAAwAEAAMABAADAAQAAwAEAAMABAADAAQAAwAEAAMABAADAAQAAwAEAAMABAADAAQAAwAE AAMABAADAAQAAwAEAAMABAADAAQAAwAEAAMABAADAAQAAwAEAAMABAADAAQABwADAChStxRmE3KZ EgD/D/8P/w//D/8P/w//D/8PEAAAJ9VfXgnEoxEA/w//D/8P/w//D/8P/w//DxAACUvoZSqvzjD/ D/8P/w//D/8P/w//D/8P/w8QAAEAAAAWAAQAAAAAAAAAAAAAAAAAAAAAABUYAAAPhAAAEYQw/RXG BQABAAAGXoQAAGCEMP01CAE2CABDShgAT0oAAFFKAABvKAAFAFsAMAAwAAAAXQABAAAABIABAAAA AAAAAAAAAAAAAAAAAAAAGAAAD4SgBRGEmP4VxgUAAaAFBl6EoAVghJj+AgABAC4AAQAAAAKCAQAA AAAAAAAAAAAAAAAAAAAAABgAAA+EcAgRhEz/FcYFAAFwCAZehHAIYIRM/wIAAgAuAAEAAAAAgAEA AAAAAAAAAAAAAAAAAAAAAAAYAAAPhEALEYSY/hXGBQABQAsGXoRAC2CEmP4CAAMALgABAAAABIAB AAAAAAAAAAAAAAAAAAAAAAAAGAAAD4QQDhGEmP4VxgUAARAOBl6EEA5ghJj+AgAEAC4AAQAAAAKC AQAAAAAAAAAAAAAAAAAAAAAAABgAAA+E4BARhEz/FcYFAAHgEAZehOAQYIRM/wIABQAuAAEAAAAA gAEAAAAAAAAAAAAAAAAAAAAAAAAYAAAPhLATEYSY/hXGBQABsBMGXoSwE2CEmP4CAAYALgABAAAA BIABAAAAAAAAAAAAAAAAAAAAAAAAGAAAD4SAFhGEmP4VxgUAAYAWBl6EgBZghJj+AgAHAC4AAQAA AAKCAQAAAAAAAAAAAAAAAAAAAAAAABgAAA+EUBkRhEz/FcYFAAFQGQZehFAZYIRM/wIACAAuAAEA AAAWAAQAAAAAAAAAAAAAAAAAAAAAABMYAAAPhNACEYQw/RXGBQABoAUGXoTQAmCEMP01CAE2CABv KACHaAAAAACISAAABQBbADAAMAAAAF0AAQAAAAQAAQAAAAAAAAAAAAAAAAAAAAAAChgAAA+EoAUR hJj+FcYFAAGgBQZehKAFYISY/odoAAAAAIhIAAACAAEALgABAAAAAoIBAAAAAAAAAAAAAAAAAAAA AAAKGAAAD4RwCBGETP8VxgUAAXAIBl6EcAhghEz/h2gAAAAAiEgAAAIAAgAuAAEAAAAAgAEAAAAA AAAAAAAAAAAAAAAAAAoYAAAPhEALEYSY/hXGBQABQAsGXoRAC2CEmP6HaAAAAACISAAAAgADAC4A AQAAAASAAQAAAAAAAAAAAAAAAAAAAAAAChgAAA+EEA4RhJj+FcYFAAEQDgZehBAOYISY/odoAAAA AIhIAAACAAQALgABAAAAAoIBAAAAAAAAAAAAAAAAAAAAAAAKGAAAD4TgEBGETP8VxgUAAeAQBl6E 4BBghEz/h2gAAAAAiEgAAAIABQAuAAEAAAAAgAEAAAAAAAAAAAAAAAAAAAAAAAoYAAAPhLATEYSY /hXGBQABsBMGXoSwE2CEmP6HaAAAAACISAAAAgAGAC4AAQAAAASAAQAAAAAAAAAAAAAAAAAAAAAA ChgAAA+EgBYRhJj+FcYFAAGAFgZehIAWYISY/odoAAAAAIhIAAACAAcALgABAAAAAoIBAAAAAAAA AAAAAAAAAAAAAAAKGAAAD4RQGRGETP8VxgUAAVAZBl6EUBlghEz/h2gAAAAAiEgAAAIACAAuAAEA AAAAAAEAAAAAAAAAAAAAAAAAAAAAAAMQAAAPhNACEYSY/l6E0AJghJj+bygAAgAAAC4AAQAAAASA AQAAAAAAAAAAAAAAAAAAAAAAChAAAA+EoAURhJj+XoSgBWCEmP6HaAAAAACISAAAAgABAC4AAQAA AAKCAQAAAAAAAAAAAAAAAAAAAAAAChAAAA+EcAgRhEz/XoRwCGCETP+HaAAAAACISAAAAgACAC4A AQAAAACAAQAAAAAAAAAAAAAAAAAAAAAAChAAAA+EQAsRhJj+XoRAC2CEmP6HaAAAAACISAAAAgAD AC4AAQAAAASAAQAAAAAAAAAAAAAAAAAAAAAAChAAAA+EEA4RhJj+XoQQDmCEmP6HaAAAAACISAAA AgAEAC4AAQAAAAKCAQAAAAAAAAAAAAAAAAAAAAAAChAAAA+E4BARhEz/XoTgEGCETP+HaAAAAACI SAAAAgAFAC4AAQAAAACAAQAAAAAAAAAAAAAAAAAAAAAAChAAAA+EsBMRhJj+XoSwE2CEmP6HaAAA AACISAAAAgAGAC4AAQAAAASAAQAAAAAAAAAAAAAAAAAAAAAAChAAAA+EgBYRhJj+XoSAFmCEmP6H aAAAAACISAAAAgAHAC4AAQAAAAKCAQAAAAAAAAAAAAAAAAAAAAAAChAAAA+EUBkRhEz/XoRQGWCE TP+HaAAAAACISAAAAgAIAC4AAwAAAAlL6GUAAAAAAAAAAAAAAAAAJ9VfAAAAAAAAAAAAAAAAKFK3 FAAAAAAAAAAAAAAAAP//////////////////AwAAAAAAAAAAAP//AwAAABIA//////////////// ////////////////////////////////EgAOHBpzGQAJBBsACQQPAAkEGQAJBBsACQQPAAkEGQAJ BBsACQQSAA8ACQQZAAkEGwAJBA8ACQQZAAkEGwAJBA8ACQQZAAkEGwAJBCAAAAAEAAAACAAAAOUA AAAAAAAAHwAAAL1fBgC9ShoAkUEeAFdlJwD2ICgAVCwtANYxLQCiJy4AOgM2ABlsPwAqFUwAnXNR ACgVVQDgbFYALG1oANEPbwCODHAASDd2AJRmegBQdH0AdDp/ALRAhwAAQYgAsAyKAHJ/jgCKZKAA bwynAKljwQA1RsUA10TYAFlF3QC0WPIAAAAAAJFGAACTRgAAAAAAAAEAAAD/QAIoAAAAAAAAAAAy AAAAMwAAkUYAAFAAABAAQAAAUAAAOgAAAABQAAB4AEAAAP//AQAAAAcAVQBuAGsAbgBvAHcAbgD/ /wEACAAAAAAAAAAAAAAA//8BAAAAAAD//wAAAgD//wAAAAD//wAAAgD//wAAAAAHAAAARx6QAQAA AgIGAwUEBQIDBIcqACAAAACACAAAAAAAAAD/AQAAAAAAAFQAaQBtAGUAcwAgAE4AZQB3ACAAUgBv AG0AYQBuAAAANR6QAQIABQUBAgEHBgIFBwAAAAAAAAAQAAAAAAAAAAAAAACAAAAAAFMAeQBtAGIA bwBsAAAAMy6QAQAAAgsGBAICAgICBIcqACAAAACACAAAAAAAAAD/AQAAAAAAAEEAcgBpAGEAbAAA AD89kAEAAAIHAwkCAgUCBASHKgAgAAAAgAgAAAAAAAAA/wEAAAAAAABDAG8AdQByAGkAZQByACAA TgBlAHcAAAA3LpABAAACDwUCAgIEAwIE7wIAoHsgAEAAAAAAAAAAAJ8AAAAAAAAAQwBhAGwAaQBi AHIAaQAAADcekAEAAAIEBQMFBAYDAgTvAgCgSwAAQAAAAAAAAAAAnwAAAAAAAABDAGEAbQBiAHIA aQBhAAAAQR6QAQAAAgQFAwUEBgMCBO8CAKDrIABCAAAAAAAAAACfAAAAAAAAAEMAYQBtAGIAcgBp AGEAIABNAGEAdABoAAAAIgAEAHEIiBgA8NACAABoAQAAAAAqtOqmY7XqpgAAAAALAPIAAACICgAA CTwAAAcAJAAAAAQAA5CAAAAAiAoAAAk8AAAHACQAAACAAAAAAAAAACEDAPAQAAAAAQAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAKAFoAW0ALQAgYESMAAAAAAAAAAAAAAAAAAAbUYAAG1GAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACAAAA AAAAAAAACDKDcQDwEAAIAPz9AQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAhIUAAAAAAJ8P8PAQgk UAAA5AQAAP///3////9/////f////3////9/////f////3+9XwYAAAQAADIAAAAAAAAAAAAAAAAA AAAAAAAAAAAhBAAAAAAAAAAAAAAAAAAAAAAAABAcAAAGAAAAAAAAAAAAeAAAAHgAAAAAAAAAAAAA AKAFAAD//xIAAAAAAAAAAAAAAAAAAAAYAEoAdQBpAG4ALQBIAHcAZQB5ACAAKABSAGEAeQBtAG8A bgBkACkAIABDAGgAZQBuABgASgB1AGkAbgAtAEgAdwBlAHkAIAAoAFIAYQB5AG0AbwBuAGQAKQAg AEMAaABlAG4AAAAAAAAAAAAAAAAAAAAAAAAAAAAYAAAABgAAAAMAAAAAAAwAAQAMAAIADAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAP7/AwoAAP////8G CQIAAAAAAMAAAAAAAABGJwAAAE1pY3Jvc29mdCBPZmZpY2UgV29yZCA5Ny0yMDAzIERvY3VtZW50 AAoAAABNU1dvcmREb2MAEAAAAFdvcmQuRG9jdW1lbnQuOAD0ObJxAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAFIAbwBvAHQAIABFAG4A dAByAHkAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAWAAUB//// ///////QAAAABgkCAAAAAADAAAAAAAAARgAAAAAAAAAAAAAAAKDDHO1wcssBTgAAAMBGAAAAAAAA RABhAHQAYQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAoAAgH///////////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AABCAAAAmZQAAAAAAABXAG8AcgBkAEQAbwBjAHUAbQBlAG4AdAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAGgACAc8AAAD//////////wAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAA0ggAAAAAAAE8AYgBqAGUAYwB0AFAAbwBvAGwAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAWAAEB0gAAAP////8EAAAAAAAAAAAA AAAAAAAAAAAAAAAAAADQ3Li8a3LLAcDqxLxrcssBAAAAAAAAAAAAAAAAAQIAAAICAAADAgAABAIA AAUCAAAGAgAABwIAAAgCAAAJAgAACgIAAAsCAAAMAgAA/v////////9MAAAA/f////3////9//// /v////7////+////FgIAABMCAAD///////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// ////////////////////////////////////////////////////////////hAEAAIMBAACFAQAA hwEAAIgBAAASAgAAiQEAAIoBAACMAQAAiwEAAI0BAACPAQAAjgEAAJABAACRAQAAkgEAAJQBAACT AQAAlQEAAJYBAACXAQAAmAEAAJkBAACbAQAAmgEAAJwBAACdAQAAngEAAKABAACfAQAAoQEAANsB AADaAQAAowEAAKQBAAClAQAApgEAAKcBAACoAQAAqQEAAKoBAACrAQAArAEAAK0BAACuAQAArwEA ALABAACxAQAAsgEAALMBAAC0AQAAtQEAALYBAAC3AQAAuAEAALkBAAC6AQAAuwEAALwBAAC9AQAA vgEAAL8BAADAAQAAwQEAAMIBAADDAQAAxAEAAMUBAADGAQAAxwEAAMgBAADJAQAAygEAAMsBAADM AQAAzQEAAM4BAADPAQAA0AEAANEBAADSAQAA0wEAANQBAADVAQAA1gEAANcBAADYAQAA2QEAAOoB AADcAQAA3QEAAN4BAADgAQAA3wEAAOEBAADjAQAA4gEAAOQBAADmAQAA5QEAAOcBAADpAQAAFAIA AP////8VAgAA6wEAAOwBAADtAQAA7gEAAO8BAADwAQAA8QEAAPIBAADzAQAA9AEAAP7///////// ///////////5AQAA+gEAAPsBAAD8AQAA/QEAAP4BAAD/AQAAAAIAAAEAAAACAAAAAwAAAAQAAAAF AAAABgAAAAcAAAAIAAAACQAAAAoAAAALAAAADAAAAA0AAAAOAAAADwAAABAAAAARAAAAEgAAABMA AAAUAAAAFQAAABYAAAAXAAAAGAAAABkAAAAaAAAAGwAAABwAAAAdAAAAHgAAAB8AAAAgAAAAIQAA ACIAAAAjAAAAJAAAACUAAAAmAAAAJwAAACgAAAApAAAAKgAAACsAAAAsAAAALQAAAC4AAAAvAAAA MAAAADEAAAAyAAAAMwAAADQAAAA1AAAANgAAADcAAAA4AAAAOQAAADoAAAA7AAAAPAAAAD0AAAA+ AAAAPwAAAEAAAABBAAAA/v///0MAAABEAAAARQAAAEYAAABHAAAASAAAAEkAAACiAQAA//////// //9qAAAAhgEAALEAAABQAAAAUQAAAFIAAABTAAAAVAAAAFUAAABWAAAAVwAAAFgAAABZAAAAWgAA AFsAAABcAAAAXQAAAF4AAABfAAAAYAAAAGEAAABiAAAAYwAAAGQAAABlAAAAZgAAAGcAAABoAAAA aQAAAP7///+wAAAA/f///20AAABuAAAAbwAAAHAAAABxAAAAcgAAAHMAAAB0AAAAdQAAAHYAAAB3 AAAAeAAAAHkAAAB6AAAAewAAAHwAAAB9AAAAfgAAAH8AAACAAAAAAQEAAAIBAAD+////BAEAAAUB AAAGAQAABwEAAAgBAAAJAQAACgEAAP7///8MAQAADQEAAA4BAAAPAQAAEgEAABEBAAD+////EwEA ABQBAAAVAQAAFgEAABcBAAAYAQAAGQEAABoBAAD+//////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////8gAHQAbwAgAGkAbQBwAHIAbwB2 AGUAIAB0AGgAZQAgAG0AbwBkAGUAIAAxACAAQwBFAEwAVAAgAGMAbwBkAGUAYwAAAB8AAAATAAAA cgBjAGgAZQBuAEAAYgByAG8AYQBkAGMAbwBtAC4AYwBvAG0AAAAAAB8AAAAZAAAAUgBhAHkAbQBv AG4AZAAgACgASgB1AGkAbgAtAEgAdwBlAHkAKQAgAEMAaABlAG4AAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAUAUwB1AG0AbQBhAHIAeQBJAG4A ZgBvAHIAbQBhAHQAaQBvAG4AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAoAAIBAgAAANEAAAD/ ////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAwEAAMQBAAAAAAAABQBEAG8A YwB1AG0AZQBuAHQAUwB1AG0AbQBhAHIAeQBJAG4AZgBvAHIAbQBhAHQAaQBvAG4AAAAAAAAAAAAA ADgAAgH///////////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAALAQAA cAMAAAAAAAABAEMAbwBtAHAATwBiAGoAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAEgACAP///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAABABAAB5AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA////////////////AAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAeSAoUmF5bW9uZCkgQ2hlbgAAAAAe AAAABAAAADExAAAeAAAAGAAAAE1pY3Jvc29mdCBPZmZpY2UgV29yZAAAAEAAAAAALJjOIQAAAEAA AAAAVI+4QnLLAUAAAAAAcg6na3LLAQMAAAAHAAAAAwAAAIgKAAADAAAACTwAAAMAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA /v8AAAUBAgAAAAAAAAAAAAAAAAAAAAAAAgAAAALVzdWcLhsQk5cIACss+a5EAAAABdXN1ZwuGxCT lwgAKyz5rkABAAD8AAAADAAAAAEAAABoAAAADwAAAHAAAAAFAAAAkAAAAAYAAACYAAAAEQAAAKAA AAAXAAAAqAAAAAsAAACwAAAAEAAAALgAAAATAAAAwAAAABYAAADIAAAADQAAANAAAAAMAAAA3QAA AAIAAADkBAAAHgAAABgAAABCcm9hZGNvbSBDb3Jwb3JhdGlvbgAAAAADAAAAgAAAAAMAAAAkAAAA AwAAAG1GAAADAAAAAAAMAAsAAAAAAAAACwAAAAAAAAALAAAAAAAAAAsAAAAAAAAAHhAAAAEAAAAB AAAAAAwQAAACAAAAHgAAAAYAAABUaXRsZQADAAAAAQAAAAABAP7/AwoAAP////8GCQIAAAAAAMAA AAAAAABGJwAAAE1pY3Jvc29mdCBPZmZpY2UgV29yZCA5Ny0yMDAzIERvY3VtZW50AAoAAABNU1dv cmREb2MAEAAAAFdvcmQuRG9jdW1lbnQuOAD0ObJxAAAAAAAAAAAAAAAAAAAAAAAAADACAAAIAAAA AAAAAEgAAAABAAAAJAEAAAAAAIAsAQAAAgAAADQBAAADAAAAPAEAAAQAAABIAQAABQAAAMQBAAAG AAAA9AEAAAUAAAACAAAAFAAAAF8AQQBkAEgAbwBjAFIAZQB2AGkAZQB3AEMAeQBjAGwAZQBJAEQA AAADAAAAEAAAAF8ATgBlAHcAUgBlAHYAaQBlAHcAQwB5AGMAbABlAAAABAAAAA4AAABfAEUAbQBh AGkAbABTAHUAYgBqAGUAYwB0AAAABQAAAA0AAABfAEEAdQB0AGgAbwByAEUAbQBhAGkAbAAAAAAA BgAAABgAAABfAEEAdQB0AGgAbwByAEUAbQBhAGkAbABEAGkAcwBwAGwAYQB5AE4AYQBtAGUAAAAC AAAAsAQAABMAAAAJBAAAAwAAALe6TiEfAAAAAQAAAAAAAAAfAAAAOgAAAEEAIABwAGkAdABjAGgA IABmAGkAbAB0AGUAcgBpAG4AZwAgAG0AZQB0AGgAbwBkAA== --_002_CB68DF4CFBEF4942881AD37AE1A7E8C75F264FFF8CIRVEXCHCCR01c_-- From jean-marc.valin@usherbrooke.ca Sat Oct 23 05:50:13 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id EB3FF3A68A9 for ; Sat, 23 Oct 2010 05:50:13 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.248 X-Spam-Level: X-Spam-Status: No, score=-2.248 tagged_above=-999 required=5 tests=[AWL=0.351, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id QhV+YuklcIPQ for ; Sat, 23 Oct 2010 05:50:12 -0700 (PDT) Received: from relais.videotron.ca (relais.videotron.ca [24.201.245.36]) by core3.amsl.com (Postfix) with ESMTP id D02383A689A for ; Sat, 23 Oct 2010 05:50:12 -0700 (PDT) MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; charset=ISO-8859-1 Received: from [192.168.1.14] ([70.81.109.112]) by vl-mo-mrz23.ip.videotron.ca (Sun Java(tm) System Messaging Server 6.3-8.01 (built Dec 16 2008; 32bit)) with ESMTP id <0LAQ000OYUE9OX60@vl-mo-mrz23.ip.videotron.ca> for codec@ietf.org; Sat, 23 Oct 2010 08:51:46 -0400 (EDT) Message-id: <4CC2DA7E.1060705@usherbrooke.ca> Date: Sat, 23 Oct 2010 08:52:14 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.14) Gecko/20101006 Thunderbird/3.0.9 To: "Raymond (Juin-Hwey) Chen" References: In-reply-to: X-Enigmail-Version: 1.0.1 Cc: "'codec@ietf.org'" Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 23 Oct 2010 12:50:14 -0000 Hi Raymond, Thank you very much for your contribution and the highly detailed description you just sent. I believe I have enough information to start implementing it. At this point, I only have one minor question. When estimating the pitch period/gain, do you recommend working on the input signal directly or should the input be "whitened" using a filter like A(z/gamma) (where A(z) is the LPC filter)? Thanks again, Jean-Marc On 10-10-23 01:13 AM, Raymond (Juin-Hwey) Chen wrote: > All, > > In the last few weeks, I have been working on a pitch prefiltering/postfiltering method to try to > improve the output audio quality of the IETF mode 1 CELT codec for certain audio signals that CELT > low-delay modes give more audible coding distortion than usual (typically nearly periodic solo > music instrument signals). I am happy to report that this method has achieved fairly significant > audio quality improvement for many of these kinds of audio signals that I tried. This is achieved > without increasing the coding delay and with only a slight increase in the codec complexity and > encoding bit-rate (a potential increase of around 2 kb/s or so; less if a range coder is used). > > I have been in discussion with Jean-Marc about this and have sent him CELT-coded audio signals at > different bit-rates obtained with or without this method. After listening, he agreed that the > method made a significant improvement when compared with the current CELT implementation, and he > would like to try to incorporate this method into his CELT C code to evaluate and verify the > performance gain and possibly further develop the method. > > Jean-Marc asked me to write a detailed description of the method in a document and send the > document to the IETF codec WG email reflector so he can read it, understand it, and then implement > it himself in his CELT C code. I am attaching such a technical description of the method. > > Note that although this document is 7 pages long, the pitch prefiltering/postfiltering method is > actually quite simple. The document is long only because (1) I tried to describe the technical > details as much as I can so Jean-Marc can implement it exactly as I had it without any ambiguity, > and (2) I also described/discussed a few alternative approaches for some of the algorithm > components in case we need to go those routes. > > Note also that this is "fresh out of the oven" as I only had time to finish a very rough first cut > for a proof-of-concept simulation and did not have time yet to refine or tune it. Thus, I am sure > there are still many rough edges that need to be smoothed out and many areas that can be improved > upon. However, I am very much encouraged that even such a rough first cut showed a lot of promise > as it made a very large audio quality improvement at several different bit-rates for a trumpet > test signal, which is a signal that CELT low-delay modes have problems with. Therefore, after > this pitch prefiltering/postfiltering method is integrated into CELT and further refined/tuned, I > think it has the potential to make the IETF mode 1 CELT codec more robust to different audio > signals. That is, it can potentially improve CELT performance for many signals that are currently > difficult for CELT to code well, and thus ensure a more uniform audio quality level across a wider > variety of different audio signals. If Jean-Marc's integration, testing, and further development > of this pitch filtering method confirms this potential, then it is proposed that this pitch > prefiltering/postfiltering method be incorporated into Mode 1 of the IETF Opus codec to make the > codec more robust to different input signal types. > > Cheers, > > Raymond > > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From rchen@broadcom.com Sat Oct 23 12:48:44 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id CB71C3A68D0 for ; Sat, 23 Oct 2010 12:48:44 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.413 X-Spam-Level: X-Spam-Status: No, score=-2.413 tagged_above=-999 required=5 tests=[AWL=0.186, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RQOFeOHk9lqK for ; Sat, 23 Oct 2010 12:48:43 -0700 (PDT) Received: from mms1.broadcom.com (mms1.broadcom.com [216.31.210.17]) by core3.amsl.com (Postfix) with ESMTP id 4F9BF3A67C2 for ; Sat, 23 Oct 2010 12:48:43 -0700 (PDT) Received: from [10.9.200.131] by mms1.broadcom.com with ESMTP (Broadcom SMTP Relay (Email Firewall v6.3.2)); Sat, 23 Oct 2010 12:50:08 -0700 X-Server-Uuid: 02CED230-5797-4B57-9875-D5D2FEE4708A Received: from IRVEXCHCCR01.corp.ad.broadcom.com ([10.252.49.30]) by IRVEXCHHUB01.corp.ad.broadcom.com ([10.9.200.131]) with mapi; Sat, 23 Oct 2010 12:50:08 -0700 From: "Raymond (Juin-Hwey) Chen" To: "Jean-Marc Valin" Importance: low X-Priority: 5 Date: Sat, 23 Oct 2010 12:50:05 -0700 Thread-Topic: [codec] A pitch filtering method to improve the mode 1 CELT codec Thread-Index: ActysRfobs1S495JQYqKBIRQIh4scwALPo5Q Message-ID: References: <4CC2DA7E.1060705@usherbrooke.ca> In-Reply-To: <4CC2DA7E.1060705@usherbrooke.ca> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US MIME-Version: 1.0 X-WSS-ID: 60DDE3FA47833145508-01-01 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Cc: "'codec@ietf.org'" Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 23 Oct 2010 19:48:45 -0000 Hi Jean-Marc, You are very welcome. Glad to contribute and glad to know that the document= is good=20 enough for you to start implementing the techniques. Regarding your question, I think it really depends on what your main emphas= is is. =20 If you just want the best possible pitch period contour and don't mind havi= ng a=20 little higher complexity, then partially whitening the input signal with th= e=20 A(z/gamma) or A(z)/A(z/gamma) or even A(z/alpha)/A(z/beta) is a common appr= oach used=20 in the pitch estimators of many speech codecs (as I am sure you are well aw= are of),=20 and it does help to reduce the chance of detecting the frequency of a stron= g first=20 formant of the input speech signal as the pitch frequency.=20 If, on the other hand, your main emphasis is to get a codec complexity as l= ow as=20 possible, I think it is possible to get a reasonably good pitch period cont= our by=20 using the input signal directly without partial whitening filtering, thus s= aving the=20 complexity of the associated LPC analysis and filtering (which can be relat= ively=20 high for the 48 kHz sampling rate). Note that when I said "Any low-complex= ity pitch=20 estimator can be used here..." in the document I sent last night, I used th= e term=20 "low-complexity" because I thought you told me you tried to keep the comple= xity of=20 CELT as low as possible. Since this pitch prefiltering/postfiltering appro= ach is=20 meant to be used with CELT, I thought it is only natural to make the comple= xity of=20 the pitch estimator as low as possible. That's why I used the term "low-com= plexity=20 pitch estimator" in that document. I think for this pitch prefiltering/postfiltering approach to work well, wh= at's most=20 important is that we need to do everything we can to ensure that the pitch = period we=20 extract is the true pitch period rather than an integer multiple or sub-mul= tiple of=20 the pitch period. Unlike speech coding or packet loss concealment applicat= ions=20 where double pitch period or triple pitch period still gives you decent pit= ch=20 prediction or replacement waveform, here double or triple pitch period will= create=20 one or two unwanted frequency response peak(s) in the valley regions betwee= n pitch=20 harmonic peaks and thus will greatly reduce the noise reduction achievable = by the=20 noise spectral shaping effect of the pitch prefilter/postfilter approach. = =20 Hence, ideally we should have some decision logic and smoothing techniques = to get a=20 smooth contour of the true pitch period. I think the pitch estimator of th= e=20 BroadVoice codecs (BV16 and BV32) that Broadcom previously submitted can se= rve as a=20 good starting point for further development, because it was optimized to ha= ve=20 relatively low complexity and it also has decision logic and smoothing buil= t-in to=20 try to get a smooth contour of the true pitch period for voiced regions of = speech. =20 It is optimized for 8 kHz or 16 kHz speech signals, so to use it for 48 kHz= speech=20 and music signals, we need to adjust the decimation factor and re-optimize = it to=20 work better for music as well. We also need to increase the range of the p= itch=20 period, or equivalently, the pitch frequency. Typical pitch frequency rang= e for=20 adult voice is somewhere around 50 to 400 Hz, but the pitch frequencies of = music=20 instruments cover a much wider range and can go up to 1600 Hz or higher. B= V16 and=20 BV32 detectable pitch ranges are around 59 Hz ~ 800 Hz and 60 Hz ~ 1600 Hz,= =20 respectively, because we previously already increased their pitch ranges to= be able=20 to detect the pitch of some of the music instruments. Of course, if there is another royalty-free and unencumbered pitch estimato= r=20 available that has even lower complexity than the BV16/BV32 pitch estimator= and=20 performs better, I am all for it. Raymond -----Original Message----- From: Jean-Marc Valin [mailto:jean-marc.valin@usherbrooke.ca]=20 Sent: Saturday, October 23, 2010 5:52 AM To: Raymond (Juin-Hwey) Chen Cc: 'codec@ietf.org' Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT co= dec Hi Raymond, Thank you very much for your contribution and the highly detailed description you just sent. I believe I have enough information to start implementing it. At this point, I only have one minor question. When estimating the pitch period/gain, do you recommend working on the input signal directly or should the input be "whitened" using a filter like A(z/gamma) (where A(z) is the LPC filter)? Thanks again, Jean-Marc On 10-10-23 01:13 AM, Raymond (Juin-Hwey) Chen wrote: > All, >=20 > In the last few weeks, I have been working on a pitch prefiltering/postfi= ltering method to try to=20 > improve the output audio quality of the IETF mode 1 CELT codec for certai= n audio signals that CELT=20 > low-delay modes give more audible coding distortion than usual (typically= nearly periodic solo=20 > music instrument signals). I am happy to report that this method has ach= ieved fairly significant=20 > audio quality improvement for many of these kinds of audio signals that I= tried. This is achieved=20 > without increasing the coding delay and with only a slight increase in th= e codec complexity and=20 > encoding bit-rate (a potential increase of around 2 kb/s or so; less if a= range coder is used). >=20 > I have been in discussion with Jean-Marc about this and have sent him CEL= T-coded audio signals at=20 > different bit-rates obtained with or without this method. After listenin= g, he agreed that the=20 > method made a significant improvement when compared with the current CELT= implementation, and he=20 > would like to try to incorporate this method into his CELT C code to eval= uate and verify the=20 > performance gain and possibly further develop the method. =20 >=20 > Jean-Marc asked me to write a detailed description of the method in a doc= ument and send the=20 > document to the IETF codec WG email reflector so he can read it, understa= nd it, and then implement=20 > it himself in his CELT C code. I am attaching such a technical descripti= on of the method. >=20 > Note that although this document is 7 pages long, the pitch prefiltering/= postfiltering method is=20 > actually quite simple. The document is long only because (1) I tried to = describe the technical=20 > details as much as I can so Jean-Marc can implement it exactly as I had i= t without any ambiguity,=20 > and (2) I also described/discussed a few alternative approaches for some = of the algorithm=20 > components in case we need to go those routes. >=20 > Note also that this is "fresh out of the oven" as I only had time to fini= sh a very rough first cut=20 > for a proof-of-concept simulation and did not have time yet to refine or = tune it. Thus, I am sure=20 > there are still many rough edges that need to be smoothed out and many ar= eas that can be improved=20 > upon. However, I am very much encouraged that even such a rough first cu= t showed a lot of promise=20 > as it made a very large audio quality improvement at several different bi= t-rates for a trumpet=20 > test signal, which is a signal that CELT low-delay modes have problems wi= th. Therefore, after=20 > this pitch prefiltering/postfiltering method is integrated into CELT and = further refined/tuned, I=20 > think it has the potential to make the IETF mode 1 CELT codec more robust= to different audio=20 > signals. That is, it can potentially improve CELT performance for many s= ignals that are currently=20 > difficult for CELT to code well, and thus ensure a more uniform audio qua= lity level across a wider=20 > variety of different audio signals. If Jean-Marc's integration, testing,= and further development=20 > of this pitch filtering method confirms this potential, then it is propos= ed that this pitch=20 > prefiltering/postfiltering method be incorporated into Mode 1 of the IETF= Opus codec to make the=20 > codec more robust to different input signal types. >=20 > Cheers, >=20 > Raymond >=20 >=20 >=20 > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From rchen@broadcom.com Sat Oct 23 14:55:33 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 860433A6784 for ; Sat, 23 Oct 2010 14:55:33 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.437 X-Spam-Level: X-Spam-Status: No, score=-2.437 tagged_above=-999 required=5 tests=[AWL=0.163, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1K+JLPQPUGwV for ; Sat, 23 Oct 2010 14:55:31 -0700 (PDT) Received: from mms1.broadcom.com (mms1.broadcom.com [216.31.210.17]) by core3.amsl.com (Postfix) with ESMTP id A040E3A67A4 for ; Sat, 23 Oct 2010 14:55:31 -0700 (PDT) Received: from [10.9.200.133] by mms1.broadcom.com with ESMTP (Broadcom SMTP Relay (Email Firewall v6.3.2)); Sat, 23 Oct 2010 14:56:59 -0700 X-Server-Uuid: 02CED230-5797-4B57-9875-D5D2FEE4708A Received: from IRVEXCHCCR01.corp.ad.broadcom.com ([10.252.49.30]) by IRVEXCHHUB02.corp.ad.broadcom.com ([10.9.200.133]) with mapi; Sat, 23 Oct 2010 14:58:17 -0700 From: "Raymond (Juin-Hwey) Chen" To: "Jean-Marc Valin" Importance: low X-Priority: 5 Date: Sat, 23 Oct 2010 14:56:53 -0700 Thread-Topic: [codec] A pitch filtering method to improve the mode 1 CELT codec Thread-Index: ActysRfobs1S495JQYqKBIRQIh4scwALPo5QAAOQ94A= Message-ID: References: <4CC2DA7E.1060705@usherbrooke.ca> In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-cr-hashedpuzzle: ADyN ApdY CC17 DQ+z FLed FqkP Jqno K0rr Od2E Pla7 QloK R6FS SllG UKll U4NN VcO5; 2; YwBvAGQAZQBjAEAAaQBlAHQAZgAuAG8AcgBnADsAagBlAGEAbgAtAG0AYQByAGMALgB2AGEAbABpAG4AQAB1AHMAaABlAHIAYgByAG8AbwBrAGUALgBjAGEA; Sosha1_v1; 7; {3725CAD5-E1C3-4966-BEA6-8BF2525E143B}; cgBjAGgAZQBuAEAAYgByAG8AYQBkAGMAbwBtAC4AYwBvAG0A; Sat, 23 Oct 2010 21:56:53 GMT; UgBFADoAIABbAGMAbwBkAGUAYwBdACAAQQAgAHAAaQB0AGMAaAAgAGYAaQBsAHQAZQByAGkAbgBnACAAbQBlAHQAaABvAGQAIAB0AG8AIABpAG0AcAByAG8AdgBlACAAdABoAGUAIABtAG8AZABlACAAMQAgAEMARQBMAFQAIABjAG8AZABlAGMA x-cr-puzzleid: {3725CAD5-E1C3-4966-BEA6-8BF2525E143B} acceptlanguage: en-US MIME-Version: 1.0 X-WSS-ID: 60DD85A147833236139-01-01 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Cc: "'codec@ietf.org'" Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 23 Oct 2010 21:55:33 -0000 Hi Jean-Marc, A few more comments about the technical description document I sent last ni= ght: (1) I should have clarified: when I said "frame size" in my document, I was referring to the frame size for pitch estimation and pitch filter tap calcu= lation, not the CELT codec frame size. For ease of implementation, the boundaries = of the pitch frames and codec frames should obviously be aligned. Given that the = CELT-only mode of the IETF codec has possible frame sizes of 2.5, 5, 10, and 20 ms, I= would recommend that we update the pitch period and pitch filter tap(s) at a rate= of once every 5 ms if the CELT frame size is 5, 10, or 20 ms, because updating them= once every 10 or 20 ms is probably too infrequent and probably will decrease the= system's effectiveness in the reduction of coding distortion (although it doesn't hu= rt for you to try updating them every 10 or 20 ms to be the same as the codec fram= e size just to see what happens). On the other hand, if the CELT frame size is 2.= 5 ms in order to achieve an ultra-low delay, then it doesn't make sense for the pit= ch prefilter/postfilter to use a 5 ms frame size since it will just increase t= he system delay. In this case, even though 2.5 ms pitch frame size may be more frequ= ent than necessary and will double the pitch side information bit-rate, it is still = a necessity just to enable CELT to achieve the ultra-low delay. (2) On a second thought, I think to calculate the normalized correlation c = or the optimal tap weight beta for a single-tap pitch predictor, rather than calcu= lating the summation over the current frame (n from 1 to L), it probably makes mor= e sense for the summation length to be the extracted pitch period or the pitch fram= e size, whichever is larger. The motivation is this: suppose the pitch frame size i= s 2.5 ms or 5 ms and the pitch period is 15 ms, then trying to measure the level of periodicity using only 1/6 to 1/3 of the pitch cycle waveform seems dangero= us or at least unreliable. Thus, if the detected pitch period is longer than the fr= ame size, then we should sum over all of the samples in the current frame plus more s= amples in the previous frame(s) until the total number of samples in the summation eq= uals the pitch period. If the pitch period is smaller than the frame size and the p= itch frame size is only 2.5 ms or 5 ms, I recommend just summing over the entire= current frame. (3) If we use multi-tap pitch filters in order to taper off the comb filter= ing effect at higher frequencies, we need to be careful about the stability of = the IIR pitch filter, especially if the IIR filter is in the decoder side (as is th= e case in my example single-tap all-pole pitch postfilter). I suppose you probably a= lready know this, but I would like to mention and discuss it anyway. Even though = the multi-tap IIR pitch postfilter can be an exact inverse filter of a correspo= nding pitch prefilter and supposedly is just "undoing" the effect of the pitch pr= efilter, the quantization errors in the CELT codec and even the effects caused by ch= annel errors can potentially make the output of the multi-tap IIR pitch postfilte= r "go wild" (have very large amplitude or even oscillate) if that filter is unsta= ble or even if it is only close to unstable (i.e. all poles are inside the unit ci= rcle but too close to the unit circle). If the IIR pitch filter is a one-tap filter= , ensuring the stability is trivial. If it is a 2-tap or 3-tap filter, Profe= ssor Peter Kabal and his student Ravi Ramachandran has a nice paper on a stabili= ty test and a stabilization technique that can be used to ensure the stability of s= uch a filter. Their paper is published in the IEEE Transactions on ASSP, July 19= 87. We can use that stabilization technique to stabilize every single candidate 2-= tap or 3- tap filter coefficient set that we will transmit and use, with enough safet= y margin to instability, to ensure that a 2-tap or 3-tap IIR pitch postfilter will n= ot amplify the output audio signal wildly. Alternatively, we can use the second form of the pitch prefilter and postfi= lter that I mentioned in my document so that the pitch postfilter is a 2-tap or 3-tap= FIR filter in the form of H(z) =3D 1 + a1 * z^(-p+1) + a2 * z^-p + a3 * z^(-p-1= ), with a1 or a3 equals to zero in the case of a 2-tap filter. In this case, we will = move the multi-tap IIR filter from the decoder side to the encoder side where it is = applied directly to the input audio signal and thus we do not have to worry about C= ELT quantization errors or channel errors driving a "nearly unstable" filter wi= ld. Of course, we do not want the IIR prefilter there to be too close to instabili= ty, either, otherwise the potential wild swing in its output may still cause pr= oblems for the CELT codec. The main advantage of this alternative arrangement is = that the pitch postfilter is now an FIR filter, which is guaranteed to be stable and= no amount of CELT coding distortion or channel error will ever "drive it crazy= and wild" like they can for a multi-tap IIR pitch postfilter that has not been stabilized or has not maintained a sufficient safety margin to instability.= The main disadvantage of this arrangement is that with an FIR pitch postfilter,= its frequency response has downward spikes located at pitch harmonics, and it w= ill tend to shape the coding noise that way, too. A noise spectral shape with downw= ard spikes is probably not as ideal as the noise spectral shape with upward spi= kes that the all-pole pitch postfilter provides. This is more so when the parameter= bmax is closer to 1 but probably won't be too big a deal if bmax is in the neighbor= hood of 0.4 to 0.6. Raymond -----Original Message----- From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of R= aymond (Juin-Hwey) Chen Sent: Saturday, October 23, 2010 12:50 PM To: Jean-Marc Valin Cc: 'codec@ietf.org' Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT co= dec Importance: Low Hi Jean-Marc, You are very welcome. Glad to contribute and glad to know that the document= is good enough for you to start implementing the techniques. Regarding your question, I think it really depends on what your main emphas= is is. If you just want the best possible pitch period contour and don't mind havi= ng a little higher complexity, then partially whitening the input signal with th= e A(z/gamma) or A(z)/A(z/gamma) or even A(z/alpha)/A(z/beta) is a common appr= oach used in the pitch estimators of many speech codecs (as I am sure you are well aw= are of), and it does help to reduce the chance of detecting the frequency of a stron= g first formant of the input speech signal as the pitch frequency. If, on the other hand, your main emphasis is to get a codec complexity as l= ow as possible, I think it is possible to get a reasonably good pitch period cont= our by using the input signal directly without partial whitening filtering, thus s= aving the complexity of the associated LPC analysis and filtering (which can be relat= ively high for the 48 kHz sampling rate). Note that when I said "Any low-complex= ity pitch estimator can be used here..." in the document I sent last night, I used th= e term "low-complexity" because I thought you told me you tried to keep the comple= xity of CELT as low as possible. Since this pitch prefiltering/postfiltering appro= ach is meant to be used with CELT, I thought it is only natural to make the comple= xity of the pitch estimator as low as possible. That's why I used the term "low-com= plexity pitch estimator" in that document. I think for this pitch prefiltering/postfiltering approach to work well, wh= at's most important is that we need to do everything we can to ensure that the pitch = period we extract is the true pitch period rather than an integer multiple or sub-mul= tiple of the pitch period. Unlike speech coding or packet loss concealment applicat= ions where double pitch period or triple pitch period still gives you decent pit= ch prediction or replacement waveform, here double or triple pitch period will= create one or two unwanted frequency response peak(s) in the valley regions betwee= n pitch harmonic peaks and thus will greatly reduce the noise reduction achievable = by the noise spectral shaping effect of the pitch prefilter/postfilter approach. Hence, ideally we should have some decision logic and smoothing techniques = to get a smooth contour of the true pitch period. I think the pitch estimator of th= e BroadVoice codecs (BV16 and BV32) that Broadcom previously submitted can se= rve as a good starting point for further development, because it was optimized to ha= ve relatively low complexity and it also has decision logic and smoothing buil= t-in to try to get a smooth contour of the true pitch period for voiced regions of = speech. It is optimized for 8 kHz or 16 kHz speech signals, so to use it for 48 kHz= speech and music signals, we need to adjust the decimation factor and re-optimize = it to work better for music as well. We also need to increase the range of the p= itch period, or equivalently, the pitch frequency. Typical pitch frequency rang= e for adult voice is somewhere around 50 to 400 Hz, but the pitch frequencies of = music instruments cover a much wider range and can go up to 1600 Hz or higher. B= V16 and BV32 detectable pitch ranges are around 59 Hz ~ 800 Hz and 60 Hz ~ 1600 Hz, respectively, because we previously already increased their pitch ranges to= be able to detect the pitch of some of the music instruments. Of course, if there is another royalty-free and unencumbered pitch estimato= r available that has even lower complexity than the BV16/BV32 pitch estimator= and performs better, I am all for it. Raymond -----Original Message----- From: Jean-Marc Valin [mailto:jean-marc.valin@usherbrooke.ca] Sent: Saturday, October 23, 2010 5:52 AM To: Raymond (Juin-Hwey) Chen Cc: 'codec@ietf.org' Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT co= dec Hi Raymond, Thank you very much for your contribution and the highly detailed description you just sent. I believe I have enough information to start implementing it. At this point, I only have one minor question. When estimating the pitch period/gain, do you recommend working on the input signal directly or should the input be "whitened" using a filter like A(z/gamma) (where A(z) is the LPC filter)? Thanks again, Jean-Marc On 10-10-23 01:13 AM, Raymond (Juin-Hwey) Chen wrote: > All, > > In the last few weeks, I have been working on a pitch prefiltering/postfi= ltering method to try to > improve the output audio quality of the IETF mode 1 CELT codec for certai= n audio signals that CELT > low-delay modes give more audible coding distortion than usual (typically= nearly periodic solo > music instrument signals). I am happy to report that this method has ach= ieved fairly significant > audio quality improvement for many of these kinds of audio signals that I= tried. This is achieved > without increasing the coding delay and with only a slight increase in th= e codec complexity and > encoding bit-rate (a potential increase of around 2 kb/s or so; less if a= range coder is used). > > I have been in discussion with Jean-Marc about this and have sent him CEL= T-coded audio signals at > different bit-rates obtained with or without this method. After listenin= g, he agreed that the > method made a significant improvement when compared with the current CELT= implementation, and he > would like to try to incorporate this method into his CELT C code to eval= uate and verify the > performance gain and possibly further develop the method. > > Jean-Marc asked me to write a detailed description of the method in a doc= ument and send the > document to the IETF codec WG email reflector so he can read it, understa= nd it, and then implement > it himself in his CELT C code. I am attaching such a technical descripti= on of the method. > > Note that although this document is 7 pages long, the pitch prefiltering/= postfiltering method is > actually quite simple. The document is long only because (1) I tried to = describe the technical > details as much as I can so Jean-Marc can implement it exactly as I had i= t without any ambiguity, > and (2) I also described/discussed a few alternative approaches for some = of the algorithm > components in case we need to go those routes. > > Note also that this is "fresh out of the oven" as I only had time to fini= sh a very rough first cut > for a proof-of-concept simulation and did not have time yet to refine or = tune it. Thus, I am sure > there are still many rough edges that need to be smoothed out and many ar= eas that can be improved > upon. However, I am very much encouraged that even such a rough first cu= t showed a lot of promise > as it made a very large audio quality improvement at several different bi= t-rates for a trumpet > test signal, which is a signal that CELT low-delay modes have problems wi= th. Therefore, after > this pitch prefiltering/postfiltering method is integrated into CELT and = further refined/tuned, I > think it has the potential to make the IETF mode 1 CELT codec more robust= to different audio > signals. That is, it can potentially improve CELT performance for many s= ignals that are currently > difficult for CELT to code well, and thus ensure a more uniform audio qua= lity level across a wider > variety of different audio signals. If Jean-Marc's integration, testing,= and further development > of this pitch filtering method confirms this potential, then it is propos= ed that this pitch > prefiltering/postfiltering method be incorporated into Mode 1 of the IETF= Opus codec to make the > codec more robust to different input signal types. > > Cheers, > > Raymond > > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec From jean-marc.valin@usherbrooke.ca Sat Oct 23 17:33:13 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 971423A6767 for ; Sat, 23 Oct 2010 17:33:12 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.27 X-Spam-Level: X-Spam-Status: No, score=-2.27 tagged_above=-999 required=5 tests=[AWL=0.329, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Mx4vAXe5kGAU for ; Sat, 23 Oct 2010 17:33:10 -0700 (PDT) Received: from relais.videotron.ca (relais.videotron.ca [24.201.245.36]) by core3.amsl.com (Postfix) with ESMTP id 8404E3A65A5 for ; Sat, 23 Oct 2010 17:33:10 -0700 (PDT) MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; charset=ISO-8859-1 Received: from [192.168.1.14] ([70.81.109.112]) by vl-mh-mrz21.ip.videotron.ca (Sun Java(tm) System Messaging Server 6.3-8.01 (built Dec 16 2008; 32bit)) with ESMTP id <0LAR00DDMQXTVBK0@vl-mh-mrz21.ip.videotron.ca> for codec@ietf.org; Sat, 23 Oct 2010 20:34:42 -0400 (EDT) Message-id: <4CC37F42.2040103@usherbrooke.ca> Date: Sat, 23 Oct 2010 20:35:14 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.14) Gecko/20101006 Thunderbird/3.0.9 To: "Raymond (Juin-Hwey) Chen" References: <4CC2DA7E.1060705@usherbrooke.ca> In-reply-to: X-Enigmail-Version: 1.0.1 Cc: "'codec@ietf.org'" Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Oct 2010 00:33:13 -0000 Hi Raymond, On 10-10-23 03:50 PM, Raymond (Juin-Hwey) Chen wrote: > Regarding your question, I think it really depends on what your main emphasis is. > If you just want the best possible pitch period contour and don't mind having a > little higher complexity, then partially whitening the input signal with the > A(z/gamma) or A(z)/A(z/gamma) or even A(z/alpha)/A(z/beta) is a common approach used > in the pitch estimators of many speech codecs (as I am sure you are well aware of), > and it does help to reduce the chance of detecting the frequency of a strong first > formant of the input speech signal as the pitch frequency. My concern wasn't that much about the period, but the gain. I'll probably have to try both approaches, but I was mainly concerned that a non-harmonic signal with one strong sinusoid could cause the comb filter to be turned on with a high gain. > Hence, ideally we should have some decision logic and smoothing techniques to get a > smooth contour of the true pitch period. I think the pitch estimator of the > BroadVoice codecs (BV16 and BV32) that Broadcom previously submitted can serve as a > good starting point for further development, because it was optimized to have > relatively low complexity and it also has decision logic and smoothing built-in to > try to get a smooth contour of the true pitch period for voiced regions of speech. CELT currently has a pitch predictor used for PLC, but I doubt it's very good for this as it is now. OTOH, it's fairly low complexity too. It's part of my plans to compare it with the one in BV16/32. I assume that Broadcom would be willing to relicense it to the simplified BSD license (it's currently LGPL) if we wanted to use it instead of the current CELT estimator. > It is optimized for 8 kHz or 16 kHz speech signals, so to use it for 48 kHz speech > and music signals, we need to adjust the decimation factor and re-optimize it to > work better for music as well. We also need to increase the range of the pitch > period, or equivalently, the pitch frequency. Typical pitch frequency range for > adult voice is somewhere around 50 to 400 Hz, but the pitch frequencies of music > instruments cover a much wider range and can go up to 1600 Hz or higher. BV16 and > BV32 detectable pitch ranges are around 59 Hz ~ 800 Hz and 60 Hz ~ 1600 Hz, > respectively, because we previously already increased their pitch ranges to be able > to detect the pitch of some of the music instruments. The pitch range is another thing that needs experimenting with, but my intuition here is that it's probably not necessary to go below 100 Hz because for these very low pitches, there is enough masking between the harmonics to make the coding noise inaudible even without a post-filter. Again worth testing. > Of course, if there is another royalty-free and unencumbered pitch estimator > available that has even lower complexity than the BV16/BV32 pitch estimator and > performs better, I am all for it. There's a few more starting with the SILK one, but it may be too good/complex for what we need. The Speex one is fairly complex and not that good, so it's not interesting for this. There's also the codec2 low-bitrate vocoder, but I don't know how well its pitch estimator is. Cheers, Jean-Marc From jean-marc.valin@usherbrooke.ca Sat Oct 23 18:04:24 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 7A8F33A67F2 for ; Sat, 23 Oct 2010 18:04:24 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.289 X-Spam-Level: X-Spam-Status: No, score=-2.289 tagged_above=-999 required=5 tests=[AWL=0.310, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1m3Imzzy6lcd for ; Sat, 23 Oct 2010 18:04:22 -0700 (PDT) Received: from relais.videotron.ca (relais.videotron.ca [24.201.245.36]) by core3.amsl.com (Postfix) with ESMTP id BB7423A67EA for ; Sat, 23 Oct 2010 18:04:22 -0700 (PDT) MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; charset=ISO-8859-1 Received: from [192.168.1.14] ([70.81.109.112]) by VL-MR-MRZ20.ip.videotron.ca (Sun Java(tm) System Messaging Server 6.3-8.01 (built Dec 16 2008; 32bit)) with ESMTP id <0LAR009D7SDYA430@VL-MR-MRZ20.ip.videotron.ca> for codec@ietf.org; Sat, 23 Oct 2010 21:05:58 -0400 (EDT) Message-id: <4CC38693.1070300@usherbrooke.ca> Date: Sat, 23 Oct 2010 21:06:27 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.14) Gecko/20101006 Thunderbird/3.0.9 To: "Raymond (Juin-Hwey) Chen" References: <4CC2DA7E.1060705@usherbrooke.ca> In-reply-to: X-Enigmail-Version: 1.0.1 Cc: "codec@ietf.org" Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Oct 2010 01:04:25 -0000 Hi Raymond, On 10-10-23 05:56 PM, Raymond (Juin-Hwey) Chen wrote: > (2) On a second thought, I think to calculate the normalized correlation c or the > optimal tap weight beta for a single-tap pitch predictor, rather than calculating > the summation over the current frame (n from 1 to L), it probably makes more sense > for the summation length to be the extracted pitch period or the pitch frame size, > whichever is larger. The motivation is this: suppose the pitch frame size is 2.5 ms > or 5 ms and the pitch period is 15 ms, then trying to measure the level of > periodicity using only 1/6 to 1/3 of the pitch cycle waveform seems dangerous or at > least unreliable. Thus, if the detected pitch period is longer than the frame size, > then we should sum over all of the samples in the current frame plus more samples in > the previous frame(s) until the total number of samples in the summation equals the > pitch period. If the pitch period is smaller than the frame size and the pitch > frame size is only 2.5 ms or 5 ms, I recommend just summing over the entire current > frame. That's indeed an issue that I hadn't thought of. It makes sense that we want at least one period. > (3) If we use multi-tap pitch filters in order to taper off the comb filtering > effect at higher frequencies, we need to be careful about the stability of the IIR > pitch filter, especially if the IIR filter is in the decoder side (as is the case in > my example single-tap all-pole pitch postfilter). I suppose you probably already > know this, but I would like to mention and discuss it anyway. Indeed, I learned this the hard way with Speex in the past. Fortunately, the gains we need tend to be small and we probably only want positive gains. As far as I know, having a sum of the absolute values of the taps being smaller than 1 is a sufficient condition to ensure stability and it's probably good enough even if stricter than necessary. Cheers, Jean-Marc From cloos@jhcloos.com Sun Oct 24 11:05:15 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 7859C3A68E4 for ; Sun, 24 Oct 2010 11:05:15 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.297 X-Spam-Level: X-Spam-Status: No, score=-1.297 tagged_above=-999 required=5 tests=[AWL=0.233, BAYES_00=-2.599, DATE_IN_PAST_06_12=1.069] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id x+PycD7iEjpG for ; Sun, 24 Oct 2010 11:05:14 -0700 (PDT) Received: from eagle.jhcloos.com (eagle.jhcloos.com [IPv6:2001:1938:12d::53]) by core3.amsl.com (Postfix) with ESMTP id 15AF13A691C for ; Sun, 24 Oct 2010 11:05:12 -0700 (PDT) Received: by eagle.jhcloos.com (Postfix, from userid 10) id 76B69400B0; Sun, 24 Oct 2010 18:06:30 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=jhcloos.com; s=eagle; t=1287943614; bh=ebLUcxv8UM6VJY9Jd6Ug5z5IcD9lrZpPXMwGeARee8Y=; h=From:To:Cc:Subject:In-Reply-To:References:Date:Message-ID: MIME-Version:Content-Type; b=UvaE0/CW0jYUZT4nMG/bOMvJypyi5ciTrnF3YvXgJ95uiJ64l5ZDJOoOhYMOSN3e+ F7kSz0knYoj9sJL3+OeQwt3h4ZEyVUOFa3Ey67VzSbT+s3TQJFcvwk+9QT84Tt9Bjh wLkUm5r5lpgWO55MbXN/V0io6INVCn7CHu+XYlPk= Received: from carbon (localhost [127.0.0.1]) by carbon.jhcloos.org (Postfix) with ESMTP id CD81A1E90A4; Sun, 24 Oct 2010 07:09:41 +0000 (UTC) From: James Cloos To: "Jean-Marc Valin" In-Reply-To: <390831ED3DF58E41A3D2FB82591E2C3605A34C4D@MAILEXCH.octasic.com> (Jean-Marc Valin's message of "Mon, 18 Oct 2010 22:26:24 -0400") References: <390831ED3DF58E41A3D2FB82591E2C3605A34C4D@MAILEXCH.octasic.com> User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/24.0.50 (gnu/linux) Face: iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAABHNCSVQICAgIfAhkiAAAAI1J REFUOE+lU9ESgCAIg64P1y+ngUdxhl5H8wFbbM0OmUiEhKkCYaZThXCo6KE5sCbA1DDX3genvO4d eBQgEMaM5qy6uWk4SfBYfdu9jvBN9nSVDOKRtwb+I3epboOsOX5pZbJNsBJFvmQQ05YMfieIBnYX FK2N6dOawd97r/e8RjkTLzmMsiVgrAoEugtviCM3v2WzjgAAAABJRU5ErkJggg== Copyright: Copyright 2009 James Cloos OpenPGP: ED7DAEA6; url=http://jhcloos.com/public_key/0xED7DAEA6.asc OpenPGP-Fingerprint: E9E9 F828 61A4 6EA9 0F2B 63E7 997A 9F17 ED7D AEA6 Date: Sun, 24 Oct 2010 03:09:41 -0400 Message-ID: Lines: 13 MIME-Version: 1.0 Content-Type: text/plain X-Hashcash: 1:30:101024:jean-marc.valin@octasic.com::U15go1S7KZb+H+1H:0000000000000000000000000000000006XfPX X-Hashcash: 1:30:101024:pochol@webfootgames.com::NBJFUe2D6WRq2Yq9:0000000000000000000000000000000000000AOADf X-Hashcash: 1:30:101024:codec@ietf.org::b8mbNELz5TZhwc9U:00L4gnN Cc: codec@ietf.org, Pochol@WebfootGames.com Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Oct 2010 18:05:15 -0000 >>>>> "JV" == Jean-Marc Valin writes: JV> Opus is available under the BSD license. As for patents, Skype has JV> pledged to make them available royalty-free once the codec is JV> accepted as an IETF standard. As a Standard? Or just published as an standards-track RFC? That is a /huge/ difference, time-wise. -JimC -- James Cloos OpenPGP: 1024D/ED7DAEA6 From hoene@uni-tuebingen.de Mon Oct 25 02:01:11 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 9202B3A69A7 for ; Mon, 25 Oct 2010 02:01:11 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.809 X-Spam-Level: X-Spam-Status: No, score=-3.809 tagged_above=-999 required=5 tests=[AWL=-0.161, BAYES_50=0.001, HELO_EQ_DE=0.35, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fFRWoxRk8GfS for ; Mon, 25 Oct 2010 02:01:09 -0700 (PDT) Received: from mx06.uni-tuebingen.de (mx06.uni-tuebingen.de [134.2.3.3]) by core3.amsl.com (Postfix) with ESMTP id 20E1B3A6992 for ; Mon, 25 Oct 2010 02:01:07 -0700 (PDT) Received: from hoeneT60 (u-173-c044.cs.uni-tuebingen.de [134.2.173.44]) (authenticated bits=0) by mx06.uni-tuebingen.de (8.13.6/8.13.6) with ESMTP id o9P92oD0009081 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Mon, 25 Oct 2010 11:02:50 +0200 From: "Christian Hoene" To: "'Koen Vos'" References: <1389617873.938373.1287516517422.JavaMail.root@lu2-zimbra> <1559386036.938403.1287516673396.JavaMail.root@lu2-zimbra> In-Reply-To: <1559386036.938403.1287516673396.JavaMail.root@lu2-zimbra> Date: Mon, 25 Oct 2010 11:02:50 +0200 Organization: =?utf-8?Q?Universit=C3=A4t_T=C3=BCbingen?= Message-ID: <000001cb7423$67074c40$3515e4c0$@de> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0001_01CB7434.2A901C40" X-Mailer: Microsoft Office Outlook 12.0 Thread-Index: ActvxDaRX38vuXL/QUiVdamKy++KYwEXQv7A Content-Language: de X-AntiVirus: NOT checked by Avira MailGate (version: 3.0.0-4; host: mx06) Cc: codec@ietf.org Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Oct 2010 09:01:11 -0000 This is a multi-part message in MIME format. ------=_NextPart_000_0001_01CB7434.2A901C40 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Hi Koen, =20 Comments inline=E2=80=A6 =20 --------------------------------------------------------------- Dr.-Ing. Christian Hoene Interactive Communication Systems (ICS), University of T=C3=BCbingen=20 Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532=20 http://www.net.uni-tuebingen.de/ =20 From: Koen Vos [mailto:koen.vos@skype.net]=20 Sent: Tuesday, October 19, 2010 9:31 PM To: Christian Hoene Cc: Stephan Wenger; codec@ietf.org Subject: Re: [codec] Opus codec licensing =20 Sorry, please ignore previous email - bad keyboard shortcut... >> Furthermore, any implementation of the codec that accidentally does = not confirm to the standard yet will require a license fee. If you modify the codec so much that it's no longer compatible to the = standard, then that modified codec is no longer "Opus". In that case = you've created a new codec and I think it's only reasonable that the = patent license doesn't apply. [Christian Hoene] Any good codec such as G.729, AMR, and GSM had = successors. Even a potentially great codec such as Opus can be enhanced = - for example - by supporting multi channel compression or spatial = audio. Also, the first version of the codec might come with bugs, = weaknesses, or high third-party IPR licensing claims that might require = updates of the standard. =20 Thus, in my opinion it is reasonable to allow standardizing a follow-up = codec, which may need a changed bit-stream and thus cannot be backwards = compatible. =20 Of course, this issue is not time pressing.=20 =20 With best regards, =20 Christian Hoene =20 best, koen. _____ =20 From: "Koen Vos" To: "Christian Hoene" Cc: "Stephan Wenger" , codec@ietf.org Sent: Tuesday, October 19, 2010 12:27:39 PM Subject: Re: [codec] Opus codec licensing If you modify the codec so much that it can't=20 _____ =20 From: "Christian Hoene" To: "Stephan Wenger" , codec@ietf.org Sent: Tuesday, October 19, 2010 11:12:48 AM Subject: Re: [codec] Opus codec licensing Hi =20 From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf = Of Stephan Wenger Sent: Tuesday, October 19, 2010 5:21 PM To: Jean-Marc Valin; Pochol@WebfootGames.com; codec@ietf.org Subject: Re: [codec] Opus codec licensing =20 Hi, With =E2=80=9Chat=E2=80=9D: It is, in almost all cases, unwise for a third party to = circumscribe=E2=80=94or worse: abbreviate=E2=80=94licensing terms = provided by a rightholder. Skype=E2=80=99s licensing declaration can be = found at https://datatracker.ietf.org/ipr/1297. I note that the = statement contains conditions beyond the acceptance of their technology = into an IETF standard. Stephan [Christian Hoene] +1 =20 It is planned to test the new codec by spreading it to users prior = standardization. However, under the terms of this licensing declaration, = this is not possible for anybody but Skype because it would require a = license from Skype. =20 Also, any enhancement of the codec, which does not conform to the = original standard anymore, will require a license. =20 Furthermore, any implementation of the codec that accidentally does not = confirm to the standard yet will require a license fee. =20 Clearly, these terms are unacceptable. Please, Koen, talk to your lawyer = again to weaken this license statements. This shall include at least a = draft versions of the standard. Also, updates of the standard shall not = require any new license anymore. =20 It is patent really essential for the opus codec?=20 =E2=80=9CA method of estimating noise in data containing voice = information and noise includes receiving the data as a sequence of input = values; transforming the data by applying a first non linear mapping to = the input values wherein the derivative function of the mapping = decreases in magnitude as the input values increase in magnitude = smoothing the transformed data; and transforming the smoothed = transformed data by applying a second non linear mapping that is = opposite to the first non linear mapping, to determine an estimate of = the noise in the inputted data.=E2=80=9D Anyhow, it should not be too difficult to circumvent the patent. Any = ideas?=20 =20 Christian =20 --------------------------------------------------------------- Dr.-Ing. Christian Hoene Interactive Communication Systems (ICS), University of T=C3=BCbingen=20 Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532=20 http://www.net.uni-tuebingen.de/ =20 On 10.18.2010 19:26 , "Jean-Marc Valin" = wrote: Opus is available under the BSD license. As for patents, Skype has = pledged to make them available royalty-free once the codec is accepted = as an IETF standard. Jean-Marc -----Original Message----- From: codec-bounces@ietf.org on behalf of Pascal Pochol Sent: Mon 10/18/2010 10:19 PM To: codec@ietf.org Subject: [codec] Opus codec licensing Hello, we've been using speex and celt for voice and music but today I heard = about Opus which sounds like a fantastic replacement for both these codecs. We're wondering if Opus will be released under the same type of license = as speex and celt? SILK's license forbid its use in commercial software and we're worried that it might carry over to Opus. If not, as soon as Opus handles lower bitrates, stereo and fixed point decoding we'll be using = it. -Pascal _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec _____ =20 _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec ------=_NextPart_000_0001_01CB7434.2A901C40 Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: quoted-printable Re: [codec] Opus codec licensing

Hi Koen,

 

Comments inline=E2=80=A6

 

----------------------------------------------------------= -----

Dr.-Ing. Christian Hoene

Interactive Communication Systems (ICS), University of = T=C3=BCbingen

Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 = 2970532

http://www.net.uni-tuebingen.de= /

 

From:= Koen Vos [mailto:koen.vos@skype.net]
Sent: Tuesday, October 19, 2010 9:31 PM
To: Christian Hoene
Cc: Stephan Wenger; codec@ietf.org
Subject: Re: [codec] Opus codec licensing

 

Sorry, please ignore previous email - bad keyboard shortcut...

>> Furthermore, any implementation of the codec = that accidentally does not confirm to the standard yet will require a license = fee.

If you modify the codec so much that it's no longer compatible to the = standard, then that modified codec is no longer "Opus".  In that = case you've created a new codec and I think it's only reasonable that the = patent license doesn't apply.

[Christian Hoene] = Any good codec such as G.729, AMR,=C2=A0 and GSM had = successors. Even a potentially great codec such as Opus can be enhanced - for example - = by supporting multi channel compression or spatial audio. Also, the first version of = the codec might come with bugs, weaknesses, or high third-party IPR = licensing claims that might require updates of the standard.

 

Thus, in my opinion it is reasonable to allow = standardizing a follow-up codec, which may need a changed bit-stream and thus cannot be backwards compatible.

 

Of course, this issue is not time pressing. =

 

With best regards,

 

=C2=A0Christian Hoene

 



best,
koen.

From: "Koen Vos" = <koen.vos@skype.net>
To: "Christian Hoene" = <hoene@uni-tuebingen.de>
Cc: "Stephan Wenger" <stewe@stewe.org>, = codec@ietf.org
Sent: Tuesday, October 19, 2010 12:27:39 PM
Subject: Re: [codec] Opus codec licensing

If you modify the codec so much that it can't

From: = "Christian Hoene" = <hoene@uni-tuebingen.de>
To: "Stephan Wenger" <stewe@stewe.org>, = codec@ietf.org
Sent: Tuesday, October 19, 2010 11:12:48 AM
Subject: Re: [codec] Opus codec licensing


Hi

 

From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] = On Behalf Of Stephan Wenger
Sent: Tuesday, October 19, 2010 5:21 PM
To: Jean-Marc Valin; Pochol@WebfootGames.com; codec@ietf.org
Subject: Re: [codec] Opus codec licensing

 

Hi,
With =E2=80=9Chat=E2=80=9D:
It is, in almost all cases, unwise for a third party to = circumscribe=E2=80=94or worse: abbreviate=E2=80=94licensing terms provided by a rightholder. =  Skype=E2=80=99s licensing declaration can be found at https://datatracker.ietf.org/ipr/1297.  I = note that the statement contains conditions beyond the acceptance of their = technology into an IETF standard.
Stephan

[Christian Hoene] = +1

 

It is planned to test the new codec by spreading it to = users prior standardization. However, under the terms of this licensing = declaration, this is not possible for anybody but Skype because it would require a = license from Skype.

 

Also, any enhancement of the codec, which does not = conform to the original standard anymore, will require a license.

 

Furthermore, any implementation of the codec that = accidentally does not confirm to the standard yet will require a license = fee.

 

Clearly, these terms are unacceptable. Please, Koen, talk = to your lawyer again to weaken this license statements. This shall include = at least a draft versions of the standard. Also, updates of the standard = shall not require any new license anymore.

 

It is patent really essential for the opus codec? =

=E2=80=9C= A method of estimating noise in data containing voice information and noise = includes receiving the data as a sequence of input values; transforming the data = by applying a first non linear mapping to the input values wherein the = derivative function of the mapping decreases in magnitude as the input values = increase in magnitude smoothing the transformed data; and transforming the smoothed transformed data by applying a second non linear mapping that is = opposite to the first non linear mapping, to determine an estimate of the noise in = the inputted data.=E2=80=9D

Anyhow, it should not be too difficult to circumvent the = patent. Any ideas?

 

Christian

 

----------------------------------------------------------= -----

Dr.-Ing. Christian Hoene

Interactive Communication Systems (ICS), University of = T=C3=BCbingen

Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 = 2970532

http://www.net.uni-tuebingen.de/

 




On 10.18.2010 19:26 , "Jean-Marc Valin" <jean-marc.valin@octasic.com> wrote:

Opus is available under = the BSD license. As for patents, Skype has pledged to make them available = royalty-free once the codec is accepted as an IETF standard.

    Jean-Marc

-----Original Message-----
From: codec-bounces@ietf.org on behalf of Pascal Pochol
Sent: Mon 10/18/2010 10:19 PM
To: codec@ietf.org
Subject: [codec] Opus codec licensing

Hello,

we've been using speex and celt for voice and music but today I heard = about
Opus which sounds like a fantastic replacement for both these = codecs.

We're wondering if Opus will be released under the same type of license = as
speex and celt? SILK's license forbid its use in commercial software = and
we're worried that it might carry over to Opus. If not, as soon as = Opus
handles lower bitrates, stereo and fixed point decoding we'll be using = it.

-Pascal

_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/mailman/listinfo/codec<= span style=3D'color:black'>

_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/mailman/listinfo/codec<= span style=3D'color:black'>


_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/mailman/listinfo/codec

------=_NextPart_000_0001_01CB7434.2A901C40-- From roman@telurix.com Mon Oct 25 11:58:03 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 243773A6874 for ; Mon, 25 Oct 2010 11:58:03 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.574 X-Spam-Level: X-Spam-Status: No, score=-1.574 tagged_above=-999 required=5 tests=[AWL=0.402, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id k3eL-cTPgHVY for ; Mon, 25 Oct 2010 11:58:00 -0700 (PDT) Received: from mail-yw0-f44.google.com (mail-yw0-f44.google.com [209.85.213.44]) by core3.amsl.com (Postfix) with ESMTP id 1219D3A69A4 for ; Mon, 25 Oct 2010 11:57:58 -0700 (PDT) Received: by ywp6 with SMTP id 6so1882552ywp.31 for ; Mon, 25 Oct 2010 11:59:44 -0700 (PDT) Received: by 10.151.50.2 with SMTP id c2mr5662895ybk.321.1288033183668; Mon, 25 Oct 2010 11:59:43 -0700 (PDT) Received: from mail-yx0-f172.google.com (mail-yx0-f172.google.com [209.85.213.172]) by mx.google.com with ESMTPS id r29sm3103126ybn.10.2010.10.25.11.59.41 (version=TLSv1/SSLv3 cipher=RC4-MD5); Mon, 25 Oct 2010 11:59:41 -0700 (PDT) Received: by yxp4 with SMTP id 4so2590926yxp.31 for ; Mon, 25 Oct 2010 11:59:40 -0700 (PDT) MIME-Version: 1.0 Received: by 10.42.183.203 with SMTP id ch11mr5102901icb.463.1288033180399; Mon, 25 Oct 2010 11:59:40 -0700 (PDT) Received: by 10.231.154.19 with HTTP; Mon, 25 Oct 2010 11:59:40 -0700 (PDT) In-Reply-To: <000001cb7423$67074c40$3515e4c0$@de> References: <1389617873.938373.1287516517422.JavaMail.root@lu2-zimbra> <1559386036.938403.1287516673396.JavaMail.root@lu2-zimbra> <000001cb7423$67074c40$3515e4c0$@de> Date: Mon, 25 Oct 2010 14:59:40 -0400 Message-ID: From: Roman Shpount To: Christian Hoene Content-Type: multipart/alternative; boundary=20cf303e9de82d39f004937598b2 Cc: codec@ietf.org Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Oct 2010 18:58:03 -0000 --20cf303e9de82d39f004937598b2 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: quoted-printable I know all of you have been very concerned about the Skype IPR, but I don't think this is the biggest problem. In this group we are currently missing a lot of traditional CODEC industry players (GIPS (now google), Qualcomm, Nokia, etc). We never checked that we are not using any of their intellectual property. If we are serious about creating a royalty free CODEC, someone (I am not implying doing this on the mailing list, since thi= s probably is ideally done offline), needs to review Opus, as well as Silk an= d CELT, and see if uses any of the third party intellectual property. Good place to start would be to see if none of the G.729, GSM, AMR, and MP3 patents apply to this code. A wide general purpose patent search is probabl= y appropriate as well. We can pretend that none of this intellectual property exists, but we might end up with having royalty associated with a new CODEC after all the effort that was invested into creating it. I understand that we cannot possibly protect ourselves against everything, but we should at least check the most obvious potential intellectual property. What this comes down to is that we need to do a patent search and CODEC IPR review. This is something that will cost a considerable amount of money. I am not sure what is the best way to approach this. We can either find a volunteer, familiar with CODEC development and patent law, or setup some type of fund and hire a company to do this. _____________________________ Roman Shpount - www.telurix.com 2010/10/25 Christian Hoene > Hi Koen, > > > > Comments inline=85 > > > > --------------------------------------------------------------- > > Dr.-Ing. Christian Hoene > > Interactive Communication Systems (ICS), University of T=FCbingen > > Sand 13, 72076 T=FCbingen, Germany, Phone +49 7071 2970532 > > http://www.net.uni-tuebingen.de/ > > > > *From:* Koen Vos [mailto:koen.vos@skype.net] > *Sent:* Tuesday, October 19, 2010 9:31 PM > *To:* Christian Hoene > *Cc:* Stephan Wenger; codec@ietf.org > *Subject:* Re: [codec] Opus codec licensing > > > > Sorry, please ignore previous email - bad keyboard shortcut... > > >> Furthermore, any implementation of the codec that accidentally does no= t > confirm to the standard yet will require a license fee. > > If you modify the codec so much that it's no longer compatible to the > standard, then that modified codec is no longer "Opus". In that case you= 've > created a new codec and I think it's only reasonable that the patent lice= nse > doesn't apply. > > *[Christian Hoene] *Any good codec such as G.729, AMR, and GSM had > successors. Even a potentially great codec such as Opus can be enhanced - > for example - by supporting multi channel compression or spatial audio. > Also, the first version of the codec might come with bugs, weaknesses, or > high third-party IPR licensing claims that might require updates of the > standard. > > > > Thus, in my opinion it is reasonable to allow standardizing a follow-up > codec, which may need a changed bit-stream and thus cannot be backwards > compatible. > > > > Of course, this issue is not time pressing. > > > > With best regards, > > > > Christian Hoene > > * * > > > > best, > koen. > > ------------------------------ > > *From: *"Koen Vos" > *To: *"Christian Hoene" > *Cc: *"Stephan Wenger" , codec@ietf.org > *Sent: *Tuesday, October 19, 2010 12:27:39 PM > *Subject: *Re: [codec] Opus codec licensing > > If you modify the codec so much that it can't > > ------------------------------ > > *From: *"Christian Hoene" > *To: *"Stephan Wenger" , codec@ietf.org > *Sent: *Tuesday, October 19, 2010 11:12:48 AM > *Subject: *Re: [codec] Opus codec licensing > > > Hi > > > > *From:* codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] *On Behalf > Of *Stephan Wenger > *Sent:* Tuesday, October 19, 2010 5:21 PM > *To:* Jean-Marc Valin; Pochol@WebfootGames.com; codec@ietf.org > *Subject:* Re: [codec] Opus codec licensing > > > > Hi, > With =93hat=94: > It is, in almost all cases, unwise for a third party to circumscribe=97or > worse: abbreviate=97licensing terms provided by a rightholder. Skype=92s > licensing declaration can be found at > https://datatracker.ietf.org/ipr/1297. I note that the statement contain= s > conditions beyond the acceptance of their technology into an IETF standar= d. > Stephan > > *[Christian Hoene] *+1 > > > > It is planned to test the new codec by spreading it to users prior > standardization. However, under the terms of this licensing declaration, > this is not possible for anybody but Skype because it would require a > license from Skype. > > > > Also, any enhancement of the codec, which does not conform to the origina= l > standard anymore, will require a license. > > > > Furthermore, any implementation of the codec that accidentally does not > confirm to the standard yet will require a license fee. > > > > Clearly, these terms are unacceptable. Please, Koen, talk to your lawyer > again to weaken this license statements. This shall include at least a dr= aft > versions of the standard. Also, updates of the standard shall not require > any new license anymore. > > > > It is patent really essential for the opus codec? > > =93A method of estimating noise in data containing voice information and > noise includes receiving the data as a sequence of input values; > transforming the data by applying a first non linear mapping to the input > values wherein the derivative function of the mapping decreases in magnit= ude > as the input values increase in magnitude smoothing the transformed data; > and transforming the smoothed transformed data by applying a second non > linear mapping that is opposite to the first non linear mapping, to > determine an estimate of the noise in the inputted data.=94 > > Anyhow, it should not be too difficult to circumvent the patent. Any idea= s? > > > > > Christian > > > > --------------------------------------------------------------- > > Dr.-Ing. Christian Hoene > > Interactive Communication Systems (ICS), University of T=FCbingen > > Sand 13, 72076 T=FCbingen, Germany, Phone +49 7071 2970532 > > http://www.net.uni-tuebingen.de/ > > * * > > > > > On 10.18.2010 19:26 , "Jean-Marc Valin" > wrote: > > Opus is available under the BSD license. As for patents, Skype has pledge= d > to make them available royalty-free once the codec is accepted as an IETF > standard. > > Jean-Marc > > -----Original Message----- > From: codec-bounces@ietf.org on behalf of Pascal Pochol > Sent: Mon 10/18/2010 10:19 PM > To: codec@ietf.org > Subject: [codec] Opus codec licensing > > Hello, > > we've been using speex and celt for voice and music but today I heard abo= ut > Opus which sounds like a fantastic replacement for both these codecs. > > We're wondering if Opus will be released under the same type of license a= s > speex and celt? SILK's license forbid its use in commercial software and > we're worried that it might carry over to Opus. If not, as soon as Opus > handles lower bitrates, stereo and fixed point decoding we'll be using it= . > > -Pascal > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > ------------------------------ > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > --20cf303e9de82d39f004937598b2 Content-Type: text/html; charset=windows-1252 Content-Transfer-Encoding: quoted-printable I know all of you have been very concerned about the Skype IPR, but I don&#= 39;t think this is the biggest problem. In this group we are currently miss= ing a lot of traditional CODEC industry players (GIPS (now google), Qualcom= m, Nokia, etc). We never checked that we are not using any of their intelle= ctual property. If we are serious about creating a royalty free CODEC, some= one (I am not implying doing this on the mailing list, since this probably = is ideally done offline), needs to review Opus, as well as Silk and CELT, a= nd see if uses any of the third party intellectual property. Good place to = start would be to see if none of the G.729, GSM, AMR, and MP3 patents apply= to this code. A wide general purpose patent search is probably appropriate= as well. We can pretend that none of this intellectual property exists, bu= t we might end up with having royalty associated with a new CODEC after all= the effort that was invested into creating it. I understand that we cannot= possibly protect ourselves against everything, but we should at least chec= k the most obvious potential intellectual property.

What this comes down to is that we need to do a patent search and CODEC= IPR review. This is something that will cost a considerable amount of mone= y. I am not sure what is the best way to approach this. We can either find = a volunteer, familiar with CODEC development and patent law, or setup some = type of fund and hire a company to do this.
_____________________________
Roman Shpount - www.telurix.com


2010/10/25 Christian Hoene <hoene@uni-tuebingen.de>

Hi Koen,

=A0

Comments inline=85

=A0

=A0

From:= Koen Vos [mailto:koen.vos@sk= ype.net]
Sent: Tuesday, October 19, 2010 9:31 PM
To: Christian Hoene
Cc: Stephan Wenger; codec@ietf.org
Subject: Re: [codec] Opus codec licensing

=A0

Sorry, please ignore previous email - bad keyboard shortcut...

>> Furthermore, any implementation of the codec that accidentally does not confirm to the standard yet will require a license fe= e.

If you modify the codec so much that it's no longer compatible to the s= tandard, then that modified codec is no longer "Opus".=A0 In that case you've created a new codec and I think it's only reasonable that th= e patent license doesn't apply.=

[Christian Hoene] Any good codec = such as G.729, AMR,=A0 and GSM had successors. Even a potentially great codec such as Opus can be enhanced - for example - by s= upporting multi channel compression or spatial audio. Also, the first version of the codec might come with bugs, weaknesses, or high third-party IPR licensing c= laims that might require updates of the standard.

=A0

Thus, in my opinion it is reasonable to allow standar= dizing a follow-up codec, which may need a changed bit-stream and thus cannot be backwards compatible.

=A0

Of course, this issue is not time pressing.

=A0

With best regards,

=A0

=A0Christian Hoene

=A0



best,
koen.

From: "Koen Vos" <koen.vos@skype.net>=
To: "Christian Hoene" <hoene@uni-tuebingen.de>
Cc: "Stephan Wenger" <stewe@stewe.org>, codec@ietf.org
Sent: Tuesday, October 19, 2010 12:27:39 PM
Subject: Re: [codec] Opus codec licensing

If you modify the codec so much that it can't

From: "Christian Hoene" <hoene@uni-tuebingen.de>=
To: "Stephan Wenger" <stewe@stewe.org>, codec@ietf.org
Sent: Tuesday, October 19, 2010 11:12:48 AM
Subject: Re: [codec] Opus codec licensing


Hi

=A0

Fr= om: codec-bounces@ietf.org [= mailto:codec-bo= unces@ietf.org] On Behalf Of Stephan Wenger
Sent: Tuesday, October 19, 2010 5:21 PM
To: Jean-Marc Valin; Pochol@WebfootGames.com; codec@ietf.org
Subject: Re: [codec] Opus codec licensing

=A0

Hi,
With =93hat=94:
It is, in almost all cases, unwise for a third party to circumscribe=97or w= orse: abbreviate=97licensing terms provided by a rightholder. =A0Skype=92s licens= ing declaration can be found at https://datatracker.ietf.org/ipr/1297. =A0I note t= hat the statement contains conditions beyond the acceptance of their technology into an IETF standard.
Stephan

[Christian Hoene] +1

=A0

It is planned to test the new codec by spreading it to= users prior standardization. However, under the terms of this licensing declarati= on, this is not possible for anybody but Skype because it would require a licen= se from Skype.

=A0

Also, any enhancement of the codec, which does not con= form to the original standard anymore, will require a license.

=A0

Furthermore, any implementation of the codec that acci= dentally does not confirm to the standard yet will require a license fee.

=A0

Clearly, these terms are unacceptable. Please, Koen, t= alk to your lawyer again to weaken this license statements. This shall include at least a draft versions of the standard. Also, updates of the standard shall= not require any new license anymore.

=A0

It is patent really essential for the opus codec?

=93A method of estimating noise in data containing voice information and noise includes receiving the data as a sequence of input values; transforming the data by applying a first non linear mapping to the input values wherein the derivat= ive function of the mapping decreases in magnitude as the input values increase= in magnitude smoothing the transformed data; and transforming the smoothed transformed data by applying a second non linear mapping that is opposite t= o the first non linear mapping, to determine an estimate of the noise in the inputted data.=94

Anyhow, it should not be too difficult to circumvent t= he patent. Any ideas?

=A0

Christian<= /p>

=A0

-----------------------------= ----------------------------------

Dr.-Ing. Christian Hoene

Interactive Communication Sys= tems (ICS), University of T=FCbingen <= /span>

Sand 13, 72076 T=FCbingen, Ge= rmany, Phone +49 7071 2970532 <= /p>

http://www.net.uni-tuebingen.de/

=A0




On 10.18.2010 19:26 , "Jean-Marc Valin" <jean-marc.valin@octasic.com> wrote:

Opus is available under the BSD license. As for patents, Skype has pledged to make them available royalty-f= ree once the codec is accepted as an IETF standard.

=A0=A0=A0=A0Jean-Marc

-----Original Message-----
From: codec-bou= nces@ietf.org on behalf of Pascal Pochol
Sent: Mon 10/18/2010 10:19 PM
To: codec@ietf.org<= br> Subject: [codec] Opus codec licensing

Hello,

we've been using speex and celt for voice and music but today I heard a= bout
Opus which sounds like a fantastic replacement for both these codecs.

We're wondering if Opus will be released under the same type of license= as
speex and celt? SILK's license forbid its use in commercial software an= d
we're worried that it might carry over to Opus. If not, as soon as Opus=
handles lower bitrates, stereo and fixed point decoding we'll be using = it.

-Pascal

_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec

_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec


_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec


_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec


--20cf303e9de82d39f004937598b2-- From bmschwar@fas.harvard.edu Mon Oct 25 12:25:04 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 6612C3A6915 for ; Mon, 25 Oct 2010 12:25:04 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.599 X-Spam-Level: X-Spam-Status: No, score=-6.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Mi0+B0IRfLIh for ; Mon, 25 Oct 2010 12:25:03 -0700 (PDT) Received: from us17.unix.fas.harvard.edu (us17.unix.fas.harvard.edu [140.247.35.197]) by core3.amsl.com (Postfix) with ESMTP id 293083A688A for ; Mon, 25 Oct 2010 12:25:03 -0700 (PDT) Received: from us17.unix.fas.harvard.edu (localhost.localdomain [127.0.0.1]) by us17.unix.fas.harvard.edu (Postfix) with ESMTP id 18B4841E481; Mon, 25 Oct 2010 15:26:48 -0400 (EDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed; d=fas.harvard.edu; h= message-id:date:from:reply-to:mime-version:to:cc:subject :references:in-reply-to:content-type; s=mail; bh=bxUzI6lfhlSaOl5 K/4B/tgmeaeZDF6MAgQ3kUXAkPO8=; b=Z7bwsKOnPZocvEmO5uWWc6LMliVevyG aib7RmJ+9sLSrymnJiHqlzC+85s3dl//ey1EqArPK30V3vjc0V3zS1v/IIo1T0UI pErhzp7k3yLy0PRnAsqg3pfTAoOctY0gO+nFdZYV+IsqrH7w/PAMw7kSsRCwP/81 ZvunNDcnkeag= DomainKey-Signature: a=rsa-sha1; c=simple; d=fas.harvard.edu; h= message-id:date:from:reply-to:mime-version:to:cc:subject :references:in-reply-to:content-type; q=dns; s=mail; b=ZLvfG8ank RP0sWosiAvxYdcGx6ZjE4VUtxETbcINvgjiF2/s6Yghgk+TSnzrVaBJXrWtvdcGa eKckvqKMom8PsmURmsdLTvwK7D//z0WReVUyuqtoeiY/eSGqvTOoJkD8Uq2qyMZt MLiENCJyVuCKDmSf+VgtEbabecIvHuKKMs= Received: from [172.23.141.145] (bwhmaincampuspat25.partners.org [170.223.207.25]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) (Authenticated sender: bmschwar@fas) by us17.unix.fas.harvard.edu (Postfix) with ESMTPSA id 11C0141E46F; Mon, 25 Oct 2010 15:26:48 -0400 (EDT) Message-ID: <4CC5D9F7.9040308@fas.harvard.edu> Date: Mon, 25 Oct 2010 15:26:47 -0400 From: "Benjamin M. Schwartz" User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1.12) Gecko/20100915 Lightning/1.0b1 Thunderbird/3.0.8 MIME-Version: 1.0 To: Roman Shpount References: <1389617873.938373.1287516517422.JavaMail.root@lu2-zimbra> <1559386036.938403.1287516673396.JavaMail.root@lu2-zimbra> <000001cb7423$67074c40$3515e4c0$@de> In-Reply-To: X-Enigmail-Version: 1.0.1 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="------------enig15AF50DFA085B11480578AE7" Cc: codec@ietf.org Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list Reply-To: bens@alum.mit.edu List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Oct 2010 19:25:04 -0000 This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enig15AF50DFA085B11480578AE7 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: quoted-printable On 10/25/2010 02:59 PM, Roman Shpount wrote: > I understand that we cannot possibly protect > ourselves against everything, but we should at least check the most > obvious potential intellectual property. Work of this nature has been going on throughout codec development, and will continue. It may even accelerate. The codec's structures have been= carefully chosen based on knowledge of existing patents, to ensure that the result is entirely non-infringing. You (and everyone here) are welcome to participate in this process, which of course is not taking place in a public forum. Contact any of the active developers with your questions if you would like to contribute to the effort. Your contribution would be appreciated. > What this comes down to is that we need to do a patent search and CODEC= > IPR review. Yes. People have been doing this, and will continue to do this. > This is something that will cost a considerable amount of > money. I don't see why this should be true. Reading patents is free. Even getting file wrappers is not expensive. The only thing that would be expensive is hiring certified legal professionals. In my view, any possible advantage to hiring lawyers is eliminated by the fact that they would be unwilling to publish a report with any level of detail beyond "the codec is non-infringing according to our research". I don't think that hiring legal professionals would bring any great advantage regarding IPR review for the codec. If anything, it is easier for a codec engineer to learn to read patents than for legal experts to learn signal processing algorithms. --Ben --------------enig15AF50DFA085B11480578AE7 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) iEYEARECAAYFAkzF2fcACgkQUJT6e6HFtqTjQQCgjvGmIjf7+SH2rYYp2GP30BNF ZvMAnRZnRFqH6wxkQRA6TVRsxpRlqNfE =46MK -----END PGP SIGNATURE----- --------------enig15AF50DFA085B11480578AE7-- From jean-marc.valin@octasic.com Mon Oct 25 12:28:53 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 412483A6809 for ; Mon, 25 Oct 2010 12:28:52 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.454 X-Spam-Level: X-Spam-Status: No, score=-2.454 tagged_above=-999 required=5 tests=[AWL=0.145, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id WeBbreKKvjcN for ; Mon, 25 Oct 2010 12:28:45 -0700 (PDT) Received: from toroondcbmts06-srv.bellnexxia.net (toroondcbmts06.bellnexxia.net [207.236.237.40]) by core3.amsl.com (Postfix) with ESMTP id 967E33A66B4 for ; Mon, 25 Oct 2010 12:28:44 -0700 (PDT) Received: from toip54-bus.srvr.bell.ca ([67.69.240.140]) by toroondcbmts06-srv.bellnexxia.net (InterMail vM.8.00.01.00 201-2244-105-20090324) with ESMTP id <20101025193029.LYBB25002.toroondcbmts06-srv.bellnexxia.net@toip54-bus.srvr.bell.ca>; Mon, 25 Oct 2010 15:30:29 -0400 Received: from toip60-bus.srvr.bell.ca ([67.69.240.187]) by toip54-bus.srvr.bell.ca with ESMTP; 25 Oct 2010 15:30:26 -0400 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: ApsEAB93xUxGNv5q/2dsb2JhbACiQb5RAoVGBI1TBg Received: from mail.octasic.com (HELO MAILEXCH.octasic.com) ([70.54.254.106]) by toip60-bus.srvr.bell.ca with ESMTP; 25 Oct 2010 15:30:26 -0400 Received: from [10.100.60.27] ([10.100.60.27]) by MAILEXCH.octasic.com with Microsoft SMTPSVC(6.0.3790.4675); Mon, 25 Oct 2010 15:22:12 -0400 Message-ID: <4CC5D8E3.1030806@octasic.com> Date: Mon, 25 Oct 2010 15:22:11 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.14) Gecko/20101006 Lightning/1.0b1 Thunderbird/3.0.9 MIME-Version: 1.0 To: Christian Hoene References: <20101012224501.B93503A6A22@core3.amsl.com> <4CB514BE.6080104@usherbrooke.ca> <001201cb6f13$2a24fd00$7e6ef700$@de> In-Reply-To: <001201cb6f13$2a24fd00$7e6ef700$@de> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit X-OriginalArrivalTime: 25 Oct 2010 19:22:12.0044 (UTC) FILETIME=[ECA740C0:01CB7479] Cc: "codec@ietf.org" Subject: Re: [codec] I-D Action:draft-ietf-codec-requirements-02.txt X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Oct 2010 19:28:54 -0000 Hi Christian, Haven't had time to go through everything, but I was able to make some of the changes before the submission deadline (see draft -08). One issue I have found with your suggestion is the use of may vs should and MUST. Note that this is an informational draft and as far as I understand -- someone please correct me if I'm wrong -- the strict meaning of MAY/SHOULD/MUST as defined in RFC2119 does not apply. So in practice, it doesn't make much difference whether we use may or should, and we can't use the capitalized MUST at all. Cheers, Jean-Marc On 10-10-18 06:23 PM, Christian Hoene wrote: > Hi, > > just a few comments from regarding the guidelines draft (see attachment). I made some minor improvements to increase the clarity. > > I would suggest to split Section 4 into two. One section on how to describe the codec and its source code and a second on suggestion what to specify and what not to specify. > > Also, I added a section on how to collaborate with other WGs and on what topics the collaboration shall take place. It might be useful to plan when go for a bar Bof with those guys... > > With best regards, > > Christian > > > --------------------------------------------------------------- > Dr.-Ing. Christian Hoene > Interactive Communication Systems (ICS), University of Tübingen > Sand 13, 72076 Tübingen, Germany, Phone +49 7071 2970532 > http://www.net.uni-tuebingen.de/ > > > -----Original Message----- > From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Jean-Marc Valin > Sent: Wednesday, October 13, 2010 4:09 AM > To: codec@ietf.org > Subject: Re: [codec] I-D Action:draft-ietf-codec-requirements-02.txt > > There's also an updated version of the guidelines draft. > > Cheers, > > Jean-Marc > > On 10-10-12 06:45 PM, Internet-Drafts@ietf.org wrote: >> A New Internet-Draft is available from the on-line Internet-Drafts directories. >> This draft is a work item of the Internet Wideband Audio Codec Working Group of the IETF. >> >> >> Title : Codec Requirements >> Author(s) : J. Valin, K. Vos >> Filename : draft-ietf-codec-requirements-02.txt >> Pages : 21 >> Date : 2010-10-12 >> >> This document provides specific requirements for an Internet audio >> codec. These requirements address quality, sampling rate, bit-rate, >> and packet loss robustness, as well as other desirable properties. >> >> A URL for this Internet-Draft is: >> http://www.ietf.org/internet-drafts/draft-ietf-codec-requirements-02.txt >> >> Internet-Drafts are also available by anonymous FTP at: >> ftp://ftp.ietf.org/internet-drafts/ >> >> Below is the data which will enable a MIME compliant mail reader >> implementation to automatically retrieve the ASCII version of the >> Internet-Draft. >> >> >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From jean-marc.valin@octasic.com Mon Oct 25 12:30:57 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 282FC3A66B4 for ; Mon, 25 Oct 2010 12:30:57 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.469 X-Spam-Level: X-Spam-Status: No, score=-2.469 tagged_above=-999 required=5 tests=[AWL=0.130, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ZHEma1J8Qbdc for ; Mon, 25 Oct 2010 12:30:55 -0700 (PDT) Received: from toroondcbmts05-srv.bellnexxia.net (toroondcbmts05.bellnexxia.net [207.236.237.39]) by core3.amsl.com (Postfix) with ESMTP id 8421F3A6AA0 for ; Mon, 25 Oct 2010 12:30:55 -0700 (PDT) Received: from toip57-bus.srvr.bell.ca ([67.69.240.184]) by toroondcbmts05-srv.bellnexxia.net (InterMail vM.8.00.01.00 201-2244-105-20090324) with ESMTP id <20101025193240.KJXK28711.toroondcbmts05-srv.bellnexxia.net@toip57-bus.srvr.bell.ca>; Mon, 25 Oct 2010 15:32:40 -0400 Received: from toip35-bus.srvr.bell.ca ([67.69.240.36]) by toip57-bus.srvr.bell.ca with ESMTP; 25 Oct 2010 15:32:29 -0400 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: ApsEADZuxUxGNv5q/2dsb2JhbACiQL5RhUgEjVMG Received: from mail.octasic.com (HELO MAILEXCH.octasic.com) ([70.54.254.106]) by toip35-bus.srvr.bell.ca with ESMTP; 25 Oct 2010 15:32:28 -0400 Received: from [10.100.60.27] ([10.100.60.27]) by MAILEXCH.octasic.com with Microsoft SMTPSVC(6.0.3790.4675); Mon, 25 Oct 2010 15:31:40 -0400 Message-ID: <4CC5DB1C.3020002@octasic.com> Date: Mon, 25 Oct 2010 15:31:40 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.14) Gecko/20101006 Lightning/1.0b1 Thunderbird/3.0.9 MIME-Version: 1.0 To: Jean-Marc Valin References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC37F42.2040103@usherbrooke.ca> In-Reply-To: <4CC37F42.2040103@usherbrooke.ca> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 25 Oct 2010 19:31:40.0720 (UTC) FILETIME=[3F9C5300:01CB747B] Cc: "'codec@ietf.org'" Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Oct 2010 19:30:57 -0000 Hi, I have checked in an initial implementation in the CELT exp_comb_filter branch. Right now the quality is very bad, but that's mainly an implementation problem. For example, the current pitch estimator I'm using has atrocious behaviour when it comes to pitch doubling/tripling/... Also, I'm still suspecting bugs in the filter's overlap handling. Cheers, Jean-Marc On 10-10-23 08:35 PM, Jean-Marc Valin wrote: > Hi Raymond, > > On 10-10-23 03:50 PM, Raymond (Juin-Hwey) Chen wrote: >> Regarding your question, I think it really depends on what your main emphasis is. >> If you just want the best possible pitch period contour and don't mind having a >> little higher complexity, then partially whitening the input signal with the >> A(z/gamma) or A(z)/A(z/gamma) or even A(z/alpha)/A(z/beta) is a common approach used >> in the pitch estimators of many speech codecs (as I am sure you are well aware of), >> and it does help to reduce the chance of detecting the frequency of a strong first >> formant of the input speech signal as the pitch frequency. > > My concern wasn't that much about the period, but the gain. I'll > probably have to try both approaches, but I was mainly concerned that a > non-harmonic signal with one strong sinusoid could cause the comb filter > to be turned on with a high gain. > >> Hence, ideally we should have some decision logic and smoothing techniques to get a >> smooth contour of the true pitch period. I think the pitch estimator of the >> BroadVoice codecs (BV16 and BV32) that Broadcom previously submitted can serve as a >> good starting point for further development, because it was optimized to have >> relatively low complexity and it also has decision logic and smoothing built-in to >> try to get a smooth contour of the true pitch period for voiced regions of speech. > > CELT currently has a pitch predictor used for PLC, but I doubt it's very > good for this as it is now. OTOH, it's fairly low complexity too. It's > part of my plans to compare it with the one in BV16/32. I assume that > Broadcom would be willing to relicense it to the simplified BSD license > (it's currently LGPL) if we wanted to use it instead of the current CELT > estimator. > >> It is optimized for 8 kHz or 16 kHz speech signals, so to use it for 48 kHz speech >> and music signals, we need to adjust the decimation factor and re-optimize it to >> work better for music as well. We also need to increase the range of the pitch >> period, or equivalently, the pitch frequency. Typical pitch frequency range for >> adult voice is somewhere around 50 to 400 Hz, but the pitch frequencies of music >> instruments cover a much wider range and can go up to 1600 Hz or higher. BV16 and >> BV32 detectable pitch ranges are around 59 Hz ~ 800 Hz and 60 Hz ~ 1600 Hz, >> respectively, because we previously already increased their pitch ranges to be able >> to detect the pitch of some of the music instruments. > > The pitch range is another thing that needs experimenting with, but my > intuition here is that it's probably not necessary to go below 100 Hz > because for these very low pitches, there is enough masking between the > harmonics to make the coding noise inaudible even without a post-filter. > Again worth testing. > >> Of course, if there is another royalty-free and unencumbered pitch estimator >> available that has even lower complexity than the BV16/BV32 pitch estimator and >> performs better, I am all for it. > > There's a few more starting with the SILK one, but it may be too > good/complex for what we need. The Speex one is fairly complex and not > that good, so it's not interesting for this. There's also the codec2 > low-bitrate vocoder, but I don't know how well its pitch estimator is. > > Cheers, > > Jean-Marc > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From koen.vos@skype.net Mon Oct 25 13:52:10 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id DC02C3A6822 for ; Mon, 25 Oct 2010 13:52:10 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.343 X-Spam-Level: X-Spam-Status: No, score=-2.343 tagged_above=-999 required=5 tests=[AWL=0.256, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id n7T-9oQLo-fw for ; Mon, 25 Oct 2010 13:52:09 -0700 (PDT) Received: from mx.skype.net (mx.skype.net [78.141.177.88]) by core3.amsl.com (Postfix) with ESMTP id EDD633A6407 for ; Mon, 25 Oct 2010 13:52:08 -0700 (PDT) Received: from mx.skype.net (localhost [127.0.0.1]) by mx.skype.net (Postfix) with ESMTP id 6A19B7FD; Mon, 25 Oct 2010 22:53:53 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=skype.net; h=date:from:to :cc:message-id:in-reply-to:subject:mime-version:content-type: content-transfer-encoding; s=mx; bh=HlmNninuDLHEbH+Zzq2POi0BBYo= ; b=Abe7bGBqWjcGW+wYntX1MCyrWr2QK60Xa190GJb/aYE60yOVfbese1tMYrfa /RkepCgox9Ajs5BDZqlbBApNI0qsxU6MV3Qf9UWvw8090kAHdcJ6LU+D99tC1nqK uEPExF1+gjFglWnuCoIhvu+1oR79SciAGG+UhcOmoG8zITQ= DomainKey-Signature: a=rsa-sha1; c=nofws; d=skype.net; h=date:from:to:cc :message-id:in-reply-to:subject:mime-version:content-type: content-transfer-encoding; q=dns; s=mx; b=OOT9Yt31OP7rpg2mh/Vn7V psCWOC+Fj576q3iFfKWkAEyO5zLeVkDaqRnKeJ+GW0jBzRAef7gEh/8tmyF795hv iLnn3kH5/cBcRNo/Kwt/n7OFjB41CORuzDeetX886qk3HP/qbbgOfJdLKfNYbUWK mdysCG+D8tRrhGI974K+k= Received: from zimbra.skype.net (zimbra.skype.net [78.141.177.82]) by mx.skype.net (Postfix) with ESMTP id 686EA7F3; Mon, 25 Oct 2010 22:53:53 +0200 (CEST) Received: from localhost (localhost [127.0.0.1]) by zimbra.skype.net (Postfix) with ESMTP id 48F2F1672687; Mon, 25 Oct 2010 22:53:53 +0200 (CEST) X-Virus-Scanned: amavisd-new at lu2-zimbra.skype.net Received: from zimbra.skype.net ([127.0.0.1]) by localhost (zimbra.skype.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id He7CtUsKvpDb; Mon, 25 Oct 2010 22:53:52 +0200 (CEST) Received: from zimbra.skype.net (lu2-zimbra.skype.net [78.141.177.82]) by zimbra.skype.net (Postfix) with ESMTP id 9B1321672684; Mon, 25 Oct 2010 22:53:52 +0200 (CEST) Date: Mon, 25 Oct 2010 22:53:52 +0200 (CEST) From: Koen Vos To: Jean-Marc Valin Message-ID: <971279185.1094841.1288040032498.JavaMail.root@lu2-zimbra> In-Reply-To: <4CC5DB1C.3020002@octasic.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [69.181.192.115] X-Mailer: Zimbra 6.0.6_GA_2330.UBUNTU8_64 (ZimbraWebClient - FF3.0 (Win)/6.0.6_GA_2330.UBUNTU8_64) Cc: "codec@ietf.org" Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Oct 2010 20:52:11 -0000 Hi, It may be a good idea to look at SILK's pitch estimator. Having duplicate pitch analysis within one codec is not ideal, and the one in SILK is quite mature/stable. It may need some changes to work with all of CELT's frame sizes, but it already works for 10 and 20 ms frames - good enough for testing. As an extra benefit, it efficiently codes the pitch evolution over time, which would benefit CELT too. koen. ----- Original Message ----- From: "Jean-Marc Valin" To: "Jean-Marc Valin" Cc: "codec@ietf.org" Sent: Monday, October 25, 2010 12:31:40 PM Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec Hi, I have checked in an initial implementation in the CELT exp_comb_filter branch. Right now the quality is very bad, but that's mainly an implementation problem. For example, the current pitch estimator I'm using has atrocious behaviour when it comes to pitch doubling/tripling/... Also, I'm still suspecting bugs in the filter's overlap handling. Cheers, Jean-Marc On 10-10-23 08:35 PM, Jean-Marc Valin wrote: > Hi Raymond, > > On 10-10-23 03:50 PM, Raymond (Juin-Hwey) Chen wrote: >> Regarding your question, I think it really depends on what your main >> emphasis is. >> If you just want the best possible pitch period contour and don't >> mind having a >> little higher complexity, then partially whitening the input signal >> with the >> A(z/gamma) or A(z)/A(z/gamma) or even A(z/alpha)/A(z/beta) is a >> common approach used >> in the pitch estimators of many speech codecs (as I am sure you are >> well aware of), >> and it does help to reduce the chance of detecting the frequency of a >> strong first >> formant of the input speech signal as the pitch frequency. > > My concern wasn't that much about the period, but the gain. I'll > probably have to try both approaches, but I was mainly concerned that > a non-harmonic signal with one strong sinusoid could cause the comb > filter to be turned on with a high gain. > >> Hence, ideally we should have some decision logic and smoothing >> techniques to get a >> smooth contour of the true pitch period. I think the pitch estimator >> of the >> BroadVoice codecs (BV16 and BV32) that Broadcom previously submitted >> can serve as a >> good starting point for further development, because it was optimized >> to have >> relatively low complexity and it also has decision logic and >> smoothing built-in to >> try to get a smooth contour of the true pitch period for voiced >> regions of speech. > > CELT currently has a pitch predictor used for PLC, but I doubt it's > very good for this as it is now. OTOH, it's fairly low complexity too. > It's part of my plans to compare it with the one in BV16/32. I assume > that Broadcom would be willing to relicense it to the simplified BSD > license (it's currently LGPL) if we wanted to use it instead of the > current CELT > estimator. > >> It is optimized for 8 kHz or 16 kHz speech signals, so to use it for >> 48 kHz speech >> and music signals, we need to adjust the decimation factor and >> re-optimize it to >> work better for music as well. We also need to increase the range of >> the pitch >> period, or equivalently, the pitch frequency. Typical pitch frequency >> range for >> adult voice is somewhere around 50 to 400 Hz, but the pitch >> frequencies of music >> instruments cover a much wider range and can go up to 1600 Hz or >> higher. BV16 and >> BV32 detectable pitch ranges are around 59 Hz ~ 800 Hz and 60 Hz ~ >> 1600 Hz, >> respectively, because we previously already increased their pitch >> ranges to be able >> to detect the pitch of some of the music instruments. > > The pitch range is another thing that needs experimenting with, but my > intuition here is that it's probably not necessary to go below 100 Hz > because for these very low pitches, there is enough masking between > the harmonics to make the coding noise inaudible even without a > post-filter. Again worth testing. > >> Of course, if there is another royalty-free and unencumbered pitch >> estimator available that has even lower complexity than the BV16/BV32 >> pitch estimator and >> performs better, I am all for it. > > There's a few more starting with the SILK one, but it may be too > good/complex for what we need. The Speex one is fairly complex and not > that good, so it's not interesting for this. There's also the codec2 > low-bitrate vocoder, but I don't know how well its pitch estimator is. > > Cheers, > > Jean-Marc > _______________________________________________ codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec From jean-marc.valin@octasic.com Mon Oct 25 14:27:37 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 5D7FF3A6407 for ; Mon, 25 Oct 2010 14:27:37 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.599 X-Spam-Level: X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id oVgXUTc8cdxR for ; Mon, 25 Oct 2010 14:27:36 -0700 (PDT) Received: from relais.videotron.ca (relais.videotron.ca [24.201.245.36]) by core3.amsl.com (Postfix) with ESMTP id 008723A63D3 for ; Mon, 25 Oct 2010 14:27:35 -0700 (PDT) MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; charset=ISO-8859-1 Received: from [192.168.1.14] ([70.81.109.112]) by VL-MR-MRZ22.ip.videotron.ca (Sun Java(tm) System Messaging Server 6.3-8.01 (built Dec 16 2008; 32bit)) with ESMTP id <0LAV00F0I7OW09J0@VL-MR-MRZ22.ip.videotron.ca> for codec@ietf.org; Mon, 25 Oct 2010 17:29:21 -0400 (EDT) Message-id: <4CC5F6CE.3000005@octasic.com> Date: Mon, 25 Oct 2010 17:29:50 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.14) Gecko/20101006 Thunderbird/3.0.9 To: Koen Vos References: <971279185.1094841.1288040032498.JavaMail.root@lu2-zimbra> In-reply-to: <971279185.1094841.1288040032498.JavaMail.root@lu2-zimbra> Cc: "codec@ietf.org" Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Oct 2010 21:27:37 -0000 Hi Koen, Yes, using SILK's estimator is definitely a possibility. My only concern was that it may be a bit too complex for what we need. Think you'd be able to add a "low complexity" option to it? Otherwise, the options are to improve the existing CELT pitch estimator (it's small, fast, just not very good), or adapt either the estimator from either Broadvoice or the codec2. I haven't experimented enough to have any preference over any of these 4 options. One other question about SILK's pitch predictor. How hard does it work to prevent pitch doubling (at the expense of finding a slightly poorer correlation). What I observed (that Raymond had already warned about) is that pitch doubling is really a problem for the postfilter. Jean-Marc On 10-10-25 04:53 PM, Koen Vos wrote: > Hi, > > It may be a good idea to look at SILK's pitch estimator. Having duplicate pitch analysis within one codec is not ideal, and the one in SILK is quite mature/stable. It may need some changes to work with all of CELT's frame sizes, but it already works for 10 and 20 ms frames - good enough for testing. As an extra benefit, it efficiently codes the pitch evolution over time, which would benefit CELT too. > > koen. > > > ----- Original Message ----- > From: "Jean-Marc Valin" > To: "Jean-Marc Valin" > Cc: "codec@ietf.org" > Sent: Monday, October 25, 2010 12:31:40 PM > Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec > > Hi, > > I have checked in an initial implementation in the CELT exp_comb_filter > branch. Right now the quality is very bad, but that's mainly an > implementation problem. For example, the current pitch estimator I'm > using has atrocious behaviour when it comes to pitch > doubling/tripling/... Also, > I'm still suspecting bugs in the filter's overlap handling. > > Cheers, > > Jean-Marc > > On 10-10-23 08:35 PM, Jean-Marc Valin wrote: >> Hi Raymond, >> >> On 10-10-23 03:50 PM, Raymond (Juin-Hwey) Chen wrote: >>> Regarding your question, I think it really depends on what your main >>> emphasis is. >>> If you just want the best possible pitch period contour and don't >>> mind having a >>> little higher complexity, then partially whitening the input signal >>> with the >>> A(z/gamma) or A(z)/A(z/gamma) or even A(z/alpha)/A(z/beta) is a >>> common approach used >>> in the pitch estimators of many speech codecs (as I am sure you are >>> well aware of), >>> and it does help to reduce the chance of detecting the frequency of a >>> strong first >>> formant of the input speech signal as the pitch frequency. >> >> My concern wasn't that much about the period, but the gain. I'll >> probably have to try both approaches, but I was mainly concerned that >> a non-harmonic signal with one strong sinusoid could cause the comb >> filter to be turned on with a high gain. >> >>> Hence, ideally we should have some decision logic and smoothing >>> techniques to get a >>> smooth contour of the true pitch period. I think the pitch estimator >>> of the >>> BroadVoice codecs (BV16 and BV32) that Broadcom previously submitted >>> can serve as a >>> good starting point for further development, because it was optimized >>> to have >>> relatively low complexity and it also has decision logic and >>> smoothing built-in to >>> try to get a smooth contour of the true pitch period for voiced >>> regions of speech. >> >> CELT currently has a pitch predictor used for PLC, but I doubt it's >> very good for this as it is now. OTOH, it's fairly low complexity too. >> It's part of my plans to compare it with the one in BV16/32. I assume >> that Broadcom would be willing to relicense it to the simplified BSD >> license (it's currently LGPL) if we wanted to use it instead of the >> current CELT >> estimator. >> >>> It is optimized for 8 kHz or 16 kHz speech signals, so to use it for >>> 48 kHz speech >>> and music signals, we need to adjust the decimation factor and >>> re-optimize it to >>> work better for music as well. We also need to increase the range of >>> the pitch >>> period, or equivalently, the pitch frequency. Typical pitch frequency >>> range for >>> adult voice is somewhere around 50 to 400 Hz, but the pitch >>> frequencies of music >>> instruments cover a much wider range and can go up to 1600 Hz or >>> higher. BV16 and >>> BV32 detectable pitch ranges are around 59 Hz ~ 800 Hz and 60 Hz ~ >>> 1600 Hz, >>> respectively, because we previously already increased their pitch >>> ranges to be able >>> to detect the pitch of some of the music instruments. >> >> The pitch range is another thing that needs experimenting with, but my >> intuition here is that it's probably not necessary to go below 100 Hz >> because for these very low pitches, there is enough masking between >> the harmonics to make the coding noise inaudible even without a >> post-filter. Again worth testing. >> >>> Of course, if there is another royalty-free and unencumbered pitch >>> estimator available that has even lower complexity than the BV16/BV32 >>> pitch estimator and >>> performs better, I am all for it. >> >> There's a few more starting with the SILK one, but it may be too >> good/complex for what we need. The Speex one is fairly complex and not >> that good, so it's not interesting for this. There's also the codec2 >> low-bitrate vocoder, but I don't know how well its pitch estimator is. >> >> Cheers, >> >> Jean-Marc >> _______________________________________________ codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec > > _______________________________________________ codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > From koen.vos@skype.net Mon Oct 25 15:27:08 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id BCE283A68F1 for ; Mon, 25 Oct 2010 15:27:08 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.372 X-Spam-Level: X-Spam-Status: No, score=-2.372 tagged_above=-999 required=5 tests=[AWL=0.227, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id UncviBb6+8gZ for ; Mon, 25 Oct 2010 15:27:07 -0700 (PDT) Received: from mx.skype.net (mx.skype.net [78.141.177.88]) by core3.amsl.com (Postfix) with ESMTP id 4149A3A677C for ; Mon, 25 Oct 2010 15:27:07 -0700 (PDT) Received: from mx.skype.net (localhost [127.0.0.1]) by mx.skype.net (Postfix) with ESMTP id 5019E16E2; Tue, 26 Oct 2010 00:28:52 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=skype.net; h=date:from:to :cc:message-id:in-reply-to:subject:mime-version:content-type: content-transfer-encoding; s=mx; bh=cYeufXme0xSRAkF54OMkhZHEtGE= ; b=BZ45+BBEIe+XOwCo6NymVPjmivu5TLw0kZ3v2y4BcSTVGhzJGmKbTW9+smKF vbC6W4l9ca0LeVN4m3AnVvO+pk9K6OV00cmvHf+/oUtpFlO7Zvs8UiGpaebKtgfH rdsTYLAoyt2FK/I8ZzHSrAJQmst0X76IEvw+HYVeLXqPGbo= DomainKey-Signature: a=rsa-sha1; c=nofws; d=skype.net; h=date:from:to:cc :message-id:in-reply-to:subject:mime-version:content-type: content-transfer-encoding; q=dns; s=mx; b=CboBAyvPapmIjHhfqntfqE XrYf4STx48OkpNLV/c01Nmc1Vhpw1pN+SdZfbCOCFtZn5bbHddAlo2jbeu1xvIj4 5PGDeU29cDGGcNn4HSFCEp4nlAZxHyigj8xzHpsr3mzawabkPj6lW05DXXCJm7yT mXpUNsL0BHV81j0BNKZXs= Received: from zimbra.skype.net (zimbra.skype.net [78.141.177.82]) by mx.skype.net (Postfix) with ESMTP id 4B2877FD; Tue, 26 Oct 2010 00:28:52 +0200 (CEST) Received: from localhost (localhost [127.0.0.1]) by zimbra.skype.net (Postfix) with ESMTP id 2415D350865C; Tue, 26 Oct 2010 00:28:52 +0200 (CEST) X-Virus-Scanned: amavisd-new at lu2-zimbra.skype.net Received: from zimbra.skype.net ([127.0.0.1]) by localhost (zimbra.skype.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id LpPECgFUZJtw; Tue, 26 Oct 2010 00:28:51 +0200 (CEST) Received: from zimbra.skype.net (lu2-zimbra.skype.net [78.141.177.82]) by zimbra.skype.net (Postfix) with ESMTP id 3000E3508659; Tue, 26 Oct 2010 00:28:51 +0200 (CEST) Date: Tue, 26 Oct 2010 00:28:51 +0200 (CEST) From: Koen Vos To: Jean-Marc Valin Message-ID: <149562940.1096167.1288045731071.JavaMail.root@lu2-zimbra> In-Reply-To: <4CC5F6CE.3000005@octasic.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [69.181.192.115] X-Mailer: Zimbra 6.0.6_GA_2330.UBUNTU8_64 (ZimbraWebClient - FF3.0 (Win)/6.0.6_GA_2330.UBUNTU8_64) Cc: codec@ietf.org Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Oct 2010 22:27:08 -0000 Hi Jean-Marc, Yes, SILK has 3 complexity settings for the pitch estimator, you probably ran it in the highest setting. Also make sure to feed it a signal sampled at 16 or 24 kHz - complexity is obviously much higher with 48 kHz data for which it wasn't designed. I'd expect SILK's quality to be roughly equally sensitive to pitch doublings as CELT with pre/postfilter. There's a bias towards shorter lags to avoid these. And (partial) whitening is also quite important for this. koen. ----- Original Message ----- From: "Jean-Marc Valin" To: "Koen Vos" Cc: codec@ietf.org Sent: Monday, October 25, 2010 2:29:50 PM Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec Hi Koen, Yes, using SILK's estimator is definitely a possibility. My only concern was that it may be a bit too complex for what we need. Think you'd be able to add a "low complexity" option to it? Otherwise, the options are to improve the existing CELT pitch estimator (it's small, fast, just not very good), or adapt either the estimator from either Broadvoice or the codec2. I haven't experimented enough to have any preference over any of these 4 options. One other question about SILK's pitch predictor. How hard does it work to prevent pitch doubling (at the expense of finding a slightly poorer correlation). What I observed (that Raymond had already warned about) is that pitch doubling is really a problem for the postfilter. Jean-Marc On 10-10-25 04:53 PM, Koen Vos wrote: > Hi, > > It may be a good idea to look at SILK's pitch estimator. Having > duplicate pitch analysis within one codec is not ideal, and the one in > SILK is quite mature/stable. It may need some changes to work with all > of CELT's frame sizes, but it already works for 10 and 20 ms frames - > good enough for testing. As an extra benefit, it efficiently codes the > pitch evolution over time, which would benefit CELT too. > > koen. > > > ----- Original Message ----- > From: "Jean-Marc Valin" > To: "Jean-Marc Valin" > Cc: "codec@ietf.org" > Sent: Monday, October 25, 2010 12:31:40 PM > Subject: Re: [codec] A pitch filtering method to improve the mode 1 > CELT codec > > Hi, > > I have checked in an initial implementation in the CELT > exp_comb_filter branch. Right now the quality is very bad, but that's > mainly an > implementation problem. For example, the current pitch estimator I'm > using has atrocious behaviour when it comes to pitch > doubling/tripling/... Also, > I'm still suspecting bugs in the filter's overlap handling. > > Cheers, > > Jean-Marc > > On 10-10-23 08:35 PM, Jean-Marc Valin wrote: >> Hi Raymond, >> >> On 10-10-23 03:50 PM, Raymond (Juin-Hwey) Chen wrote: >>> Regarding your question, I think it really depends on what your main >>> emphasis is. >>> If you just want the best possible pitch period contour and don't >>> mind having a >>> little higher complexity, then partially whitening the input signal >>> with the >>> A(z/gamma) or A(z)/A(z/gamma) or even A(z/alpha)/A(z/beta) is a >>> common approach used >>> in the pitch estimators of many speech codecs (as I am sure you are >>> well aware of), >>> and it does help to reduce the chance of detecting the frequency of >>> a strong first >>> formant of the input speech signal as the pitch frequency. >> >> My concern wasn't that much about the period, but the gain. I'll >> probably have to try both approaches, but I was mainly concerned that >> a non-harmonic signal with one strong sinusoid could cause the comb >> filter to be turned on with a high gain. >> >>> Hence, ideally we should have some decision logic and smoothing >>> techniques to get a >>> smooth contour of the true pitch period. I think the pitch estimator >>> of the >>> BroadVoice codecs (BV16 and BV32) that Broadcom previously submitted >>> can serve as a >>> good starting point for further development, because it was >>> optimized to have >>> relatively low complexity and it also has decision logic and >>> smoothing built-in to >>> try to get a smooth contour of the true pitch period for voiced >>> regions of speech. >> >> CELT currently has a pitch predictor used for PLC, but I doubt it's >> very good for this as it is now. OTOH, it's fairly low complexity >> too. It's part of my plans to compare it with the one in BV16/32. I >> assume that Broadcom would be willing to relicense it to the >> simplified BSD >> license (it's currently LGPL) if we wanted to use it instead of the >> current CELT >> estimator. >> >>> It is optimized for 8 kHz or 16 kHz speech signals, so to use it for >>> 48 kHz speech >>> and music signals, we need to adjust the decimation factor and >>> re-optimize it to >>> work better for music as well. We also need to increase the range of >>> the pitch >>> period, or equivalently, the pitch frequency. Typical pitch >>> frequency range for >>> adult voice is somewhere around 50 to 400 Hz, but the pitch >>> frequencies of music >>> instruments cover a much wider range and can go up to 1600 Hz or >>> higher. BV16 and >>> BV32 detectable pitch ranges are around 59 Hz ~ 800 Hz and 60 Hz ~ >>> 1600 Hz, >>> respectively, because we previously already increased their pitch >>> ranges to be able >>> to detect the pitch of some of the music instruments. >> >> The pitch range is another thing that needs experimenting with, but >> my intuition here is that it's probably not necessary to go below 100 >> Hz because for these very low pitches, there is enough masking >> between the harmonics to make the coding noise inaudible even without >> a post-filter. Again worth testing. >> >>> Of course, if there is another royalty-free and unencumbered pitch >>> estimator available that has even lower complexity than the >>> BV16/BV32 pitch estimator and >>> performs better, I am all for it. >> >> There's a few more starting with the SILK one, but it may be too >> good/complex for what we need. The Speex one is fairly complex and >> not that good, so it's not interesting for this. There's also the >> codec2 low-bitrate vocoder, but I don't know how well its pitch >> estimator is. >> >> Cheers, >> >> Jean-Marc >> _______________________________________________ codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec > > _______________________________________________ codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > _______________________________________________ codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > From jean-marc.valin@usherbrooke.ca Mon Oct 25 19:11:09 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 5DA573A6C71 for ; Mon, 25 Oct 2010 19:11:09 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.224 X-Spam-Level: X-Spam-Status: No, score=-1.224 tagged_above=-999 required=5 tests=[AWL=-0.791, BAYES_00=-2.599, FF_IHOPE_YOU_SINK=2.166] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id BbbtDpw6mNKL for ; Mon, 25 Oct 2010 19:11:05 -0700 (PDT) Received: from relais.videotron.ca (relais.videotron.ca [24.201.245.36]) by core3.amsl.com (Postfix) with ESMTP id B10E23A6C1B for ; Mon, 25 Oct 2010 19:10:57 -0700 (PDT) MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; charset=ISO-8859-1 Received: from [192.168.1.14] ([70.81.109.112]) by vl-mo-mrz23.ip.videotron.ca (Sun Java(tm) System Messaging Server 6.3-8.01 (built Dec 16 2008; 32bit)) with ESMTP id <0LAV00C0FKSZLW00@vl-mo-mrz23.ip.videotron.ca> for codec@ietf.org; Mon, 25 Oct 2010 22:12:36 -0400 (EDT) Message-id: <4CC63939.9030300@usherbrooke.ca> Date: Mon, 25 Oct 2010 22:13:13 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.14) Gecko/20101006 Thunderbird/3.0.9 To: "Raymond (Juin-Hwey) Chen" References: <4CC2DA7E.1060705@usherbrooke.ca> In-reply-to: X-Enigmail-Version: 1.0.1 Cc: "'codec@ietf.org'" Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Oct 2010 02:11:09 -0000 Hi Raymond, I've just realised that changes in the pitch and/or gain are causing significant artefacts. While the pitch estimator should definitely mimimise these changes, I still think they'll be unavoidable. At this point, I'm not sure whether it's an implementation bug or an issue with the algorithm. I've copied below the function I use the the pre-filter and post-filter. In the case of the pre-filter, the gains are negative, while for the post-filter, x and y point to the same array. Can you see something obviously wrong with this code? Otherwise, any idea what the problem may be? Note that I *do* get perfect reconstruction when the pitch and gain are constant, or when there's no codec in the middle. void comb_filter(float *y, float *x, int T0, int T1, int N, float g0, float g1, int overlap) { int i; for (i=0;i Hi Jean-Marc, > > A few more comments about the technical description document I sent last night: > > (1) I should have clarified: when I said "frame size" in my document, I was > referring to the frame size for pitch estimation and pitch filter tap calculation, > not the CELT codec frame size. For ease of implementation, the boundaries of the > pitch frames and codec frames should obviously be aligned. Given that the CELT-only > mode of the IETF codec has possible frame sizes of 2.5, 5, 10, and 20 ms, I would > recommend that we update the pitch period and pitch filter tap(s) at a rate of once > every 5 ms if the CELT frame size is 5, 10, or 20 ms, because updating them once > every 10 or 20 ms is probably too infrequent and probably will decrease the system's > effectiveness in the reduction of coding distortion (although it doesn't hurt for > you to try updating them every 10 or 20 ms to be the same as the codec frame size > just to see what happens). On the other hand, if the CELT frame size is 2.5 ms in > order to achieve an ultra-low delay, then it doesn't make sense for the pitch > prefilter/postfilter to use a 5 ms frame size since it will just increase the system > delay. In this case, even though 2.5 ms pitch frame size may be more frequent than > necessary and will double the pitch side information bit-rate, it is still a > necessity just to enable CELT to achieve the ultra-low delay. > > (2) On a second thought, I think to calculate the normalized correlation c or the > optimal tap weight beta for a single-tap pitch predictor, rather than calculating > the summation over the current frame (n from 1 to L), it probably makes more sense > for the summation length to be the extracted pitch period or the pitch frame size, > whichever is larger. The motivation is this: suppose the pitch frame size is 2.5 ms > or 5 ms and the pitch period is 15 ms, then trying to measure the level of > periodicity using only 1/6 to 1/3 of the pitch cycle waveform seems dangerous or at > least unreliable. Thus, if the detected pitch period is longer than the frame size, > then we should sum over all of the samples in the current frame plus more samples in > the previous frame(s) until the total number of samples in the summation equals the > pitch period. If the pitch period is smaller than the frame size and the pitch > frame size is only 2.5 ms or 5 ms, I recommend just summing over the entire current > frame. > > (3) If we use multi-tap pitch filters in order to taper off the comb filtering > effect at higher frequencies, we need to be careful about the stability of the IIR > pitch filter, especially if the IIR filter is in the decoder side (as is the case in > my example single-tap all-pole pitch postfilter). I suppose you probably already > know this, but I would like to mention and discuss it anyway. Even though the > multi-tap IIR pitch postfilter can be an exact inverse filter of a corresponding > pitch prefilter and supposedly is just "undoing" the effect of the pitch prefilter, > the quantization errors in the CELT codec and even the effects caused by channel > errors can potentially make the output of the multi-tap IIR pitch postfilter "go > wild" (have very large amplitude or even oscillate) if that filter is unstable or > even if it is only close to unstable (i.e. all poles are inside the unit circle but > too close to the unit circle). If the IIR pitch filter is a one-tap filter, > ensuring the stability is trivial. If it is a 2-tap or 3-tap filter, Professor > Peter Kabal and his student Ravi Ramachandran has a nice paper on a stability test > and a stabilization technique that can be used to ensure the stability of such a > filter. Their paper is published in the IEEE Transactions on ASSP, July 1987. We > can use that stabilization technique to stabilize every single candidate 2-tap or 3- > tap filter coefficient set that we will transmit and use, with enough safety margin > to instability, to ensure that a 2-tap or 3-tap IIR pitch postfilter will not > amplify the output audio signal wildly. > > Alternatively, we can use the second form of the pitch prefilter and postfilter that > I mentioned in my document so that the pitch postfilter is a 2-tap or 3-tap FIR > filter in the form of H(z) = 1 + a1 * z^(-p+1) + a2 * z^-p + a3 * z^(-p-1), with a1 > or a3 equals to zero in the case of a 2-tap filter. In this case, we will move the > multi-tap IIR filter from the decoder side to the encoder side where it is applied > directly to the input audio signal and thus we do not have to worry about CELT > quantization errors or channel errors driving a "nearly unstable" filter wild. Of > course, we do not want the IIR prefilter there to be too close to instability, > either, otherwise the potential wild swing in its output may still cause problems > for the CELT codec. The main advantage of this alternative arrangement is that the > pitch postfilter is now an FIR filter, which is guaranteed to be stable and no > amount of CELT coding distortion or channel error will ever "drive it crazy and > wild" like they can for a multi-tap IIR pitch postfilter that has not been > stabilized or has not maintained a sufficient safety margin to instability. The > main disadvantage of this arrangement is that with an FIR pitch postfilter, its > frequency response has downward spikes located at pitch harmonics, and it will tend > to shape the coding noise that way, too. A noise spectral shape with downward > spikes is probably not as ideal as the noise spectral shape with upward spikes that > the all-pole pitch postfilter provides. This is more so when the parameter bmax is > closer to 1 but probably won't be too big a deal if bmax is in the neighborhood of > 0.4 to 0.6. > > Raymond > > -----Original Message----- > From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Raymond (Juin-Hwey) Chen > Sent: Saturday, October 23, 2010 12:50 PM > To: Jean-Marc Valin > Cc: 'codec@ietf.org' > Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec > Importance: Low > > Hi Jean-Marc, > > You are very welcome. Glad to contribute and glad to know that the document is good > enough for you to start implementing the techniques. > > Regarding your question, I think it really depends on what your main emphasis is. > If you just want the best possible pitch period contour and don't mind having a > little higher complexity, then partially whitening the input signal with the > A(z/gamma) or A(z)/A(z/gamma) or even A(z/alpha)/A(z/beta) is a common approach used > in the pitch estimators of many speech codecs (as I am sure you are well aware of), > and it does help to reduce the chance of detecting the frequency of a strong first > formant of the input speech signal as the pitch frequency. > > If, on the other hand, your main emphasis is to get a codec complexity as low as > possible, I think it is possible to get a reasonably good pitch period contour by > using the input signal directly without partial whitening filtering, thus saving the > complexity of the associated LPC analysis and filtering (which can be relatively > high for the 48 kHz sampling rate). Note that when I said "Any low-complexity pitch > estimator can be used here..." in the document I sent last night, I used the term > "low-complexity" because I thought you told me you tried to keep the complexity of > CELT as low as possible. Since this pitch prefiltering/postfiltering approach is > meant to be used with CELT, I thought it is only natural to make the complexity of > the pitch estimator as low as possible. That's why I used the term "low-complexity > pitch estimator" in that document. > > I think for this pitch prefiltering/postfiltering approach to work well, what's most > important is that we need to do everything we can to ensure that the pitch period we > extract is the true pitch period rather than an integer multiple or sub-multiple of > the pitch period. Unlike speech coding or packet loss concealment applications > where double pitch period or triple pitch period still gives you decent pitch > prediction or replacement waveform, here double or triple pitch period will create > one or two unwanted frequency response peak(s) in the valley regions between pitch > harmonic peaks and thus will greatly reduce the noise reduction achievable by the > noise spectral shaping effect of the pitch prefilter/postfilter approach. > > Hence, ideally we should have some decision logic and smoothing techniques to get a > smooth contour of the true pitch period. I think the pitch estimator of the > BroadVoice codecs (BV16 and BV32) that Broadcom previously submitted can serve as a > good starting point for further development, because it was optimized to have > relatively low complexity and it also has decision logic and smoothing built-in to > try to get a smooth contour of the true pitch period for voiced regions of speech. > It is optimized for 8 kHz or 16 kHz speech signals, so to use it for 48 kHz speech > and music signals, we need to adjust the decimation factor and re-optimize it to > work better for music as well. We also need to increase the range of the pitch > period, or equivalently, the pitch frequency. Typical pitch frequency range for > adult voice is somewhere around 50 to 400 Hz, but the pitch frequencies of music > instruments cover a much wider range and can go up to 1600 Hz or higher. BV16 and > BV32 detectable pitch ranges are around 59 Hz ~ 800 Hz and 60 Hz ~ 1600 Hz, > respectively, because we previously already increased their pitch ranges to be able > to detect the pitch of some of the music instruments. > > Of course, if there is another royalty-free and unencumbered pitch estimator > available that has even lower complexity than the BV16/BV32 pitch estimator and > performs better, I am all for it. > > Raymond > > -----Original Message----- > From: Jean-Marc Valin [mailto:jean-marc.valin@usherbrooke.ca] > Sent: Saturday, October 23, 2010 5:52 AM > To: Raymond (Juin-Hwey) Chen > Cc: 'codec@ietf.org' > Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec > > Hi Raymond, > > Thank you very much for your contribution and the highly detailed > description you just sent. I believe I have enough information to start > implementing it. At this point, I only have one minor question. When > estimating the pitch period/gain, do you recommend working on the input > signal directly or should the input be "whitened" using a filter like > A(z/gamma) (where A(z) is the LPC filter)? > > Thanks again, > > Jean-Marc > > On 10-10-23 01:13 AM, Raymond (Juin-Hwey) Chen wrote: >> All, >> >> In the last few weeks, I have been working on a pitch prefiltering/postfiltering method to try to >> improve the output audio quality of the IETF mode 1 CELT codec for certain audio signals that CELT >> low-delay modes give more audible coding distortion than usual (typically nearly periodic solo >> music instrument signals). I am happy to report that this method has achieved fairly significant >> audio quality improvement for many of these kinds of audio signals that I tried. This is achieved >> without increasing the coding delay and with only a slight increase in the codec complexity and >> encoding bit-rate (a potential increase of around 2 kb/s or so; less if a range coder is used). >> >> I have been in discussion with Jean-Marc about this and have sent him CELT-coded audio signals at >> different bit-rates obtained with or without this method. After listening, he agreed that the >> method made a significant improvement when compared with the current CELT implementation, and he >> would like to try to incorporate this method into his CELT C code to evaluate and verify the >> performance gain and possibly further develop the method. >> >> Jean-Marc asked me to write a detailed description of the method in a document and send the >> document to the IETF codec WG email reflector so he can read it, understand it, and then implement >> it himself in his CELT C code. I am attaching such a technical description of the method. >> >> Note that although this document is 7 pages long, the pitch prefiltering/postfiltering method is >> actually quite simple. The document is long only because (1) I tried to describe the technical >> details as much as I can so Jean-Marc can implement it exactly as I had it without any ambiguity, >> and (2) I also described/discussed a few alternative approaches for some of the algorithm >> components in case we need to go those routes. >> >> Note also that this is "fresh out of the oven" as I only had time to finish a very rough first cut >> for a proof-of-concept simulation and did not have time yet to refine or tune it. Thus, I am sure >> there are still many rough edges that need to be smoothed out and many areas that can be improved >> upon. However, I am very much encouraged that even such a rough first cut showed a lot of promise >> as it made a very large audio quality improvement at several different bit-rates for a trumpet >> test signal, which is a signal that CELT low-delay modes have problems with. Therefore, after >> this pitch prefiltering/postfiltering method is integrated into CELT and further refined/tuned, I >> think it has the potential to make the IETF mode 1 CELT codec more robust to different audio >> signals. That is, it can potentially improve CELT performance for many signals that are currently >> difficult for CELT to code well, and thus ensure a more uniform audio quality level across a wider >> variety of different audio signals. If Jean-Marc's integration, testing, and further development >> of this pitch filtering method confirms this potential, then it is proposed that this pitch >> prefiltering/postfiltering method be incorporated into Mode 1 of the IETF Opus codec to make the >> codec more robust to different input signal types. >> >> Cheers, >> >> Raymond >> >> >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > > > From jean-marc.valin@usherbrooke.ca Mon Oct 25 20:27:15 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 5B8723A67DF for ; Mon, 25 Oct 2010 20:27:15 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.182 X-Spam-Level: X-Spam-Status: No, score=-1.182 tagged_above=-999 required=5 tests=[AWL=-0.749, BAYES_00=-2.599, FF_IHOPE_YOU_SINK=2.166] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id BtOZK61qZXWr for ; Mon, 25 Oct 2010 20:27:11 -0700 (PDT) Received: from relais.videotron.ca (relais.videotron.ca [24.201.245.36]) by core3.amsl.com (Postfix) with ESMTP id 97E253A67A8 for ; Mon, 25 Oct 2010 20:27:11 -0700 (PDT) MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; charset=ISO-8859-1 Received: from [192.168.1.14] ([70.81.109.112]) by vl-mo-mrz23.ip.videotron.ca (Sun Java(tm) System Messaging Server 6.3-8.01 (built Dec 16 2008; 32bit)) with ESMTP id <0LAV00CE9OC0LWA0@vl-mo-mrz23.ip.videotron.ca> for codec@ietf.org; Mon, 25 Oct 2010 23:28:50 -0400 (EDT) Message-id: <4CC64B17.70203@usherbrooke.ca> Date: Mon, 25 Oct 2010 23:29:27 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.14) Gecko/20101006 Thunderbird/3.0.9 To: Jean-Marc Valin References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC63939.9030300@usherbrooke.ca> In-reply-to: <4CC63939.9030300@usherbrooke.ca> X-Enigmail-Version: 1.0.1 Cc: "'codec@ietf.org'" Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Oct 2010 03:27:15 -0000 Sorry for replying to myself, but before anyone wastes time on this, I'm fairly sure I found what my error was. It wasn't related to the filter itself, but the fact that I forgot to compensate for the 2.5 ms delay introduced by the codec. Jean-Marc On 10-10-25 10:13 PM, Jean-Marc Valin wrote: > Hi Raymond, > > I've just realised that changes in the pitch and/or gain are causing > significant artefacts. While the pitch estimator should definitely > mimimise these changes, I still think they'll be unavoidable. At this > point, I'm not sure whether it's an implementation bug or an issue with > the algorithm. > > I've copied below the function I use the the pre-filter and post-filter. > In the case of the pre-filter, the gains are negative, while for the > post-filter, x and y point to the same array. Can you see something > obviously wrong with this code? Otherwise, any idea what the problem may > be? Note that I *do* get perfect reconstruction when the pitch and gain > are constant, or when there's no codec in the middle. > > > void comb_filter(float *y, float *x, int T0, int T1, int N, > float g0, float g1, int overlap) > { > int i; > for (i=0;i { > float f; > f = i/(float)overlap; > y[i] = x[i] + (1.-f)*g0*x[i-T0] + f*g1*x[i-T1]; > } > for (i=overlap;i y[i] = x[i] + g1*x[i-T1]; > } > > > Cheers, > > Jean-Marc > > > On 10-10-23 05:56 PM, Raymond (Juin-Hwey) Chen wrote: >> Hi Jean-Marc, >> >> A few more comments about the technical description document I sent last night: >> >> (1) I should have clarified: when I said "frame size" in my document, I was >> referring to the frame size for pitch estimation and pitch filter tap calculation, >> not the CELT codec frame size. For ease of implementation, the boundaries of the >> pitch frames and codec frames should obviously be aligned. Given that the CELT-only >> mode of the IETF codec has possible frame sizes of 2.5, 5, 10, and 20 ms, I would >> recommend that we update the pitch period and pitch filter tap(s) at a rate of once >> every 5 ms if the CELT frame size is 5, 10, or 20 ms, because updating them once >> every 10 or 20 ms is probably too infrequent and probably will decrease the system's >> effectiveness in the reduction of coding distortion (although it doesn't hurt for >> you to try updating them every 10 or 20 ms to be the same as the codec frame size >> just to see what happens). On the other hand, if the CELT frame size is 2.5 ms in >> order to achieve an ultra-low delay, then it doesn't make sense for the pitch >> prefilter/postfilter to use a 5 ms frame size since it will just increase the system >> delay. In this case, even though 2.5 ms pitch frame size may be more frequent than >> necessary and will double the pitch side information bit-rate, it is still a >> necessity just to enable CELT to achieve the ultra-low delay. >> >> (2) On a second thought, I think to calculate the normalized correlation c or the >> optimal tap weight beta for a single-tap pitch predictor, rather than calculating >> the summation over the current frame (n from 1 to L), it probably makes more sense >> for the summation length to be the extracted pitch period or the pitch frame size, >> whichever is larger. The motivation is this: suppose the pitch frame size is 2.5 ms >> or 5 ms and the pitch period is 15 ms, then trying to measure the level of >> periodicity using only 1/6 to 1/3 of the pitch cycle waveform seems dangerous or at >> least unreliable. Thus, if the detected pitch period is longer than the frame size, >> then we should sum over all of the samples in the current frame plus more samples in >> the previous frame(s) until the total number of samples in the summation equals the >> pitch period. If the pitch period is smaller than the frame size and the pitch >> frame size is only 2.5 ms or 5 ms, I recommend just summing over the entire current >> frame. >> >> (3) If we use multi-tap pitch filters in order to taper off the comb filtering >> effect at higher frequencies, we need to be careful about the stability of the IIR >> pitch filter, especially if the IIR filter is in the decoder side (as is the case in >> my example single-tap all-pole pitch postfilter). I suppose you probably already >> know this, but I would like to mention and discuss it anyway. Even though the >> multi-tap IIR pitch postfilter can be an exact inverse filter of a corresponding >> pitch prefilter and supposedly is just "undoing" the effect of the pitch prefilter, >> the quantization errors in the CELT codec and even the effects caused by channel >> errors can potentially make the output of the multi-tap IIR pitch postfilter "go >> wild" (have very large amplitude or even oscillate) if that filter is unstable or >> even if it is only close to unstable (i.e. all poles are inside the unit circle but >> too close to the unit circle). If the IIR pitch filter is a one-tap filter, >> ensuring the stability is trivial. If it is a 2-tap or 3-tap filter, Professor >> Peter Kabal and his student Ravi Ramachandran has a nice paper on a stability test >> and a stabilization technique that can be used to ensure the stability of such a >> filter. Their paper is published in the IEEE Transactions on ASSP, July 1987. We >> can use that stabilization technique to stabilize every single candidate 2-tap or 3- >> tap filter coefficient set that we will transmit and use, with enough safety margin >> to instability, to ensure that a 2-tap or 3-tap IIR pitch postfilter will not >> amplify the output audio signal wildly. >> >> Alternatively, we can use the second form of the pitch prefilter and postfilter that >> I mentioned in my document so that the pitch postfilter is a 2-tap or 3-tap FIR >> filter in the form of H(z) = 1 + a1 * z^(-p+1) + a2 * z^-p + a3 * z^(-p-1), with a1 >> or a3 equals to zero in the case of a 2-tap filter. In this case, we will move the >> multi-tap IIR filter from the decoder side to the encoder side where it is applied >> directly to the input audio signal and thus we do not have to worry about CELT >> quantization errors or channel errors driving a "nearly unstable" filter wild. Of >> course, we do not want the IIR prefilter there to be too close to instability, >> either, otherwise the potential wild swing in its output may still cause problems >> for the CELT codec. The main advantage of this alternative arrangement is that the >> pitch postfilter is now an FIR filter, which is guaranteed to be stable and no >> amount of CELT coding distortion or channel error will ever "drive it crazy and >> wild" like they can for a multi-tap IIR pitch postfilter that has not been >> stabilized or has not maintained a sufficient safety margin to instability. The >> main disadvantage of this arrangement is that with an FIR pitch postfilter, its >> frequency response has downward spikes located at pitch harmonics, and it will tend >> to shape the coding noise that way, too. A noise spectral shape with downward >> spikes is probably not as ideal as the noise spectral shape with upward spikes that >> the all-pole pitch postfilter provides. This is more so when the parameter bmax is >> closer to 1 but probably won't be too big a deal if bmax is in the neighborhood of >> 0.4 to 0.6. >> >> Raymond >> >> -----Original Message----- >> From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Raymond (Juin-Hwey) Chen >> Sent: Saturday, October 23, 2010 12:50 PM >> To: Jean-Marc Valin >> Cc: 'codec@ietf.org' >> Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec >> Importance: Low >> >> Hi Jean-Marc, >> >> You are very welcome. Glad to contribute and glad to know that the document is good >> enough for you to start implementing the techniques. >> >> Regarding your question, I think it really depends on what your main emphasis is. >> If you just want the best possible pitch period contour and don't mind having a >> little higher complexity, then partially whitening the input signal with the >> A(z/gamma) or A(z)/A(z/gamma) or even A(z/alpha)/A(z/beta) is a common approach used >> in the pitch estimators of many speech codecs (as I am sure you are well aware of), >> and it does help to reduce the chance of detecting the frequency of a strong first >> formant of the input speech signal as the pitch frequency. >> >> If, on the other hand, your main emphasis is to get a codec complexity as low as >> possible, I think it is possible to get a reasonably good pitch period contour by >> using the input signal directly without partial whitening filtering, thus saving the >> complexity of the associated LPC analysis and filtering (which can be relatively >> high for the 48 kHz sampling rate). Note that when I said "Any low-complexity pitch >> estimator can be used here..." in the document I sent last night, I used the term >> "low-complexity" because I thought you told me you tried to keep the complexity of >> CELT as low as possible. Since this pitch prefiltering/postfiltering approach is >> meant to be used with CELT, I thought it is only natural to make the complexity of >> the pitch estimator as low as possible. That's why I used the term "low-complexity >> pitch estimator" in that document. >> >> I think for this pitch prefiltering/postfiltering approach to work well, what's most >> important is that we need to do everything we can to ensure that the pitch period we >> extract is the true pitch period rather than an integer multiple or sub-multiple of >> the pitch period. Unlike speech coding or packet loss concealment applications >> where double pitch period or triple pitch period still gives you decent pitch >> prediction or replacement waveform, here double or triple pitch period will create >> one or two unwanted frequency response peak(s) in the valley regions between pitch >> harmonic peaks and thus will greatly reduce the noise reduction achievable by the >> noise spectral shaping effect of the pitch prefilter/postfilter approach. >> >> Hence, ideally we should have some decision logic and smoothing techniques to get a >> smooth contour of the true pitch period. I think the pitch estimator of the >> BroadVoice codecs (BV16 and BV32) that Broadcom previously submitted can serve as a >> good starting point for further development, because it was optimized to have >> relatively low complexity and it also has decision logic and smoothing built-in to >> try to get a smooth contour of the true pitch period for voiced regions of speech. >> It is optimized for 8 kHz or 16 kHz speech signals, so to use it for 48 kHz speech >> and music signals, we need to adjust the decimation factor and re-optimize it to >> work better for music as well. We also need to increase the range of the pitch >> period, or equivalently, the pitch frequency. Typical pitch frequency range for >> adult voice is somewhere around 50 to 400 Hz, but the pitch frequencies of music >> instruments cover a much wider range and can go up to 1600 Hz or higher. BV16 and >> BV32 detectable pitch ranges are around 59 Hz ~ 800 Hz and 60 Hz ~ 1600 Hz, >> respectively, because we previously already increased their pitch ranges to be able >> to detect the pitch of some of the music instruments. >> >> Of course, if there is another royalty-free and unencumbered pitch estimator >> available that has even lower complexity than the BV16/BV32 pitch estimator and >> performs better, I am all for it. >> >> Raymond >> >> -----Original Message----- >> From: Jean-Marc Valin [mailto:jean-marc.valin@usherbrooke.ca] >> Sent: Saturday, October 23, 2010 5:52 AM >> To: Raymond (Juin-Hwey) Chen >> Cc: 'codec@ietf.org' >> Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec >> >> Hi Raymond, >> >> Thank you very much for your contribution and the highly detailed >> description you just sent. I believe I have enough information to start >> implementing it. At this point, I only have one minor question. When >> estimating the pitch period/gain, do you recommend working on the input >> signal directly or should the input be "whitened" using a filter like >> A(z/gamma) (where A(z) is the LPC filter)? >> >> Thanks again, >> >> Jean-Marc >> >> On 10-10-23 01:13 AM, Raymond (Juin-Hwey) Chen wrote: >>> All, >>> >>> In the last few weeks, I have been working on a pitch prefiltering/postfiltering method to try to >>> improve the output audio quality of the IETF mode 1 CELT codec for certain audio signals that CELT >>> low-delay modes give more audible coding distortion than usual (typically nearly periodic solo >>> music instrument signals). I am happy to report that this method has achieved fairly significant >>> audio quality improvement for many of these kinds of audio signals that I tried. This is achieved >>> without increasing the coding delay and with only a slight increase in the codec complexity and >>> encoding bit-rate (a potential increase of around 2 kb/s or so; less if a range coder is used). >>> >>> I have been in discussion with Jean-Marc about this and have sent him CELT-coded audio signals at >>> different bit-rates obtained with or without this method. After listening, he agreed that the >>> method made a significant improvement when compared with the current CELT implementation, and he >>> would like to try to incorporate this method into his CELT C code to evaluate and verify the >>> performance gain and possibly further develop the method. >>> >>> Jean-Marc asked me to write a detailed description of the method in a document and send the >>> document to the IETF codec WG email reflector so he can read it, understand it, and then implement >>> it himself in his CELT C code. I am attaching such a technical description of the method. >>> >>> Note that although this document is 7 pages long, the pitch prefiltering/postfiltering method is >>> actually quite simple. The document is long only because (1) I tried to describe the technical >>> details as much as I can so Jean-Marc can implement it exactly as I had it without any ambiguity, >>> and (2) I also described/discussed a few alternative approaches for some of the algorithm >>> components in case we need to go those routes. >>> >>> Note also that this is "fresh out of the oven" as I only had time to finish a very rough first cut >>> for a proof-of-concept simulation and did not have time yet to refine or tune it. Thus, I am sure >>> there are still many rough edges that need to be smoothed out and many areas that can be improved >>> upon. However, I am very much encouraged that even such a rough first cut showed a lot of promise >>> as it made a very large audio quality improvement at several different bit-rates for a trumpet >>> test signal, which is a signal that CELT low-delay modes have problems with. Therefore, after >>> this pitch prefiltering/postfiltering method is integrated into CELT and further refined/tuned, I >>> think it has the potential to make the IETF mode 1 CELT codec more robust to different audio >>> signals. That is, it can potentially improve CELT performance for many signals that are currently >>> difficult for CELT to code well, and thus ensure a more uniform audio quality level across a wider >>> variety of different audio signals. If Jean-Marc's integration, testing, and further development >>> of this pitch filtering method confirms this potential, then it is proposed that this pitch >>> prefiltering/postfiltering method be incorporated into Mode 1 of the IETF Opus codec to make the >>> codec more robust to different input signal types. >>> >>> Cheers, >>> >>> Raymond >>> >>> >>> >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >> >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >> >> >> >> > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > From rchen@broadcom.com Mon Oct 25 23:03:11 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 0153D3A681F for ; Mon, 25 Oct 2010 23:03:09 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.372 X-Spam-Level: X-Spam-Status: No, score=-1.372 tagged_above=-999 required=5 tests=[AWL=-0.939, BAYES_00=-2.599, FF_IHOPE_YOU_SINK=2.166] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id GG67nn29qn6I for ; Mon, 25 Oct 2010 23:03:02 -0700 (PDT) Received: from mms1.broadcom.com (mms1.broadcom.com [216.31.210.17]) by core3.amsl.com (Postfix) with ESMTP id 71DA23A6845 for ; Mon, 25 Oct 2010 23:03:01 -0700 (PDT) Received: from [10.9.200.131] by mms1.broadcom.com with ESMTP (Broadcom SMTP Relay (Email Firewall v6.3.2)); Mon, 25 Oct 2010 23:04:36 -0700 X-Server-Uuid: 02CED230-5797-4B57-9875-D5D2FEE4708A Received: from IRVEXCHCCR01.corp.ad.broadcom.com ([10.252.49.30]) by IRVEXCHHUB01.corp.ad.broadcom.com ([10.9.200.131]) with mapi; Mon, 25 Oct 2010 23:04:36 -0700 From: "Raymond (Juin-Hwey) Chen" To: "Jean-Marc Valin" Importance: low X-Priority: 5 Date: Mon, 25 Oct 2010 23:04:31 -0700 Thread-Topic: [codec] A pitch filtering method to improve the mode 1 CELT codec Thread-Index: Act0vfUbniWjzxFiTEm8uAthj3hCFwAFAbiw Message-ID: References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC63939.9030300@usherbrooke.ca> <4CC64B17.70203@usherbrooke.ca> In-Reply-To: <4CC64B17.70203@usherbrooke.ca> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-cr-hashedpuzzle: AVK3 Ah4G BJdb CT/Q CW3u J3SK Kc85 M8rF NxSw QZbZ RJhW RVYE Rs9J SyCo UFVq VWA/; 2; YwBvAGQAZQBjAEAAaQBlAHQAZgAuAG8AcgBnADsAagBlAGEAbgAtAG0AYQByAGMALgB2AGEAbABpAG4AQAB1AHMAaABlAHIAYgByAG8AbwBrAGUALgBjAGEA; Sosha1_v1; 7; {E177A84F-6402-4183-91C9-D79AC6D865C3}; cgBjAGgAZQBuAEAAYgByAG8AYQBkAGMAbwBtAC4AYwBvAG0A; Tue, 26 Oct 2010 06:04:31 GMT; UgBFADoAIABbAGMAbwBkAGUAYwBdACAAQQAgAHAAaQB0AGMAaAAgAGYAaQBsAHQAZQByAGkAbgBnACAAbQBlAHQAaABvAGQAIAB0AG8AIABpAG0AcAByAG8AdgBlACAAdABoAGUAIABtAG8AZABlACAAMQAgAEMARQBMAFQAIABjAG8AZABlAGMA x-cr-puzzleid: {E177A84F-6402-4183-91C9-D79AC6D865C3} acceptlanguage: en-US MIME-Version: 1.0 X-WSS-ID: 60D8B0FE47836251552-01-01 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Cc: "'codec@ietf.org'" Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Oct 2010 06:03:11 -0000 Hi Jean-Marc, Sorry for the late reply. As I mentioned to you before, I have been and wo= uld be in an off-site conference for a few days and I generally don't have = Internet access there during the day, so I could not respond earlier. Sorr= y. Glad to hear that you found the error. To verify the performance of your p= refilter/postfilter implementation, I suggest that you try to use the same = trumpet signal that I used before and test the prefilter followed by postfi= lter with and without CELT coding, and then listen to them and compare them= with the output signals I sent you previously to see if you get similar au= dio quality for the same condition. Raymond -----Original Message----- From: Jean-Marc Valin [mailto:jean-marc.valin@usherbrooke.ca] Sent: Monday, October 25, 2010 8:29 PM To: Jean-Marc Valin Cc: Raymond (Juin-Hwey) Chen; 'codec@ietf.org' Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT co= dec Sorry for replying to myself, but before anyone wastes time on this, I'm fairly sure I found what my error was. It wasn't related to the filter itself, but the fact that I forgot to compensate for the 2.5 ms delay introduced by the codec. Jean-Marc On 10-10-25 10:13 PM, Jean-Marc Valin wrote: > Hi Raymond, > > I've just realised that changes in the pitch and/or gain are causing > significant artefacts. While the pitch estimator should definitely > mimimise these changes, I still think they'll be unavoidable. At this > point, I'm not sure whether it's an implementation bug or an issue with > the algorithm. > > I've copied below the function I use the the pre-filter and post-filter. > In the case of the pre-filter, the gains are negative, while for the > post-filter, x and y point to the same array. Can you see something > obviously wrong with this code? Otherwise, any idea what the problem may > be? Note that I *do* get perfect reconstruction when the pitch and gain > are constant, or when there's no codec in the middle. > > > void comb_filter(float *y, float *x, int T0, int T1, int N, > float g0, float g1, int overlap) > { > int i; > for (i=3D0;i { > float f; > f =3D i/(float)overlap; > y[i] =3D x[i] + (1.-f)*g0*x[i-T0] + f*g1*x[i-T1]; > } > for (i=3Doverlap;i y[i] =3D x[i] + g1*x[i-T1]; > } > > > Cheers, > > Jean-Marc > > > On 10-10-23 05:56 PM, Raymond (Juin-Hwey) Chen wrote: >> Hi Jean-Marc, >> >> A few more comments about the technical description document I sent last= night: >> >> (1) I should have clarified: when I said "frame size" in my document, I = was >> referring to the frame size for pitch estimation and pitch filter tap ca= lculation, >> not the CELT codec frame size. For ease of implementation, the boundari= es of the >> pitch frames and codec frames should obviously be aligned. Given that t= he CELT-only >> mode of the IETF codec has possible frame sizes of 2.5, 5, 10, and 20 ms= , I would >> recommend that we update the pitch period and pitch filter tap(s) at a r= ate of once >> every 5 ms if the CELT frame size is 5, 10, or 20 ms, because updating t= hem once >> every 10 or 20 ms is probably too infrequent and probably will decrease = the system's >> effectiveness in the reduction of coding distortion (although it doesn't= hurt for >> you to try updating them every 10 or 20 ms to be the same as the codec f= rame size >> just to see what happens). On the other hand, if the CELT frame size is= 2.5 ms in >> order to achieve an ultra-low delay, then it doesn't make sense for the = pitch >> prefilter/postfilter to use a 5 ms frame size since it will just increas= e the system >> delay. In this case, even though 2.5 ms pitch frame size may be more fr= equent than >> necessary and will double the pitch side information bit-rate, it is sti= ll a >> necessity just to enable CELT to achieve the ultra-low delay. >> >> (2) On a second thought, I think to calculate the normalized correlation= c or the >> optimal tap weight beta for a single-tap pitch predictor, rather than ca= lculating >> the summation over the current frame (n from 1 to L), it probably makes = more sense >> for the summation length to be the extracted pitch period or the pitch f= rame size, >> whichever is larger. The motivation is this: suppose the pitch frame siz= e is 2.5 ms >> or 5 ms and the pitch period is 15 ms, then trying to measure the level = of >> periodicity using only 1/6 to 1/3 of the pitch cycle waveform seems dang= erous or at >> least unreliable. Thus, if the detected pitch period is longer than the= frame size, >> then we should sum over all of the samples in the current frame plus mor= e samples in >> the previous frame(s) until the total number of samples in the summation= equals the >> pitch period. If the pitch period is smaller than the frame size and th= e pitch >> frame size is only 2.5 ms or 5 ms, I recommend just summing over the ent= ire current >> frame. >> >> (3) If we use multi-tap pitch filters in order to taper off the comb fil= tering >> effect at higher frequencies, we need to be careful about the stability = of the IIR >> pitch filter, especially if the IIR filter is in the decoder side (as is= the case in >> my example single-tap all-pole pitch postfilter). I suppose you probabl= y already >> know this, but I would like to mention and discuss it anyway. Even thou= gh the >> multi-tap IIR pitch postfilter can be an exact inverse filter of a corre= sponding >> pitch prefilter and supposedly is just "undoing" the effect of the pitch= prefilter, >> the quantization errors in the CELT codec and even the effects caused by= channel >> errors can potentially make the output of the multi-tap IIR pitch postfi= lter "go >> wild" (have very large amplitude or even oscillate) if that filter is un= stable or >> even if it is only close to unstable (i.e. all poles are inside the unit= circle but >> too close to the unit circle). If the IIR pitch filter is a one-tap fil= ter, >> ensuring the stability is trivial. If it is a 2-tap or 3-tap filter, Pr= ofessor >> Peter Kabal and his student Ravi Ramachandran has a nice paper on a stab= ility test >> and a stabilization technique that can be used to ensure the stability o= f such a >> filter. Their paper is published in the IEEE Transactions on ASSP, July= 1987. We >> can use that stabilization technique to stabilize every single candidate= 2-tap or 3- >> tap filter coefficient set that we will transmit and use, with enough sa= fety margin >> to instability, to ensure that a 2-tap or 3-tap IIR pitch postfilter wil= l not >> amplify the output audio signal wildly. >> >> Alternatively, we can use the second form of the pitch prefilter and pos= tfilter that >> I mentioned in my document so that the pitch postfilter is a 2-tap or 3-= tap FIR >> filter in the form of H(z) =3D 1 + a1 * z^(-p+1) + a2 * z^-p + a3 * z^(-= p-1), with a1 >> or a3 equals to zero in the case of a 2-tap filter. In this case, we wi= ll move the >> multi-tap IIR filter from the decoder side to the encoder side where it = is applied >> directly to the input audio signal and thus we do not have to worry abou= t CELT >> quantization errors or channel errors driving a "nearly unstable" filter= wild. Of >> course, we do not want the IIR prefilter there to be too close to instab= ility, >> either, otherwise the potential wild swing in its output may still cause= problems >> for the CELT codec. The main advantage of this alternative arrangement = is that the >> pitch postfilter is now an FIR filter, which is guaranteed to be stable = and no >> amount of CELT coding distortion or channel error will ever "drive it cr= azy and >> wild" like they can for a multi-tap IIR pitch postfilter that has not be= en >> stabilized or has not maintained a sufficient safety margin to instabili= ty. The >> main disadvantage of this arrangement is that with an FIR pitch postfilt= er, its >> frequency response has downward spikes located at pitch harmonics, and i= t will tend >> to shape the coding noise that way, too. A noise spectral shape with do= wnward >> spikes is probably not as ideal as the noise spectral shape with upward = spikes that >> the all-pole pitch postfilter provides. This is more so when the parame= ter bmax is >> closer to 1 but probably won't be too big a deal if bmax is in the neigh= borhood of >> 0.4 to 0.6. >> >> Raymond >> >> -----Original Message----- >> From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf O= f Raymond (Juin-Hwey) Chen >> Sent: Saturday, October 23, 2010 12:50 PM >> To: Jean-Marc Valin >> Cc: 'codec@ietf.org' >> Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT= codec >> Importance: Low >> >> Hi Jean-Marc, >> >> You are very welcome. Glad to contribute and glad to know that the docum= ent is good >> enough for you to start implementing the techniques. >> >> Regarding your question, I think it really depends on what your main emp= hasis is. >> If you just want the best possible pitch period contour and don't mind h= aving a >> little higher complexity, then partially whitening the input signal with= the >> A(z/gamma) or A(z)/A(z/gamma) or even A(z/alpha)/A(z/beta) is a common a= pproach used >> in the pitch estimators of many speech codecs (as I am sure you are well= aware of), >> and it does help to reduce the chance of detecting the frequency of a st= rong first >> formant of the input speech signal as the pitch frequency. >> >> If, on the other hand, your main emphasis is to get a codec complexity a= s low as >> possible, I think it is possible to get a reasonably good pitch period c= ontour by >> using the input signal directly without partial whitening filtering, thu= s saving the >> complexity of the associated LPC analysis and filtering (which can be re= latively >> high for the 48 kHz sampling rate). Note that when I said "Any low-comp= lexity pitch >> estimator can be used here..." in the document I sent last night, I used= the term >> "low-complexity" because I thought you told me you tried to keep the com= plexity of >> CELT as low as possible. Since this pitch prefiltering/postfiltering ap= proach is >> meant to be used with CELT, I thought it is only natural to make the com= plexity of >> the pitch estimator as low as possible. That's why I used the term "low-= complexity >> pitch estimator" in that document. >> >> I think for this pitch prefiltering/postfiltering approach to work well,= what's most >> important is that we need to do everything we can to ensure that the pit= ch period we >> extract is the true pitch period rather than an integer multiple or sub-= multiple of >> the pitch period. Unlike speech coding or packet loss concealment appli= cations >> where double pitch period or triple pitch period still gives you decent = pitch >> prediction or replacement waveform, here double or triple pitch period w= ill create >> one or two unwanted frequency response peak(s) in the valley regions bet= ween pitch >> harmonic peaks and thus will greatly reduce the noise reduction achievab= le by the >> noise spectral shaping effect of the pitch prefilter/postfilter approach= . >> >> Hence, ideally we should have some decision logic and smoothing techniqu= es to get a >> smooth contour of the true pitch period. I think the pitch estimator of= the >> BroadVoice codecs (BV16 and BV32) that Broadcom previously submitted can= serve as a >> good starting point for further development, because it was optimized to= have >> relatively low complexity and it also has decision logic and smoothing b= uilt-in to >> try to get a smooth contour of the true pitch period for voiced regions = of speech. >> It is optimized for 8 kHz or 16 kHz speech signals, so to use it for 48 = kHz speech >> and music signals, we need to adjust the decimation factor and re-optimi= ze it to >> work better for music as well. We also need to increase the range of th= e pitch >> period, or equivalently, the pitch frequency. Typical pitch frequency r= ange for >> adult voice is somewhere around 50 to 400 Hz, but the pitch frequencies = of music >> instruments cover a much wider range and can go up to 1600 Hz or higher.= BV16 and >> BV32 detectable pitch ranges are around 59 Hz ~ 800 Hz and 60 Hz ~ 1600 = Hz, >> respectively, because we previously already increased their pitch ranges= to be able >> to detect the pitch of some of the music instruments. >> >> Of course, if there is another royalty-free and unencumbered pitch estim= ator >> available that has even lower complexity than the BV16/BV32 pitch estima= tor and >> performs better, I am all for it. >> >> Raymond >> >> -----Original Message----- >> From: Jean-Marc Valin [mailto:jean-marc.valin@usherbrooke.ca] >> Sent: Saturday, October 23, 2010 5:52 AM >> To: Raymond (Juin-Hwey) Chen >> Cc: 'codec@ietf.org' >> Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT= codec >> >> Hi Raymond, >> >> Thank you very much for your contribution and the highly detailed >> description you just sent. I believe I have enough information to start >> implementing it. At this point, I only have one minor question. When >> estimating the pitch period/gain, do you recommend working on the input >> signal directly or should the input be "whitened" using a filter like >> A(z/gamma) (where A(z) is the LPC filter)? >> >> Thanks again, >> >> Jean-Marc >> >> On 10-10-23 01:13 AM, Raymond (Juin-Hwey) Chen wrote: >>> All, >>> >>> In the last few weeks, I have been working on a pitch prefiltering/post= filtering method to try to >>> improve the output audio quality of the IETF mode 1 CELT codec for cert= ain audio signals that CELT >>> low-delay modes give more audible coding distortion than usual (typical= ly nearly periodic solo >>> music instrument signals). I am happy to report that this method has a= chieved fairly significant >>> audio quality improvement for many of these kinds of audio signals that= I tried. This is achieved >>> without increasing the coding delay and with only a slight increase in = the codec complexity and >>> encoding bit-rate (a potential increase of around 2 kb/s or so; less if= a range coder is used). >>> >>> I have been in discussion with Jean-Marc about this and have sent him C= ELT-coded audio signals at >>> different bit-rates obtained with or without this method. After listen= ing, he agreed that the >>> method made a significant improvement when compared with the current CE= LT implementation, and he >>> would like to try to incorporate this method into his CELT C code to ev= aluate and verify the >>> performance gain and possibly further develop the method. >>> >>> Jean-Marc asked me to write a detailed description of the method in a d= ocument and send the >>> document to the IETF codec WG email reflector so he can read it, unders= tand it, and then implement >>> it himself in his CELT C code. I am attaching such a technical descrip= tion of the method. >>> >>> Note that although this document is 7 pages long, the pitch prefilterin= g/postfiltering method is >>> actually quite simple. The document is long only because (1) I tried t= o describe the technical >>> details as much as I can so Jean-Marc can implement it exactly as I had= it without any ambiguity, >>> and (2) I also described/discussed a few alternative approaches for som= e of the algorithm >>> components in case we need to go those routes. >>> >>> Note also that this is "fresh out of the oven" as I only had time to fi= nish a very rough first cut >>> for a proof-of-concept simulation and did not have time yet to refine o= r tune it. Thus, I am sure >>> there are still many rough edges that need to be smoothed out and many = areas that can be improved >>> upon. However, I am very much encouraged that even such a rough first = cut showed a lot of promise >>> as it made a very large audio quality improvement at several different = bit-rates for a trumpet >>> test signal, which is a signal that CELT low-delay modes have problems = with. Therefore, after >>> this pitch prefiltering/postfiltering method is integrated into CELT an= d further refined/tuned, I >>> think it has the potential to make the IETF mode 1 CELT codec more robu= st to different audio >>> signals. That is, it can potentially improve CELT performance for many= signals that are currently >>> difficult for CELT to code well, and thus ensure a more uniform audio q= uality level across a wider >>> variety of different audio signals. If Jean-Marc's integration, testin= g, and further development >>> of this pitch filtering method confirms this potential, then it is prop= osed that this pitch >>> prefiltering/postfiltering method be incorporated into Mode 1 of the IE= TF Opus codec to make the >>> codec more robust to different input signal types. >>> >>> Cheers, >>> >>> Raymond >>> >>> >>> >>> _______________________________________________ >>> codec mailing list >>> codec@ietf.org >>> https://www.ietf.org/mailman/listinfo/codec >> >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >> >> >> >> > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > From rchen@broadcom.com Mon Oct 25 23:39:16 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 027813A6846 for ; Mon, 25 Oct 2010 23:39:16 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.361 X-Spam-Level: X-Spam-Status: No, score=-2.361 tagged_above=-999 required=5 tests=[AWL=0.238, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Qzb38yLVzdtG for ; Mon, 25 Oct 2010 23:39:14 -0700 (PDT) Received: from mms1.broadcom.com (mms1.broadcom.com [216.31.210.17]) by core3.amsl.com (Postfix) with ESMTP id 47C013A67FC for ; Mon, 25 Oct 2010 23:39:14 -0700 (PDT) Received: from [10.9.200.131] by mms1.broadcom.com with ESMTP (Broadcom SMTP Relay (Email Firewall v6.3.2)); Mon, 25 Oct 2010 23:40:44 -0700 X-Server-Uuid: 02CED230-5797-4B57-9875-D5D2FEE4708A Received: from IRVEXCHCCR01.corp.ad.broadcom.com ([10.252.49.30]) by IRVEXCHHUB01.corp.ad.broadcom.com ([10.9.200.131]) with mapi; Mon, 25 Oct 2010 23:40:43 -0700 From: "Raymond (Juin-Hwey) Chen" To: "Jean-Marc Valin" Importance: low X-Priority: 5 Date: Mon, 25 Oct 2010 23:40:39 -0700 Thread-Topic: [codec] A pitch filtering method to improve the mode 1 CELT codec Thread-Index: ActzE0mHciHPoitmTXyvYNanyFEPrQBbzXsw Message-ID: References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC37F42.2040103@usherbrooke.ca> In-Reply-To: <4CC37F42.2040103@usherbrooke.ca> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-cr-hashedpuzzle: CMir EiXe E0WK E2qJ GE9S G4g4 HMkZ Id13 IxnC Jn0g J0JX Kc59 NlDC NzFK PI5J SNSy; 2; YwBvAGQAZQBjAEAAaQBlAHQAZgAuAG8AcgBnADsAagBlAGEAbgAtAG0AYQByAGMALgB2AGEAbABpAG4AQAB1AHMAaABlAHIAYgByAG8AbwBrAGUALgBjAGEA; Sosha1_v1; 7; {2368FC93-790D-48F8-8ADC-CA5733CBB442}; cgBjAGgAZQBuAEAAYgByAG8AYQBkAGMAbwBtAC4AYwBvAG0A; Tue, 26 Oct 2010 06:40:39 GMT; UgBFADoAIABbAGMAbwBkAGUAYwBdACAAQQAgAHAAaQB0AGMAaAAgAGYAaQBsAHQAZQByAGkAbgBnACAAbQBlAHQAaABvAGQAIAB0AG8AIABpAG0AcAByAG8AdgBlACAAdABoAGUAIABtAG8AZABlACAAMQAgAEMARQBMAFQAIABjAG8AZABlAGMA x-cr-puzzleid: {2368FC93-790D-48F8-8ADC-CA5733CBB442} acceptlanguage: en-US MIME-Version: 1.0 X-WSS-ID: 60D8A86647836320872-01-01 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Cc: "'codec@ietf.org'" Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Oct 2010 06:39:16 -0000 Hi Jean-Marc, See my in-line comments below. You wrote: > My concern wasn't that much about the period, but the gain. I'll > probably have to try both approaches, but I was mainly concerned that a > non-harmonic signal with one strong sinusoid could cause the comb filter > to be turned on with a high gain. [Raymond]: I see your concern with introducing artificial harmonics for a s= ingle sinusoid. On the other hand, my concern is that if you derive the ga= in (or taps) based on partially whitened input signal, the gain or taps may= be too small for legitimate periodic music or speech signals. To address = both your and my concerns, I think it is probably best to use a 3-tap filte= r where the taps are derived directly from the input signal without whiteni= ng. Since a 3-tap filter has the ability to taper off the degree of comb f= iltering with increasing frequency, a set of 3 filter taps that is chosen t= o match the input spectrum can probably have a reasonable spectral match fo= r either a single tone or music/speech with a harmonic structure. The Broa= dVoice codecs use a 3-tap pitch predictor. The way those 3 taps are jointl= y searched and vector quantized might provide a way to quantize the 3 taps = and do a spectral match at the same time for our current pitch prefilter/po= stfilter approach. We need to experiment to find out. The BroadVoice code= books for the 3 taps were designed using mainly speech. If we adopt this a= pproach, we need to redesign the codebook with sufficient music signals in = the training set. If you are particularly concerned with a single sinusoid= , we may want to train one or a few 3-tap codevectors using just sinusoids.= If this doesn't work well, in the worst case, we could potentially design= a detector that detect (nearly) periodic time-domain waveform that does no= t have a spectrum with a harmonic structure (like a single sinusoid) and tu= rn off the prefilter/postfilter accordingly. > CELT currently has a pitch predictor used for PLC, but I doubt it's very > good for this as it is now. OTOH, it's fairly low complexity too. It's > part of my plans to compare it with the one in BV16/32. I assume that > Broadcom would be willing to relicense it to the simplified BSD license > (it's currently LGPL) if we wanted to use it instead of the current CELT > estimator. [Raymond]: Broadcom's legal counsel will formally answer that question soon= . My assumption is that if we were previously willing to allow royalty-fre= e use of BroadVoice in the IETF codec, then allowing certain components of = it to be used in the IETF codec under IETF's licensing terms is probably no= t a problem, but this is just my personal thought and not an official Broad= com position yet. I will talk to our lawyer and let him give the formal an= swer. > The pitch range is another thing that needs experimenting with, but my > intuition here is that it's probably not necessary to go below 100 Hz > because for these very low pitches, there is enough masking between the > harmonics to make the coding noise inaudible even without a post-filter. > Again worth testing. [Raymond]: I agree that lower pitch frequency generally has better noise ma= sking, so time-domain predictive codecs often sound noisier for female spee= ch than for male speech. However, my previous work with pitch postfilterin= g during my Ph.D. study showed that pitch postfiltering alone could still g= ive audible quality improvement to low-bit-rate-coded male speech. Does it= improve quality when the pitch frequency is below 100 Hz? I don't remember= the pitch frequencies of my pitch postfilter tests. I agree it is worth t= esting. From jonas.svedberg@ericsson.com Tue Oct 26 04:38:42 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id C57AC3A67DB for ; Tue, 26 Oct 2010 04:38:42 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.454 X-Spam-Level: X-Spam-Status: No, score=-4.454 tagged_above=-999 required=5 tests=[AWL=2.145, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id kevUF2LcvIlT for ; Tue, 26 Oct 2010 04:38:41 -0700 (PDT) Received: from mailgw9.se.ericsson.net (mailgw9.se.ericsson.net [193.180.251.57]) by core3.amsl.com (Postfix) with ESMTP id 0C6DD3A6942 for ; Tue, 26 Oct 2010 04:38:37 -0700 (PDT) X-AuditID: c1b4fb39-b7b54ae000003464-34-4cc6be28dcac Received: from esessmw0191.eemea.ericsson.se (Unknown_Domain [153.88.253.125]) by mailgw9.se.ericsson.net (Symantec Mail Security) with SMTP id E6.93.13412.82EB6CC4; Tue, 26 Oct 2010 13:40:24 +0200 (CEST) Received: from ESESSCMS0356.eemea.ericsson.se ([169.254.1.175]) by esessmw0191.eemea.ericsson.se ([153.88.115.84]) with mapi; Tue, 26 Oct 2010 13:40:23 +0200 From: Jonas Svedberg To: "codec@ietf.org" Date: Tue, 26 Oct 2010 13:40:22 +0200 Thread-Topic: Proposed Guidelines update , was: Re:Opus codec licensing Thread-Index: Act0epUvmgTNgbXuRTy9lbdQGOwkVwAhZVPA Message-ID: References: <1389617873.938373.1287516517422.JavaMail.root@lu2-zimbra> <1559386036.938403.1287516673396.JavaMail.root@lu2-zimbra> <000001cb7423$67074c40$3515e4c0$@de> <4CC5D9F7.9040308@fas.harvard.edu> In-Reply-To: <4CC5D9F7.9040308@fas.harvard.edu> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: acceptlanguage: en-US Content-Type: multipart/mixed; boundary="_002_DEAE495523C8F140A875D22C7C59D31902FC0921ESESSCMS0356eem_" MIME-Version: 1.0 X-Brightmail-Tracker: AAAAAA== Cc: Jean-Marc Valin Subject: [codec] Proposed Guidelines update , was: Re:Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Oct 2010 11:38:42 -0000 --_002_DEAE495523C8F140A875D22C7C59D31902FC0921ESESSCMS0356eem_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Hi all=20 from my point of view it is beneficial if we can boil down the current di= scussion on licensing and IPR disclosures=20 to what should be included in our WGs process document (guidelines).=20 =20 With regard to the current proposed guidelines text v07, it is in secti= on 5 refering to BCP79:=20 "In general, IETF working groups prefer technologies with no known IPR c= laims or, for technologies with=20 claims against them, an offer of royalty-free licensing."=20 This BCP79 reference to does not state a preference between the two bas= ic forms of Royalty Free=20 A) 'RF' where 'No license is required' (no royalities, no entering = into any agreement )=20 B) 'RF'-licensing, where obtaining a license and entering into an ag= reement is still required.=20 In form B) the licensing terms may include limitations and/or compensa= tions which are identified as encumbrances by=20 the charter of this WG. Reading the charter, it defines encumbrance a= s : "... need to request a license, enter into a business agreement, pay= licensing fees or royalties,=20 or attempt to adhere to other special conditions or restrictions".= =20 The charter further expresses as one of the goals:=20 "The working group cannot explicitly rule out the possibility of ado= pting encumbered technologies; =20 however, the working group will try to avoid encumbered technologie= s that require royalties or=20 other encumbrances that would prevent such technologies from being = easy to redistribute and use." Hence, in the spirit of the charter and the creation of the WG,=20 we should make it clear in the guidelines that royalty-free licensin= g by no means is identical with unencumbrance=20 and that proposed technology adhereing to the 'No lic. required optio= n' should be prefered if the WG is presented=20 with such a choice. Also in light of the discussion, it seems reasonable to state in the g= uidelines that any inclusion of technology=20 into the codec will not be considered unless proper IP disclosure for = the contribution has been made.=20 =20 Attached is an proposed updated guidelines (section 5, page ~10), with = change marks. =20 //BR Jonas=20 --_002_DEAE495523C8F140A875D22C7C59D31902FC0921ESESSCMS0356eem_ Content-Type: application/msword; name="draft-valin-codec-guidelines-07-JSv.doc" Content-Description: draft-valin-codec-guidelines-07-JSv.doc Content-Disposition: attachment; filename="draft-valin-codec-guidelines-07-JSv.doc"; size=157184; creation-date="Fri, 22 Oct 2010 11:10:35 GMT"; modification-date="Tue, 26 Oct 2010 13:30:49 GMT" Content-Transfer-Encoding: base64 0M8R4KGxGuEAAAAAAAAAAAAAAAAAAAAAPgADAP7/CQAGAAAAAAAAAAAAAAADAAAALgEAAAAAAAAA EAAAMAEAAAEAAAD+////AAAAACsBAAAsAQAALQEAAP////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// ///////////////////////////////////////////////////////////////////////////s pcEAJWAJBAAA8BK/AAAAAAAAEAAAAAAABgAAKIUAAA4AYmpiaq71rvUAAAAAAAAAAAAAAAAAAAAA AAAJBBYANNYAAMyfAADMnwAAKH0AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD//w8AAAAA AAAAAAD//w8AAAAAAAAAAAD//w8AAAAAAAAAAAAAAAAAAAAAAKQAAAAAAOoGAAAAAAAA6gYAAOoG AAAAAAAA6gYAAAAAAADqBgAAAAAAAOoGAAAAAAAA6gYAABQAAAAAAAAAAAAAAP4GAAAAAAAAhhEB AAAAAACGEQEAAAAAAIYRAQAAAAAAhhEBAAwAAACSEQEANAEAAP4GAAAAAAAA0kgBANoBAADSEgEA AAAAANISAQAAAAAA0hIBAAAAAADSEgEAAAAAANISAQAAAAAArRMBAAAAAACtEwEAAAAAAK0TAQAA AAAAPUgBAAIAAAA/SAEAAAAAAD9IAQAAAAAAP0gBAAAAAAA/SAEAAAAAAD9IAQAAAAAAP0gBACQA AACsSgEAaAIAABRNAQDCAAAAY0gBABUAAAAAAAAAAAAAAAAAAAAAAAAA6gYAAAAAAACtEwEAAAAA AAAAAAAAAAAAAAAAAAAAAACtEwEAAAAAAK0TAQAAAAAArRMBAAAAAACtEwEAAAAAAGNIAQAAAAAA AAAAAAAAAADqBgAAAAAAAOoGAAAAAAAA0hIBAAAAAAAAAAAAAAAAANISAQDbAAAAeEgBAB4AAACL RAEAAAAAAItEAQAAAAAAi0QBAAAAAACtEwEANhkAAOoGAAAAAAAA0hIBAAAAAADqBgAAAAAAANIS AQAAAAAAPUgBAAAAAAAAAAAAAAAAAItEAQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAArRMBAAAAAAA9SAEAAAAAAAAAAAAAAAAAi0QBAAAAAACLRAEA HgAAAC1HAQAYAAAA6gYAAAAAAADqBgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAeUcBAAAAAADSEgEAAAAAAMYSAQAMAAAAYNggHQF1 ywEAAAAAAAAAAIYRAQAAAAAA4ywBAJ4XAABFRwEACAAAAAAAAAAAAAAAKUgBABQAAACWSAEAPAAA ANJIAQAAAAAATUcBACwAAADWTQEAAAAAAIFEAQAAAAAA1k0BABAAAAB5RwEAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAB5 RwEAFAAAANZNAQAAAAAAAAAAAAAAAADqBgAAAAAAAI1HAQCcAAAArRMBAAAAAACtEwEAAAAAAItE AQAAAAAArRMBAAAAAACtEwEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAArRMB AAAAAACtEwEAAAAAAK0TAQAAAAAAY0gBAAAAAABjSAEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAgUQBAAoAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAK0TAQAA AAAArRMBAAAAAACtEwEAAAAAANJIAQAAAAAArRMBAAAAAACtEwEAAAAAAK0TAQAAAAAArRMBAAAA AAAAAAAAAAAAAP4GAAAAAAAA/gYAAAAAAAD+BgAARIUAAEKMAABEhQAA/gYAAAAAAAD+BgAAAAAA AP4GAAAAAAAAQowAAAAAAAD+BgAAAAAAAP4GAAAAAAAA/gYAAAAAAADqBgAAAAAAAOoGAAAAAAAA 6gYAAAAAAADqBgAAAAAAAOoGAAAAAAAA6gYAAAAAAAD/////AAAAAAIADAEAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA0NDU5l dHdvcmsgV29ya2luZyBHcm91cCAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgIEpNLiBWYWxpbg1JbnRlcm5ldC1EcmFmdCAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgICBPY3Rhc2ljIEluYy4NSW50ZW5kZWQgc3RhdHVzOiBTdGFuZGFyZHMg VHJhY2sgICAgICAgICAgICAgICAgICAgICAgICAgICAgICBTLiBCb3JpbGluDUV4cGlyZXM6IEFw cmlsIDE1LCAyMDExICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgU1BJUklU IERTUA0gICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICBLLiBWb3MNICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgIFNreXBlIFRlY2hub2xvZ2llcyBTLkEuDSAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgQy4gTW9udGdvbWVyeQ0gICAg ICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIFhpcGguT3Jn IEZvdW5kYXRpb24NICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICBSLiBDaGVuDSAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICAgICBCcm9hZGNvbSBDb3Jwb3JhdGlvbg0gICAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIE9jdG9iZXIgMTIsIDIw MTANDQ0gICAgICAgICAgR3VpZGVsaW5lcyBmb3IgdGhlIENvZGVjIERldmVsb3BtZW50IFdpdGhp biB0aGUgSUVURg0gICAgICAgICAgICAgICAgICAgIGRyYWZ0LXZhbGluLWNvZGVjLWd1aWRlbGlu ZXMtMDcNDUFic3RyYWN0DQ0gICBUaGlzIGRvY3VtZW50IHByb3ZpZGVzIGdlbmVyYWwgZ3VpZGVs aW5lcyBmb3Igd29yayBvbiBkZXZlbG9waW5nIGFuZA0gICBzcGVjaWZ5aW5nIGEgY29kZWMgd2l0 aGluIHRoZSBJRVRGLiAgVGhlc2UgZ3VpZGVsaW5lcyBjb3ZlciB0aGUNICAgZGV2ZWxvcG1lbnQg cHJvY2VzcywgZXZhbHVhdGlvbiwgcmVxdWlyZW1lbnRzIGNvbmZvcm1hbmNlLCBhbmQNICAgaW50 ZWxsZWN0dWFsIHByb3BlcnR5IGlzc3Vlcy4NDVN0YXR1cyBvZiB0aGlzIE1lbW8NDSAgIFRoaXMg SW50ZXJuZXQtRHJhZnQgaXMgc3VibWl0dGVkIGluIGZ1bGwgY29uZm9ybWFuY2Ugd2l0aCB0aGUN ICAgcHJvdmlzaW9ucyBvZiBCQ1AgNzggYW5kIEJDUCA3OS4NDSAgIEludGVybmV0LURyYWZ0cyBh cmUgd29ya2luZyBkb2N1bWVudHMgb2YgdGhlIEludGVybmV0IEVuZ2luZWVyaW5nDSAgIFRhc2sg Rm9yY2UgKElFVEYpLiAgTm90ZSB0aGF0IG90aGVyIGdyb3VwcyBtYXkgYWxzbyBkaXN0cmlidXRl DSAgIHdvcmtpbmcgZG9jdW1lbnRzIGFzIEludGVybmV0LURyYWZ0cy4gIFRoZSBsaXN0IG9mIGN1 cnJlbnQgSW50ZXJuZXQtDSAgIERyYWZ0cyBpcyBhdCBodHRwOi8vZGF0YXRyYWNrZXIuaWV0Zi5v cmcvZHJhZnRzL2N1cnJlbnQvLg0NICAgSW50ZXJuZXQtRHJhZnRzIGFyZSBkcmFmdCBkb2N1bWVu dHMgdmFsaWQgZm9yIGEgbWF4aW11bSBvZiBzaXggbW9udGhzDSAgIGFuZCBtYXkgYmUgdXBkYXRl ZCwgcmVwbGFjZWQsIG9yIG9ic29sZXRlZCBieSBvdGhlciBkb2N1bWVudHMgYXQgYW55DSAgIHRp bWUuICBJdCBpcyBpbmFwcHJvcHJpYXRlIHRvIHVzZSBJbnRlcm5ldC1EcmFmdHMgYXMgcmVmZXJl bmNlDSAgIG1hdGVyaWFsIG9yIHRvIGNpdGUgdGhlbSBvdGhlciB0aGFuIGFzICJ3b3JrIGluIHBy b2dyZXNzLiINDSAgIFRoaXMgSW50ZXJuZXQtRHJhZnQgd2lsbCBleHBpcmUgb24gQXByaWwgMTUs IDIwMTEuDQ1Db3B5cmlnaHQgTm90aWNlDQ0gICBDb3B5cmlnaHQgKGMpIDIwMTAgSUVURiBUcnVz dCBhbmQgdGhlIHBlcnNvbnMgaWRlbnRpZmllZCBhcyB0aGUNICAgZG9jdW1lbnQgYXV0aG9ycy4g IEFsbCByaWdodHMgcmVzZXJ2ZWQuDQ0gICBUaGlzIGRvY3VtZW50IGlzIHN1YmplY3QgdG8gQkNQ IDc4IGFuZCB0aGUgSUVURiBUcnVzdCdzIExlZ2FsDSAgIFByb3Zpc2lvbnMgUmVsYXRpbmcgdG8g SUVURiBEb2N1bWVudHMNICAgKGh0dHA6Ly90cnVzdGVlLmlldGYub3JnL2xpY2Vuc2UtaW5mbykg aW4gZWZmZWN0IG9uIHRoZSBkYXRlIG9mDQ0NDVZhbGluLCBldCBhbC4gICAgICAgICAgICBFeHBp cmVzIEFwcmlsIDE1LCAyMDExICAgICAgICAgICAgICAgICBbUGFnZSAxXQ0MDUludGVybmV0LURy YWZ0ICAgICAgICAgICAgICBDb2RlYyBHdWlkZWxpbmVzICAgICAgICAgICAgICAgIE9jdG9iZXIg MjAxMA0NDSAgIHB1YmxpY2F0aW9uIG9mIHRoaXMgZG9jdW1lbnQuICBQbGVhc2UgcmV2aWV3IHRo ZXNlIGRvY3VtZW50cw0gICBjYXJlZnVsbHksIGFzIHRoZXkgZGVzY3JpYmUgeW91ciByaWdodHMg YW5kIHJlc3RyaWN0aW9ucyB3aXRoIHJlc3BlY3QNICAgdG8gdGhpcyBkb2N1bWVudC4gIENvZGUg Q29tcG9uZW50cyBleHRyYWN0ZWQgZnJvbSB0aGlzIGRvY3VtZW50IG11c3QNICAgaW5jbHVkZSBT aW1wbGlmaWVkIEJTRCBMaWNlbnNlIHRleHQgYXMgZGVzY3JpYmVkIGluIFNlY3Rpb24gNC5lIG9m DSAgIHRoZSBUcnVzdCBMZWdhbCBQcm92aXNpb25zIGFuZCBhcmUgcHJvdmlkZWQgd2l0aG91dCB3 YXJyYW50eSBhcw0gICBkZXNjcmliZWQgaW4gdGhlIFNpbXBsaWZpZWQgQlNEIExpY2Vuc2UuDQ0N VGFibGUgb2YgQ29udGVudHMNDSAgIDEuICBJbnRyb2R1Y3Rpb24gLiAuIC4gLiAuIC4gLiAuIC4g LiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAgMw0gICAyLiAgRGV2ZWxvcG1lbnQgUHJv Y2VzcyAgLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gIDQNICAgMy4g IEV2YWx1YXRpb24sIFRlc3RpbmcsIGFuZCBDaGFyYWN0ZXJpemF0aW9uICAuIC4gLiAuIC4gLiAu IC4gLiAuICA3DSAgIDQuICBSZXF1aXJlbWVudHMgQ29uZm9ybWFuY2UgLiAuIC4gLiAuIC4gLiAu IC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAgOA0gICA1LiAgSW50ZWxsZWN0dWFsIFByb3BlcnR5ICAu IC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gMTANICAgNi4gIFJlbGF0aW9u c2hpcCB3aXRoIE90aGVyIFNET3MgLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIDEy DSAgIDcuICBTZWN1cml0eSBDb25zaWRlcmF0aW9ucyAgLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4g LiAuIC4gLiAuIC4gLiAxNA0gICA4LiAgSUFOQSBDb25zaWRlcmF0aW9ucyAgLiAuIC4gLiAuIC4g LiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gMTUNICAgOS4gIEFja25vd2xlZGdtZW50cyAg LiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIDE2DSAgIDEwLiBS ZWZlcmVuY2VzIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAu IC4gLiAxNw0gICAgIDEwLjEuICBOb3JtYXRpdmUgUmVmZXJlbmNlcyAgLiAuIC4gLiAuIC4gLiAu IC4gLiAuIC4gLiAuIC4gLiAuIC4gMTcNICAgICAxMC4yLiAgSW5mb3JtYXRpdmUgUmVmZXJlbmNl cyAgLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIDE3DSAgIEF1dGhvcnMnIEFkZHJl c3NlcyAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAxOQ0N DQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NVmFsaW4sIGV0IGFsLiAgICAgICAgICAgIEV4cGly ZXMgQXByaWwgMTUsIDIwMTEgICAgICAgICAgICAgICAgIFtQYWdlIDJdDQwNSW50ZXJuZXQtRHJh ZnQgICAgICAgICAgICAgIENvZGVjIEd1aWRlbGluZXMgICAgICAgICAgICAgICAgT2N0b2JlciAy MDEwDQ0NMS4gIEludHJvZHVjdGlvbg0NICAgVGhpcyBkb2N1bWVudCBkZXNjcmliZXMgYSBzdWdn ZXN0ZWQgcHJvY2VzcyBmb3Igd29yayBhdCB0aGUgSUVURiBvbg0gICBzdGFuZGFyZGl6YXRpb24g b2YgYSBjb2RlYyB0aGF0IGlzIG9wdGltaXplZCBmb3IgdXNlIGluIGludGVyYWN0aXZlDSAgIElu dGVybmV0IGFwcGxpY2F0aW9ucyBhbmQgdGhhdCBjYW4gYmUgd2lkZWx5IGltcGxlbWVudGVkIGFu ZCBlYXNpbHkNICAgZGlzdHJpYnV0ZWQgYW1vbmcgYXBwbGljYXRpb24gZGV2ZWxvcGVycywgc2Vy dmljZSBvcGVyYXRvcnMsIGFuZCBlbmQNICAgdXNlcnMuDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0N DQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NVmFsaW4sIGV0IGFsLiAgICAgICAgICAgIEV4cGlyZXMgQXBy aWwgMTUsIDIwMTEgICAgICAgICAgICAgICAgIFtQYWdlIDNdDQwNSW50ZXJuZXQtRHJhZnQgICAg ICAgICAgICAgIENvZGVjIEd1aWRlbGluZXMgICAgICAgICAgICAgICAgT2N0b2JlciAyMDEwDQ0N Mi4gIERldmVsb3BtZW50IFByb2Nlc3MNDSAgIFRoZSBwcm9jZXNzIG91dGxpbmVkIGhlcmUgaXMg aW50ZW5kZWQgdG8gbWF4aW1pemUgdGhlIHRyYW5zcGFyZW5jeSBvZg0gICB3b3JrIG9uIGFuIGF1 ZGlvIGNvZGVjIHdpdGhpbiB0aGUgSUVURi4gIFN1Y2ggd29yayBtaWdodCBpbnZvbHZlDSAgIGRl dmVsb3BtZW50IG9mIGEgY29tcGxldGVseSBuZXcgY29kZWMsIGFkYXB0YXRpb24gb2YgYW4gZXhp c3RpbmcNICAgY29kZWMgdG8gbWVldCB0aGUgcmVxdWlyZW1lbnRzLCBvciBpbnRlZ3JhdGlvbiBi ZXR3ZWVuIHR3byBvciBtb3JlDSAgIGV4aXN0aW5nIGNvZGVjcyB0aGF0IHJlc3VsdHMgaW4gYW4g aW1wcm92ZWQgY29kZWMgY29tYmluaW5nIHRoZSBiZXN0DSAgIGFzcGVjdHMgb2YgZWFjaCBjb2Rl Yy4gIFRvIGVuYWJsZSBzdWNoIHByb2NlZHVyYWwgdHJhbnNwYXJlbmN5LCB0aGUNICAgY29udHJp YnV0b3Igb2YgYW4gZXhpc3RpbmcgY29kZWMgbXVzdCBiZSB3aWxsaW5nIHRvIGNlZGUgY2hhbmdl DSAgIGNvbnRyb2wgdG8gdGhlIElFVEYgYW5kIHNob3VsZCBoYXZlIHN1ZmZpY2llbnQga25vd2xl ZGdlIG9mIHRoZSBjb2RlYw0gICB0byBhc3Npc3QgaW4gdGhlIHdvcmsgb2YgYWRhcHRpbmcgaXQg b3IgYXBwbHlpbmcgc29tZSBvZiBpdHMNICAgdGVjaG5vbG9neSB0byB0aGUgZGV2ZWxvcG1lbnQg b3IgaW1wcm92ZW1uZXQgb2Ygb3RoZXIgY29kZWNzLg0gICBGdXJ0aGVybW9yZSwgY29udHJpYnV0 b3JzIG5lZWQgdG8gYmUgYXdhcmUgdGhhdCBhbnkgY29kZWMgdGhhdA0gICByZXN1bHRzIGZyb20g d29yayB3aXRoaW4gdGhlIElFVEYgaXMgbGlrZWx5IHRvIGJlIGRpZmZlcmVudCBmcm9tIGFueQ0g ICBleGlzdGluZyBjb2RlYyB0aGF0IHdhcyBjb250cmlidXRlZCB0byB0aGUgSW50ZXJuZXQgU3Rh bmRhcmRzDSAgIFByb2Nlc3MuDQ0gICBXb3JrIG9uIGNvZGVjIGRldmVsb3BtZW50IGlzIGV4cGVj dGVkIHRvIHByb2NlZWQgYXMgZm9sbG93czoNDSAgIDEuICBJRVRGIHBhcnRpY2lwYW50cyB3aWxs IGlkZW50aWZ5IHRoZSByZXF1aXJlbWVudHMgdG8gYmUgbWV0IGJ5IGFuDSAgICAgICBJbnRlcm5l dCBjb2RlYywgaW4gdGhlIGZvcm0gb2YgYW4gSW50ZXJuZXQtRHJhZnQuDQ0gICAyLiAgSW50ZXJl c3RlZCBwYXJ0aWVzIGFyZSBlbmNvdXJhZ2VkIHRvIG1ha2UgY29udHJpYnV0aW9ucyBwcm9wb3Np bmcNICAgICAgIGV4aXN0aW5nIG9yIG5ldyBjb2RlY3MsIG9yIGVsZW1lbnRzIHRoZXJlb2YsIHRv IHRoZSBjb2RlYyBXRyBhcw0gICAgICAgbG9uZyBhcyB0aGVzZSBjb250cmlidXRpb25zIGFyZSB3 aXRoaW4gdGhlIHNjb3BlIG9mIHRoZSBXRy4NICAgICAgIElkZWFsbHksIHRoZXNlIGNvbnRyaWJ1 dGlvbnMgc2hvdWxkIGJlIGluIHRoZSBmb3JtIG9mIEludGVybmV0DSAgICAgICBEcmFmdHMsIGFs dGhvdWdoIG90aGVyIGZvcm1zIG9mIGNvbnRyaWJ1dGlvbnMgYXJlIGFsc28gcG9zc2libGUNICAg ICAgIGFzIGRpc2N1c3NlZCBpbiBbUFJPQ0VTU10gYW5kIGluIHRoZSBJRVRGJ3MgTm90ZSBXZWxs LiAgQXMNICAgICAgIGFsd2F5cywgY29udHJpYnV0aW9ucyB0byB0aGUgSUVURiBhcmUgc3ViamVj dCwgYW1vbmcgb3RoZXINICAgICAgIHByb2Nlc3Mgb3JpZW50ZWQgUkZDcywgdG8gW1BST0NFU1Nd LCBbVFJVU1RdLCBhbmQgW0lQUl0uDSAgICAgICBDb25zaWRlcmluZyB0aGUgZmllbGQgb2YgdGVj aG5vbG9neSwgSVBSIHRyYW5zcGFyZW5jeSBtYXkgYmUNICAgICAgIHBhcnRpY3VsYXJseSBoaWdo IG9uIHRoZSBwcmlvcml0eSBsaXN0IG9mIG1hbnkgY29kZWMgV0cNICAgICAgIHBhcnRpY2lwYW50 cy4gIEFjY29yZGluZ2x5LCBjb250cmlidXRvcnMgYXJlIHNwZWNpZmljYWxseQ0gICAgICAgcmVt aW5kZWQgb2YgdGhlaXIgSVBSIGRpc2Nsb3N1cmUgcmVxdWlyZW1lbnQsIGFuZCBhbGwNICAgICAg IHBhcnRpY2lwYW50cyBhcmUgcmVtaW5kZWQgb2YgdGhlIHNvbGljaXRhdGlvbiBvZiB0aGUgZGlz Y2xvc3VyZQ0gICAgICAgb2YgdGhpcmQgcGFydHkgSVBSLCBib3RoIGFzIGNvZGlmaWVkIGluIFtJ UFJdLg0NICAgMy4gIEFzIGNvbnRyaWJ1dGlvbnMgYXJlIHJlY2VpdmVkIGFuZCBkaXNjdXNzZWQg d2l0aGluIHRoZSB3b3JraW5nDSAgICAgICBncm91cCwgdGhlIGdyb3VwIHNob3VsZCBnYWluIGEg Y2xlYXJlciB1bmRlcnN0YW5kaW5nIG9mIHdoYXQgaXMNICAgICAgIGFjaGlldmFibGUgd2l0aGlu IHRoZSBkZXNpZ24gc3BhY2UuICBBcyBhIHJlc3VsdCwgdGhlIGF1dGhvcnMgb2YNICAgICAgIHRo ZSByZXF1aXJlbWVudHMgZG9jdW1lbnQgc2hvdWxkIGl0ZXJhdGl2ZWx5IGNsYXJpZnkgYW5kIGlt cHJvdmUNICAgICAgIHRoZWlyIGRvY3VtZW50IHRvIHJlZmxlY3QgdGhlIGVtZXJnaW5nIHdvcmtp bmcgZ3JvdXAgY29uc2Vuc3VzLg0gICAgICAgVGhpcyBpcyBsaWtlbHkgdG8gaW52b2x2ZSBjb2xs YWJvcmF0aW9uIHdpdGggSUVURiB3b3JraW5nIGdyb3Vwcw0gICAgICAgaW4gb3RoZXIgYXJlYXMs IHN1Y2ggYXMgY29sbGFib3JhdGlvbiB3aXRoIHdvcmtpbmcgZ3JvdXBzIGluIHRoZQ0gICAgICAg VHJhbnNwb3J0IGFyZWEgdG8gaWRlbnRpZnkgaW1wb3J0YW50IGFzcGVjdHMgb2YgcGFja2V0DSAg ICAgICB0cmFuc21pc3Npb24gb3ZlciB0aGUgSW50ZXJuZXQsIHdpdGggd29ya2luZyBncm91cHMg aW4gdGhlIFJBSQ0gICAgICAgYXJlYSB0byBlbnN1cmUgdGhhdCBpbmZvcm1hdGlvbiBhYm91dCBh bmQgbmVnb3RpYXRpb24gb2YgdGhlDSAgICAgICBjb2RlYyBjYW4gYmUgZWFzaWx5IHJlcHJlc2Vu dGVkIGF0IHRoZSBzaWduYWxsaW5nIGxheWVyLCBhbmQgd2l0aA0NDQ1WYWxpbiwgZXQgYWwuICAg ICAgICAgICAgRXhwaXJlcyBBcHJpbCAxNSwgMjAxMSAgICAgICAgICAgICAgICAgW1BhZ2UgNF0N DA1JbnRlcm5ldC1EcmFmdCAgICAgICAgICAgICAgQ29kZWMgR3VpZGVsaW5lcyAgICAgICAgICAg ICAgICBPY3RvYmVyIDIwMTANDQ0gICAgICAgd29ya2luZyBncm91cHMgaW4gdGhlIFRyYW5zcG9y dCBhcmVhIHRvIHVuZGVyc3RhbmQgdGhlIGRlZ3JlZSBvZg0gICAgICAgcmF0ZSBhZGFwdGF0aW9u IGRlc2lyYWJsZS4gIEluIHBhcmFsbGVsIHdpdGggdGhpcyB3b3JrLA0gICAgICAgaW50ZXJlc3Rl ZCBwYXJ0aWVzIHNob3VsZCBldmFsdWF0ZSB0aGUgY29udHJpYnV0aW9ucyBhdCBhIGhpZ2hlcg0g ICAgICAgbGV2ZWwgdG8gc2VlIHdoaWNoIHJlcXVpcmVtZW50cyBtaWdodCBiZSBtZXQgYnkgZWFj aCBjb2RlYy4NDSAgIDQuICBPbmNlIGEgc3VmZmljaWVudCBudW1iZXIgb2YgcHJvcG9zYWxzIGhh cyBiZWVuIHJlY2VpdmVkLCB0aGUNICAgICAgIGludGVyZXN0ZWQgcGFydGllcyB3aWxsIGlkZW50 aWZ5IHRoZSBzdHJlbmd0aHMsIHdlYWtuZXNzZXMsIGFuZA0gICAgICAgaW5ub3ZhdGl2ZSBhc3Bl Y3RzIG9mIHRoZSBjb250cmlidXRlZCBjb2RlY3MuICBUaGlzIHN0ZXAgd2lsbA0gICAgICAgY29u c2lkZXIgbm90IG9ubHkgdGhlIGNvZGVjcyBhcyBhIHdob2xlLCBidXQgYWxzbyBrZXkgZmVhdHVy ZXMgb2YNICAgICAgIHRoZSBpbmRpdmlkdWFsIGFsZ29yaXRobXMgKHByZWRpY3RvcnMsIHF1YW50 aXplcnMsIHRyYW5zZm9ybXMsDSAgICAgICBldGMuKS4NDSAgIDUuICBJdCBpcyBleHBlY3RlZCB0 aGF0IG5vbmUgb2YgdGhlIGNvbnRyaWJ1dGVkIGNvZGVjcyB3aWxsIG1lZXQgYWxsDSAgICAgICBv ZiB0aGUgZGVmaW5lZCByZXF1aXJlbWVudHMuICBUaGVyZWZvcmUsIGl0IGlzIGV4cGVjdGVkIHRo YXQgSUVURg0gICAgICAgcGFydGljaXBhbnRzIHdpbGwgYWNjZXB0IGEgX2Jhc2VsaW5lIGNvZGVj XyBhcyBhIFdHIGl0ZW0gdG8NICAgICAgIGZhY2lsaXRhdGUgdGhlIGRldmVsb3BtZW50IHByb2Nl c3MuICBUaGlzIGJhc2VsaW5lIGNvZGVjIHdpbGwNICAgICAgIG1lZXQgYXMgbWFueSBvZiB0aGUg cmVxdWlyZW1lbnRzIGFzIHBvc3NpYmxlLCBidXQgcHJvYmFibHkgd2lsbA0gICAgICAgbmVlZCB0 byBiZSBhZGp1c3RlZCB0aHJvdWdoIGFuIGl0ZXJhdGl2ZSBkZXZlbG9wbWVudCBwcm9jZXNzIGlu DSAgICAgICBvcmRlciB0byBtZWV0IGFsbCBvZiB0aGUgcmVxdWlyZW1lbnRzIChvciBhcyBtYW55 IHJlcXVpcmVtZW50cyBhcw0gICAgICAgcG9zc2libGUpLiAgVGhlIGJhc2VsaW5lIGNvZGVjIG1p Z2h0IGJlIG9uZSBvZiB0aGUgY29udHJpYnV0ZWQNICAgICAgIGNvZGVjcyAoZXNwZWNpYWxseSBp ZiBpdCBpcyB0aGUgb25seSBjb2RlYyB0aGF0IG1lZXRzIG1vc3Qgb2YgdGhlDSAgICAgICByZXF1 aXJlbWVudHMpLCBhIGNvbWJpbmF0aW9uIG9mIHR3byBvciBtb3JlIG9mIHRoZSBjb250cmlidXRl ZA0gICAgICAgY29kZWNzLCBvciBhbiBlbnRpcmVseSBuZXcgY29kZWMuICBOb25lIG9mIHRoZSBk ZWNpc2lvbnMgdGFrZW4gYXQNICAgICAgIHRoaXMgc3RlcCB3aWxsIGJlIGRlZmluaXRpdmUuICBJ biBwYXJ0aWN1bGFyLCBJRVRGIHBhcnRpY2lwYW50cw0gICAgICAgd2lsbCBub3QgcHJvdmlkZSBh ICJydWJiZXIgc3RhbXAiIGZvciBhbnkgY29udHJpYnV0ZWQgY29kZWMuDQ0gICA2LiAgSUVURiBw YXJ0aWNpcGFudHMgc2hvdWxkIHRoZW4gYXR0ZW1wdCB0byBpdGVyYXRpdmVseSBpbXByb3ZlIGVh Y2gNICAgICAgIGNvbXBvbmVudCBvZiB0aGUgYmFzZWxpbmUgY29kZWMgcmVmZXJlbmNlIGltcGxl bWVudGF0aW9uLCB3aGVyZQ0gICAgICAgYnkgImNvbXBvbmVudCIgd2UgbWVhbiBpbmRpdmlkdWFs IGFsZ29yaXRobXMgc3VjaCBhcyBwcmVkaWN0b3JzLA0gICAgICAgdHJhbnNmb3JtcywgcXVhbnRp emVycywgYW5kIGVudHJvcHkgY29kZXJzLiAgVGhlIHBhcnRpY2lwYW50cw0gICAgICAgc2hvdWxk IHByb2NlZWQgYnkgdHJ5aW5nIG5ldyBkZXNpZ25zLCBhcHBseWluZyBpZGVhcyBmcm9tIHRoZQ0g ICAgICAgY29udHJpYnV0ZWQgY29kZWNzLCBldmFsdWF0aW5nICJwcm9vZiBvZiBjb25jZXB0IiBp ZGVhcywgYW5kDSAgICAgICB1c2luZyB0aGVpciBleHBlcnRpc2UgaW4gY29kZWMgZGV2ZWxvcG1l bnQgdG8gaW1wcm92ZSB0aGUNICAgICAgIGJhc2VsaW5lIGNvZGVjLiAgQW55IGFzcGVjdCBvZiB0 aGUgYmFzZWxpbmUgY29kZWMgbWlnaHQgYmUNICAgICAgIGNoYW5nZWQgKGV2ZW4gdGhlIGZ1bmRh bWVudGFsIHByaW5jaXBsZXMgb2YgdGhlIGNvZGVjKSBvciB0aGUNICAgICAgIHBhcnRpY2lwYW50 cyBtaWdodCBzdGFydCBvdmVyIGVudGlyZWx5IGJ5IHNjcmFwcGluZyB0aGUgYmFzZWxpbmUNICAg ICAgIGNvZGVjIGFuZCBkZXNpZ25pbmcgYSBjb21wbGV0ZWx5IG5ldyBvbmUuICBUaGUgb3ZlcnJp ZGluZyBnb2FsDSAgICAgICBzaGFsbCBiZSB0byBkZXNpZ24gYSBjb2RlYyB0aGF0IHdpbGwgbWVl dCB0aGUgcmVxdWlyZW1lbnRzDSAgICAgICBkZWZpbmVkIGluIHRoZSByZXF1aXJlbWVudHMgZG9j dW1lbnQuICBHaXZlbiB0aGUgSUVURidzIG9wZW4NICAgICAgIHN0YW5kYXJkcyBwcm9jZXNzLCBh bnkgaW50ZXJlc3RlZCBwYXJ0eSB3aWxsIGJlIGFibGUgdG8NICAgICAgIGNvbnRyaWJ1dGUgdG8g dGhpcyB3b3JrLCB3aGV0aGVyIG9yIG5vdCB0aGV5IHN1Ym1pdHRlZCBhbg0gICAgICAgSW50ZXJu ZXQtRHJhZnQgZm9yIG9uZSBvZiB0aGUgY29udHJpYnV0ZWQgY29kZWNzLiAgVGhlIGNvZGVjDSAg ICAgICBpdHNlbGYgc2hvdWxkIGJlIG5vcm1hdGl2ZWx5IHNwZWNpZmllZCB3aXRoIGNvZGUgaW4g YW4gSW50ZXJuZXQtDSAgICAgICBEcmFmdC4NDSAgIDcuICBJbiBwYXJhbGxlbCB3aXRoIHdvcmsg b24gdGhlIGNvZGVjIHJlZmVyZW5jZSBpbXBsZW1lbnRhdGlvbiwNICAgICAgIGRldmVsb3BlcnMg YW5kIG90aGVyIGludGVyZXN0ZWQgcGFydGllcyBzaG91bGQgcGVyZm9ybSBldmFsdWF0aW9uDSAg ICAgICBvZiB0aGUgY29kZWMgYXMgZGVzY3JpYmVkIHVuZGVyIFNlY3Rpb24gMywgSUVURiBwYXJ0 aWNpcGFudHMNDQ0NVmFsaW4sIGV0IGFsLiAgICAgICAgICAgIEV4cGlyZXMgQXByaWwgMTUsIDIw MTEgICAgICAgICAgICAgICAgIFtQYWdlIDVdDQwNSW50ZXJuZXQtRHJhZnQgICAgICAgICAgICAg IENvZGVjIEd1aWRlbGluZXMgICAgICAgICAgICAgICAgT2N0b2JlciAyMDEwDQ0NICAgICAgIHNo b3VsZCBkZWZpbmUgKHdpdGhpbiB0aGUgQVZUIFdvcmtpbmcgR3JvdXApIHRoZSBjb2RlYydzIHBh eWxvYWQNICAgICAgIGZvcm1hdCBmb3IgdXNlIHdpdGggdGhlIFJlYWwtdGltZSBUcmFuc3BvcnQg UHJvdG9jb2wgW1JUUF0sIGFuZA0gICAgICAgYXBwbGljYXRpb24gZGV2ZWxvcGVycyBzaG91bGQg c3RhcnQgdGVzdGluZyB0aGUgY29kZWMgYnkNICAgICAgIGltcGxlbWVudGluZyBpdCBpbiBjb2Rl IGFuZCBkZXBsb3lpbmcgaXQgaW4gYWN0dWFsIEludGVybmV0DSAgICAgICBhcHBsaWNhdGlvbnMg dG8gaWRlbnRpZnkgYW55IHBvdGVudGlhbCBwcm9ibGVtcy4NDSAgIDguICBPbmNlIElFVEYgcGFy dGljaXBhbnRzIGFncmVlIHRoYXQgdGhlIGNvZGVjIGJlaW5nIGRldmVsb3BlZCBtZWV0cw0gICAg ICAgdGhlIHJlcXVpcmVtZW50cywgSUVURiBwYXJ0aWNpcGFudHMgY2FuIGJlZ2luIHRoZSB0YXNr IG9mDSAgICAgICBjaGFyYWN0ZXJpemluZyB0aGUgY29kZWMuICBUaGUgY2hhcmFjdGVyaXphdGlv biBwcm9jZXNzIGlzDSAgICAgICBkZXNjcmliZWQgdW5kZXIgU2VjdGlvbiAzLg0NDQ0NDQ0NDQ0N DQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDVZhbGluLCBldCBhbC4gICAgICAgICAgICBF eHBpcmVzIEFwcmlsIDE1LCAyMDExICAgICAgICAgICAgICAgICBbUGFnZSA2XQ0MDUludGVybmV0 LURyYWZ0ICAgICAgICAgICAgICBDb2RlYyBHdWlkZWxpbmVzICAgICAgICAgICAgICAgIE9jdG9i ZXIgMjAxMA0NDTMuICBFdmFsdWF0aW9uLCBUZXN0aW5nLCBhbmQgQ2hhcmFjdGVyaXphdGlvbg0N ICAgTGFiIGV2YWx1YXRpb24gb2YgdGhlIGNvZGVjIGJlaW5nIGRldmVsb3BlZCBzaG91bGQgaGFw cGVuIHRocm91Z2hvdXQNICAgdGhlIGRldmVsb3BtZW50IHByb2Nlc3MgYmVjYXVzZSBpdCB3aWxs IGhlbHAgZW5zdXJlIHRoYXQgcHJvZ3Jlc3MgaXMNICAgYmVpbmcgbWFkZSB0b3dhcmQgZnVsZmls bG1lbnQgb2YgdGhlIHJlcXVpcmVtZW50cy4gIFRoZXJlIGFyZSBtYW55DSAgIHdheXMgaW4gd2hp Y2ggY29udGludW91cyBldmFsdWF0aW9uIGNhbiBiZSBwZXJmb3JtZWQuICBGb3IgbWlub3IsDSAg IHVuY29udHJvdmVyc2lhbCBjaGFuZ2VzIHRvIHRoZSBjb2RlYyBpdCBzaG91bGQgdXN1YWxseSBi ZSBzdWZmaWNpZW50DSAgIHRvIHVzZSBvYmplY3RpdmUgbWVhc3VyZW1lbnRzIChlLmcuLCBQRVNR LCBQRUFRLCBhbmQgU2VnU05SKQ0gICB2YWxpZGF0ZWQgYnkgaW5mb3JtYWwgc3ViamVjdGl2ZSBl dmFsdWF0aW9uLiAgRm9yIG1vcmUgY29tcGxleA0gICBjaGFuZ2VzIChlLmcuLCB3aGVuIHBzeWNo b2Fjb3VzdGljIGFzcGVjdHMgYXJlIGludm9sdmVkKSBvciBmb3INICAgY29udHJvdmVyc2lhbCBp c3N1ZXMsIGludGVybmFsIHRlc3Rpbmcgc2hvdWxkIGJlIHBlcmZvcm1lZC4gIEFuDSAgIGV4YW1w bGUgb2YgaW50ZXJuYWwgdGVzdGluZyB3b3VsZCBiZSB0byBoYXZlIGluZGl2aWR1YWwgcGFydGlj aXBhbnRzDSAgIHJhdGUgdGhlIGRlY29kZWQgc2FtcGxlcyB1c2luZyBvbmUgb2YgdGhlIGVzdGFi bGlzaGVkIHRlc3RpbmcNICAgbWV0aG9kb2xvZ2llcywgc3VjaCBhcyBJVFUtUiBCUy4xNTM0IChN VVNIUkEpLg0NICAgVGhyb3VnaG91dCB0aGUgcHJvY2VzcywgaXQgd2lsbCBiZSBpbXBvcnRhbnQg dG8gbWFrZSB1c2Ugb2YgdGhlDSAgIEludGVybmV0IGNvbW11bml0eSBhdCBsYXJnZSBmb3IgcmVh bC13b3JsZCBkaXN0cmlidXRlZCB0ZXN0aW5nLiAgVGhpcw0gICB3aWxsIGVuYWJsZSBtYW55IGRp ZmZlcmVudCBwZW9wbGUgd2l0aCBkaWZmZXJlbnQgZXF1aXBtZW50IGFuZCB1c2UNICAgY2FzZXMg dG8gdGVzdCB0aGUgY29kZWMgYW5kIHJlcG9ydCBhbnkgcHJvYmxlbXMgdGhleSBleHBlcmllbmNl LiAgSW4NICAgdGhlIHNhbWUgd2F5LCB0aGlyZC1wYXJ0eSBzb2Z0d2FyZSBkZXZlbG9wZXJzIHdp bGwgYmUgZW5jb3VyYWdlZCB0bw0gICBpbnRlZ3JhdGUgdGhlIGNvZGVjICh3aXRoIGEgd2Fybmlu ZyBhYm91dCB0aGUgYml0LXN0cmVhbSBub3QgYmVpbmcNICAgZmluYWwpIGFuZCBwcm92aWRlIGZl ZWRiYWNrIG9uIGl0cyBwZXJmb3JtYW5jZSBpbiByZWFsLXdvcmxkIHVzZQ0gICBjYXNlcy4NDSAg IENoYXJhY3Rlcml6YXRpb24gb2YgdGhlIGZpbmFsIGNvZGVjIG11c3QgYmUgYmFzZWQgb24gdGhl IHJlZmVyZW5jZQ0gICBpbXBsZW1lbnRhdGlvbiBvbmx5IChhbmQgbm90IG9uIGFueSAicHJpdmF0 ZSBpbXBsZW1lbnRhdGlvbiIpLiAgVGhpcw0gICBjYW4gYmUgcGVyZm9ybWVkIGJ5IGluZGVwZW5k ZW50IHRlc3RpbmcgbGFicyBvciwgaWYgdGhpcyBpcyBub3QNICAgcG9zc2libGUsIHVzaW5nIHRo ZSB0ZXN0aW5nIGxhYnMgb2YgdGhlIG9yZ2FuaXphdGlvbnMgdGhhdCBjb250cmlidXRlDSAgIHRv IHRoZSBJbnRlcm5ldCBTdGFuZGFyZHMgUHJvY2Vzcy4gIFBhY2tldCBsb3NzIHJvYnVzdG5lc3Mg c2hvdWxkIGJlDSAgIGV2YWx1YXRlZCB1c2luZyBhY3R1YWwgbG9zcyBwYXR0ZXJucyBjb2xsZWN0 ZWQgZnJvbSB1c2Ugb3ZlciB0aGUNICAgSW50ZXJuZXQsIHJhdGhlciB0aGFuIHRoZW9yZXRpY2Fs IG1vZGVscy4gIFRoZSBnb2FscyBvZiB0aGUNICAgY2hhcmFjdGVyaXphdGlvbiBwaGFzZSBhcmUg dG86DQ0gICBvICBlbnN1cmUgdGhhdCB0aGUgcmVxdWlyZW1lbnRzIGhhdmUgYmVlbiBmdWxmaWxs ZWQNDSAgIG8gIGd1aWRlIHRoZSBJRVNHIGluIGl0cyBldmFsdWF0aW9uIG9mIHRoZSByZXN1bHRp bmcgd29yaw0NICAgbyAgYXNzaXN0IGFwcGxpY2F0aW9uIGRldmVsb3BlcnMgaW4gdW5kZXJzdGFu ZGluZyB3aGV0aGVyIHRoZSBjb2RlYw0gICAgICBpcyBzdWl0YWJsZSBmb3IgYSBwYXJ0aWN1bGFy IGFwcGxpY2F0aW9uDQ0gICBUaGUgZXhhY3QgbWV0aG9kb2xvZ3kgZm9yIHRoZSBjaGFyYWN0ZXJp emF0aW9uIHBoYXNlIGlzIHN0aWxsIHN1YmplY3QNICAgdG8gZGlzY3Vzc2lvbiB3aXRoaW4gdGhl IHdvcmtpbmcgZ3JvdXAuDQ0NDQ0NDQ0NDVZhbGluLCBldCBhbC4gICAgICAgICAgICBFeHBpcmVz IEFwcmlsIDE1LCAyMDExICAgICAgICAgICAgICAgICBbUGFnZSA3XQ0MDUludGVybmV0LURyYWZ0 ICAgICAgICAgICAgICBDb2RlYyBHdWlkZWxpbmVzICAgICAgICAgICAgICAgIE9jdG9iZXIgMjAx MA0NDTQuICBSZXF1aXJlbWVudHMgQ29uZm9ybWFuY2UNDSAgIEl0IGlzIHRoZSByZXNwb25zaWJp bGl0eSBvZiB0aGUgd29ya2luZyBncm91cCB0byBkZWZpbmUgY3JpdGVyaWEgZm9yDSAgIGV2YWx1 YXRpbmcgY29uZm9ybWFuY2UsIGluY2x1ZGluZyBidXQgbm90IGxpbWl0ZWQgdG8gY29tcGFyaXNv biB0b29scw0gICBhbmQgdGVzdCB2ZWN0b3JzLiAgVGhlIGZvbGxvd2luZyB0ZXh0IHByb3ZpZGVz IHN1Z2dlc3Rpb25zIGZvcg0gICBjb25zaWRlcmF0aW9uIGJ5IHRoZSB3b3JraW5nIGdyb3VwOg0N ICAgMS4gIEFueSBjb2RlYyBzcGVjaWZpZWQgYnkgdGhlIElFVEYgbXVzdCBpbmNsdWRlIHNvdXJj ZSBjb2RlIGZvciBhDSAgICAgICBub3JtYXRpdmUgQzg5IGltcGxlbWVudGF0aW9uLCBkb2N1bWVu dGVkIGluIGFuIEludGVybmV0IERyYWZ0DSAgICAgICBkZXN0aW5lZCBmb3Igc3RhbmRhcmRzIHRy YWNrIFJGQy4gIFRoaXMgaW1wbGVtZW50YXRpb24gd2lsbCBiZQ0gICAgICAgdXNlZCB0byB2ZXJp ZnkgY29uZm9ybWFuY2Ugb2YgYW4gaW1wbGVtZW50YXRpb24uICBBbHRob3VnaCBhIHRleHQNICAg ICAgIGRlc2NyaXB0aW9uIG9mIHRoZSBhbGdvcml0aG0gc2hvdWxkIGJlIHByb3ZpZGVkLCBpdHMg dXNlIHNob3VsZA0gICAgICAgYmUgbGltaXRlZCB0byBoZWxwaW5nIHRoZSByZWFkZXIgaW4gdW5k ZXJzdGFuZGluZyB0aGUgc291cmNlDSAgICAgICBjb2RlLiAgU2hvdWxkIHRoZSBkZXNjcmlwdGlv biBjb250cmFkaWN0IHRoZSBzb3VyY2UgY29kZSwgdGhlDSAgICAgICBsYXR0ZXIgc2hhbGwgdGFr ZSBwcmVjZWRlbmNlLiAgRm9yIGNvbnZlbmllbmNlLCB0aGUgc291cmNlIGNvZGUNICAgICAgIG1h eSBiZSBwcm92aWRlZCBpbiBjb21wcmVzc2VkIGZvcm0sIHdpdGggYmFzZTY0IGVuY29kaW5nLg0N ICAgMi4gIEl0IGlzIHRoZSBpbnRlbnRpb24gb2YgdGhlIGdyb3VwIHRvIGFsbG93IHRoZSBncmVh dGVzdCBwb3NzaWJsZQ0gICAgICAgY2hvaWNlIG9mIGZyZWVkb20gaW4gaW1wbGVtZW50aW5nIHRo ZSBzcGVjaWZpY2F0aW9uLg0gICAgICAgQWNjb3JkaW5nbHksIHRoZSBudW1iZXIgb2YgYmluZGlu ZyBSRkMyMTE5IGtleXdvcmRzIGlzIGdvaW5nIHRvDSAgICAgICBiZSB0aGUgbWluaW11bSBzdGls bCBhbGxvd2luZyBmb3IgaW50ZXJvcGFibGUgaW1wbGVtZW50YXRpb25zLg0gICAgICAgSW4gcHJh Y3RpY2UgdGhpcyBnZW5lcmFsbHkgbWVhbnMgdGhhdCBvbmx5IHRoZSBkZWNvZGVyIG5lZWRzIHRv DSAgICAgICBiZSBub3JtYXRpdmUsIHNvIHRoYXQgdGhlIGVuY29kZXIgY2FuIGltcHJvdmUgb3Zl ciB0aW1lLiAgVGhpcw0gICAgICAgYWxzbyBlbmFibGVzIGRpZmZlcmVudCB0cmFkZW9mZnMgYmV0 d2VlbiBxdWFsaXR5IGFuZCBjb21wbGV4aXR5Lg0NICAgMy4gIFRvIHJlZHVjZSB0aGUgcmlzayBv ZiBiaWFzIHRvd2FyZHMgY2VydGFpbiBDUFUvRFNQIGFyY2hpdGVjdHVyZXMsDSAgICAgICBpZGVh bGx5IHRoZSBkZWNvZGVyIHNwZWNpZmljYXRpb24gc2hvdWxkIG5vdCByZXF1aXJlICJiaXQtZXhh Y3QiDSAgICAgICBjb25mb3JtYW5jZSB3aXRoIHRoZSByZWZlcmVuY2UgaW1wbGVtZW50YXRpb24u ICBUaGUgb3V0cHV0IG9mIGENICAgICAgIGRlY29kZXIgaW1wbGVtZW50YXRpb24gc2hvdWxkIG9u bHkgYmUgImNsb3NlIGVub3VnaCIgdG8gdGhlDSAgICAgICBvdXRwdXQgb2YgdGhlIHJlZmVyZW5j ZSBkZWNvZGVyLiAgQSBjb21wYXJpc29uIHRvb2wgc2hvdWxkIGJlDSAgICAgICBwcm92aWRlZCBh bG9uZyB3aXRoIHRoZSBjb2RlYyB0byB2ZXJpZnkgb2JqZWN0aXZlbHkgdGhhdCB0aGUNICAgICAg IG91dHB1dCBvZiBhIGRlY29kZXIgaXMgbGlrZWx5IHRvIGJlIHBlcmNlcHR1YWxseQ0gICAgICAg aW5kaXN0aW5ndWlzaGFibGUgZnJvbSB0aGF0IG9mIHRoZSByZWZlcmVuY2UgZGVjb2Rlci4gIEhv d2V2ZXIsDSAgICAgICBhbiBpbXBsZW1lbnRhdGlvbiBtYXkgc3RpbGwgd2lzaCB0byBwcm9kdWNl IGFuIG91dHB1dCB0aGF0IGlzDSAgICAgICBiaXQtZXhhY3Qgd2l0aCB0aGUgcmVmZXJlbmNlIGlt cGxlbWVudGF0aW9uIHRvIHNpbXBsaWZ5IHRoZQ0gICAgICAgdGVzdGluZyBwcm9jZWR1cmUuDQ0g ICA0LiAgVG8gZW5zdXJlIGZyZWVkb20gb2YgaW1wbGVtZW50YXRpb24sIGRlY29kZXItc2lkZSBv bmx5IGVycm9yDSAgICAgICBjb25jZWFsbWVudCBkb2VzIG5vdCBuZWVkIHRvIGJlIHNwZWNpZmll ZCwgYWx0aG91Z2ggdGhlIHJlZmVyZW5jZQ0gICAgICAgaW1wbGVtZW50YXRpb24gc2hvdWxkIGlu Y2x1ZGUgdGhlIHNhbWUgUExDIGFsZ29yaXRobSBhcyB1c2VkIGluDSAgICAgICB0aGUgdGVzdGlu ZyBwaGFzZS4gIElzIGl0IHVwIHRvIHRoZSB3b3JraW5nIGdyb3VwIHRvIGRlY2lkZQ0gICAgICAg d2hldGhlciBtaW5pbXVtIHJlcXVpcmVtZW50cyBvbiBQTEMgcXVhbGl0eSB3aWxsIGJlIHJlcXVp cmVkIGZvcg0gICAgICAgY29tcGxpYW5jZSB3aXRoIHRoZSBzcGVjaWZpY2F0aW9uLiAgT2J2aW91 c2x5LCBhbnkgaW5mb3JtYXRpb24NICAgICAgIHNpZ25hbGVkIGluIHRoZSBiaXRzdHJlYW0gaW50 ZW5kZWQgdG8gYWlkIFBMQyBuZWVkcyB0byBiZQ0gICAgICAgc3BlY2lmaWVkLg0NICAgNS4gIEFu IGVuY29kZXIgaW1wbGVtZW50YXRpb24gc2hvdWxkIG5vdCBiZSByZXF1aXJlZCB0byBtYWtlIHVz ZSBvZg0gICAgICAgYWxsIHRoZSAiZmVhdHVyZXMiICh0b29scykgaW4gdGhlIGJpdC1zdHJlYW0g ZGVmaW5pdGlvbi4NDQ0NVmFsaW4sIGV0IGFsLiAgICAgICAgICAgIEV4cGlyZXMgQXByaWwgMTUs IDIwMTEgICAgICAgICAgICAgICAgIFtQYWdlIDhdDQwNSW50ZXJuZXQtRHJhZnQgICAgICAgICAg ICAgIENvZGVjIEd1aWRlbGluZXMgICAgICAgICAgICAgICAgT2N0b2JlciAyMDEwDQ0NICAgICAg IEhvd2V2ZXIsIHRoZSBjb2RlYyBzcGVjaWZpY2F0aW9uIG1heSByZXF1aXJlIHRoYXQgYW4gZW5j b2Rlcg0gICAgICAgaW1wbGVtZW50YXRpb24gYmUgYWJsZSB0byBnZW5lcmF0ZSBhbnkgcG9zc2li bGUgYml0LXJhdGUuICBVbmxlc3MNICAgICAgIGEgcGFydGljdWxhciAicHJvZmlsZSIgaXMgZGVm aW5lZCBpbiB0aGUgc3BlY2lmaWNhdGlvbiwgdGhlDSAgICAgICBkZWNvZGVyIG11c3QgYmUgYWJs ZSB0byBkZWNvZGUgYWxsIGZlYXR1cmVzIG9mIHRoZSBiaXQtc3RyZWFtLg0gICAgICAgVGhlIGRl Y29kZXIgbXVzdCBhbHNvIGJlIGFibGUgdG8gaGFuZGxlIGFueSBjb21iaW5hdGlvbiBvZiBiaXRz LA0gICAgICAgZXZlbiBjb21iaW5hdGlvbnMgdGhhdCBjYW5ub3QgYmUgZ2VuZXJhdGVkIGJ5IHRo ZSByZWZlcmVuY2UNICAgICAgIGVuY29kZXIuICBJdCBpcyByZWNvbW1lbmRlZCB0aGF0IHRoZSBk ZWNvZGVyIHNwZWNpZmljYXRpb24gc2hhbGwNICAgICAgIGRlZmluZSBleGFjdGx5IGhvdyB0aGUg ZGVjb2RlciBzaG91bGQgcmVhY3QgdG8gImltcG9zc2libGUiDSAgICAgICBwYWNrZXRzLiAgSG93 ZXZlciwgYW4gZW5jb2RlciBtdXN0IG5ldmVyIGdlbmVyYXRlIHN1Y2ggcGFja2V0cw0gICAgICAg dGhhdCBkbyBub3QgY29uZm9ybSB0byB0aGUgYml0LXN0cmVhbSBkZWZpbml0aW9uLg0NICAgNi4g IENvbXByZXNzZWQgdGVzdCB2ZWN0b3JzIHNob3VsZCBiZSBwcm92aWRlZCBhcyBhIG1lYW5zIHRv IHZlcmlmeQ0gICAgICAgY29uZm9ybWFuY2Ugd2l0aCB0aGUgZGVjb2RlciBzcGVjaWZpY2F0aW9u LiAgVGhlc2UgdGVzdCB2ZWN0b3JzDSAgICAgICBzaG91bGQgZXhlcmNpc2UgYWxsIHBhdGhzIGlu IHRoZSBkZWNvZGVyICgxMDAlIGNvZGUgY292ZXJhZ2UpLg0NICAgNy4gIFdoaWxlIHRoZSBleGFj dCBlbmNvZGVyIHdpbGwgbm90IGJlIHNwZWNpZmllZCwgaXQgaXMgcmVjb21tZW5kZWQNICAgICAg IHRvIHNwZWNpZnkgb2JqZWN0aXZlIG1lYXN1cmVtZW50IHRhcmdldHMgZm9yIGFuIGVuY29kZXIs IGJlbG93DSAgICAgICB3aGljaCB1c2Ugb2YgYSBwYXJ0aWN1bGFyIGVuY29kZXIgaW1wbGVtZW50 YXRpb24gaXMgbm90DSAgICAgICByZWNvbW1lbmRlZC4gIEZvciBleGFtcGxlLCBvbmUgc3VjaCBz cGVjaWZpY2F0aW9uIGNvdWxkIGJlOiAidGhlDSAgICAgICB1c2Ugb2YgYW4gZW5jb2RlciB3aG9z ZSBQRVNRIE1PUyBpcyBsZXNzIHRoYW4gMC4xIGJlbG93IHRoZQ0gICAgICAgcmVmZXJlbmNlIGVu Y29kZXIgaW4gdGhlIGZvbGxvd2luZyBjb25kaXRpb25zIGlzIG5vdA0gICAgICAgcmVjb21tZW5k ZWQiLg0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDVZhbGluLCBldCBhbC4gICAgICAgICAg ICBFeHBpcmVzIEFwcmlsIDE1LCAyMDExICAgICAgICAgICAgICAgICBbUGFnZSA5XQ0MDUludGVy bmV0LURyYWZ0ICAgICAgICAgICAgICBDb2RlYyBHdWlkZWxpbmVzICAgICAgICAgICAgICAgIE9j dG9iZXIgMjAxMA0NDTUuICBJbnRlbGxlY3R1YWwgUHJvcGVydHkNDSAgIFByb2R1Y2luZyBhbiB1 bmVuY3VtYmVyZWQgY29kZWMgaXMgZGVzaXJhYmxlIGZvciB0aGUgZm9sbG93aW5nDSAgIHJlYXNv bnM6DQ0gICBvICBJdCBpcyB0aGUgZXhwZXJpZW5jZSBvZiBhIHdpZGUgdmFyaWV0eSBvZiBhcHBs aWNhdGlvbiBkZXZlbG9wZXJzDSAgICAgIGFuZCBzZXJ2aWNlIHByb3ZpZGVycyB0aGF0IGVuY3Vt YnJhbmNlcyBzdWNoIGFzIGxpY2Vuc2luZyBhbmQNICAgICAgcm95YWx0aWVzIG1ha2UgaXQgZGlm ZmljdWx0IHRvIGltcGxlbWVudCwgZGVwbG95LCBhbmQgZGlzdHJpYnV0ZQ0gICAgICBhdWRpbyBh cHBsaWNhdGlvbnMgZm9yIHVzZSBieSB0aGUgSW50ZXJuZXQgY29tbXVuaXR5Lg0NICAgbyAgSXQg aXMgYmVuZWZpY2lhbCB0byBoYXZlIGxvdy1jb3N0IG9wdGlvbnMgd2hlbmV2ZXIgcG9zc2libGUN ICAgICAgYmVjYXVzZSBzdGFuZGFsb25lIHZvaWNlIHNlcnZpY2VzIGFyZSBiZWluZyBjb21tb2Rp dGl6ZWQgYW5kDSAgICAgIHNtYWxsLCBpbm5vdmF0aXZlIGRldmVsb3BtZW50IHRlYW1zIG9mdGVu IGNhbm5vdCBhZmZvcmQgdG8gcGF5DSAgICAgIHBlci1jaGFubmVsIGxpY2Vuc2luZyBmZWVzIGFu ZCByb3lhbHRpZXMuDQ0gICBvICBNYW55IG1hcmtldCBzZWdtZW50cyBhcmUgbW92aW5nIGF3YXkg ZnJvbSBzZWxsaW5nIGhhcmQtY29kZWQNICAgICAgaGFyZHdhcmUgZGV2aWNlcyBhbmQgdG93YXJk IGZyZWVseSBkaXN0cmlidXRpbmcgZW5kLXVzZXIgc29mdHdhcmU7DSAgICAgIHRoaXMgaXMgdHJ1 ZSBvZiBudW1lcm91cyBsYXJnZSBhcHBsaWNhdGlvbiBwcm92aWRlcnMgYW5kIGV2ZW4NICAgICAg dGVsY29zIHRoZW1zZWx2ZXMuDQ0gICBvICBDb21wYXRpYmlsaXR5IHdpdGggdGhlIGxpY2Vuc2lu ZyBvZiB0eXBpY2FsIG9wZW4gc291cmNlDSAgICAgIGFwcGxpY2F0aW9ucyBpbXBsaWVzIHRoZSBu ZWVkIHRvIGF2b2lkIGVuY3VtYnJhbmNlcywgaW5jbHVkaW5nDSAgICAgIGV2ZW4gdGhlIHJlcXVp cmVtZW50IHRvIG9idGFpbiBhIGxpY2Vuc2UgZm9yIGltcGxlbWVudGF0aW9uLA0gICAgICBkZXBs b3ltZW50LCBvciB1c2UgKGV2ZW4gaWYgdGhlIGxpY2Vuc2UgZG9lcyBub3QgcmVxdWlyZSB0aGUN ICAgICAgcGF5bWVudCBvZiBhIGZlZSkuDQ0gICBUaGVyZWZvcmUsIGEgY29kZWMgdGhhdCBjYW4g YmUgd2lkZWx5IGltcGxlbWVudGVkIGFuZCBlYXNpbHkNICAgZGlzdHJpYnV0ZWQgYW1vbmcgYXBw bGljYXRpb24gZGV2ZWxvcGVycywgc2VydmljZSBvcGVyYXRvcnMsIGFuZCBlbmQNICAgdXNlcnMg aXMgcHJlZmVycmVkLiAgTWFueSBleGlzdGluZyBjb2RlY3MgdGhhdCBtaWdodCBmdWxmaWxsIHNv bWUgb3INICAgbW9zdCBvZiB0aGUgdGVjaG5pY2FsIGF0dHJpYnV0ZXMgbGlzdGVkIGFib3ZlIGFy ZSBlbmN1bWJlcmVkIGluDSAgIHZhcmlvdXMgd2F5cy4gIEZvciBleGFtcGxlLCBwYXRlbnQgaG9s ZGVycyBtaWdodCByZXF1aXJlIHRoYXQgdGhvc2UNICAgd2lzaGluZyB0byBpbXBsZW1lbnQgdGhl IGNvZGVjIGluIHNvZnR3YXJlLCBkZXBsb3kgdGhlIGNvZGVjIGluIGENICAgc2VydmljZSwgb3Ig ZGlzdHJpYnV0ZSB0aGUgY29kZWMgaW4gc29mdHdhcmUgb3IgaGFyZHdhcmUgbmVlZCB0bw0gICBy ZXF1ZXN0IGEgbGljZW5zZSwgZW50ZXIgaW50byBhIGJ1c2luZXNzIGFncmVlbWVudCwgcGF5IGxp Y2Vuc2luZw0gICBmZWVzIG9yIHJveWFsdGllcywgb3IgYWRoZXJlIHRvIG90aGVyIHNwZWNpYWwg Y29uZGl0aW9ucyBvcg0gICByZXN0cmljdGlvbnMuICBCZWNhdXNlIHN1Y2ggZW5jdW1icmFuY2Vz IGhhdmUgbWFkZSBpdCBkaWZmaWN1bHQgdG8NICAgd2lkZWx5IGltcGxlbWVudCBhbmQgZWFzaWx5 IGRpc3RyaWJ1dGUgaGlnaC1xdWFsaXR5IGF1ZGlvIGNvZGVjcw0gICBhY3Jvc3MgdGhlIGVudGly ZSBJbnRlcm5ldCBjb21tdW5pdHksIHRoZSB3b3JraW5nIGdyb3VwIHByZWZlcnMNICAgdW5lbmN1 bWJlcmVkIHRlY2hub2xvZ2llcyBpbiBhIHdheSB0aGF0IGlzIGNvbnNpc3RlbnQgd2l0aCBCQ1Ag NzggYW5kDSAgIEJDUCA3OS4gIEluIHBhcnRpY3VsYXIsIHRoZSB3b3JraW5nIGdyb3VwIHNoYWxs IGhlZWQgdGhlIHByZWZlcmVuY2UNICAgc3RhdGVkIGluIEJDUCA3OTogIkluIGdlbmVyYWwsIElF VEYgd29ya2luZyBncm91cHMgcHJlZmVyDSAgIHRlY2hub2xvZ2llcyB3aXRoIG5vIGtub3duIElQ UiBjbGFpbXMgb3IsIGZvciB0ZWNobm9sb2dpZXMgd2l0aA0gICBjbGFpbXMgYWdhaW5zdCB0aGVt LCBhbiBvZmZlciBvZiByb3lhbHR5LWZyZWUgbGljZW5zaW5nLiIgIEFsdGhvdWdoDSAgIHRoaXMg cHJlZmVyZW5jZSBjYW5ub3QgZ3VhcmFudGVlIHRoYXQgdGhlIHdvcmtpbmcgZ3JvdXAgd2lsbCBw cm9kdWNlDSAgIGFuIHVuZW5jdW1iZXJlZCBjb2RlYywgdGhlIHdvcmtpbmcgZ3JvdXAgc2hhbGwg Zm9sbG93IEJDUCA3OSwgYW5kDSAgIGFkaGVyZSB0byB0aGUgc3Bpcml0IG9mIEJDUCA3OS4gIFRo ZSB3b3JraW5nIGdyb3VwIGNhbm5vdCBleHBsaWNpdGx5DSAgIHJ1bGUgb3V0IHRoZSBwb3NzaWJp bGl0eSBvZiBhZG9wdGluZyBlbmN1bWJlcmVkIHRlY2hub2xvZ2llczsNICAgaG93ZXZlciwgdGhl IHdvcmtpbmcgZ3JvdXAgd2lsbCB0cnkgdG8gYXZvaWQgZW5jdW1iZXJlZCB0ZWNobm9sb2dpZXMN DQ0NVmFsaW4sIGV0IGFsLiAgICAgICAgICAgIEV4cGlyZXMgQXByaWwgMTUsIDIwMTEgICAgICAg ICAgICAgICAgW1BhZ2UgMTBdDQwNSW50ZXJuZXQtRHJhZnQgICAgICAgICAgICAgIENvZGVjIEd1 aWRlbGluZXMgICAgICAgICAgICAgICAgT2N0b2JlciAyMDEwDQ0NICAgdGhhdCByZXF1aXJlIHJv eWFsdGllcyBvciBvdGhlciBlbmN1bWJyYW5jZXMgdGhhdCB3b3VsZCBwcmV2ZW50IHN1Y2gNdGVj aG5vbG9naWVzIGZyb20gYmVpbmcgZWFzeSB0byByZWRpc3RyaWJ1dGUgYW5kIHVzZS4gIEZvciB0 aGUgY29kZWMgdG8gYmUgdHJ1bHkgdW5lbmN1bWJlcmVkLCB0aGUgY29udHJpYnV0b3JzIHNob3Vs ZCBhZGhlcmUgdG8gdGhlICdObyBsaWNlbnNlIHJlcXVpcmVkJyBvcHRpb24sIGFzIGV4cHJlc3Nl ZCBpbiBCQ1AgNzkgc2VjdGlvbiA2LjU6ICIuLmMpIHdpdGhvdXQgdGhlIG5lZWQgdG8gb2J0YWlu IGEgbGljZW5zZSBmcm9tIHRoZSBJUFIgaG9sZGVyLiINDQ0gICBUaGUgZm9sbG93aW5nIGd1aWRl bGluZXMgd2lsbCBoZWxwIHRvIG1heGltaXplIHRoZSBvZGRzIHRoYXQgdGhlDSAgIGNvZGVjIHdp bGwgYmUgdW5lbmN1bWJlcmVkOg0NICAgMS4gIEluIGFjY29yZGFuY2Ugd2l0aCBCQ1AgNzkgW0lQ Ul0sIGNvbnRyaWJ1dGVkIGNvZGVjcyBzaG91bGQNICAgIHByZWZlcmFibHkgdXNlIHRlY2hub2xv Z2llcyB3aXRoIG5vIGtub3duIElQUiBjbGFpbXMgb3INICAgIHRlY2hub2xvZ2llcyB3aXRoIGFu IG9mZmVyIG9mIHJveWFsdHktZnJlZSAoUkYpIGxpY2Vuc2luZy4gC0lmIHRoZSB3b3JraW5nIGdy b3VwIGlzIHByZXNlbnRlZCB3aXRoIGEgY2hvaWNlIGEgdHJ1bHkgdW5lbmN1bWJlcmVkIGNvZGVj IHdpbGwgYmUgcHJlZmVyZWQuICAgICANDSAgIDIuICBXaGVuZXZlciBwb3NzaWJsZSwgdGhlIHdv cmtpbmcgZ3JvdXAgc2hvdWxkIHVzZSB0ZWNobm9sb2dpZXMgdGhhdA0gICAgICAgYXJlIHBlcmNl aXZlZCBieSB0aGUgcGFydGljaXBhbnRzIHRvIGJlIHNhZmVyIHdpdGggcmVnYXJkIHRvIElQUg0g ICAgICAgaXNzdWVzLg0NMy4gIENvbnRyaWJ1dG9ycyBtdXN0IGRpc2Nsb3NlIElQUiBhcyBzcGVj aWZpZWQgaW4gQkNQIDc5LiBDb250cmlidXRlZCB0ZWNobm9sb2d5IHdpbGwgbm90IGJlIGNvbnNp ZGVyZWQgZm9yIGludGVncmF0aW9uIGludG8gdGhlIGNvZGVjIGJlZm9yZSBhbiBJUFIgZGlzY2xv c3VyZSBoYXMgYmVlbiBtYWRlLg0NICAgNC4gIEluIGNhc2VzIHdoZXJlIG5vIFJGIGxpY2Vuc2Ug Y2FuIGJlIG9idGFpbmVkIHJlZ2FyZGluZyBhIHBhdGVudCwNICAgICAgIHRoZSBncm91cCBzaG91 bGQgY29uc2lkZXIgYWx0ZXJuYXRpdmUgYWxnb3JpdGhtcyBvciBtZXRob2RzLCBldmVuDSAgICAg ICBpZiB0aGV5IHJlc3VsdCBpbiBsb3dlciBxdWFsaXR5LCBoaWdoZXIgY29tcGxleGl0eSwgb3Ig b3RoZXJ3aXNlDSAgICAgICBsZXNzIGRlc2lyYWJsZSBjaGFyYWN0ZXJpc3RpY3MgKGluIG1vc3Qg Y2FzZXMsIHRoZSBkZWdyYWRhdGlvbg0gICAgICAgd2lsbCBsaWtlbHkgYmUgc21hbGwgb25jZSB0 aGUgYmVzdCBhbHRlcm5hdGl2ZSBoYXMgYmVlbg0gICAgICAgaWRlbnRpZmllZCkuDQ0gICA1LiAg SW4gYWNjb3JkYW5jZSB3aXRoIEJDUCA3OCBbVFJVU1RdLCB0aGUgc291cmNlIGNvZGUgZm9yIHRo ZQ0gICAgICAgcmVmZXJlbmNlIGltcGxlbWVudGF0aW9uIG11c3QgYmUgbWFkZSBhdmFpbGFibGUg dW5kZXIgYSBCU0Qtc3R5bGUNICAgICAgIGxpY2Vuc2UgKG9yIHdoYXRldmVyIGxpY2Vuc2UgaXMg ZGVmaW5lZCBhcyBhY2NlcHRhYmxlIGJ5IHRoZSBJRVRGDSAgICAgICBUcnVzdCB3aGVuIHRoZSBJ bnRlcm5ldC1EcmFmdCBkZWZpbmluZyB0aGUgcmVmZXJlbmNlDSAgICAgICBpbXBsZW1lbnRhdGlv biBpcyBwdWJsaXNoZWQpLg0NICAgSUVURiBwYXJ0aWNpcGFudHMgc2hvdWxkIGJlIGF3YXJlIHRo YXQsIGdpdmVuIHRoZSB3YXkgcGF0ZW50cyB3b3JrIGluDSAgIG1vc3QgY291bnRyaWVzLCB0aGUg cmVzdWx0aW5nIGNvZGVjIGNhbiBuZXZlciBiZSBndWFyYW50ZWVkIHRvIGJlDSAgIGZyZWUgb2Yg cGF0ZW50IGNsYWltcyBiZWNhdXNlIHNvbWUgcGF0ZW50cyBtYXkgbm90IGJlIGtub3duIHRvIHRo ZQ0gICBjb250cmlidXRvcnMsIHNvbWUgcGF0ZW50IGFwcGxpY2F0aW9ucyBtYXkgbm90IGJlIGRp c2Nsb3NlZCBhdCB0aGUNICAgdGltZSB0aGUgY29kZWMgaXMgZGV2ZWxvcGVkLCBhbmQgb25seSBj b3VydHMgb2YgbGF3IGNhbiBkZXRlcm1pbmUgdGhlDSAgIHZhbGlkaXR5IGFuZCBicmVhZHRoIG9m IHBhdGVudCBjbGFpbXMuICBIb3dldmVyLCB0aGVzZSBvYnNlcnZhdGlvbnMNICAgYXJlIG5vIGRp ZmZlcmVudCB3aXRoaW4gdGhlIEludGVybmV0IFN0YW5kYXJkcyBQcm9jZXNzIHRoYW4gdGhleSBh cmUNICAgZm9yIHN0YW5kYXJkaXphdGlvbiBvZiBjb2RlY3Mgd2l0aGluIG90aGVyIFNET3MgKG9y IGRldmVsb3BtZW50IG9mDSAgIGNvZGVjcyBvdXRzaWRlIHRoZSBjb250ZXh0IG9mIGFueSBTRE8p LCBhbmQgZnVydGhlcm1vcmUgYXJlIG5vDSAgIGRpZmZlcmVudCBmb3IgY29kZWNzIHRoYW4gZm9y IG90aGVyIHRlY2hub2xvZ2llcyB3b3JrZWQgb24gd2l0aGluIHRoZQ0gICBJRVRGLiAgSW4gYWxs IHRoZXNlIGNhc2VzLCB0aGUgYmVzdCBhcHByb2FjaCBpcyB0byBtaW5pbWl6ZSB0aGUgcmlzaw0g ICBvZiB1bmtub3dpbmdseSBpbmN1cnJpbmcgZW5jcnVtYnJhbmNlIG9uIGV4aXN0aW5nIHBhdGVu dHMuICBEZXNwaXRlDSAgIHRoZXNlIHByZWNhdXRpb25zLCBwYXJ0aWNpcGFudHMgbmVlZCB0byB1 bmRlcnN0YW5kIHRoYXQsIHByYWN0aWNhbGx5DSAgIHNwZWFraW5nLCBpdCBpcyBuZWFybHkgaW1w b3NzaWJsZSB0byBfZ3VhcmFudGVlXyB0aGF0IGltcGxlbWVudG9ycw0gICB3aWxsIG5vdCBpbmN1 ciBlbmN1bWJyYW5jZSBvbiBleGlzdGluZyBwYXRlbnRzLg0NDQ0NDQ0NVmFsaW4sIGV0IGFsLiAg ICAgICAgICAgIEV4cGlyZXMgQXByaWwgMTUsIDIwMTEgICAgICAgICAgICAgICAgW1BhZ2UgMTFd DQwNSW50ZXJuZXQtRHJhZnQgICAgICAgICAgICAgIENvZGVjIEd1aWRlbGluZXMgICAgICAgICAg ICAgICAgT2N0b2JlciAyMDEwDQ0NNi4gIFJlbGF0aW9uc2hpcCB3aXRoIE90aGVyIFNET3MNDSAg IEl0IGlzIHVuZGVyc3Rvb2QgdGhhdCBvdGhlciBTRE9zIGFyZSBhbHNvIGludm9sdmVkIGluIHRo ZSBjb2RlYw0gICBkZXZlbG9wbWVudCBhbmQgc3RhbmRhcmRpemF0aW9uLCBpbmNsdWRpbmcgYnV0 IG5vdCBuZWNlc3NhcmlseQ0gICBsaW1pdGVkIHRvOg0NICAgbyAgVGhlIFRlbGVjb21tdW5pY2F0 aW9uIFN0YW5kYXJkaXphdGlvbiBTZWN0b3IgKElUVS1UKSBvZiB0aGUNICAgICAgSW50ZXJuYXRp b25hbCBUZWxlY29tbXVuaWNhdGlvbiBVbmlvbiAoSVRVKSwgaW4gcGFydGljdWxhciBTdHVkeQ0g ICAgICBHcm91cCAxNg0NICAgbyAgVGhlIE1vdmluZyBQaWN0dXJlIEV4cGVydHMgR3JvdXAgKE1Q RUcpDQ0gICBvICBUaGUgRXVyb3BlYW4gVGVsZWNvbW11bmljYXRpb25zIFN0YW5kYXJkcyBJbnN0 aXR1dGUgKEVUU0kpDQ0gICBvICBUaGUgM3JkIEdlbmVyYXRpb24gUGFydG5lcnNoaXAgUHJvamVj dCAoM0dQUCkNDSAgIG8gIFRoZSAzcmQgR2VuZXJhdGlvbiBQYXJ0bmVyc2hpcCBQcm9qZWN0IDIg KDNHUFAyKQ0NICAgSXQgaXMgaW1wb3J0YW50IHRvIGVuc3VyZSB0aGF0IHN1Y2ggd29yayBkb2Vz IG5vdCBjb25zdGl0dXRlDSAgIHVuY29vcmRpbmF0ZWQgcHJvdG9jb2wgZGV2ZWxvcG1lbnQsIG9m IHRoZSBraW5kIGRlc2NyaWJlZCBpbg0gICBbVU5DT09SRF0gaW4gdGhlIGZvbGxvd2luZyBwcmlu Y2lwbGU6DQ0gICAgICBbVF1oZSBJQUIgY29uc2lkZXJzIGFuIGVzc2VudGlhbCBwcmluY2lwbGUg b2YgdGhlIHByb3RvY29sDSAgICAgIGRldmVsb3BtZW50IHByb2Nlc3MgdGhhdCBvbmx5IG9uZSBT RE8gbWFpbnRhaW5zIGRlc2lnbiBhdXRob3JpdHkNICAgICAgZm9yIGEgZ2l2ZW4gcHJvdG9jb2ws IHdpdGggdGhhdCBTRE8gaGF2aW5nIHVsdGltYXRlIGF1dGhvcml0eSBvdmVyDSAgICAgIHRoZSBh bGxvY2F0aW9uIG9mIHByb3RvY29sIHBhcmFtZXRlciBjb2RlLXBvaW50czsgZGVmaW5pbmcgdGhl DSAgICAgIGludGVuZGVkIHNlbWFudGljcywgaW50ZXJwcmV0YXRpb24sIGFuZCBhY3Rpb25zIGFz c29jaWF0ZWQgd2l0aA0gICAgICB0aG9zZSBjb2RlLXBvaW50cy4NDSAgIFRoZSB3b3JrIGVudmlz aW9uZWQgYnkgdGhpcyBndWlkZWxpbmVzIGRvY3VtZW50IGlzIG5vdA0gICAidW5jb29yZGluYXRl ZCIgaW4gdGhlIHNlbnNlIGRlc2NyaWJlZCBpbiB0aGUgZm9yZWdvaW5nIHF1b3RlLCBmb3INICAg dGhlIGZvbGxvd2luZyByZWFzb25zOg0NICAgbyAgSW50ZXJuZXQgc2lnbmFsbGluZyB0ZWNobm9s b2dpZXMgYXJlIGRlc2lnbmVkIHRvIGVuYWJsZSB0aGUNICAgICAgbmVnb3RpYXRpb24gb2YgYW55 IGNvZGVjcyB0aGF0IGFyZSBzdXBwb3J0ZWQgaW4gYSBwYXJ0aWN1bGFyDSAgICAgIGFwcGxpY2F0 aW9uIChzdWNoIHNpZ25hbGxpbmcgdGVjaG5vbG9naWVzIGluY2x1ZGUgdGhlIFNlc3Npb24NICAg ICAgSW5pdGlhdGlvbiBQcm90b2NvbCBbU0lQXSwgU2Vzc2lvbiBEZXNjcmlwdGlvbiBQcm90b2Nv bCBbU0RQXSwgYW5kDSAgICAgIHRoZSBFeHRlbnNpYmxlIE1lc3NhZ2luZyBhbmQgUHJlc2VuY2Ug UHJvdG9jb2wgW1hNUFBdIGV4dGVuc2lvbnMNICAgICAgZm9yIG1lZGlhIG5lZ290aWF0aW9uIGFz IHNwZWNpZmllZCBpbiBbSmluZ2xlXSkuDQ0gICBvICBJbnRlcm5ldCB0cmFuc3BvcnQgdGVjaG5v bG9naWVzIHN1Y2ggYXMgdGhlIFJlYWwtdGltZSBUcmFuc3BvcnQNICAgICAgUHJvdG9jb2wgW1JU UF0gKGluY2x1ZGluZyBzZWN1cmUgdHJhbnNwb3J0IGFzIGRlc2NyaWJlZCBpbiBbU1JUUF0pDSAg ICAgIGFyZSBkZXNpZ25lZCB0byBzdXBwb3J0IGFueSBjb2RlYyBmb3Igd2hpY2ggUlRQIHBhY2tl dGl6YXRpb24NICAgICAgcnVsZXMgaGF2ZSBiZWVuIGRlZmluZWQuDQ0gICBvICBUaGUgSUVURiBj b2RlYyB3b3JraW5nIGdyb3VwIHdpbGwgZm9jdXMgb24gaXNzdWVzIHRoYXQgYXJlDSAgICAgIHNw ZWNpZmljIHRvIHRoZSBJbnRlcm5ldCwgaW5jbHVkaW5nIHJvYnVzdG5lc3MgdG8gcGFja2V0IGxv c3MgYW5kDSAgICAgIG90aGVyIGFzcGVjdHMgb2YgcGFja2V0IHRyYW5zbWlzc2lvbiBvdmVyIHRo ZSBJbnRlcm5ldC4gIElzc3Vlcw0NDQ1WYWxpbiwgZXQgYWwuICAgICAgICAgICAgRXhwaXJlcyBB cHJpbCAxNSwgMjAxMSAgICAgICAgICAgICAgICBbUGFnZSAxMl0NDA1JbnRlcm5ldC1EcmFmdCAg ICAgICAgICAgICAgQ29kZWMgR3VpZGVsaW5lcyAgICAgICAgICAgICAgICBPY3RvYmVyIDIwMTAN DQ0gICAgICB0aGF0IGFyZSBzcGVjaWZpYyB0byBub24tSW50ZXJuZXQgdHJhbnNwb3J0cyAoZS5n LiwgcmFkaW8NICAgICAgY29tbXVuaWNhdGlvbiBhbmQgY2lyY3VpdC1zd2l0Y2hlZCBuZXR3b3Jr cykgYXJlIHNwZWNpZmljYWxseSBvdXQNICAgICAgb2Ygc2NvcGUuDQ0gICBBbHRob3VnaCB0aGVy ZSBpcyBhbHJlYWR5IHN1ZmZpY2llbnQgY29kZWMgZXhwZXJ0aXNlIGF2YWlsYWJsZSBhbW9uZw0g ICBJRVRGIHBhcnRpY2lwYW50cyB0byBjb21wbGV0ZSB0aGUgZW52aXNpb25lZCB3b3JrLCBhZGRp dGlvbmFsDSAgIGNvbnRyaWJ1dGlvbnMgYXJlIHdlbGNvbWUgd2l0aGluIHRoZSBmcmFtZXdvcmsg b2YgdGhlIEludGVybmV0DSAgIFN0YW5kYXJkcyBQcm9jZXNzLCBpbiB0aGUgZm9sbG93aW5nIHdh eXM6DQ0gICBvICBJbmRpdmlkdWFscyB3aG8gYXJlIHRlY2huaWNhbCBjb250cmlidXRvcnMgdG8g Y29kZWMgd29yayB3aXRoaW4NICAgICAgb3RoZXIgU0RPcyBjYW4gcGFydGljaXBhdGUgZGlyZWN0 bHkgaW4gY29kZWMgd29yayB3aXRoaW4gdGhlIElFVEYuDQ0gICBvICBPdGhlciBTRE9zIGNhbiBj b250cmlidXRlIHRoZWlyIGV4cGVydGlzZSAoZS5nLiwgY29kZWMNICAgICAgY2hhcmFjdGVyaXph dGlvbiBhbmQgZXZhbHVhdGlvbiB0ZWNobmlxdWVzKSBhbmQgdGh1cyBmYWNpbGl0YXRlDSAgICAg IHRoZSB0ZXN0aW5nIG9mIGEgY29kZWMgcHJvZHVjZWQgYnkgdGhlIElFVEYuDQ0gICBvICBBbnkg U0RPIGNhbiBwcm92aWRlIGlucHV0IHRvIElFVEYgd29yayB0aHJvdWdoIGxpYWlzb24gc3RhdGVt ZW50cy4NDSAgIEhvd2V2ZXIsIGl0IGlzIGltcG9ydGFudCB0byBub3RlIHRoYXQgZmluYWwgcmVz cG9uc2liaWxpdHkgZm9yIHRoZQ0gICBkZXZlbG9wbWVudCBwcm9jZXNzIGFuZCB0aGUgcmVzdWx0 aW5nIGNvZGVjIHdpbGwgcmVtYWluIHdpdGggdGhlIElFVEYNICAgYXMgZ292ZXJuZWQgYnkgQkNQ IDkgW1BST0NFU1NdLg0NICAgRmluYWxseSwgdGhlcmUgaXMgcHJlY2VkZW50IGZvciB0aGUgY29u dHJpYnV0aW9uIG9mIGNvZGVjcyBkZXZlbG9wZWQNICAgZWxzZXdoZXJlIHRvIHRoZSBJVFUtVCAo ZS5nLiwgQU1SIFdpZGViYW5kIHdhcyBzdGFuZGFyZGl6ZWQNICAgb3JpZ2luYWxseSB3aXRoaW4g M0dQUCkuICBUaGlzIGlzIGEgbW9kZWwgdG8gZXhwbG9yZSBhcyB0aGUgSUVURg0gICBjb29yZGlu YXRlcyBmdXJ0aGVyIHdpdGggdGhlIElUVS1UIGluIGFjY29yZGFuY2Ugd2l0aCB0aGUNICAgY29s bGFib3JhdGlvbiBndWlkZWxpbmVzIGRlZmluZWQgaW4gW0NPTExBQl0uDQ0NDQ0NDQ0NDQ0NDQ0N DQ0NDQ0NDQ0NDVZhbGluLCBldCBhbC4gICAgICAgICAgICBFeHBpcmVzIEFwcmlsIDE1LCAyMDEx ICAgICAgICAgICAgICAgIFtQYWdlIDEzXQ0MDUludGVybmV0LURyYWZ0ICAgICAgICAgICAgICBD b2RlYyBHdWlkZWxpbmVzICAgICAgICAgICAgICAgIE9jdG9iZXIgMjAxMA0NDTcuICBTZWN1cml0 eSBDb25zaWRlcmF0aW9ucw0NICAgVGhlIHByb2NlZHVyYWwgZ3VpZGVsaW5lcyBmb3IgY29kZWMg ZGV2ZWxvcG1lbnQgZG8gbm90IGhhdmUgc2VjdXJpdHkNICAgY29uc2lkZXJhdGlvbnMuICBIb3dl dmVyLCB0aGUgcmVzdWx0aW5nIGNvZGVjIG5lZWRzIHRvIHRha2UNICAgYXBwcm9wcmlhdGUgc2Vj dXJpdHkgY29uc2lkZXJhdGlvbnMgaW50byBhY2NvdW50LCBmb3IgZXhhbXBsZSBhcw0gICBvdXRs aW5lZCBpbiBbRE9TXSBhbmQgW1NFQ0dVSURFXS4NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0N DQ0NDQ0NDQ0NDQ0NDQ0NDQ0NVmFsaW4sIGV0IGFsLiAgICAgICAgICAgIEV4cGlyZXMgQXByaWwg MTUsIDIwMTEgICAgICAgICAgICAgICAgW1BhZ2UgMTRdDQwNSW50ZXJuZXQtRHJhZnQgICAgICAg ICAgICAgIENvZGVjIEd1aWRlbGluZXMgICAgICAgICAgICAgICAgT2N0b2JlciAyMDEwDQ0NOC4g IElBTkEgQ29uc2lkZXJhdGlvbnMNDSAgIFRoaXMgZG9jdW1lbnQgaGFzIG5vIGFjdGlvbnMgZm9y IElBTkEuDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDVZh bGluLCBldCBhbC4gICAgICAgICAgICBFeHBpcmVzIEFwcmlsIDE1LCAyMDExICAgICAgICAgICAg ICAgIFtQYWdlIDE1XQ0MDUludGVybmV0LURyYWZ0ICAgICAgICAgICAgICBDb2RlYyBHdWlkZWxp bmVzICAgICAgICAgICAgICAgIE9jdG9iZXIgMjAxMA0NDTkuICBBY2tub3dsZWRnbWVudHMNDSAg IFdlIHdvdWxkIGxpa2UgdG8gdGhhbmsgYWxsIHRoZSBvdGhlciBwZW9wbGUgd2hvIGNvbnRyaWJ1 dGVkIGRpcmVjdGx5DSAgIG9yIGluZGlyZWN0bHkgdG8gdGhpcyBkb2N1bWVudCwgaW5jbHVkaW5n IEphc29uIEZpc2NobCwgR3JlZ29yeQ0gICBNYXh3ZWxsLCBBbGFuIER1cmljLCBKb25hdGhhbiBD aHJpc3RlbnNlbiwgSnVsaWFuIFNwaXR0a2EsIE1pY2hhZWwNICAgS25hcHBlLCBUaW1vdGh5IFRl cnJpYmVycnksIENocmlzdGlhbiBIb2VuZSwgU3RlcGhhbiBXZW5nZXIgYW5kIEhlbnJ5DSAgIFNp bm5yZWljaC4gIFdlIGFsc28gbGlrZSB0byB0aGFuayBDdWxsZW4gSmVubmluZ3MgYW5kIEdyZWdv cnkNICAgTGVib3ZpdHogZm9yIHRoZWlyIGFkdmljZS4gIFNwZWNpYWwgdGhhbmtzIHRvIFBldGVy IFNhaW50LUFuZHJlLCB3aG8NICAgb3JpZ2luYWxseSBjby1hdXRob3JlZCB0aGlzIGRvY3VtZW50 Lg0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ1WYWxpbiwgZXQgYWwu ICAgICAgICAgICAgRXhwaXJlcyBBcHJpbCAxNSwgMjAxMSAgICAgICAgICAgICAgICBbUGFnZSAx Nl0NDA1JbnRlcm5ldC1EcmFmdCAgICAgICAgICAgICAgQ29kZWMgR3VpZGVsaW5lcyAgICAgICAg ICAgICAgICBPY3RvYmVyIDIwMTANDQ0xMC4gIFJlZmVyZW5jZXMNDTEwLjEuICBOb3JtYXRpdmUg UmVmZXJlbmNlcw0NICAgW0lQUl0gICAgICBCcmFkbmVyLCBTLiwgIkludGVsbGVjdHVhbCBQcm9w ZXJ0eSBSaWdodHMgaW4gSUVURg0gICAgICAgICAgICAgIFRlY2hub2xvZ3kiLCBCQ1AgNzksIFJG QyAzOTc5LCBNYXJjaCAyMDA1Lg0NICAgW1BST0NFU1NdICBCcmFkbmVyLCBTLiwgIlRoZSBJbnRl cm5ldCBTdGFuZGFyZHMgUHJvY2VzcyAtLSBSZXZpc2lvbg0gICAgICAgICAgICAgIDMiLCBCQ1Ag OSwgUkZDIDIwMjYsIE9jdG9iZXIgMTk5Ni4NDSAgIFtUUlVTVF0gICAgQnJhZG5lciwgUy4gYW5k IEouIENvbnRyZXJhcywgIlJpZ2h0cyBDb250cmlidXRvcnMgUHJvdmlkZQ0gICAgICAgICAgICAg IHRvIHRoZSBJRVRGIFRydXN0IiwgQkNQIDc4LCBSRkMgNTM3OCwgTm92ZW1iZXIgMjAwOC4NDTEw LjIuICBJbmZvcm1hdGl2ZSBSZWZlcmVuY2VzDQ0gICBbQ09MTEFCXSAgIEZpc2htYW4sIEcuIGFu ZCBTLiBCcmFkbmVyLCAiSW50ZXJuZXQgRW5naW5lZXJpbmcgVGFzaw0gICAgICAgICAgICAgIEZv cmNlIGFuZCBJbnRlcm5hdGlvbmFsIFRlbGVjb21tdW5pY2F0aW9uIFVuaW9uIC0NICAgICAgICAg ICAgICBUZWxlY29tbXVuaWNhdGlvbnMgU3RhbmRhcmRpemF0aW9uIFNlY3RvciBDb2xsYWJvcmF0 aW9uDSAgICAgICAgICAgICAgR3VpZGVsaW5lcyIsIFJGQyAzMzU2LCBBdWd1c3QgMjAwMi4NDSAg IFtET1NdICAgICAgSGFuZGxleSwgTS4sIFJlc2NvcmxhLCBFLiwgYW5kIElBQiwgIkludGVybmV0 IERlbmlhbC1vZi0NICAgICAgICAgICAgICBTZXJ2aWNlIENvbnNpZGVyYXRpb25zIiwgUkZDIDQ3 MzIsIERlY2VtYmVyIDIwMDYuDQ0gICBbSmluZ2xlXSAgIEx1ZHdpZywgUy4sIFNhaW50LUFuZHJl LCBQLiwgRWdhbiwgUy4sIE1jUXVlZW4sIFIuLCBhbmQgRC4NICAgICAgICAgICAgICBDaW9ub2l1 LCAiSmluZ2xlIFJUUCBTZXNzaW9ucyIsIFhTRiBYRVAgMDE2NywgSnVuZSAyMDA5Lg0NICAgW1JU UF0gICAgICBTY2h1bHpyaW5uZSwgSC4sIENhc25lciwgUy4sIEZyZWRlcmljaywgUi4sIGFuZCBW Lg0gICAgICAgICAgICAgIEphY29ic29uLCAiUlRQOiBBIFRyYW5zcG9ydCBQcm90b2NvbCBmb3Ig UmVhbC1UaW1lDSAgICAgICAgICAgICAgQXBwbGljYXRpb25zIiwgU1REIDY0LCBSRkMgMzU1MCwg SnVseSAyMDAzLg0NICAgW1NEUF0gICAgICBIYW5kbGV5LCBNLiwgSmFjb2Jzb24sIFYuLCBhbmQg Qy4gUGVya2lucywgIlNEUDogU2Vzc2lvbg0gICAgICAgICAgICAgIERlc2NyaXB0aW9uIFByb3Rv Y29sIiwgUkZDIDQ1NjYsIEp1bHkgMjAwNi4NDSAgIFtTRUNHVUlERV0NICAgICAgICAgICAgICBS ZXNjb3JsYSwgRS4gYW5kIEIuIEtvcnZlciwgIkd1aWRlbGluZXMgZm9yIFdyaXRpbmcgUkZDDSAg ICAgICAgICAgICAgVGV4dCBvbiBTZWN1cml0eSBDb25zaWRlcmF0aW9ucyIsIEJDUCA3MiwgUkZD IDM1NTIsDSAgICAgICAgICAgICAgSnVseSAyMDAzLg0NICAgW1NJUF0gICAgICBSb3NlbmJlcmcs IEouLCBTY2h1bHpyaW5uZSwgSC4sIENhbWFyaWxsbywgRy4sIEpvaG5zdG9uLA0gICAgICAgICAg ICAgIEEuLCBQZXRlcnNvbiwgSi4sIFNwYXJrcywgUi4sIEhhbmRsZXksIE0uLCBhbmQgRS4NICAg ICAgICAgICAgICBTY2hvb2xlciwgIlNJUDogU2Vzc2lvbiBJbml0aWF0aW9uIFByb3RvY29sIiwg UkZDIDMyNjEsDSAgICAgICAgICAgICAgSnVuZSAyMDAyLg0NICAgW1NSVFBdICAgICBCYXVnaGVy LCBNLiwgTWNHcmV3LCBELiwgTmFzbHVuZCwgTS4sIENhcnJhcmEsIEUuLCBhbmQgSy4NICAgICAg ICAgICAgICBOb3JybWFuLCAiVGhlIFNlY3VyZSBSZWFsLXRpbWUgVHJhbnNwb3J0IFByb3RvY29s IChTUlRQKSIsDSAgICAgICAgICAgICAgUkZDIDM3MTEsIE1hcmNoIDIwMDQuDQ0gICBbVU5DT09S RF0gIEJyeWFudCwgUy4gYW5kIE0uIE1vcnJvdywgIlVuY29vcmRpbmF0ZWQgUHJvdG9jb2wNDQ0N VmFsaW4sIGV0IGFsLiAgICAgICAgICAgIEV4cGlyZXMgQXByaWwgMTUsIDIwMTEgICAgICAgICAg ICAgICAgW1BhZ2UgMTddDQwNSW50ZXJuZXQtRHJhZnQgICAgICAgICAgICAgIENvZGVjIEd1aWRl bGluZXMgICAgICAgICAgICAgICAgT2N0b2JlciAyMDEwDQ0NICAgICAgICAgICAgICBEZXZlbG9w bWVudCBDb25zaWRlcmVkIEhhcm1mdWwiLCBSRkMgNTcwNCwgTm92ZW1iZXIgMjAwOS4NDSAgIFtY TVBQXSAgICAgU2FpbnQtQW5kcmUsIFAuLCBFZC4sICJFeHRlbnNpYmxlIE1lc3NhZ2luZyBhbmQg UHJlc2VuY2UNICAgICAgICAgICAgICBQcm90b2NvbCAoWE1QUCk6IENvcmUiLCBSRkMgMzkyMCwg T2N0b2JlciAyMDA0Lg0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0NDQ0N DQ0NDVZhbGluLCBldCBhbC4gICAgICAgICAgICBFeHBpcmVzIEFwcmlsIDE1LCAyMDExICAgICAg ICAgICAgICAgIFtQYWdlIDE4XQ0MDUludGVybmV0LURyYWZ0ICAgICAgICAgICAgICBDb2RlYyBH dWlkZWxpbmVzICAgICAgICAgICAgICAgIE9jdG9iZXIgMjAxMA0NDUF1dGhvcnMnIEFkZHJlc3Nl cw0NICAgSmVhbi1NYXJjIFZhbGluDSAgIE9jdGFzaWMgSW5jLg0gICA0MTAxLCBNb2xzb24gU3Ry ZWV0DSAgIE1vbnRyZWFsLCBRdWViZWMNICAgQ2FuYWRhDQ0gICBFbWFpbDogamVhbi1tYXJjLnZh bGluQG9jdGFzaWMuY29tDQ0NICAgU2xhdmEgQm9yaWxpbg0gICBTUElSSVQgRFNQDQ0NICAgRW1h aWw6IGJvcmlsaW5Ac3Bpcml0ZHNwLm5ldA0NDSAgIEtvZW4gVm9zDSAgIFNreXBlIFRlY2hub2xv Z2llcyBTLkEuDSAgIFN0YWRzZ2FyZGVuIDYNICAgU3RvY2tob2xtLCAgIDExNjQ1DSAgIFN3ZWRl bg0NICAgRW1haWw6IGtvZW4udm9zQHNreXBlLm5ldA0NDSAgIENocmlzdG9waGVyIE1vbnRnb21l cnkNICAgWGlwaC5PcmcgRm91bmRhdGlvbg0NDSAgIEVtYWlsOiB4aXBobW9udEB4aXBoLm9yZw0N DSAgIFJheW1vbmQgKEp1aW4tSHdleSkgQ2hlbg0gICBCcm9hZGNvbSBDb3Jwb3JhdGlvbg0NDSAg IEVtYWlsOiByY2hlbkBicm9hZGNvbS5jb20NDQ0NDQ0NDQ0NDQ0NVmFsaW4sIGV0IGFsLiAgICAg ICAgICAgIEV4cGlyZXMgQXByaWwgMTUsIDIwMTEgICAgICAgICAgICAgICAgW1BhZ2UgMTldDQwN DQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAGAABXPgAAWD4A AMtAAADMQAAA7lkAAO9ZAADwWQAAwFoAAMFaAADjWwAA5VsAAERcAABGXAAAIl0AACNdAACYXQAA 2HIAANlyAAAohQAA8+nz3PPStZuI835uVfNIbvM+8wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAS Fmg2UZoAT0oDAFFKAwBeSgMAABgVaDQ/BgAWaHIPpwBPSgMAUUoDAF5KAwAAMQAIgRZoayrZABdo cg+nAE9KAwBRSgMAXkoDAGNIAQBkaAAAAABkaAAAAABkaFzT6kYfAQiBBEgBAAVoXNPqRhZocg+n AE9KAwBRSgMAXkoDABIWaGsq2QBPSgMAUUoDAF5KAwAAJQEIgQRIAQAFaFzT6kYVaDQ/BgAWaHIP pwBPSgMAUUoDAF5KAwAzAQiBBEgBAAVoXNPqRhZocg+nAENKFABPSgMAUEoAAFFKAwBeSgMAYUoU AG1IHQRzSB0EOQEIgQRIAQAFaFzT6kYVaHIPpwAWaHIPpwBDShQAT0oDAFBKAABRSgMAXkoDAGFK FABtSB0Ec0gdBBIWaDwQYABPSgMAUUoDAF5KAwAAGBVoND8GABZoEDPoAE9KAwBRSgMAXkoDAAAS FmjFNcgAT0oDAFFKAwBeSgMAABgVaDQ/BgAWaMU1yABPSgMAUUoDAF5KAwATAAYAAAEIAAACCAAA AwgAAEwIAACVCAAA3ggAACcJAABwCQAAuQkAAAIKAABLCgAAlAoAAN0KAAAmCwAAJwsAACgLAABn CwAAmwsAAJwLAAClCwAApgsAAO4LAAAxDAAAcwwAAJQMAACVDAAAqQwAAKoMAADrDAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAAAAAAABA8AZ2Q0PwYAAB0ABgAAKIUAAP0AAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQEAAEBAesMAAAPDQAAEA0AAFUN AACXDQAA3w0AABwOAAAdDgAAZg4AAK4OAADwDgAALg8AAC8PAABlDwAAZg8AAHcPAAB4DwAAuw8A AOYPAADnDwAAKBAAAFEQAACUEAAAlRAAAJYQAACXEAAA4BAAAOIQAAArEQAALBEAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAAAAAAAAQPAGdkND8GAAAdLBEAAC0RAABtEQAAthEA AP4RAABEEgAAhxIAALMSAAC0EgAAtRIAAMcSAADIEgAAERMAAFoTAACjEwAA7BMAADUUAAB+FAAA xxQAABAVAABZFQAAohUAAOsVAAA0FgAAfRYAAH4WAAB/FgAAgBYAAIEWAACCFgAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAAAAAAABA8AZ2Q0PwYAAB2CFgAAgxYAAIQWAACFFgAA hhYAAIcWAACIFgAAiRYAAIoWAACLFgAAjBYAAI0WAACOFgAAjxYAAJAWAACRFgAAkhYAAJMWAACU FgAAlRYAAJYWAACXFgAAmBYAAJkWAADiFgAA5BYAAC0XAAAuFwAALxcAAEAXAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAAAAAAAEDwBnZDQ/BgAAHUAXAABBFwAAiBcAAM8XAAAW GAAAXhgAAGgYAABpGAAAahgAAGsYAABsGAAAbRgAAG4YAABvGAAAcBgAAHEYAAByGAAAcxgAAHQY AAB1GAAAdhgAAHcYAAB4GAAAeRgAAHoYAAB7GAAAfBgAAH0YAAB+GAAAfxgAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAAAAAAAAQPAGdkND8GAAAdfxgAAIAYAACBGAAAghgAAIMY AACEGAAAhRgAAIYYAACHGAAAiBgAAIkYAACKGAAAixgAAIwYAACNGAAAjhgAAI8YAACQGAAAkRgA AJIYAACTGAAAlBgAAN0YAADfGAAAKBkAACkZAAAqGQAAQhkAAEMZAACMGQAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAAAAAAABA8AZ2Q0PwYAAB2MGQAA0BkAABQaAABaGgAAohoA AOkaAAAsGwAAdRsAALUbAAD2GwAAOBwAAIAcAADBHAAAzRwAAM4cAAAOHQAADx0AAFcdAACQHQAA kR0AANodAAAhHgAAZB4AAKoeAADxHgAAMx8AAHUfAAC1HwAA+R8AADggAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAAAAAAAEDwBnZDQ/BgAAHTggAAB5IAAAtiAAAP0gAAAzIQAA NCEAAHohAADBIQAACSIAAFEiAACYIgAA4CIAACgjAABmIwAArCMAAPAjAAA5JAAAOiQAADskAAA8 JAAAhSQAAIckAADQJAAA0SQAANIkAAAaJQAAWSUAAKElAADkJQAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA9QAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAAAAAAAAAAAAAAAABA8AZ2Q4MUkAAAQPAGdkND8GAAAc5CUAAOUlAAApJgAAcCYAALUmAAD+ JgAARCcAAFInAABTJwAAmycAAOQnAAAnKAAAbCgAALMoAAD6KAAAQykAAIkpAADSKQAAGCoAAGEq AACoKgAA7CoAAO0qAAA2KwAAfSsAAMUrAAAKLAAATywAAJMsAADULAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAAAAAAABA8AZ2Q0PwYAAB3ULAAAFi0AAFstAACjLQAA6S0AACsu AABvLgAAri4AAO8uAAAzLwAAey8AAIkvAACKLwAAzi8AABcwAABbMAAAXDAAAF0wAABeMAAApzAA AKkwAADyMAAA8zAAAPQwAAA8MQAAgzEAAMMxAAAGMgAAPjIAAD8yAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAAAAAAAEDwBnZDQ/BgAAHT8yAACIMgAAyTIAAAszAAAtMwAALjMA AC8zAAAwMwAAMTMAADIzAAAzMwAANDMAADUzAAA2MwAANzMAADgzAAA5MwAAOjMAADszAAA8MwAA PTMAAD4zAAA/MwAAQDMAAEEzAABCMwAAQzMAAEQzAABFMwAARjMAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAAAAAAAAQPAGdkND8GAAAdRjMAAEczAABIMwAASTMAAEozAABLMwAA TDMAAE0zAABOMwAATzMAAFAzAABRMwAAUjMAAFMzAABUMwAAVTMAAFYzAACfMwAAoTMAAOozAADr MwAA7DMAABo0AAAbNAAAYzQAAKs0AADxNAAANjUAAH41AAC+NQAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAAAAAAABA8AZ2Q0PwYAAB2+NQAAADYAAEM2AACGNgAAzjYAAA83AABB NwAAQjcAAIU3AADONwAAFDgAAFw4AACjOAAA6TgAAC05AAA3OQAAODkAAH45AADGOQAACToAAFI6 AACaOgAA3joAAB07AAA/OwAAQDsAAHc7AAB4OwAAtTsAALY7AAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAAAAAAAEDwBnZDQ/BgAAHbY7AAD9OwAALDwAAC08AAB2PAAAoTwAAKI8 AACjPAAApDwAAKU8AACmPAAApzwAAKg8AACpPAAAqjwAAPM8AAD1PAAAPj0AAD89AABAPQAAXT0A AF49AACmPQAA7z0AADE+AABYPgAAWT4AAJ8+AADkPgAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAAAA AAAAAAAAAAAABA8AZ2RcDngAAAQPAGdkND8GAAAc5D4AACo/AABzPwAAuj8AAP4/AABDQAAAikAA AMtAAADMQAAAE0EAAE9BAACWQQAA3EEAACNCAABpQgAAsUIAALJCAAD7QgAAQ0MAAIpDAADNQwAA EkQAAFZEAACORAAA1UQAABpFAABdRQAAd0UAAHhFAAD6AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAA APUAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD1AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAEDwBnZEo/OwAABA8AZ2Q0PwYAABx4RQAAvEUAAAVGAABMRgAAj0YAANdGAAAdRwAA XkcAAHBHAABxRwAAuEcAAPhHAAD5RwAA+kcAAPtHAABESAAARkgAAI9IAACQSAAAkUgAANVIAAAe SQAAYUkAAKdJAADvSQAAMkoAAHpKAAC9SgAAA0sAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAQPAGdk8H8HAAAEDwBnZDQ/BgAAHANLAAA8SwAAPUsAAIRLAADLSwAAEUwAABJMAABa TAAAoEwAAN9MAAAnTQAAak0AAKZNAAC7TQAAvE0AAL1NAAC+TQAAv00AAMBNAADBTQAAwk0AAMNN AADETQAAxU0AAMZNAADHTQAAyE0AAMlNAADKTQAA+gAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD1 AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA9QAAAAAA AAAAAAAAAPUAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAPUAAAAAAAAAAAAA AAD1AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA9QAA AAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAPUAAAAAAAAA AAAAAAD1AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA 9QAAAAAAAAAAAAAAAPUAAAAAAAAAAAAAAAD1AAAAAAAAAAAAAAAA9QAAAAAAAAAAAAAAAAAAAAAA AAAAAAAABA8AZ2Q0PwYAAAQPAGdk8H8HAAAcyk0AAMtNAADMTQAAzU0AAM5NAADPTQAA0E0AANFN AADSTQAA000AANRNAADVTQAA1k0AANdNAADYTQAAIU4AACNOAABsTgAAbU4AAG5OAACITgAAiU4A AMtOAADXTgAA2E4AAB9PAABjTwAAqk8AAOZPAADnTwAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAAAAAAABA8AZ2Q0PwYAAB3nTwAAKVAAAGxQAACxUAAA4VAAAOJQAAAlUQAAblEA ALJRAADLUQAAzFEAAApSAABPUgAAklIAANVSAADuUgAA71IAAC9TAAB3UwAAv1MAAAJUAABJVAAA jlQAANJUAAAXVQAAVlUAAJxVAADgVQAAI1YAAGxWAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAAAAAAAEDwBnZDQ/BgAAHWxWAACzVgAA8FYAADNXAAB6VwAAwlcAAAdYAABPWAAA kFgAANhYAADZWAAA2lgAANtYAAAkWQAAJlkAAG9ZAABwWQAAcVkAALlZAADBWgAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAArgAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABLDwAPhGgBQyQBRcaAAAABAFzT6kYAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AABehGgBZ2RyD6cAAAQPAGdkND8GAAATwVoAAMJaAADDWgAAB1sAACZbAAAnWwAAaVsAAKVbAABH XAAASFwAAJFcAADZXAAA6FwAAOlcAACZXQAAml0AAOJdAAArXgAAc14AALleAAD4XgAADF8AAA1f AABPXwAAmF8AAOFfAAD2AAAAAAAAAAAAAAAA8QAAAAAAAAAAAAAAAPEAAAAAAAAAAAAAAADxAAAA AAAAAAAAAAAA8QAAAAAAAAAAAAAAAPEAAAAAAAAAAAAAAADkAAAAAAAAAAAAAAAA5AAAAAAAAAAA AAAAAPEAAAAAAAAAAAAAAADxAAAAAAAAAAAAAAAA8QAAAAAAAAAAAAAAAPEAAAAAAAAAAAAAAADx AAAAAAAAAAAAAAAA1wAAAAAAAAAAAAAAAPEAAAAAAAAAAAAAAADxAAAAAAAAAAAAAAAA8QAAAAAA AAAAAAAAAPEAAAAAAAAAAAAAAADxAAAAAAAAAAAAAAAA8QAAAAAAAAAAAAAAAPEAAAAAAAAAAAAA AADxAAAAAAAAAAAAAAAA8QAAAAAAAAAAAAAAAPEAAAAAAAAAAAAAAADxAAAAAAAAAAAAAAAAAAAM DwAPhIQDEYTk/V6EhANghOT9Z2Q8EGAAAAwPAA+EhAMRhOT9XoSEA2CE5P1nZGsq2QAABA8AZ2Q0 PwYAAAgPAA+EaAFehGgBZ2Q8EGAAABnhXwAAHWAAAEJgAABDYAAAjGAAANFgAAAXYQAAXWEAAKZh AADtYQAANWIAAHtiAAC9YgAABmMAAE5jAACVYwAA3WMAACNkAABWZAAAV2QAAFhkAABZZAAAWmQA AFtkAABcZAAAXWQAAKZkAACoZAAA8WQAAPJkAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAAAAAAAEDwBnZDQ/BgAAHfJkAADzZAAAFGUAABVlAABYZQAAmmUAAKllAACqZQAA7GUA ADNmAABCZgAAQ2YAAHFmAAByZgAAs2YAALRmAADoZgAA6WYAACBnAAAhZwAAYWcAAKFnAADKZwAA y2cAAAxoAABTaAAAnGgAAOFoAAAnaQAAQGkAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAAAAAAAAQPAGdkND8GAAAdQGkAAEFpAAB7aQAAwWkAANtpAADcaQAAHmoAAGFqAAClagAA 7moAADVrAABsawAAbWsAALNrAAD8awAAQGwAAF9sAABgbAAAoWwAAOlsAAAvbQAAMG0AADFtAAAy bQAAe20AAH1tAADGbQAAx20AAMhtAAAIbgAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAAAAAAABA8AZ2Q0PwYAAB0IbgAAUG4AAGBuAABhbgAAqW4AAOpuAAAsbwAAWW8AAFpvAACg bwAA6W8AAOpvAAAncAAAbXAAAKBwAAChcAAA6nAAAOtwAAAxcQAAenEAAJ1xAACecQAA5nEAACVy AABpcgAApnIAANdyAADYcgAA2XIAANpyAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAAAAAAAEDwBnZDQ/BgAAHdpyAADbcgAA3HIAAN1yAADecgAA33IAAOByAADhcgAA4nIAAONy AADkcgAA5XIAAOZyAADncgAA6HIAAOlyAADqcgAA63IAAOxyAADtcgAA7nIAAO9yAAA4cwAAOnMA AINzAACEcwAAhXMAAKFzAACicwAA6nMAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAAAAAAAAQPAGdkND8GAAAd6nMAACl0AABtdAAAknQAAJN0AACUdAAAlXQAAJZ0AACXdAAAmHQA AJl0AACadAAAm3QAAJx0AACddAAAnnQAAJ90AACgdAAAoXQAAKJ0AACjdAAApHQAAKV0AACmdAAA p3QAAKh0AACpdAAAqnQAAKt0AACsdAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAAAAAAABA8AZ2Q0PwYAAB2sdAAArXQAAK50AACvdAAAsHQAALF0AACydAAAs3QAALR0AAC1dAAA tnQAALd0AAC4dAAAuXQAALp0AAC7dAAAvHQAAL10AAC+dAAAv3QAAAh1AAAKdQAAU3UAAFR1AABV dQAAbXUAAG51AACYdQAAmXUAAJp1AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA AAAAAAAEDwBnZDQ/BgAAHZp1AACbdQAAnHUAAJ11AACedQAAn3UAAKB1AAChdQAAonUAAKN1AACk dQAApXUAAKZ1AACndQAAqHUAAKl1AACqdQAAq3UAAKx1AACtdQAArnUAAK91AACwdQAAsXUAALJ1 AACzdQAAtHUAALV1AAC2dQAAt3UAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA AAAAAAQPAGdkND8GAAAdt3UAALh1AAC5dQAAunUAALt1AAC8dQAAvXUAAL51AAC/dQAAwHUAAMF1 AADCdQAAw3UAAMR1AADFdQAAxnUAAMd1AADIdQAAEXYAABN2AABcdgAAXXYAAF52AABydgAAc3YA ALt2AAD+dgAARHcAAI13AADOdwAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAAA AAAABA8AZ2Q0PwYAAB3OdwAAFngAAD94AABAeAAAQXgAAEJ4AABDeAAARHgAAEV4AABGeAAAR3gA AEh4AABJeAAASngAAEt4AABMeAAATXgAAE54AABPeAAAUHgAAFF4AABSeAAAU3gAAFR4AABVeAAA VngAAFd4AABYeAAAWXgAAFp4AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAAAA AAAEDwBnZDQ/BgAAHVp4AABbeAAAXHgAAF14AABeeAAAX3gAAGB4AABheAAAYngAAGN4AABkeAAA ZXgAAGZ4AABneAAAaHgAAGl4AACyeAAAtHgAAP14AAD+eAAA/3gAAA95AAAQeQAALHkAAC15AABu eQAAp3kAAKh5AADveQAAIHoAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAAAAAA AAQPAGdkND8GAAAdIHoAACF6AABqegAArXoAAK56AADMegAAzXoAABJ7AABSewAAmHsAAMp7AADL ewAAEnwAAFJ8AABTfAAAnHwAAON8AADkfAAAJX0AAGZ9AACgfQAAoX0AAOh9AAAifgAAI34AADF+ AAB3fgAAuX4AANJ+AADTfgAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAAAAAAA BA8AZ2Q0PwYAAB3TfgAAGn8AAFp/AACgfwAAuX8AALp/AAACgAAAS4AAAG+AAABwgAAAsIAAALGA AACygAAAs4AAAPyAAAD+gAAAR4EAAEiBAABJgQAAkYEAAJKBAADZgQAAF4IAABiCAAAZggAAGoIA ABuCAAAcggAAHYIAAB6CAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAAAAAAAE DwBnZDQ/BgAAHR6CAAAfggAAIIIAACGCAAAiggAAI4IAACSCAAAlggAAJoIAACeCAAAoggAAKYIA ACqCAAArggAALIIAAC2CAAAuggAAL4IAADCCAAAxggAAMoIAADOCAAA0ggAANYIAADaCAAA3ggAA OIIAADmCAAA6ggAAO4IAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAA AAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAA APoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAAAAAAAAQP AGdkND8GAAAdO4IAADyCAAA9ggAAPoIAAD+CAABAggAAQYIAAEKCAABDggAARIIAAEWCAABGggAA j4IAAJGCAADaggAA24IAANyCAADvggAA8IIAAAODAAATgwAAKoMAAD6DAABIgwAASYMAAG+DAABw gwAAcYMAAIKDAACQgwAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA +gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAA AAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAA AAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoA AAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAAAAAAABA8A Z2Q0PwYAAB2QgwAAkYMAAJKDAACygwAAs4MAALSDAADAgwAA24MAAOyDAAAChAAADIQAAA2EAAAq hAAAK4QAACyEAABGhAAAXYQAAF6EAABfhAAAe4QAAHyEAAB9hAAAmYQAALGEAACyhAAAs4QAANCE AADRhAAA0oQAANOEAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAA AAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAA AAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6 AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAA AAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAA AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAAAAAAAEDwBn ZDQ/BgAAHdOEAADUhAAA1YQAANaEAADXhAAA2IQAANmEAADahAAA24QAANyEAAAlhQAAJ4UAACiF AAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAA AAAAAAAAAAAAAPoAAAAAAAAAAAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAA AAAAAAD6AAAAAAAAAAAAAAAA+gAAAAAAAAAAAAAAAPoAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQPAGdk ND8GAAAMMgAxkGgBOnA0PwYAH7CCLiCwxkEhsDYFIrA3BSOQiQUkkG4EJbAAABewxAIYsMQCDJDE AgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAhgIb ABIAAQCcAA8ABAAEAAQABAAAAAQACAAAAJgAAACeAAAAngAAAJ4AAACeAAAAngAAAJ4AAACeAAAA ngAAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAAHYCAAB2AgAAdgIAAHYCAAB2 AgAAdgIAAHYCAAB2AgAAdgIAADYGAAA2BgAANgYAADYGAAA2BgAANgYAAD4CAAA2BgAANgYAADYG AAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYA ADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAACoAAAANgYAADYGAAAWAAAA NgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAC4AAAANgYAADYGAAA2BgAANgYAADYGAAA2 BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAAaAEAAEgBAAA2BgAANgYAADYGAAA2BgAANgYAADYG AAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYA ADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAA NgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2 BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYGAAA2BgAANgYAADYG AAA2BgAANgYAALADAAA2BgAASgAAQPH/AgBKAAwQAAByWKwAAAAGAE4AbwByAG0AYQBsAAAADAAA ABJkFAEBABSkyAAYAENKFgBfSAEEYUoWAG1IBwRzSAcEdEgJBAAAAAAAAAAAAAAAAAAAAAAAAEQA QUDy/6EARAAMDQAAAAAAABAAFgBEAGUAZgBhAHUAbAB0ACAAUABhAHIAYQBnAHIAYQBwAGgAIABG AG8AbgB0AAAAAABSAGlA8/+zAFIADB0AAAAAAAAwBgwAVABhAGIAbABlACAATgBvAHIAbQBhAGwA AAAcABf2AwAANNYGAAEKA2wANNYGAAEFAwAAYfYDAAACAAsAAAAoAGsA9P/BACgAAA0AAAAAAAAw BgcATgBvACAATABpAHMAdAAAAAIADAAAAAAATgBaQAEA8gBOAAwMEAA0PwYAMAYKAFAAbABhAGkA bgAgAFQAZQB4AHQAAAAMAA8AEmTwAAEAFKQAABQAQ0oVAE9KBQBRSgUAXkoFAGFKFQBCAP5PogAB AUIADAAPADQ/BgAwBgsAIABDAGgAYQByACAAQwBoAGEAcgAzAAAAFABDShUAT0oFAFFKBQBeSgUA YUoVAEIAJwCiABEBQgAMDQAAODFJADAGEQBDAG8AbQBtAGUAbgB0ACAAUgBlAGYAZQByAGUAbgBj AGUAAAAIAENKEABhShAAPAAeQAEAIgE8AAwNEwA4MUkAMAYMAEMAbwBtAG0AZQBuAHQAIABUAGUA eAB0AAAAAgASAAgAQ0oUAGFKFAAyAP4PogAxATIADAESADgxSQAwBgsAIABDAGgAYQByACAAQwBo AGEAcgAyAAAABAB0SAkEQABqQCEBIgFAAAwNFQA4MUkAMAYPAEMAbwBtAG0AZQBuAHQAIABTAHUA YgBqAGUAYwB0AAAAAgAUAAYANQiBXAiBNAD+DzIBUQE0AAwBFAA4MUkAMAYLACAAQwBoAGEAcgAg AEMAaABhAHIAMQAAAAYANQgBXAgBUgCZQAEAYgFSAAwNFwA4MUkAMAYMAEIAYQBsAGwAbwBvAG4A IABUAGUAeAB0AAAADAAWABJk8AABABSkAAAUAENKEABPSgYAUUoGAF5KBgBhShAARAD+D6IAcQFE AAwBFgA4MUkAMAYKACAAQwBoAGEAcgAgAEMAaABhAHIAAAAYAENKEABPSgYAUUoGAF5KBgBhShAA dEgJBGAAXkABAIIBYAAMCQAANlGaADAGDABOAG8AcgBtAGEAbAAgACgAVwBlAGIAKQAAABYAGAAS ZPAAAQATpGQAFKRkAFskAVwkARgAQ0oYAE9KAABQSgAAUUoAAGFKGAB0SAcEQgD+T6IAkQFCAAwA AAA2UZoAAAAVAGEAcABwAGwAZQAtAGMAbwBuAHYAZQByAHQAZQBkAC0AcwBwAGEAYwBlAAAAAAA4 AP5PogChATgADAAAADZRmgAAABAAYQBwAHAAbABlAC0AcwB0AHkAbABlAC0AcwBwAGEAbgAAAAAA AAAAACh9AAAFAADWAAAAAP////8AAAAAAQAAAAIAAAADAAAATAAAAJUAAADeAAAAJwEAAHABAAC5 AQAAAgIAAEsCAACUAgAA3QIAACYDAAAnAwAAKAMAAGcDAACbAwAAnAMAAKUDAACmAwAA7gMAADEE AABzBAAAlAQAAJUEAACpBAAAqgQAAOsEAAAPBQAAEAUAAFUFAACXBQAA3wUAABwGAAAdBgAAZgYA AK4GAADwBgAALgcAAC8HAABlBwAAZgcAAHcHAAB4BwAAuwcAAOYHAADnBwAAKAgAAFEIAACUCAAA lQgAAJYIAACXCAAA4AgAAOIIAAArCQAALAkAAC0JAABtCQAAtgkAAP4JAABECgAAhwoAALMKAAC0 CgAAtQoAAMcKAADICgAAEQsAAFoLAACjCwAA7AsAADUMAAB+DAAAxwwAABANAABZDQAAog0AAOsN AAA0DgAAfQ4AAH4OAAB/DgAAgA4AAIEOAACCDgAAgw4AAIQOAACFDgAAhg4AAIcOAACIDgAAiQ4A AIoOAACLDgAAjA4AAI0OAACODgAAjw4AAJAOAACRDgAAkg4AAJMOAACUDgAAlQ4AAJYOAACXDgAA mA4AAJkOAADiDgAA5A4AAC0PAAAuDwAALw8AAEAPAABBDwAAiA8AAM8PAAAWEAAAXhAAAGgQAABp EAAAahAAAGsQAABsEAAAbRAAAG4QAABvEAAAcBAAAHEQAAByEAAAcxAAAHQQAAB1EAAAdhAAAHcQ AAB4EAAAeRAAAHoQAAB7EAAAfBAAAH0QAAB+EAAAfxAAAIAQAACBEAAAghAAAIMQAACEEAAAhRAA AIYQAACHEAAAiBAAAIkQAACKEAAAixAAAIwQAACNEAAAjhAAAI8QAACQEAAAkRAAAJIQAACTEAAA lBAAAN0QAADfEAAAKBEAACkRAAAqEQAAQhEAAEMRAACMEQAA0BEAABQSAABaEgAAohIAAOkSAAAs EwAAdRMAALUTAAD2EwAAOBQAAIAUAADBFAAAzRQAAM4UAAAOFQAADxUAAFcVAACQFQAAkRUAANoV AAAhFgAAZBYAAKoWAADxFgAAMxcAAHUXAAC1FwAA+RcAADgYAAB5GAAAthgAAP0YAAAzGQAANBkA AHoZAADBGQAACRoAAFEaAACYGgAA4BoAACgbAABmGwAArBsAAPAbAAA5HAAAOhwAADscAAA8HAAA hRwAAIccAADQHAAA0RwAANIcAAAaHQAAWR0AAKEdAADkHQAA5R0AACkeAABwHgAAtR4AAP4eAABE HwAAUh8AAFMfAACbHwAA5B8AACcgAABsIAAAsyAAAPogAABDIQAAiSEAANIhAAAYIgAAYSIAAKgi AADsIgAA7SIAADYjAAB9IwAAxSMAAAokAABPJAAAkyQAANQkAAAWJQAAWyUAAKMlAADpJQAAKyYA AG8mAACuJgAA7yYAADMnAAB7JwAAiScAAIonAADOJwAAFygAAFsoAABcKAAAXSgAAF4oAACnKAAA qSgAAPIoAADzKAAA9CgAADwpAACDKQAAwykAAAYqAAA+KgAAPyoAAIgqAADJKgAACysAAC0rAAAu KwAALysAADArAAAxKwAAMisAADMrAAA0KwAANSsAADYrAAA3KwAAOCsAADkrAAA6KwAAOysAADwr AAA9KwAAPisAAD8rAABAKwAAQSsAAEIrAABDKwAARCsAAEUrAABGKwAARysAAEgrAABJKwAASisA AEsrAABMKwAATSsAAE4rAABPKwAAUCsAAFErAABSKwAAUysAAFQrAABVKwAAVisAAJ8rAAChKwAA 6isAAOsrAADsKwAAGiwAABssAABjLAAAqywAAPEsAAA2LQAAfi0AAL4tAAAALgAAQy4AAIYuAADO LgAADy8AAEEvAABCLwAAhS8AAM4vAAAUMAAAXDAAAKMwAADpMAAALTEAADcxAAA4MQAAfjEAAMYx AAAJMgAAUjIAAJoyAADeMgAAHTMAAD8zAABAMwAAdzMAAHgzAAC1MwAAtjMAAP0zAAAsNAAALTQA AHY0AAChNAAAojQAAKM0AACkNAAApTQAAKY0AACnNAAAqDQAAKk0AACqNAAA8zQAAPU0AAA+NQAA PzUAAEA1AABdNQAAXjUAAKY1AADvNQAAMTYAAFg2AABZNgAAnzYAAOQ2AAAqNwAAczcAALo3AAD+ NwAAQzgAAIo4AADLOAAAzDgAABM5AABPOQAAljkAANw5AAAjOgAAaToAALE6AACyOgAA+zoAAEM7 AACKOwAAzTsAABI8AABWPAAAjjwAANU8AAAaPQAAXT0AAHc9AAB4PQAAvD0AAAU+AABMPgAAjz4A ANc+AAAdPwAAXj8AAHA/AABxPwAAuD8AAPg/AAD5PwAA+j8AAPs/AABEQAAARkAAAI9AAACQQAAA kUAAANVAAAAeQQAAYUEAAKdBAADvQQAAMkIAAHpCAAC9QgAAA0MAADxDAAA9QwAAhEMAAMtDAAAR RAAAEkQAAFpEAACgRAAA30QAACdFAABqRQAApkUAALtFAAC8RQAAvUUAAL5FAAC/RQAAwEUAAMFF AADCRQAAw0UAAMRFAADFRQAAxkUAAMdFAADIRQAAyUUAAMpFAADLRQAAzEUAAM1FAADORQAAz0UA ANBFAADRRQAA0kUAANNFAADURQAA1UUAANZFAADXRQAA2EUAACFGAAAjRgAAbEYAAG1GAABuRgAA iEYAAIlGAADLRgAA10YAANhGAAAfRwAAY0cAAKpHAADmRwAA50cAAClIAABsSAAAsUgAAOFIAADi SAAAJUkAAG5JAACySQAAy0kAAMxJAAAKSgAAT0oAAJJKAADVSgAA7koAAO9KAAAvSwAAd0sAAL9L AAACTAAASUwAAI5MAADSTAAAF00AAFZNAACcTQAA4E0AACNOAABsTgAAs04AAPBOAAAzTwAAek8A AMJPAAAHUAAAT1AAAJBQAADYUAAA2VAAANpQAADbUAAAJFEAACZRAABvUQAAcFEAAHFRAAC5UQAA wVIAAMJSAADDUgAAB1MAACZTAAAnUwAAaVMAAKVTAABHVAAASFQAAJFUAADZVAAA6FQAAOlUAACZ VQAAmlUAAOJVAAArVgAAc1YAALlWAAD4VgAADFcAAA1XAABPVwAAmFcAAOFXAAAdWAAAQlgAAENY AACMWAAA0VgAABdZAABdWQAAplkAAO1ZAAA1WgAAe1oAAL1aAAAGWwAATlsAAJVbAADdWwAAI1wA AFZcAABXXAAAWFwAAFlcAABaXAAAW1wAAFxcAABdXAAAplwAAKhcAADxXAAA8lwAAPNcAAAUXQAA FV0AAFhdAACaXQAAqV0AAKpdAADsXQAAM14AAEJeAABDXgAAcV4AAHJeAACzXgAAtF4AAOheAADp XgAAIF8AACFfAABhXwAAoV8AAMpfAADLXwAADGAAAFNgAACcYAAA4WAAACdhAABAYQAAQWEAAHth AADBYQAA22EAANxhAAAeYgAAYWIAAKViAADuYgAANWMAAGxjAABtYwAAs2MAAPxjAABAZAAAX2QA AGBkAAChZAAA6WQAAC9lAAAwZQAAMWUAADJlAAB7ZQAAfWUAAMZlAADHZQAAyGUAAAhmAABQZgAA YGYAAGFmAACpZgAA6mYAACxnAABZZwAAWmcAAKBnAADpZwAA6mcAACdoAABtaAAAoGgAAKFoAADq aAAA62gAADFpAAB6aQAAnWkAAJ5pAADmaQAAJWoAAGlqAACmagAA12oAANhqAADZagAA2moAANtq AADcagAA3WoAAN5qAADfagAA4GoAAOFqAADiagAA42oAAORqAADlagAA5moAAOdqAADoagAA6WoA AOpqAADragAA7GoAAO1qAADuagAA72oAADhrAAA6awAAg2sAAIRrAACFawAAoWsAAKJrAADqawAA KWwAAG1sAACSbAAAk2wAAJRsAACVbAAAlmwAAJdsAACYbAAAmWwAAJpsAACbbAAAnGwAAJ1sAACe bAAAn2wAAKBsAAChbAAAomwAAKNsAACkbAAApWwAAKZsAACnbAAAqGwAAKlsAACqbAAAq2wAAKxs AACtbAAArmwAAK9sAACwbAAAsWwAALJsAACzbAAAtGwAALVsAAC2bAAAt2wAALhsAAC5bAAAumwA ALtsAAC8bAAAvWwAAL5sAAC/bAAACG0AAAptAABTbQAAVG0AAFVtAABtbQAAbm0AAJhtAACZbQAA mm0AAJttAACcbQAAnW0AAJ5tAACfbQAAoG0AAKFtAACibQAAo20AAKRtAAClbQAApm0AAKdtAACo bQAAqW0AAKptAACrbQAArG0AAK1tAACubQAAr20AALBtAACxbQAAsm0AALNtAAC0bQAAtW0AALZt AAC3bQAAuG0AALltAAC6bQAAu20AALxtAAC9bQAAvm0AAL9tAADAbQAAwW0AAMJtAADDbQAAxG0A AMVtAADGbQAAx20AAMhtAAARbgAAE24AAFxuAABdbgAAXm4AAHJuAABzbgAAu24AAP5uAABEbwAA jW8AAM5vAAAWcAAAP3AAAEBwAABBcAAAQnAAAENwAABEcAAARXAAAEZwAABHcAAASHAAAElwAABK cAAAS3AAAExwAABNcAAATnAAAE9wAABQcAAAUXAAAFJwAABTcAAAVHAAAFVwAABWcAAAV3AAAFhw AABZcAAAWnAAAFtwAABccAAAXXAAAF5wAABfcAAAYHAAAGFwAABicAAAY3AAAGRwAABlcAAAZnAA AGdwAABocAAAaXAAALJwAAC0cAAA/XAAAP5wAAD/cAAAD3EAABBxAAAscQAALXEAAG5xAACncQAA qHEAAO9xAAAgcgAAIXIAAGpyAACtcgAArnIAAMxyAADNcgAAEnMAAFJzAACYcwAAynMAAMtzAAAS dAAAUnQAAFN0AACcdAAA43QAAOR0AAAldQAAZnUAAKB1AAChdQAA6HUAACJ2AAAjdgAAMXYAAHd2 AAC5dgAA0nYAANN2AAAadwAAWncAAKB3AAC5dwAAuncAAAJ4AABLeAAAb3gAAHB4AACweAAAsXgA ALJ4AACzeAAA/HgAAP54AABHeQAASHkAAEl5AACReQAAknkAANl5AAAXegAAGHoAABl6AAAaegAA G3oAABx6AAAdegAAHnoAAB96AAAgegAAIXoAACJ6AAAjegAAJHoAACV6AAAmegAAJ3oAACh6AAAp egAAKnoAACt6AAAsegAALXoAAC56AAAvegAAMHoAADF6AAAyegAAM3oAADR6AAA1egAANnoAADd6 AAA4egAAOXoAADp6AAA7egAAPHoAAD16AAA+egAAP3oAAEB6AABBegAAQnoAAEN6AABEegAARXoA AEZ6AACPegAAkXoAANp6AADbegAA3HoAAO96AADwegAAA3sAABN7AAAqewAAPnsAAEh7AABJewAA b3sAAHB7AABxewAAgnsAAJB7AACRewAAknsAALJ7AACzewAAtHsAAMB7AADbewAA7HsAAAJ8AAAM fAAADXwAACp8AAArfAAALHwAAEZ8AABdfAAAXnwAAF98AAB7fAAAfHwAAH18AACZfAAAsXwAALJ8 AACzfAAA0HwAANF8AADSfAAA03wAANR8AADVfAAA1nwAANd8AADYfAAA2XwAANp8AADbfAAA3HwA ACV9AAAnfQAAKn0AAJoAAAAPMAAAAAAAAAEAAAADAAAAAAAAAAAAgAdakQAwATAAAAAAAAABAAAA swAAAAAAAAAAAAAHWpEAMAEwAAAAAAAAAgAAALEAAAAAAAAAAAAAB1qRADABMAAAAAAAAAEAAAAG AAAAAAAAAAAAgAdakQAwBDAAAAAAAAABAAAAIAMAAAAAAAAAAIAHWpEAMAQwAAAAAAAAAgAAAB4D AAAAAAAAAACAB1qRADAEMAAAAAAAAAEAAAAZAwAAAAAAAAAAgAdakQAwBzAAAAAAAAABAAAAIwMA AAAAAAAAAIAHWpEAMAcwAAAAAAAAAgAAACEDAAAAAAAAAACAB1qRADAEMAAAAAAAAAEAAAAZAwAA AAAAAAAAgAdakQAwCjAAAAAAAAABAAAAJgMAAAAAAAAAAIAHWpEAMAowAAAAAAAAAgAAACQDAAAA AAAAAACAB1qRADAEMAAAAAAAAAEAAAAZAwAAAAAAAAAAgAdakQAwDTAAAAAAAAABAAAAKQMAAAAA AAAAAIAHWpEAMA0wAAAAAAAAAgAAACcDAAAAAAAAAACAB1qRADAEMAAAAAAAAAEAAAAZAwAAAAAA AAAAgAdakQAwEDAAAAAAAAABAAAALAMAAAAAAAAAAIAHWpEAMBAwAAAAAAAAAgAAACoDAAAAAAAA AACAB1qRADAEMAAAAAAAAAEAAAAZAwAAAAAAAAAAgAdakQAwEzAAAAAAAAABAAAALQMAAAAAAAAA AIAHWpEAMBMwAAAAAAAAAgAAACsDAAAAAAAAAACAB1qRADAEMAAAAAAAAAEAAAAZAwAAAAAAAAAA gAdakQAwFjAAAAAAAAABAAAALgMAAAAAAAAAAIAHWpEAMBYwAAAAAAAAAQAAAC0DAAAAAAAAAACA B1qRADAWMAAAAAAAAAIAAAArAwAAAAAAAAAAgAdakQAwBDAAAAAAAAABAAAAGQMAAAAAAAAAAIAH WpEAMBowAAAAAAAAAQAAAC0DAAAAAAAAAACAB1qRADAaMAAAAAAAAAIAAAArAwAAAAAAAAAAgAda kQAwBDAAAAAAAAABAAAAGQMAAAAAAAAAAIAHWpEAMB0wAAAAAAAAAQAAAGYDAAAAAAAAAACAB1qR ADAdMAAAAAAAAAIAAABkAwAAAAAAAAAAgAdakQAwBDAAAAAAAAABAAAAGQMAAAAAAAAAAIAHWpEA MCAwAAAAAAAAAQAAAGYDAAAAAAAAAACAB1qRADAgMAAAAAAAAAIAAABkAwAAAAAAAAAAgAdakQAw BDAAAAAAAAABAAAAGQMAAAAAAAAAAIAHWpEAMAQwAAAAAAAAAQAAABoDAAAAAAAAAACAB1qRADAE MAAAAAAAAAEAAAAZAwAAAAAAAAAAgAdakQAwBDAAAAAAAAACAAAAFwMAAAAAAAAAAIAHWpEAMAEw AAAAAAAAAQAAAAYAAAAAAAAAAACAB1qRADABMAAAAAAAAAEAAAAHAAAAAAAAAAAAgAdakQAwATAA AAAAAAABAAAABwAAAAAAAAAAAIAHWpEAMAEwAAAAAAAAAQAAAAcAAAAAAAAAAACAB1qRADABMAAA AAAAAAEAAAAHAAAAAAAAAAAAgAdakQAwATAAAAAAAAABAAAABwAAAAAAAAAAAIAHWpEAMAEwAAAA AAAAAQAAAAcAAAAAAAAAAACAB1qRADABMAAAAAAAAAEAAAAHAAAAAAAAAAAAgAdakQAwATAAAAAA AAABAAAABwAAAAAAAAAAAIAHWpEAMAEwAAAAAAAAAQAAAAcAAAAAAAAAAACAB1qRADABMAAAAAAA AAEAAAAHAAAAAAAAAAAAgAdakQAwATAAAAAAAAABAAAABwAAAAAAAAAAAIAHWpEAMAEwAAAAAAAA AQAAAAcAAAAAAAAAAACAB1qRADABMAAAAAAAAAEAAAAHAAAAAAAAAAAAgAdakQAwATAAAAAAAAAB AAAABwAAAAAAAAAAAIAHWpEAMAEwAAAAAAAAAQAAAAcAAAAAAAAAAACAB1qRADABMAAAAAAAAAEA AAAHAAAAAAAAAAAAgAdakQAwATAAAAAAAAABAAAABwAAAAAAAAAAAIAHWpEAMAEwAAAAAAAAAQAA AAcAAAAAAAAAAACAB1qRADABMAAAAAAAAAEAAAAHAAAAAAAAAAAAgAdakQAwATAAAAAAAAABAAAA BwAAAAAAAAAAAIAHWpEAMAEwAAAAAAAAAQAAAAcAAAAAAAAAAACAB1qRADABMAAAAAAAAAEAAAAH AAAAAAAAAAAAgAdakQAwATAAAAAAAAABAAAABwAAAAAAAAAAAIAHWpEAMAEwAAAAAAAAAQAAAAcA AAAAAAAAAACAB1qRADABMAAAAAAAAAEAAAAHAAAAAAAAAAAAgAdakQAwATAAAAAAAAABAAAABwAA AAAAAAAAAIAHWpEAMAEwAAAAAAAAAQAAAAcAAAAAAAAAAACAB1qRADABMAAAAAAAAAEAAAAHAAAA AAAAAAAAgAdakQAwATAAAAAAAAABAAAABwAAAAAAAAAAAIAHWpEAMAEwAAAAAAAAAQAAAAcAAAAA AAAAAACAB1qRADABMAAAAAAAAAEAAAAHAAAAAAAAAAAAgAdakQAwATAAAAAAAAABAAAABwAAAAAA AAAAAIAHWpEAMAEwAAAAAAAAAQAAAAcAAAAAAAAAAACAB1qRADABMAAAAAAAAAEAAAAHAAAAAAAA AAAAgAdakQAwATAAAAAAAAABAAAABwAAAAAAAAAAAIAHWpEAMAEwAAAAAAAAAQAAAAcAAAAAAAAA AACAB1qRADABMAAAAAAAAAEAAAAHAAAAAAAAAAAAgAdakQAwATAAAAAAAAABAAAABwAAAAAAAAAA AIAHWpEAMAEwAAAAAAAAAQAAAAcAAAAAAAAAAACAB1qRADABMAAAAAAAAAEAAAAHAAAAAAAAAAAA gAdakQAwATAAAAAAAAABAAAABwAAAAAAAAAAAIAHWpEAMAEwAAAAAAAAAQAAAAcAAAAAAAAAAACA B1qRADABMAAAAAAAAAEAAAAHAAAAAAAAAAAAgAdakQAwATAAAAAAAAABAAAABwAAAAAAAAAAAIAH WpEAMAEwAAAAAAAAAQAAAAcAAAAAAAAAAACAB1qRADABMAAAAAAAAAEAAAAHAAAAAAAAAAAAgAda kQAwATAAAAAAAAABAAAABwAAAAAAAAAAAIAHWpEAMAEwAAAAAAAAAQAAAAcAAAAAAAAAAACAB1qR ADABMAAAAAAAAAEAAAAHAAAAAAAAAAAAgAdakQAwATAAAAAAAAABAAAABwAAAAAAAAAAAIAHWpEA MAEwAAAAAAAAAQAAAAcAAAAAAAAAAACAB1qRADABMAAAAAAAAAEAAAAHAAAAAAAAAAAAgAdakQAw ATAAAAAAAAABAAAABwAAAAAAAAAAAIAHWpEAMAEwAAAAAAAAAQAAAAcAAAAAAAAAAACAB1qRADAB MAAAAAAAAAEAAAAHAAAAAAAAAAAAgAdakQAwATAAAAAAAAABAAAABwAAAAAAAAAAAIAHWpEAMAEw AAAAAAAAAQAAAAcAAAAAAAAAAACAB1qRADABMAAAAAAAAAEAAAAHAAAAAAAAAAAAgAdakQAwATAA AAAAAAABAAAABwAAAAAAAAAAAIAHWpEAMAEwAAAAAAAAAQAAAAcAAAAAAAAAAACAB1qRADABMAAA AAAAAAEAAAAHAAAAAAAAAAAAgAdakQAwATAAAAAAAAABAAAABwAAAAAAAAAAAIAHWpEAMAEwAAAA AAAAAQAAAAcAAAAAAAAAAACAB1qRADABMAAAAAAAAAEAAAAHAAAAAAAAAAAAgAdakQAwATAAAAAA AAABAAAABwAAAAAAAAAAAIAHWpEAMAEwAAAAAAAAAQAAAAcAAAAAAAAAAACAB1qRADABMAAAAAAA AAEAAAAHAAAAAAAAAAAAgAdakQAwATAAAAAAAAABAAAABwAAAAAAAAAAAIAHWpEAMAEwAAAAAAAA AQAAAAcAAAAAAAAAAACAB1qRADABMAAAAAAAAAEAAAAHAAAAAAAAAAAAgAdakQAwATAAAAAAAAAB AAAABwAAAAAAAAAAAIAHWpEAMAEwAAAAAAAAAQAAAAcAAAAAAAAAAACAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAB5pAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAEAeaAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA B5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAH mgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAea AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAeaAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAHmgAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAB5oAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAAcAAAAAAQAA AAIAAAADAAAATAAAAJUAAADeAAAAJwEAAHABAAC5AQAAAgIAAEsCAACUAgAA3QIAACYDAAAnAwAA KAMAAGcDAACbAwAAnAMAAKUDAACmAwAA7gMAADEEAABzBAAAlAQAAJUEAACpBAAAqgQAAOsEAAAP BQAAEAUAAFUFAACXBQAA3wUAABwGAAAdBgAAZgYAAK4GAADwBgAALgcAAC8HAABlBwAAZgcAAHcH AAB4BwAAuwcAAOYHAADnBwAAKAgAAFEIAACUCAAAlQgAAJYIAACXCAAA4AgAAOIIAAArCQAALAkA AC0JAABtCQAAtgkAAP4JAABECgAAhwoAALMKAAC0CgAAtQoAAMcKAADICgAAEQsAAFoLAACjCwAA 7AsAADUMAAB+DAAAxwwAABANAABZDQAAog0AAOsNAAA0DgAAfQ4AAH4OAAB/DgAAgA4AAIEOAACC DgAAgw4AAIQOAACFDgAAhg4AAIcOAACIDgAAiQ4AAIoOAACLDgAAjA4AAI0OAACODgAAjw4AAJAO AACRDgAAkg4AAJMOAACUDgAAlQ4AAJYOAACXDgAAmA4AAJkOAADiDgAA5A4AAC0PAAAuDwAALw8A AEAPAABBDwAAiA8AAM8PAAAWEAAAXhAAAGgQAABpEAAAahAAAGsQAABsEAAAbRAAAG4QAABvEAAA cBAAAHEQAAByEAAAcxAAAHQQAAB1EAAAdhAAAHcQAAB4EAAAeRAAAHoQAAB7EAAAfBAAAH0QAAB+ EAAAfxAAAIAQAACBEAAAghAAAIMQAACEEAAAhRAAAIYQAACHEAAAiBAAAIkQAACKEAAAixAAAIwQ AACNEAAAjhAAAI8QAACQEAAAkRAAAJIQAACTEAAAlBAAAN0QAADfEAAAKBEAACkRAAAqEQAAQhEA AEMRAACMEQAA0BEAABQSAABaEgAAohIAAOkSAAAsEwAAdRMAALUTAAD2EwAAOBQAAIAUAADBFAAA zRQAAM4UAAAOFQAADxUAAFcVAACQFQAAkRUAANoVAAAhFgAAZBYAAKoWAADxFgAAMxcAAHUXAAC1 FwAA+RcAADgYAAB5GAAAthgAAP0YAAAzGQAANBkAAHoZAADBGQAACRoAAFEaAACYGgAA4BoAACgb AABmGwAArBsAAPAbAAA5HAAAOhwAADscAAA8HAAAhRwAAIccAADQHAAA0RwAANIcAAAaHQAAWR0A AKEdAADkHQAA5R0AACkeAABwHgAAtR4AAP4eAABEHwAAUh8AAFMfAACbHwAA5B8AACcgAABsIAAA syAAAPogAABDIQAAiSEAANIhAAAYIgAAYSIAAKgiAADsIgAA7SIAADYjAAB9IwAAxSMAAAokAABP JAAAkyQAANQkAAAWJQAAWyUAAKMlAADpJQAAKyYAAG8mAACuJgAA7yYAADMnAAB7JwAAiScAAIon AADOJwAAFygAAFsoAABcKAAAXSgAAF4oAACnKAAAqSgAAPIoAADzKAAA9CgAADwpAACDKQAAwykA AAYqAAA+KgAAPyoAAIgqAADJKgAACysAAC0rAAAuKwAALysAADArAAAxKwAAMisAADMrAAA0KwAA NSsAADYrAAA3KwAAOCsAADkrAAA6KwAAOysAADwrAAA9KwAAPisAAD8rAABAKwAAQSsAAEIrAABD KwAARCsAAEUrAABGKwAARysAAEgrAABJKwAASisAAEsrAABMKwAATSsAAE4rAABPKwAAUCsAAFEr AABSKwAAUysAAFQrAABVKwAAVisAAJ8rAAChKwAA6isAAOsrAADsKwAAGiwAABssAABjLAAAqywA APEsAAA2LQAAfi0AAL4tAAAALgAAQy4AAIYuAADOLgAADy8AAEEvAABCLwAAhS8AAM4vAAAUMAAA XDAAAKMwAADpMAAALTEAADcxAAA4MQAAfjEAAMYxAAAJMgAAUjIAAJoyAADeMgAAHTMAAD8zAABA MwAAdzMAAHgzAAC1MwAAtjMAAP0zAAAsNAAALTQAAHY0AAChNAAAojQAAKM0AACkNAAApTQAAKY0 AACnNAAAqDQAAKk0AACqNAAA8zQAAPU0AAA+NQAAPzUAAEA1AABdNQAAXjUAAKY1AADvNQAAMTYA AFg2AABZNgAAnzYAAOQ2AAAqNwAAczcAALo3AAD+NwAAQzgAAIo4AADLOAAAzDgAABM5AABPOQAA ljkAANw5AAAjOgAAaToAALE6AACyOgAA+zoAAEM7AACKOwAAzTsAABI8AABWPAAAjjwAANU8AAAa PQAAXT0AAHc9AAB4PQAAvD0AAAU+AABMPgAAjz4AANc+AAAdPwAAXj8AAHA/AABxPwAAuD8AAPg/ AAD5PwAA+j8AAPs/AABEQAAARkAAAI9AAACQQAAAkUAAANVAAAAeQQAAYUEAAKdBAADvQQAAMkIA AHpCAAC9QgAAA0MAADxDAAA9QwAAhEMAAMtDAAARRAAAEkQAAFpEAACgRAAA30QAACdFAABqRQAA pkUAALtFAAC8RQAAvUUAAL5FAAC/RQAAwEUAAMFFAADCRQAAw0UAAMRFAADFRQAAxkUAAMdFAADI RQAAyUUAAMpFAADLRQAAzEUAAM1FAADORQAAz0UAANBFAADRRQAA0kUAANNFAADURQAA1UUAANZF AADXRQAA2EUAACFGAAAjRgAAbEYAAG1GAABuRgAAiEYAAIlGAADLRgAA10YAANhGAAAfRwAAY0cA AKpHAADmRwAA50cAAClIAABsSAAAsUgAAOFIAADiSAAAJUkAAG5JAACySQAAy0kAAMxJAAAKSgAA T0oAAJJKAADVSgAA7koAAO9KAAAvSwAAd0sAAL9LAAACTAAASUwAAI5MAADSTAAAF00AAFZNAACc TQAA4E0AACNOAABsTgAAs04AAPBOAAAzTwAAek8AAMJPAAAHUAAAT1AAAJBQAADYUAAA2VAAANpQ AADbUAAAJFEAACZRAABvUQAAcFEAAHFRAAC5UQAAwVIAAMJSAADDUgAAB1MAACZTAAAnUwAAaVMA AKVTAABHVAAASFQAAJFUAADZVAAA6FQAAOlUAACZVQAAmlUAAOJVAAArVgAAc1YAALlWAAD4VgAA DFcAAA1XAABPVwAAmFcAAOFXAAAdWAAAQlgAAENYAACMWAAA0VgAABdZAABdWQAAplkAAO1ZAAA1 WgAAe1oAAL1aAAAGWwAATlsAAJVbAADdWwAAI1wAAFZcAABXXAAAWFwAAFlcAABaXAAAW1wAAFxc AABdXAAAplwAAKhcAADxXAAA8lwAAPNcAAAUXQAAFV0AAFhdAACaXQAAqV0AAKpdAADsXQAAM14A AEJeAABDXgAAcV4AAHJeAACzXgAAtF4AAOheAADpXgAAIF8AACFfAABhXwAAoV8AAMpfAADLXwAA DGAAAFNgAACcYAAA4WAAACdhAABAYQAAQWEAAHthAADBYQAA22EAANxhAAAeYgAAYWIAAKViAADu YgAANWMAAGxjAABtYwAAs2MAAPxjAABAZAAAX2QAAGBkAAChZAAA6WQAAC9lAAAwZQAAMWUAADJl AAB7ZQAAfWUAAMZlAADHZQAAyGUAAAhmAABQZgAAYGYAAGFmAACpZgAA6mYAACxnAABZZwAAWmcA AKBnAADpZwAA6mcAACdoAABtaAAAoGgAAKFoAADqaAAA62gAADFpAAB6aQAAnWkAAJ5pAADmaQAA JWoAAGlqAACmagAA12oAANhqAADZagAA2moAANtqAADcagAA3WoAAN5qAADfagAA4GoAAOFqAADi agAA42oAAORqAADlagAA5moAAOdqAADoagAA6WoAAOpqAADragAA7GoAAO1qAADuagAA72oAADhr AAA6awAAg2sAAIRrAACFawAAoWsAAKJrAADqawAAKWwAAG1sAACSbAAAk2wAAJRsAACVbAAAlmwA AJdsAACYbAAAmWwAAJpsAACbbAAAnGwAAJ1sAACebAAAn2wAAKBsAAChbAAAomwAAKNsAACkbAAA pWwAAKZsAACnbAAAqGwAAKlsAACqbAAAq2wAAKxsAACtbAAArmwAAK9sAACwbAAAsWwAALJsAACz bAAAtGwAALVsAAC2bAAAt2wAALhsAAC5bAAAumwAALtsAAC8bAAAvWwAAL5sAAC/bAAACG0AAApt AABTbQAAVG0AAFVtAABtbQAAbm0AAJhtAACZbQAAmm0AAJttAACcbQAAnW0AAJ5tAACfbQAAoG0A AKFtAACibQAAo20AAKRtAAClbQAApm0AAKdtAACobQAAqW0AAKptAACrbQAArG0AAK1tAACubQAA r20AALBtAACxbQAAsm0AALNtAAC0bQAAtW0AALZtAAC3bQAAuG0AALltAAC6bQAAu20AALxtAAC9 bQAAvm0AAL9tAADAbQAAwW0AAMJtAADDbQAAxG0AAMVtAADGbQAAx20AAMhtAAARbgAAE24AAFxu AABdbgAAXm4AAHJuAABzbgAAu24AAP5uAABEbwAAjW8AAM5vAAAWcAAAP3AAAEBwAABBcAAAQnAA AENwAABEcAAARXAAAEZwAABHcAAASHAAAElwAABKcAAAS3AAAExwAABNcAAATnAAAE9wAABQcAAA UXAAAFJwAABTcAAAVHAAAFVwAABWcAAAV3AAAFhwAABZcAAAWnAAAFtwAABccAAAXXAAAF5wAABf cAAAYHAAAGFwAABicAAAY3AAAGRwAABlcAAAZnAAAGdwAABocAAAaXAAALJwAAC0cAAA/XAAAP5w AAD/cAAAD3EAABBxAAAscQAALXEAAG5xAACncQAAqHEAAO9xAAAgcgAAIXIAAGpyAACtcgAArnIA AMxyAADNcgAAEnMAAFJzAACYcwAAynMAAMtzAAASdAAAUnQAAFN0AACcdAAA43QAAOR0AAAldQAA ZnUAAKB1AAChdQAA6HUAACJ2AAAjdgAAMXYAAHd2AAC5dgAA0nYAANN2AAAadwAAWncAAKB3AAC5 dwAAuncAAAJ4AABLeAAAb3gAAHB4AACweAAAsXgAALJ4AACzeAAA/HgAAP54AABHeQAASHkAAEl5 AACReQAAknkAANl5AAAXegAAGHoAABl6AAAaegAAG3oAABx6AAAdegAAHnoAAB96AAAgegAAIXoA ACJ6AAAjegAAJHoAACV6AAAmegAAJ3oAACh6AAApegAAKnoAACt6AAAsegAALXoAAC56AAAvegAA MHoAADF6AAAyegAAM3oAADR6AAA1egAANnoAADd6AAA4egAAOXoAADp6AAA7egAAPHoAAD16AAA+ egAAP3oAAEB6AABBegAAQnoAAEN6AABEegAARXoAAEZ6AACPegAAkXoAANp6AADbegAA3HoAAO96 AADwegAAA3sAABN7AAAqewAAPnsAAEh7AABJewAAb3sAAHB7AABxewAAgnsAAJB7AACRewAAknsA ALJ7AACzewAAtHsAAMB7AADbewAA7HsAAAJ8AAAMfAAADXwAACp8AAArfAAALHwAAEZ8AABdfAAA XnwAAF98AAB7fAAAfHwAAH18AACZfAAAsXwAALJ8AACzfAAA0HwAANF8AADSfAAA03wAANR8AADV fAAA1nwAANd8AADYfAAA2XwAANp8AADbfAAA3HwAACV9AAAnfQAAKn0AAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA gACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA gACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAgACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAIAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAACAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAEACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8w AAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAA AAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAA AAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAA AAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAA AAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAA AACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAA AIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAA gAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACA AAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAA AACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAA AIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAA gAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACA AAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAA AAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAA AAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAA AAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAA AAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAA AAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAA AAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAA AAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAA AACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAA AJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAA mAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACY AAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgA AAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAADzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAA AA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAPMAAAAAAAAACAAAAAgAAAAAAAAAAAAACYAAAA DzAAAAAAAAAAgAAAAIAAAAAAAAAAAAAAmAAAAA8wAAAAAAAAAIAAAACAAAAAAAAAAAAAAJgAAAAP MAAAAAAAAACAAAAAgAAAAAAAAAAAAAAABgAAKIUAAEMAAAAABgAA6wwAACwRAACCFgAAQBcAAH8Y AACMGQAAOCAAAOQlAADULAAAPzIAAEYzAAC+NQAAtjsAAOQ+AAB4RQAAA0sAAMpNAADnTwAAbFYA AMFaAADhXwAA8mQAAEBpAAAIbgAA2nIAAOpzAACsdAAAmnUAALd1AADOdwAAWngAACB6AADTfgAA HoIAADuCAACQgwAA04QAACiFAABEAAAARgAAAEcAAABIAAAASQAAAEoAAABLAAAATAAAAE0AAABO AAAATwAAAFAAAABRAAAAUgAAAFMAAABUAAAAVQAAAFYAAABXAAAAWAAAAFkAAABaAAAAWwAAAFwA AABdAAAAXgAAAF8AAABgAAAAYQAAAGIAAABjAAAAZAAAAGUAAABmAAAAZwAAAGgAAABpAAAAagAA AAAGAAAohQAARQAAAA8AAPA4AAAAAAAG8BgAAAACCAAAAgAAAAEAAAABAAAAAQAAAAIAAABAAB7x EAAAAP//AAAAAP8AgICAAPcAABAADwAC8JIAAAAQAAjwCAAAAAEAAAABBAAADwAD8DAAAAAPAATw KAAAAAEACfAQAAAAAAAAAAAAAAAAAAAAAAAAAAIACvAIAAAAAAQAAAUAAAAPAATwQgAAABIACvAI AAAAAQQAAAAOAABTAAvwHgAAAL8BAAAQAMsBAAAAAP8BAAAIAAQDCQAAAD8DAQABAAAAEfAEAAAA AQAAAAAAAAADAAAAAAEAABwBAAAnAQAAaQEAAHABAAChAQAAuQEAAPQBAAACAgAANwIAAEsCAACM AgAAlAIAAMgCAADdAgAAFQMAACYDAAAyAwAAZwMAAHsDAACbAwAAnAMAAKUDAACpAwAA7gMAAPED AAAxBAAANAQAAHMEAAB2BAAAlAQAAJUEAACpBAAArQQAAOsEAADuBAAADwUAABMFAABVBQAAWAUA AJcFAACaBQAA3wUAAOIFAAAcBgAAIAYAAGYGAABpBgAArgYAALEGAADwBgAA8wYAAC4HAAAyBwAA ZQcAAGYHAAB3BwAAewcAALsHAAC+BwAA5gcAAOoHAAAoCAAAKwgAAFEIAABVCAAAlAgAAJcIAADg CAAA4ggAAPIIAAD+CAAAKwkAADAJAABtCQAAcAkAALYJAAC5CQAA/gkAAAEKAABECgAARwoAAIcK AACKCgAAswoAALUKAADHCgAAywoAAPIKAAAPCwAAEQsAABQLAABaCwAAXQsAAKMLAACmCwAA7AsA AO8LAADyCwAA8wsAADUMAAA4DAAAfgwAAIEMAADHDAAAygwAAPIMAAANDQAAEA0AABMNAABZDQAA XA0AAKINAACnDQAA6w0AAPANAADyDQAA8w0AADQOAAA3DgAAfQ4AAJkOAADiDgAA5A4AAPIOAAAA DwAALQ8AAC8PAABADwAARA8AAIgPAACLDwAAzw8AANIPAAAWEAAAGRAAAF4QAABhEAAAaBAAAJQQ AADdEAAA3xAAAPIQAAD7EAAAKBEAACoRAABCEQAARhEAAHsRAAB8EQAAjBEAAI8RAADQEQAA0xEA ABQSAAAXEgAAWhIAAF0SAACiEgAApRIAAOkSAADsEgAALBMAAC8TAAB1EwAAeBMAALUTAAC4EwAA 9hMAAPkTAAA4FAAAOxQAAIAUAACDFAAAwRQAAMQUAADNFAAA0RQAAA4VAAASFQAAVxUAAF4VAACQ FQAAlBUAANoVAADhFQAAIRYAACgWAABkFgAAaxYAAKoWAACxFgAA8RYAAPgWAAAzFwAAOhcAAHUX AAB8FwAAtRcAALwXAAD5FwAAABgAADgYAAA/GAAAeRgAAIAYAAC2GAAAvRgAAP0YAAAEGQAAMxkA ADcZAAB6GQAAgRkAAMEZAADIGQAACRoAABAaAABRGgAAWBoAAJgaAACfGgAA4BoAAOcaAAAoGwAA LxsAAGYbAABtGwAAixsAAI0bAACsGwAAsxsAAPAbAAD3GwAALhwAADAcAAA4HAAAPBwAAEEcAABD HAAAeBwAAH0cAACFHAAAhxwAANAcAADZHAAA/RwAAP4cAAAaHQAAIR0AADodAAA9HQAAWR0AAGAd AAChHQAAqB0AAOQdAADoHQAAKR4AADAeAABwHgAAdx4AALUeAAC8HgAA/h4AAAUfAABEHwAASx8A AFIfAABWHwAAmx8AAKIfAADkHwAA6x8AACcgAAAuIAAAbCAAAHMgAACzIAAAuiAAAPogAAABIQAA QyEAAEohAAB4IQAAeSEAAIkhAACQIQAA0iEAANkhAAAYIgAAHyIAAGEiAABoIgAAqCIAAK8iAADs IgAA8CIAADYjAAA9IwAAfSMAAIQjAADFIwAAzCMAAAokAAARJAAATyQAAFYkAACTJAAAmiQAANQk AADbJAAAFiUAAB0lAABbJQAAYiUAAKMlAACqJQAA6SUAAPAlAAArJgAAMiYAAG8mAAB2JgAAriYA ALUmAADvJgAA9iYAADMnAAA6JwAAeycAAIInAACJJwAAjScAAM4nAADVJwAAFygAAB4oAABbKAAA XigAAKcoAACpKAAA8igAAPsoAAA8KQAAQykAAIMpAACKKQAAwykAAMopAAAGKgAADSoAAD4qAABC KgAAiCoAAI8qAADJKgAA0CoAAAsrAAASKwAALSsAAFYrAACfKwAAoSsAAOorAADsKwAAGiwAAB4s AABjLAAAZiwAAKssAACuLAAA8SwAAPQsAAA2LQAAOS0AAH4tAACBLQAAvi0AAMEtAAAALgAAAy4A AEMuAABGLgAAhi4AAIkuAADOLgAA0S4AAA8vAAASLwAAQS8AAEUvAACFLwAAiC8AAM4vAADRLwAA FDAAABcwAABcMAAAXzAAAKMwAACmMAAA6TAAAOwwAAAtMQAAMDEAADcxAAA7MQAAfjEAAIExAADG MQAAyTEAAAkyAAAMMgAAUjIAAFUyAACaMgAAnTIAAN4yAADhMgAAHTMAACAzAAA/MwAAQzMAAHcz AAB7MwAAtTMAALkzAAD9MwAAAzQAACw0AAAwNAAAdjQAAHk0AAChNAAAqjQAAPM0AAD1NAAAPjUA AEA1AABdNQAAYTUAAKY1AACpNQAA7zUAAPI1AAAxNgAANDYAAFc2AABcNgAAnzYAAKY2AADkNgAA 6zYAACo3AAAxNwAAczcAAHo3AAC6NwAAwTcAAP43AAAFOAAAQzgAAEo4AACKOAAAkTgAAMs4AADP OAAAADkAAAE5AAATOQAAGjkAAE85AABWOQAAljkAAJ05AADcOQAA4zkAACM6AAAqOgAAaToAAHA6 AACxOgAAtToAAPs6AAACOwAAQzsAAEo7AACKOwAAkTsAAM07AADUOwAAEjwAABk8AABWPAAAXTwA AI48AACVPAAA1TwAANw8AAAaPQAAIT0AAF09AABkPQAAdz0AAHs9AAC8PQAAwz0AAAU+AAAMPgAA TD4AAFM+AACPPgAAlj4AANc+AADePgAAAD8AAAI/AAAdPwAAJD8AAF4/AABlPwAAcD8AAHQ/AAC4 PwAAvz8AAPg/AAD7PwAAAEAAAAJAAABEQAAARkAAAI9AAACYQAAA1UAAANxAAAAeQQAAJUEAAGFB AABoQQAAp0EAAK5BAADvQQAA9kEAADJCAAA5QgAAekIAAIFCAAC9QgAAxEIAAANDAAAKQwAAPEMA AEBDAACEQwAAi0MAAMtDAADSQwAAAEQAAAFEAAARRAAAFUQAAFpEAABhRAAAoEQAAKdEAADfRAAA 5kQAAABFAAABRQAAJ0UAAC5FAABqRQAAcUUAAKZFAACtRQAAu0UAANhFAAAhRgAAI0YAAGxGAABu RgAAiEYAAIxGAADLRgAAzkYAANdGAADbRgAAH0cAACVHAABjRwAAaUcAAKpHAACwRwAA5kcAAOpH AAAASAAAAUgAAClIAAAvSAAAbEgAAHJIAACxSAAAt0gAAOFIAADlSAAAAEkAAAFJAAAlSQAAK0kA AG5JAAB0SQAAskkAALhJAADLSQAAz0kAAApKAAAQSgAAT0oAAFVKAACSSgAAmEoAANVKAADbSgAA 7koAAPJKAAAvSwAAMksAAHdLAAB6SwAAv0sAAMJLAAACTAAABUwAAElMAABMTAAAjkwAAJFMAADS TAAA1UwAABdNAAAaTQAAVk0AAFlNAACcTQAAn00AAOBNAADjTQAAI04AACZOAABsTgAAb04AALNO AAC2TgAA8E4AAPNOAAAzTwAANk8AAHpPAAB9TwAAwk8AAMVPAAAHUAAAClAAAE9QAABSUAAAkFAA AJNQAADYUAAA21AAACRRAAAmUQAAb1EAAHRRAAC4UQAAuVEAAO5RAADwUQAAwFIAAMZSAAACUwAA A1MAAAdTAAAKUwAAJlMAACpTAABpUwAAbVMAAKVTAACpUwAA41MAAOVTAADrUwAA7FMAABxUAAAd VAAAL1QAADBUAABBVAAAS1QAAJFUAACYVAAA2VQAAOBUAADoVAAA6VQAACJVAAAkVQAAOlUAADtV AABwVQAAcVUAAHdVAAB4VQAAelUAAHtVAACYVQAAnVUAAOJVAADpVQAAK1YAADJWAABzVgAAelYA ALlWAADAVgAA+FYAAP9WAAAMVwAAEFcAAE9XAABWVwAAmFcAAJ9XAADhVwAA6FcAAB1YAAAkWAAA QlgAAEZYAACMWAAAj1gAANFYAADUWAAAF1kAABpZAABdWQAAYFkAAKZZAACpWQAA7VkAAPBZAAA1 WgAAOFoAAHtaAAB+WgAAn1oAAKBaAAC9WgAAwFoAAAZbAAAJWwAATlsAAFFbAACVWwAAmFsAAN1b AADgWwAAI1wAACZcAABWXAAAXVwAAKZcAACoXAAA8VwAAPNcAAAUXQAAGF0AAFhdAABbXQAAml0A AJ1dAACpXQAArV0AAOxdAADyXQAAM14AADleAABCXgAARl4AAHFeAAB1XgAAs14AALdeAADoXgAA 7F4AACBfAAAkXwAAYV8AAGRfAAChXwAApV8AAMpfAADSXwAADGAAABJgAABTYAAAWWAAAJxgAACi YAAA4WAAAOdgAAAnYQAALWEAAEBhAABEYQAAe2EAAH9hAADBYQAAxGEAANthAADfYQAAHmIAACRi AABhYgAAZ2IAAKViAACrYgAA7mIAAPRiAAA1YwAAO2MAAGxjAABwYwAAs2MAALljAAD8YwAAAmQA AEBkAABGZAAAX2QAAGNkAAChZAAAp2QAAOlkAADvZAAAL2UAADJlAAB7ZQAAfWUAAMZlAADOZQAA CGYAAA5mAABQZgAAVmYAAGBmAABkZgAAqWYAAKxmAADqZgAA7WYAACxnAAAvZwAAWWcAAF1nAACf ZwAApmcAAOlnAADtZwAAJ2gAAC1oAABtaAAAc2gAAJ9oAACkaAAA6mgAAO5oAAAxaQAANGkAAHpp AAB9aQAAnWkAAKFpAADmaQAA6WkAACVqAAAoagAAaWoAAGxqAACmagAAqWoAANdqAADvagAAOGsA ADprAACDawAAhWsAAKFrAAClawAA6msAAO1rAAApbAAALGwAAG1sAABwbAAAkmwAAL9sAAAIbQAA Cm0AAFNtAABVbQAAbW0AAHFtAACYbQAAyG0AABFuAAATbgAAXG4AAF5uAABybgAAdm4AALtuAAC+ bgAA/m4AAAFvAABEbwAAR28AAI1vAACQbwAAzm8AANFvAAAWcAAAGXAAAD9wAABpcAAAn3AAAKlw AACycAAAtHAAAP1wAAD/cAAAD3EAABBxAAAscQAAMXEAAG5xAAB8cQAAp3EAAKxxAADvcQAA/XEA ACByAAAlcgAAanIAAHhyAACtcgAArnIAAMxyAADRcgAAEnMAACBzAABScwAAYHMAAJhzAACmcwAA ynMAAM9zAAASdAAAIHQAAFJ0AABXdAAAnHQAAKp0AADjdAAA6HQAACV1AAAzdQAAZnUAAHR1AACf dQAApXUAAOh1AAD2dQAAInYAACd2AAAxdgAAP3YAAHd2AACFdgAAuXYAAMd2AADSdgAA13YAABp3 AAAodwAAWncAAGh3AACfdwAArncAALl3AAC+dwAAAngAABB4AABLeAAAWXgAAG94AAB0eAAAsHgA ALN4AAD8eAAA/ngAAEd5AABXeQAAkXkAAJZ5AACfeQAAoHkAANl5AADneQAAF3oAAEZ6AACPegAA kXoAAJ96AACtegAA2noAANx6AADvegAA83oAAAN7AAAGewAAE3sAABZ7AAAqewAALXsAAD57AABB ewAASHsAAEx7AABvewAAdHsAAIJ7AACFewAAkHsAAJV7AACyewAAt3sAAMB7AADDewAA23sAAN57 AADsewAA73sAAAJ8AAAFfAAADHwAABB8AAAqfAAAL3wAAEZ8AABJfAAAXXwAAGJ8AAB7fAAAgHwA AJl8AACcfAAAsXwAALZ8AADQfAAA3HwAACV9AAAqfQAABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAAAAAAAwAAAEsAAABMAAAAlAAAAJUAAADdAAAA3gAAACYBAABpAQAA bwEAAKEBAAC4AQAA9AEAAAECAAA3AgAASgIAAIwCAACTAgAAyAIAANwCAAAVAwAAJQMAADIDAABm AwAAewMAAJoDAACcAwAApAMAAKkDAADtAwAA8QMAADAEAAA0BAAAcgQAAHYEAACTBAAAlQQAAKgE AACtBAAA6gQAAO4EAAAOBQAAEwUAAFQFAABYBQAAlgUAAJoFAADeBQAA4gUAABsGAAAgBgAAZQYA AGkGAACtBgAAsQYAAO8GAADzBgAALQcAADIHAABkBwAAZgcAAHYHAAB7BwAAugcAAL4HAADlBwAA 6gcAACcIAAArCAAAUAgAAFQIAACTCAAAlwgAAN8IAADiCAAAKgkAADAJAABsCQAAcAkAALUJAAC5 CQAA/QkAAAEKAABDCgAARwoAAIYKAACKCgAAsgoAALUKAADGCgAAywoAABELAAAUCwAAWgsAAF0L AACjCwAApgsAAOwLAADvCwAANAwAADgMAAB9DAAAgQwAAMYMAADKDAAADw0AABMNAABYDQAAXA0A AKENAACnDQAA6g0AAPANAAAzDgAANw4AAHwOAACZDgAA4Q4AAOQOAAAsDwAALw8AAD8PAABEDwAA hw8AAIsPAADODwAA0g8AABUQAAAZEAAAXRAAAGEQAABnEAAAlBAAANwQAADfEAAAJxEAACoRAABB EQAARhEAAIsRAACPEQAAzxEAANMRAAATEgAAFxIAAFkSAABdEgAAoRIAAKUSAADoEgAA7BIAACsT AAAvEwAAdBMAAHgTAAC0EwAAuBMAAPUTAAD5EwAANxQAADsUAAB/FAAAgxQAAMAUAADEFAAAzBQA ANEUAAANFQAAEhUAAFYVAABeFQAAjxUAAJQVAADZFQAA4RUAACAWAAAoFgAAYxYAAGsWAACpFgAA sRYAAPAWAAD4FgAAMhcAADoXAAB0FwAAfBcAALQXAAC8FwAA+BcAAAAYAAA3GAAAPxgAAHgYAACA GAAAtRgAAL0YAAD8GAAABBkAADIZAAA3GQAAeRkAAIEZAADAGQAAyBkAAAgaAAAQGgAAUBoAAFga AACXGgAAnxoAAN8aAADnGgAAJxsAAC8bAABlGwAAbRsAAKsbAACzGwAA7xsAAPcbAAA4HAAAPBwA AIQcAACHHAAAzxwAANkcAAAZHQAAIR0AAFgdAABgHQAAoB0AAKgdAADjHQAA6B0AACgeAAAwHgAA bx4AAHceAAC0HgAAvB4AAP0eAAAFHwAAQx8AAEsfAABRHwAAVh8AAJofAACiHwAA4x8AAOsfAAAm IAAALiAAAGsgAABzIAAAsiAAALogAAD5IAAAASEAAEIhAABKIQAAiCEAAJAhAADRIQAA2SEAABci AAAfIgAAYCIAAGgiAACnIgAAryIAAOsiAADwIgAANSMAAD0jAAB8IwAAhCMAAMQjAADMIwAACSQA ABEkAABOJAAAViQAAJIkAACaJAAA0yQAANskAAAVJQAAHSUAAFolAABiJQAAoiUAAKolAADoJQAA 8CUAAComAAAyJgAAbiYAAHYmAACtJgAAtSYAAO4mAAD2JgAAMicAADonAAB6JwAAgicAAIgnAACN JwAAzScAANUnAAAWKAAAHigAAFooAABeKAAApigAAKkoAADxKAAA+ygAADspAABDKQAAgikAAIop AADCKQAAyikAAAUqAAANKgAAPSoAAEIqAACHKgAAjyoAAMgqAADQKgAACisAABIrAAAsKwAAVisA AJ4rAAChKwAA6SsAAOwrAAAZLAAAHiwAAGIsAABmLAAAqiwAAK4sAADwLAAA9CwAADUtAAA5LQAA fS0AAIEtAAC9LQAAwS0AAP8tAAADLgAAQi4AAEYuAACFLgAAiS4AAM0uAADRLgAADi8AABIvAABA LwAARS8AAIQvAACILwAAzS8AANEvAAATMAAAFzAAAFswAABfMAAAojAAAKYwAADoMAAA7DAAACwx AAAwMQAANjEAADsxAAB9MQAAgTEAAMUxAADJMQAACDIAAAwyAABRMgAAVTIAAJkyAACdMgAA3TIA AOEyAAAcMwAAIDMAAD4zAABDMwAAdjMAAHszAAC0MwAAuTMAAPwzAAADNAAAKzQAADA0AAB1NAAA eTQAAKA0AACqNAAA8jQAAPU0AAA9NQAAQDUAAFw1AABhNQAApTUAAKk1AADuNQAA8jUAADA2AAA0 NgAAVzYAAFw2AACeNgAApjYAAOM2AADrNgAAKTcAADE3AAByNwAAejcAALk3AADBNwAA/TcAAAU4 AABCOAAASjgAAIk4AACROAAAyjgAAM84AAASOQAAGjkAAE45AABWOQAAlTkAAJ05AADbOQAA4zkA ACI6AAAqOgAAaDoAAHA6AACwOgAAtToAAPo6AAACOwAAQjsAAEo7AACJOwAAkTsAAMw7AADUOwAA ETwAABk8AABVPAAAXTwAAI08AACVPAAA1DwAANw8AAAZPQAAIT0AAFw9AABkPQAAdj0AAHs9AAC7 PQAAwz0AAAQ+AAAMPgAASz4AAFM+AACOPgAAlj4AANY+AADePgAAHD8AACQ/AABdPwAAZT8AAG8/ AAB0PwAAtz8AAL8/AAD3PwAA+z8AAENAAABGQAAAjkAAAJhAAADUQAAA3EAAAB1BAAAlQQAAYEEA AGhBAACmQQAArkEAAO5BAAD2QQAAMUIAADlCAAB5QgAAgUIAALxCAADEQgAAAkMAAApDAAA7QwAA QEMAAINDAACLQwAAykMAANJDAAAQRAAAFUQAAFlEAABhRAAAn0QAAKdEAADeRAAA5kQAACZFAAAu RQAAaUUAAHFFAAClRQAArUUAALpFAADYRQAAIEYAACNGAABrRgAAbkYAAIdGAACMRgAAykYAAM5G AADWRgAA20YAAB5HAAAlRwAAYkcAAGlHAACpRwAAsEcAAOVHAADqRwAAKEgAAC9IAABrSAAAckgA ALBIAAC3SAAA4EgAAOVIAAAkSQAAK0kAAG1JAAB0SQAAsUkAALhJAADKSQAAz0kAAAlKAAAQSgAA TkoAAFVKAACRSgAAmEoAANRKAADbSgAA7UoAAPJKAAAuSwAAMksAAHZLAAB6SwAAvksAAMJLAAAB TAAABUwAAEhMAABMTAAAjUwAAJFMAADRTAAA1UwAABZNAAAaTQAAVU0AAFlNAACbTQAAn00AAN9N AADjTQAAIk4AACZOAABrTgAAb04AALJOAAC2TgAA704AAPNOAAAyTwAANk8AAHlPAAB9TwAAwU8A AMVPAAAGUAAAClAAAE5QAABSUAAAj1AAAJNQAADXUAAA21AAACNRAAAmUQAAblEAAHRRAAC4UQAA uVEAAMBSAADGUgAABlMAAApTAAAlUwAAKlMAAGhTAABtUwAApFMAAKlTAADkUwAA5VMAAEZUAABL VAAAkFQAAJhUAADYVAAA4FQAAOdUAADpVAAAmFUAAJ1VAADhVQAA6VUAACpWAAAyVgAAclYAAHpW AAC4VgAAwFYAAPdWAAD/VgAAC1cAABBXAABOVwAAVlcAAJdXAACfVwAA4FcAAOhXAAAcWAAAJFgA AEFYAABGWAAAi1gAAI9YAADQWAAA1FgAABZZAAAaWQAAXFkAAGBZAAClWQAAqVkAAOxZAADwWQAA NFoAADhaAAB6WgAAfloAALxaAADAWgAABVsAAAlbAABNWwAAUVsAAJRbAACYWwAA3FsAAOBbAAAi XAAAJlwAAFVcAABdXAAApVwAAKhcAADwXAAA81wAABNdAAAYXQAAV10AAFtdAACZXQAAnV0AAKhd AACtXQAA610AAPJdAAAyXgAAOV4AAEFeAABGXgAAcF4AAHVeAACyXgAAt14AAOdeAADsXgAAH18A ACRfAABgXwAAZF8AAKBfAACkXwAAyV8AANFfAAALYAAAEmAAAFJgAABZYAAAm2AAAKJgAADgYAAA 52AAACZhAAAtYQAAP2EAAERhAAB6YQAAfmEAAMBhAADEYQAA2mEAAN9hAAAdYgAAJGIAAGBiAABn YgAApGIAAKtiAADtYgAA9GIAADRjAAA7YwAAa2MAAHBjAACyYwAAuWMAAPtjAAACZAAAP2QAAEZk AABeZAAAY2QAAKBkAACnZAAA6GQAAO9kAAAuZQAAMmUAAHplAAB9ZQAAxWUAAM5lAAAHZgAADmYA AE9mAABWZgAAX2YAAGRmAACoZgAArGYAAOlmAADtZgAAK2cAAC9nAABYZwAAXWcAAJ9nAACmZwAA 6GcAAO1nAAAmaAAALWgAAGxoAABzaAAAn2gAAKRoAADpaAAA7mgAADBpAAA0aQAAeWkAAH1pAACc aQAAoWkAAOVpAADpaQAAJGoAAChqAABoagAAbGoAAKVqAACpagAA1moAAO9qAAA3awAAOmsAAIJr AACFawAAoGsAAKVrAADpawAA7WsAAChsAAAsbAAAbGwAAHBsAACRbAAAv2wAAAdtAAAKbQAAUm0A AFVtAABsbQAAcW0AAJdtAADIbQAAEG4AABNuAABbbgAAXm4AAHFuAAB2bgAAum4AAL5uAAD9bgAA AW8AAENvAABHbwAAjG8AAJBvAADNbwAA0W8AABVwAAAZcAAAPnAAAGlwAACxcAAAtHAAAPxwAAD/ cAAADnEAABBxAAArcQAAMHEAAG1xAAB8cQAApnEAAKtxAADucQAA/XEAAB9yAAAkcgAAaXIAAHhy AACscgAArnIAAMtyAADQcgAAEXMAACBzAABRcwAAYHMAAJdzAACmcwAAyXMAAM5zAAARdAAAIHQA AFF0AABWdAAAm3QAAKp0AADidAAA53QAACR1AAAzdQAAZXUAAHR1AACfdQAApHUAAOd1AAD2dQAA IXYAACZ2AAAwdgAAP3YAAHZ2AACFdgAAuHYAAMd2AADRdgAA1nYAABl3AAAodwAAWXcAAGh3AACf dwAArncAALh3AAC9dwAAAXgAABB4AABKeAAAWXgAAG54AABzeAAAr3gAALN4AAD7eAAA/ngAAEZ5 AABXeQAAkHkAAJV5AADYeQAA53kAABZ6AABGegAAjnoAAJF6AADZegAA3HoAAO56AADzegAAAnsA AAZ7AAASewAAFnsAACl7AAAtewAAPXsAAEF7AABHewAATHsAAG57AAB0ewAAgXsAAIV7AACPewAA lXsAALF7AAC3ewAAv3sAAMN7AADaewAA3nsAAOt7AADvewAAAXwAAAV8AAALfAAAEHwAACl8AAAv fAAARXwAAEl8AABcfAAAYnwAAHp8AACAfAAAmHwAAJx8AACwfAAAtnwAAM98AADcfAAAJH0AACp9 AAAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcA BQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAF AAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUA BwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAHAAUABwAFAAcABQAH AAUABwAFAAcABQAHAAUABwAFAAcAAAAAACp9AAAHAAEAhGWdCuIXLOf/D/8P/w//D/8P/w//D/8P /w8QAAcAAAAXAAAAAAAAAAAAAAAAAAAAAAAAACAQAAAPhN8CEYSJ/l6E3wJghIn+QioAQ0oVAE9K AQBQSgQAUUoBAF5KAwBvKABwaAAAAP8BALfwAQAAABeAAAAAAAAAAAAAAAAAAAAAAAAAGRAAAA+E oAURhJj+XoSgBWCEmP5PSgMAUUoDAF5KAwBvKACHaAAAAACISAAAAQBvAAEAAAAXgAAAAAAAAAAA AAAAAAAAAAAAABUQAAAPhHAIEYSY/l6EcAhghJj+T0oHAFFKBwBvKACHaAAAAACISAAAAQCn8AEA AAAXgAAAAAAAAAAAAAAAAAAAAAAAABUQAAAPhEALEYSY/l6EQAtghJj+T0oBAFFKAQBvKACHaAAA AACISAAAAQC38AEAAAAXgAAAAAAAAAAAAAAAAAAAAAAAABkQAAAPhBAOEYSY/l6EEA5ghJj+T0oD AFFKAwBeSgMAbygAh2gAAAAAiEgAAAEAbwABAAAAF4AAAAAAAAAAAAAAAAAAAAAAAAAVEAAAD4Tg EBGEmP5ehOAQYISY/k9KBwBRSgcAbygAh2gAAAAAiEgAAAEAp/ABAAAAF4AAAAAAAAAAAAAAAAAA AAAAAAAVEAAAD4SwExGEmP5ehLATYISY/k9KAQBRSgEAbygAh2gAAAAAiEgAAAEAt/ABAAAAF4AA AAAAAAAAAAAAAAAAAAAAAAAZEAAAD4SAFhGEmP5ehIAWYISY/k9KAwBRSgMAXkoDAG8oAIdoAAAA AIhIAAABAG8AAQAAABeAAAAAAAAAAAAAAAAAAAAAAAAAFRAAAA+EUBkRhJj+XoRQGWCEmP5PSgcA UUoHAG8oAIdoAAAAAIhIAAABAKfwAQAAAIRlnQoAAAAAAAAAAAAAAAD///////8BAAAAAAD//wEA AAASAFKYDEADAAcEBQAHBAEABwQDAAcEBQAHBAEABwQDAAcEBQAHBAEA9TuJHQAAAAAAAAAAAAEC AAIAIQAAAAQAAAAIAAAA5QAAAAAAAAAbAAAAND8GAPB/BwAESBMA1D0UAL86GQANGywASj87AC8B QQA4MUkAHVhMAIwLUwA8EGAAxF9lAHF2cgALYHUAXA54AJVXeQDwWHoALxmVADZRmgDRZ6EAvSOm AHIPpwByWKwAI06vAOdouADFNcgAayrZABAz6AD5eesALyvxALIW+gBiCPsAAAAAACd9AAAqfQAA AAAAAAEAAAD/QAOAAQDCUgAAwlIAADy+nwsBAAAAwlIAAAAAAADCUgAAAAAAAAIQAAAAAAAAACh9 AABQAAAQAEAAAP//AgAAAAcAVQBuAGsAbgBvAHcAbgADAEoAUwB2AP//AgAIAAAAAAAAAAAAAAAA AAAAAAAAAAEA//8CAAAAAAAAAP//AAACAP//AAAAAP//AAACAP//AAAAAAgAAABHFpABAAACAgYD BQQFAgME7zoA4EF4AMAJAAAAAAAAAP8BAAAAAAAAVABpAG0AZQBzACAATgBlAHcAIABSAG8AbQBh AG4AAAA1FpABAgAFBQECAQcGAgUHAAAAAAAAABAAAAAAAAAAAAAAAIAAAAAAUwB5AG0AYgBvAGwA AAAzJpABAAACCwYEAgICAgIE/zoA4EN4AMAJAAAAAAAAAP8BAAAAAAAAQQByAGkAYQBsAAAAPzWQ AQAAAgcDCQICBQIEBP86AOBDeADACQAAAAAAAAD/AQAAAAAAAEMAbwB1AHIAaQBlAHIAIABOAGUA dwAAADcmkAEAAAIPBQICAgQDAgTvAgCgeyAAQAAAAAAAAAAAnwAAAAAAAABDAGEAbABpAGIAcgBp AAAAOTWQAQAAAgsGCQICBAMCBO8CAKBLIABAAAAAAAAAAACfAAAAAAAAAEMAbwBuAHMAbwBsAGEA cwAAADUmkAEAAAILBgQDBQQEAgT/OgDhW2AAwCkAAAAAAAAA/wEBAAAAAABUAGEAaABvAG0AYQAA ADsGkAECAAUAAAAAAAAAAAAAAAAAAAAAEAAAAAAAAAAAAAAAgAAAAABXAGkAbgBnAGQAaQBuAGcA cwAAACIABADxiYwYAPDEAgAAqQEAAAAARtPqRl3T6kYAAAAAAwAFAAAArhIAAHpqAAABAD8AAAAE AAOQ4wAAAK4SAAB6agAAAQA/AAAA4wAAAAAAAAAhAwDwEAAAAAEAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAA2BYkFtAC0AIGBcjQAABAAGQBkAAAAGQAAAOl8AADpfAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgAAAAAAAAAAAAgyg1EA 8BAACAD8/QEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASFAAAAAAKfD/DwEAJFAAAAAAAADvUQAA ////f////3////9/////f////3////9/clisAAAEAAAyAAAAAAAAAAAAAAAAAAAAAAD//xIAAAAA AAAAOwBDAE8ARABFAEMAIABXAEcAIABHAHUAaQBkAGUAbABpAG4AZQBzACAAIAAgACAAIAAgACAA IAAgACAAIAAgACAAIAAgACAAIAAgACAAIAAgACAAIAAgACAAIAAgACAAIAAgACAAIAAgACAAIAAg ACAAIABKAE0AAAAAAAAADgBKAG8AbgBhAHMAIABTAHYAZQBkAGIAZQByAGcAAwBKAFMAdgAAAAAA AAAAAAAAAAAAAAAAAAAAABAAAAAGAAAAAQAAAAAADAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AP7/AAAGAAIAAAAAAAAAAAAAAAAAAAAAAAEAAADghZ/y+U9oEKuRCAArJ7PZMAAAAIgBAAAPAAAA AQAAAIgAAAACAAAAkAAAAAMAAADUAAAABAAAAOAAAAAFAAAA+AAAAAcAAAAEAQAACQAAABgBAAAS AAAAJAEAAAoAAABEAQAADAAAAFABAAANAAAAXAEAAA4AAABoAQAADwAAAHABAAAQAAAAeAEAABMA AACAAQAAAAAAAAAAAAACAAAA5AQAAB4AAAA8AAAAQ09ERUMgV0cgR3VpZGVsaW5lcyAgICAgICAg ICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgSk0AHgAAAAQAAAAAAAAAHgAAABAAAABKb25h cyBTdmVkYmVyZwAAHgAAAAQAAAAAAAAAHgAAAAwAAABOb3JtYWwuZG90AAAeAAAABAAAADMAAAAe AAAAGAAAAE1pY3Jvc29mdCBPZmZpY2UgV29yZAAAAEAAAAAAXtCyAAAAAEAAAAAAnIvF/XTLAUAA AAAA5hb8AHXLAQMAAAABAAAAAwAAAK4SAAADAAAAemoAAAMAAAAAAAAAemoAAAMAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD+/wAABgAC AAAAAAAAAAAAAAAAAAAAAAABAAAAAtXN1ZwuGxCTlwgAKyz5rjAAAAAkAQAADAAAAAEAAABoAAAA DwAAAHAAAAAFAAAAfAAAAAYAAACEAAAAEQAAAIwAAAAXAAAAlAAAAAsAAACcAAAAEAAAAKQAAAAT AAAArAAAABYAAAC0AAAADQAAALwAAAAMAAAABAEAAAIAAADkBAAAHgAAAAQAAAAAAAAAAwAAAOMA AAADAAAAPwAAAAMAAADpfAAAAwAAAA8nCwALAAAAAAAAAAsAAAAAAAAACwAAAAAAAAALAAAAAAAA AB4QAAABAAAAPAAAAENPREVDIFdHIEd1aWRlbGluZXMgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgIEpNAAwQAAACAAAAHgAAAAYAAABUaXRsZQADAAAAAQAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQAAAAIAAAADAAAABAAA AAUAAAAGAAAABwAAAAgAAAAJAAAACgAAAAsAAAAMAAAADQAAAA4AAAAPAAAAEAAAABEAAAASAAAA EwAAABQAAAAVAAAAFgAAABcAAAAYAAAAGQAAABoAAAAbAAAAHAAAAB0AAAAeAAAAHwAAACAAAAAh AAAAIgAAACMAAAAkAAAAJQAAACYAAAAnAAAAKAAAACkAAAAqAAAAKwAAACwAAAAtAAAALgAAAC8A AAAwAAAAMQAAADIAAAAzAAAANAAAADUAAAA2AAAANwAAADgAAAA5AAAAOgAAADsAAAA8AAAAPQAA AD4AAAA/AAAAQAAAAEEAAABCAAAAQwAAAEQAAABFAAAARgAAAEcAAABIAAAASQAAAEoAAABLAAAA TAAAAE0AAABOAAAATwAAAFAAAABRAAAAUgAAAFMAAABUAAAAVQAAAFYAAABXAAAAWAAAAFkAAABa AAAAWwAAAFwAAABdAAAAXgAAAF8AAABgAAAAYQAAAGIAAABjAAAAZAAAAGUAAABmAAAAZwAAAGgA AABpAAAAagAAAGsAAAD+////bQAAAG4AAABvAAAAcAAAAHEAAAByAAAAcwAAAP7///91AAAAdgAA AHcAAAB4AAAAeQAAAHoAAAB7AAAAfAAAAH0AAAB+AAAAfwAAAIAAAACBAAAAggAAAIMAAACEAAAA hQAAAIYAAACHAAAAiAAAAIkAAACKAAAAiwAAAIwAAACNAAAAjgAAAI8AAACQAAAAkQAAAJIAAACT AAAAlAAAAJUAAACWAAAAlwAAAJgAAACZAAAAmgAAAJsAAACcAAAAnQAAAJ4AAACfAAAAoAAAAKEA AACiAAAAowAAAKQAAAClAAAApgAAAKcAAACoAAAAqQAAAKoAAACrAAAArAAAAK0AAACuAAAArwAA ALAAAACxAAAAsgAAALMAAAC0AAAAtQAAALYAAAC3AAAAuAAAALkAAAC6AAAAuwAAALwAAAC9AAAA vgAAAL8AAADAAAAAwQAAAMIAAADDAAAAxAAAAMUAAADGAAAAxwAAAMgAAADJAAAAygAAAMsAAADM AAAAzQAAAM4AAADPAAAA0AAAANEAAADSAAAA0wAAANQAAADVAAAA1gAAANcAAADYAAAA2QAAANoA AADbAAAA3AAAAN0AAADeAAAA3wAAAOAAAADhAAAA4gAAAOMAAADkAAAA5QAAAOYAAADnAAAA6AAA AOkAAADqAAAA6wAAAOwAAADtAAAA7gAAAO8AAADwAAAA8QAAAPIAAADzAAAA9AAAAPUAAAD2AAAA 9wAAAPgAAAD5AAAA+gAAAPsAAAD8AAAA/QAAAP4AAAD/AAAAAAEAAAEBAAACAQAAAwEAAAQBAAAF AQAABgEAAAcBAAAIAQAACQEAAAoBAAALAQAADAEAAA0BAAAOAQAADwEAABABAAARAQAAEgEAABMB AAAUAQAAFQEAABYBAAAXAQAAGAEAABkBAAAaAQAA/v////////////////////////////////// ////////////JAEAACUBAAAmAQAAJwEAACgBAAApAQAAKgEAAP7////9/////f////3///8vAQAA /v////7////+//////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// ////////////////////////////////////////////////////UgBvAG8AdAAgAEUAbgB0AHIA eQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABYABQH///////// /wMAAAAGCQIAAAAAAMAAAAAAAABGAAAAAAAAAAAAAAAAAAdBPQF1ywExAQAAAAIAAAAAAABEAGEA dABhAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAACgACAf///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAGwA AAAAEAAAAAAAADEAVABhAGIAbABlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAOAAIBAQAAAAYAAAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAdAAAAOZNAQAAAAAAVwBvAHIAZABEAG8AYwB1AG0AZQBuAHQAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABoAAgECAAAABQAAAP////8AAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAANNYAAAAAAAAFAFMAdQBtAG0AYQByAHkASQBu AGYAbwByAG0AYQB0AGkAbwBuAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAKAACAf////////// /////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIAAABUAQAAAAAAAAUARABv AGMAdQBtAGUAbgB0AFMAdQBtAG0AYQByAHkASQBuAGYAbwByAG0AYQB0AGkAbwBuAAAAAAAAAAAA AAA4AAIBBAAAAP//////////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIwEA AAAQAAAAAAAAAQBDAG8AbQBwAE8AYgBqAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAABIAAgD///////////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAcQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP///////////////wAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEAAAD+////AwAAAAQAAAAFAAAA BgAAAAcAAAD+//////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// ////////////////////////////////////////////////AQD+/wMKAAD/////BgkCAAAAAADA AAAAAAAARh8AAABNaWNyb3NvZnQgT2ZmaWNlIFdvcmQgRG9jdW1lbnQACgAAAE1TV29yZERvYwAQ AAAAV29yZC5Eb2N1bWVudC44APQ5snEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD+/wAABgAC AAAAAAAAAAAAAAAAAAAAAAABAAAA4IWf8vlPaBCrkQgAKyez2TAAAAAkAQAADQAAAAEAAACAAAAA AwAAAIgAAAAFAAAAlAAAAAcAAACgAAAACQAAALQAAAASAAAAwAAAAAoAAADgAAAADAAAAOwAAAAN AAAA+AAAAA4AAAAEAQAADwAAAAwBAAAQAAAAFAEAABMAAAAcAQAAAAAAAAAAAAAAAAAAAAAAAAIA AADkBAAAHgAAAAQAAAAAAAAAHgAAAAQAAAAAAAAAHgAAAAwAAABOb3JtYWwuZG90AAAeAAAABAAA ADMAAAAeAAAAGAAAAE1pY3Jvc29mdCBPZmZpY2UgV29yZAAAAEAAAAAAXtCyAAAAAEAAAAAAnIvF /XTLAUAAAAAA5hb8AHXLAQMAAAABAAAAAwAAAK4SAAADAAAAemoAAAMAAAAAAAAAAwAAAAEAAAAD AAAArhIAAAMAAAB6agAAAwAAAAAAAABAAAAAAF7QsgAAAAA= --_002_DEAE495523C8F140A875D22C7C59D31902FC0921ESESSCMS0356eem_-- From mknappe@juniper.net Tue Oct 26 10:42:16 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id EFBFF3A69CB for ; Tue, 26 Oct 2010 10:42:15 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -106.092 X-Spam-Level: X-Spam-Status: No, score=-106.092 tagged_above=-999 required=5 tests=[AWL=0.507, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id YGHomg6ganqq for ; Tue, 26 Oct 2010 10:42:15 -0700 (PDT) Received: from exprod7og122.obsmtp.com (exprod7og122.obsmtp.com [64.18.2.22]) by core3.amsl.com (Postfix) with ESMTP id A16CB3A69C1 for ; Tue, 26 Oct 2010 10:42:14 -0700 (PDT) Received: from source ([66.129.224.36]) (using TLSv1) by exprod7ob122.postini.com ([64.18.6.12]) with SMTP ID DSNKTMcTWeyYE6O1LyW3uxFInoczF5NBEKQI@postini.com; Tue, 26 Oct 2010 10:44:02 PDT Received: from EMBX02-HQ.jnpr.net ([fe80::18fe:d666:b43e:f97e]) by P-EMHUB02-HQ.jnpr.net ([fe80::88f9:77fd:dfc:4d51%11]) with mapi; Tue, 26 Oct 2010 10:40:15 -0700 From: Michael Knappe To: Christian Hoene , "roman@telurix.com" , "mlubo@dolby.com" , "jan.skoglund@gipscorp.com" , "Stephan@vidyo.com" Date: Tue, 26 Oct 2010 10:40:13 -0700 Thread-Topic: [codec] semi-formal audio quality testing discussion Thread-Index: Act1NNfUnGtw+BAPkE+VRXlFk6jxVA== Message-ID: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-Entourage/13.3.0.091002 acceptlanguage: en-US Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Cc: "codec@ietf.org" Subject: [codec] semi-formal audio quality testing discussion X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Oct 2010 17:42:16 -0000 Those on the To: list have expressed interest in performing semi-formal aud= io subjective testing on the candidate codec. If possible, let=92s have a c= onference call sometime next week so we can begin discussion of this testin= g prior to the Beijing meeting. Let me know what times will work for you. A= nd of course, anyone on the codec list interested in joining in, please let= me know. Some items to discuss: * choosing representative source material (languages, music genres, etc) * orthogonality of test cases (bit rates / internal sample rates, packet= loss conditions, etc) * representative packet loss scenarios * subjective pool size * how often to test * etc etc Cheers, Mike From mknappe@juniper.net Tue Oct 26 10:43:30 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id AC0443A69E2 for ; Tue, 26 Oct 2010 10:43:30 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -106.115 X-Spam-Level: X-Spam-Status: No, score=-106.115 tagged_above=-999 required=5 tests=[AWL=0.484, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id AMJIAQz9GjG0 for ; Tue, 26 Oct 2010 10:43:29 -0700 (PDT) Received: from exprod7og103.obsmtp.com (exprod7og103.obsmtp.com [64.18.2.159]) by core3.amsl.com (Postfix) with ESMTP id B1BBB3A69E1 for ; Tue, 26 Oct 2010 10:43:29 -0700 (PDT) Received: from source ([66.129.224.36]) (using TLSv1) by exprod7ob103.postini.com ([64.18.6.12]) with SMTP ID DSNKTMcTq8gH/wrGZomEoVRdRbEiQkPPTLRV@postini.com; Tue, 26 Oct 2010 10:45:17 PDT Received: from EMBX02-HQ.jnpr.net ([fe80::18fe:d666:b43e:f97e]) by P-EMHUB01-HQ.jnpr.net ([fe80::fc92:eb1:759:2c72%11]) with mapi; Tue, 26 Oct 2010 10:44:48 -0700 From: Michael Knappe To: "codec@ietf.org" Date: Tue, 26 Oct 2010 10:44:45 -0700 Thread-Topic: [codec] Monday codec WG dinner and drinks in Beijing Thread-Index: Act1NXn0Vwr7U2naN0S0Dsc9MIgShA== Message-ID: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-Entourage/13.3.0.091002 acceptlanguage: en-US Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: [codec] Monday codec WG dinner and drinks in Beijing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Oct 2010 17:43:30 -0000 Hi everyone! I would like to invite everyone as an informal activity to joi= n me after the Monday technical plenary at a nearby restaurant for dinner a= nd drinks. Email back if you=92re interested so I can tally a rough number = for reservations. Suggestions for the restaurant are welcome, and I=92ll se= nd out logistics for meeting up after the plenary once the restaurant is de= cided. It=92s a great opportunity to meet and chat with fellow codec WG participan= ts in an informal setting, hope to see you there! Cheers, Mike From jean-marc.valin@octasic.com Tue Oct 26 11:26:44 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id E4C9C3A6958 for ; Tue, 26 Oct 2010 11:26:44 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.481 X-Spam-Level: X-Spam-Status: No, score=-2.481 tagged_above=-999 required=5 tests=[AWL=0.118, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iw9ZCZzy4o6z for ; Tue, 26 Oct 2010 11:26:43 -0700 (PDT) Received: from toroondcbmts05-srv.bellnexxia.net (toroondcbmts05.bellnexxia.net [207.236.237.39]) by core3.amsl.com (Postfix) with ESMTP id C53BD3A696B for ; Tue, 26 Oct 2010 11:26:42 -0700 (PDT) Received: from toip58-bus.srvr.bell.ca ([67.69.240.185]) by toroondcbmts05-srv.bellnexxia.net (InterMail vM.8.00.01.00 201-2244-105-20090324) with ESMTP id <20101026182830.BRZK28711.toroondcbmts05-srv.bellnexxia.net@toip58-bus.srvr.bell.ca>; Tue, 26 Oct 2010 14:28:30 -0400 Received: from toip40-bus.srvr.bell.ca ([67.69.240.41]) by toip58-bus.srvr.bell.ca with ESMTP; 26 Oct 2010 14:28:15 -0400 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: ApsEAAu3xkxGNv5q/2dsb2JhbACiP8EQhUgEjVsG Received: from mail.octasic.com (HELO MAILEXCH.octasic.com) ([70.54.254.106]) by toip40-bus.srvr.bell.ca with ESMTP; 26 Oct 2010 14:28:15 -0400 Received: from [10.100.60.27] ([10.100.60.27]) by MAILEXCH.octasic.com with Microsoft SMTPSVC(6.0.3790.4675); Tue, 26 Oct 2010 14:22:50 -0400 Message-ID: <4CC71C7A.3090002@octasic.com> Date: Tue, 26 Oct 2010 14:22:50 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.14) Gecko/20101006 Lightning/1.0b1 Thunderbird/3.0.9 MIME-Version: 1.0 To: "Raymond (Juin-Hwey) Chen" References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC37F42.2040103@usherbrooke.ca> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 26 Oct 2010 18:22:50.0820 (UTC) FILETIME=[CC695840:01CB753A] Cc: "'codec@ietf.org'" , Jean-Marc Valin Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Oct 2010 18:26:45 -0000 Hi Raymond, I've been able to reproduce results that are mostly similar to what you got earlier on the trumpet file. Unfortunately, it doesn't work as well on other files I'm trying. For example on this one (http://jmvalin.ca/misc_stuff/s02a_102.sw) there is simply too much pitch doubling and I'm not actually sure it's possible to do a good job. Could you try with your setup and see if you can improve things. I think one of the tricky thing with the postfilter will be to make it work in all circumstances and -- most importantly -- prevent it from causing harm in cases where it cannot help. Regarding the periods we'd need to support, I think the important question is not that much "on what range does it help", but "on what range is it necessary". From what I observed, CELT copes well with regular male and even female speech and even the vega sample tends to sound OK. It's only on strongly tonal samples with higher pitches that problems occur. The main reason I want to limit the maximum period allowed is that there's a memory footprint issue that's directly related. In any case, we'll need to experiment as with anything else. Cheers, Jean-Marc On 10-10-26 02:40 AM, Raymond (Juin-Hwey) Chen wrote: > Hi Jean-Marc, > > See my in-line comments below. > > You wrote: >> My concern wasn't that much about the period, but the gain. I'll >> probably have to try both approaches, but I was mainly concerned that >> a non-harmonic signal with one strong sinusoid could cause the comb >> filter to be turned on with a high gain. > > [Raymond]: I see your concern with introducing artificial harmonics for > a single sinusoid. On the other hand, my concern is that if you derive > the gain (or taps) based on partially whitened input signal, the gain or > taps may be too small for legitimate periodic music or speech signals. > To address both your and my concerns, I think it is probably best to use > a 3-tap filter where the taps are derived directly from the input signal > without whitening. Since a 3-tap filter has the ability to taper off > the degree of comb filtering with increasing frequency, a set of 3 > filter taps that is chosen to match the input spectrum can probably have > a reasonable spectral match for either a single tone or music/speech > with a harmonic structure. The BroadVoice codecs use a 3-tap pitch > predictor. The way those 3 taps are jointly searched and vector > quantized might provide a way to quantize the 3 taps and do a spectral > match at the same time for our current pitch prefilter/postfilter > approa ch. We need to experiment to find out. The BroadVoice codebooks > for the 3 taps were designed using mainly speech. If we adopt this > approach, we need to redesign the codebook with sufficient music signals > in the training set. If you are particularly concerned with a single > sinusoid, we may want to train one or a few 3-tap codevectors using just > sinusoids. If this doesn't work well, in the worst case, we could > potentially design a detector that detect (nearly) periodic time-domain > waveform that does not have a spectrum with a harmonic structure (like a > single sinusoid) and turn off the prefilter/postfilter accordingly. > >> CELT currently has a pitch predictor used for PLC, but I doubt it's >> very good for this as it is now. OTOH, it's fairly low complexity too. >> It's part of my plans to compare it with the one in BV16/32. I assume >> that Broadcom would be willing to relicense it to the simplified BSD >> license (it's currently LGPL) if we wanted to use it instead of the >> current CELT estimator. > > [Raymond]: Broadcom's legal counsel will formally answer that question > soon. My assumption is that if we were previously willing to allow > royalty-free use of BroadVoice in the IETF codec, then allowing certain > components of it to be used in the IETF codec under IETF's licensing > terms is probably not a problem, but this is just my personal thought > and not an official Broadcom position yet. I will talk to our lawyer > and let him give the formal answer. > >> The pitch range is another thing that needs experimenting with, but >> my intuition here is that it's probably not necessary to go below 100 >> Hz because for these very low pitches, there is enough masking between >> the harmonics to make the coding noise inaudible even without a >> post-filter. Again worth testing. > > [Raymond]: I agree that lower pitch frequency generally has better noise > masking, so time-domain predictive codecs often sound noisier for female > speech than for male speech. However, my previous work with pitch > postfiltering during my Ph.D. study showed that pitch postfiltering > alone could still give audible quality improvement to low-bit-rate-coded > male speech. Does it improve quality when the pitch frequency is below > 100 Hz? I don't remember the pitch frequencies of my pitch postfilter > tests. I agree it is worth testing. > > > _______________________________________________ codec mailing list > codec@ietf.org https://www.ietf.org/mailman/listinfo/codec From stewe@stewe.org Tue Oct 26 11:56:26 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 239273A68F5 for ; Tue, 26 Oct 2010 11:56:26 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.893 X-Spam-Level: X-Spam-Status: No, score=-1.893 tagged_above=-999 required=5 tests=[AWL=0.706, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RPEClrfOrQz3 for ; Tue, 26 Oct 2010 11:56:25 -0700 (PDT) Received: from stewe.org (stewe.org [85.214.122.234]) by core3.amsl.com (Postfix) with ESMTP id A50883A6827 for ; Tue, 26 Oct 2010 11:56:24 -0700 (PDT) Received: from [192.168.1.108] (unverified [24.5.132.232]) by stewe.org (SurgeMail 3.9e) with ESMTP id 5919-1743317 for ; Tue, 26 Oct 2010 20:58:09 +0200 User-Agent: Microsoft-MacOutlook/14.0.0.100825 Date: Tue, 26 Oct 2010 11:57:35 -0700 From: Stephan Wenger To: Message-ID: Thread-Topic: [codec] Monday codec WG dinner and drinks in Beijing In-Reply-To: Mime-version: 1.0 Content-type: text/plain; charset="ISO-8859-1" Content-transfer-encoding: quoted-printable X-Originating-IP: 24.5.132.232 X-Authenticated-User: stewe@stewe.org X-ORBS-Stamp: Your IP (24.5.132.232) was found in the spamhaus database. http://www.spamhaus.net Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Oct 2010 18:56:26 -0000 I'll not be in China, but I wish you good luck and good progress for the meeting. Stephan On 10.26.2010 10:44 , "Michael Knappe" wrote: >Hi everyone! I would like to invite everyone as an informal activity to >join me after the Monday technical plenary at a nearby restaurant for >dinner and drinks. Email back if you=B9re interested so I can tally a rough >number for reservations. Suggestions for the restaurant are welcome, and >I=B9ll send out logistics for meeting up after the plenary once the >restaurant is decided. > >It=B9s a great opportunity to meet and chat with fellow codec WG >participants in an informal setting, hope to see you there! > >Cheers, > >Mike > > > >_______________________________________________ >codec mailing list >codec@ietf.org >https://www.ietf.org/mailman/listinfo/codec From mknappe@juniper.net Tue Oct 26 12:11:01 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 534BF3A6949 for ; Tue, 26 Oct 2010 12:11:00 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -106.135 X-Spam-Level: X-Spam-Status: No, score=-106.135 tagged_above=-999 required=5 tests=[AWL=0.464, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id PvRdS5ibRzI6 for ; Tue, 26 Oct 2010 12:10:42 -0700 (PDT) Received: from exprod7og108.obsmtp.com (exprod7og108.obsmtp.com [64.18.2.169]) by core3.amsl.com (Postfix) with ESMTP id 103683A6946 for ; Tue, 26 Oct 2010 12:10:39 -0700 (PDT) Received: from source ([66.129.224.36]) (using TLSv1) by exprod7ob108.postini.com ([64.18.6.12]) with SMTP ID DSNKTMcoE2E/R0f9CWebFp2nG/q5KsWbaPaW@postini.com; Tue, 26 Oct 2010 12:12:28 PDT Received: from EMBX02-HQ.jnpr.net ([fe80::18fe:d666:b43e:f97e]) by P-EMHUB02-HQ.jnpr.net ([fe80::88f9:77fd:dfc:4d51%11]) with mapi; Tue, 26 Oct 2010 12:12:11 -0700 From: Michael Knappe To: "stewe@stewe.org" , "codec@ietf.org" Date: Tue, 26 Oct 2010 12:12:10 -0700 Thread-Topic: [codec] Monday codec WG dinner and drinks in Beijing Thread-Index: Act1P97zhqCX484/RUm845ZQGyN1vQAAdCRl Message-ID: <05542EC42316164383B5180707A489EE1D6D8FBC90@EMBX02-HQ.jnpr.net> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Oct 2010 19:11:02 -0000 Stephan, This brings up a good point overall - who is planning to be at IETF79? Mike ----- Original Message ----- From: codec-bounces@ietf.org To: codec@ietf.org Sent: Tue Oct 26 14:57:35 2010=0A= Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing I'll not be in China, but I wish you good luck and good progress for the meeting. Stephan On 10.26.2010 10:44 , "Michael Knappe" wrote: >Hi everyone! I would like to invite everyone as an informal activity to >join me after the Monday technical plenary at a nearby restaurant for >dinner and drinks. Email back if you=B9re interested so I can tally a roug= h >number for reservations. Suggestions for the restaurant are welcome, and >I=B9ll send out logistics for meeting up after the plenary once the >restaurant is decided. > >It=B9s a great opportunity to meet and chat with fellow codec WG >participants in an informal setting, hope to see you there! > >Cheers, > >Mike > > > >_______________________________________________ >codec mailing list >codec@ietf.org >https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec From jean-marc.valin@octasic.com Tue Oct 26 12:14:37 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 199F03A6951 for ; Tue, 26 Oct 2010 12:14:37 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.49 X-Spam-Level: X-Spam-Status: No, score=-2.49 tagged_above=-999 required=5 tests=[AWL=0.109, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6+UD2TB9u1FO for ; Tue, 26 Oct 2010 12:14:35 -0700 (PDT) Received: from toroondcbmts04-srv.bellnexxia.net (toroondcbmts04.bellnexxia.net [207.236.237.38]) by core3.amsl.com (Postfix) with ESMTP id 859273A6946 for ; Tue, 26 Oct 2010 12:14:35 -0700 (PDT) Received: from toip58-bus.srvr.bell.ca ([67.69.240.185]) by toroondcbmts04-srv.bellnexxia.net (InterMail vM.8.00.01.00 201-2244-105-20090324) with ESMTP id <20101026191622.BCIL27003.toroondcbmts04-srv.bellnexxia.net@toip58-bus.srvr.bell.ca>; Tue, 26 Oct 2010 15:16:22 -0400 Received: from toip40-bus.srvr.bell.ca ([67.69.240.41]) by toip58-bus.srvr.bell.ca with ESMTP; 26 Oct 2010 15:16:11 -0400 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: ApsEABvFxkxGNv5q/2dsb2JhbACiQMFXhUgEjVsG Received: from mail.octasic.com (HELO MAILEXCH.octasic.com) ([70.54.254.106]) by toip40-bus.srvr.bell.ca with ESMTP; 26 Oct 2010 15:16:11 -0400 Received: from [10.100.60.27] ([10.100.60.27]) by MAILEXCH.octasic.com with Microsoft SMTPSVC(6.0.3790.4675); Tue, 26 Oct 2010 15:14:13 -0400 Message-ID: <4CC72885.4060903@octasic.com> Date: Tue, 26 Oct 2010 15:14:13 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.14) Gecko/20101006 Lightning/1.0b1 Thunderbird/3.0.9 MIME-Version: 1.0 To: Michael Knappe References: <05542EC42316164383B5180707A489EE1D6D8FBC90@EMBX02-HQ.jnpr.net> In-Reply-To: <05542EC42316164383B5180707A489EE1D6D8FBC90@EMBX02-HQ.jnpr.net> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit X-OriginalArrivalTime: 26 Oct 2010 19:14:13.0310 (UTC) FILETIME=[F9B7FDE0:01CB7541] Cc: "codec@ietf.org" Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Oct 2010 19:14:37 -0000 On 10-10-26 03:12 PM, Michael Knappe wrote: > Stephan, > > This brings up a good point overall - who is planning to be at IETF79? I am. Jean-Marc > Mike > > ----- Original Message ----- > From: codec-bounces@ietf.org > To: codec@ietf.org > Sent: Tue Oct 26 14:57:35 2010 > Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing > > I'll not be in China, but I wish you good luck and good progress for the > meeting. > Stephan > > > On 10.26.2010 10:44 , "Michael Knappe" wrote: > >> Hi everyone! I would like to invite everyone as an informal activity to >> join me after the Monday technical plenary at a nearby restaurant for >> dinner and drinks. Email back if you¹re interested so I can tally a rough >> number for reservations. Suggestions for the restaurant are welcome, and >> I¹ll send out logistics for meeting up after the plenary once the >> restaurant is decided. >> >> It¹s a great opportunity to meet and chat with fellow codec WG >> participants in an informal setting, hope to see you there! >> >> Cheers, >> >> Mike >> >> >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From stephen.botzko@gmail.com Tue Oct 26 12:17:11 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 2EAB63A6946 for ; Tue, 26 Oct 2010 12:17:10 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.686 X-Spam-Level: X-Spam-Status: No, score=-2.686 tagged_above=-999 required=5 tests=[AWL=-0.088, BAYES_00=-2.599, HTML_MESSAGE=0.001] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id An-1eZSAJ12g for ; Tue, 26 Oct 2010 12:16:59 -0700 (PDT) Received: from mail-bw0-f44.google.com (mail-bw0-f44.google.com [209.85.214.44]) by core3.amsl.com (Postfix) with ESMTP id 255013A6827 for ; Tue, 26 Oct 2010 12:16:55 -0700 (PDT) Received: by bwz12 with SMTP id 12so4864618bwz.31 for ; Tue, 26 Oct 2010 12:18:43 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type; bh=9lbGnhwaFdZBHLtF3KRimfB6BTuQOBMZtfGlDyJVUxY=; b=uP1qVLA/lZ6Ts108veWQ1JjT5FdVTjtriE/tTnQ3/vor0mUNkuF7rwOxJqOia/Tg2b WRdP8qHUt+zlurcJoGod67grVmmTJYMnYFVKMvm+l3SeAkkNy2o5RENWZbtsNDXFHa6q FXArZE+c5MjTiygqSmpEes+Bo2oMY4mlmruGU= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=d5oH/Naxkxh9wElcx13qaMrSP9pa8GZwT1wnv+2Z1dAhG0w5JQ3HetRC3MJ+ZPvaEU gVpjAtNLCHz8PO6MlEYz1tu+3pGS9Ltk7/OBpzQPMSptYhoFGWfeWcHsUlegfPN1q0CU eG0s/jhDgmbko/liIatzCHF+M90zWrXb9UJR4= MIME-Version: 1.0 Received: by 10.204.2.75 with SMTP id 11mr3757816bki.63.1288120722890; Tue, 26 Oct 2010 12:18:42 -0700 (PDT) Received: by 10.204.80.213 with HTTP; Tue, 26 Oct 2010 12:18:42 -0700 (PDT) In-Reply-To: <05542EC42316164383B5180707A489EE1D6D8FBC90@EMBX02-HQ.jnpr.net> References: <05542EC42316164383B5180707A489EE1D6D8FBC90@EMBX02-HQ.jnpr.net> Date: Tue, 26 Oct 2010 15:18:42 -0400 Message-ID: From: Stephen Botzko To: Michael Knappe Content-Type: multipart/alternative; boundary=0015174c33ce1da345049389face Cc: "codec@ietf.org" Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Oct 2010 19:17:11 -0000 --0015174c33ce1da345049389face Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable I will be there, and would be happy to join you on Monday evening. Stephen Botzko On Tue, Oct 26, 2010 at 3:12 PM, Michael Knappe wrote= : > Stephan, > > This brings up a good point overall - who is planning to be at IETF79? > > Mike > > ----- Original Message ----- > From: codec-bounces@ietf.org > To: codec@ietf.org > Sent: Tue Oct 26 14:57:35 2010 > Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing > > I'll not be in China, but I wish you good luck and good progress for the > meeting. > Stephan > > > On 10.26.2010 10:44 , "Michael Knappe" wrote: > > >Hi everyone! I would like to invite everyone as an informal activity to > >join me after the Monday technical plenary at a nearby restaurant for > >dinner and drinks. Email back if you=B9re interested so I can tally a ro= ugh > >number for reservations. Suggestions for the restaurant are welcome, and > >I=B9ll send out logistics for meeting up after the plenary once the > >restaurant is decided. > > > >It=B9s a great opportunity to meet and chat with fellow codec WG > >participants in an informal setting, hope to see you there! > > > >Cheers, > > > >Mike > > > > > > > >_______________________________________________ > >codec mailing list > >codec@ietf.org > >https://www.ietf.org/mailman/listinfo/codec > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > --0015174c33ce1da345049389face Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable I will be there, and would be happy to join you on Monday evening.

S= tephen Botzko

On Tue, Oct 26, 2010 at 3:1= 2 PM, Michael Knappe <mknappe@juniper.net> wrote:
Stephan,

This brings up a good point overall - who is planning to be at IETF79?

Mike

----- Original Message -----
From: codec-bounces@ietf.org = <codec-bounces@ietf.org>= ;
To: codec@ietf.org <codec@ietf.org>
Sent: Tue Oct 26 14:57:35 2010
Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing

I'll not be in China, but I wish you good luck and good progress for th= e
meeting.
Stephan


On 10.26.2010 10:44 , "Michael Knappe" <mknappe@juniper.net> wrote:

>Hi everyone! I would like to invite everyone as an informal activity to=
>join me after the Monday technical plenary at a nearby restaurant for >dinner and drinks. Email back if you=B9re interested so I can tally a r= ough
>number for reservations. Suggestions for the restaurant are welcome, an= d
>I=B9ll send out logistics for meeting up after the plenary once the
>restaurant is decided.
>
>It=B9s a great opportunity to meet and chat with fellow codec WG
>participants in an informal setting, hope to see you there!
>
>Cheers,
>
>Mike
>
>
>
>_______________________________________________
>codec mailing list
>codec@ietf.org
>https://www.ietf.org/mailman/listinfo/codec


_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec
_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec

--0015174c33ce1da345049389face-- From gmaxwell@juniper.net Tue Oct 26 12:57:02 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 6F0C93A679C for ; Tue, 26 Oct 2010 12:57:02 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.304 X-Spam-Level: X-Spam-Status: No, score=-6.304 tagged_above=-999 required=5 tests=[AWL=0.295, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id G3Ta7TBlQ8X6 for ; Tue, 26 Oct 2010 12:57:01 -0700 (PDT) Received: from exprod7og112.obsmtp.com (exprod7og112.obsmtp.com [64.18.2.177]) by core3.amsl.com (Postfix) with ESMTP id 688223A6405 for ; Tue, 26 Oct 2010 12:56:50 -0700 (PDT) Received: from source ([66.129.224.36]) (using TLSv1) by exprod7ob112.postini.com ([64.18.6.12]) with SMTP ID DSNKTMcy6lYIZA0YRdF/jfgYI4sc0eC5lA70@postini.com; Tue, 26 Oct 2010 12:58:49 PDT Received: from EMBX01-HQ.jnpr.net ([fe80::c821:7c81:f21f:8bc7]) by P-EMHUB02-HQ.jnpr.net ([fe80::88f9:77fd:dfc:4d51%11]) with mapi; Tue, 26 Oct 2010 12:57:37 -0700 From: Gregory Maxwell To: Jean-Marc Valin , "Raymond (Juin-Hwey) Chen" Date: Tue, 26 Oct 2010 12:57:37 -0700 Thread-Topic: [codec] A pitch filtering method to improve the mode 1 CELT codec Thread-Index: Act1O5p5xKnr4lpER5ao8HT966KdVAAC1ga1 Message-ID: References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC37F42.2040103@usherbrooke.ca> , <4CC71C7A.3090002@octasic.com> In-Reply-To: <4CC71C7A.3090002@octasic.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Cc: "'codec@ietf.org'" , Jean-Marc Valin Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Oct 2010 19:57:02 -0000 Jean-Marc Valin [jean-marc.valin@octasic.com] wrote: > Regarding the periods we'd need to support, I think the important questio= n > is not that much "on what range does it help", but "on what range is it > necessary". From what I observed, CELT copes well with regular male and > even female speech and even the vega sample tends to sound OK. It's only = on You should be mindful of the low delay modes. At moderate rates in the 5ms = mode, for example. I would expect the need for this to be stronger in these cases, so the bits= tream should support them. I think we're currently at an order of magnitude less memory than anyone ha= s suggested was a requirement on the list=97 and decoders will typically wa= nt a fair amount of memory for PLC purposes=97 so I think we have room to b= reath on the design right now and can worry about putting it on a diet afte= r the technique's usefulness has been demonstrated in an implementation. = From hoene@uni-tuebingen.de Tue Oct 26 18:58:14 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 8EDBA3A68F1 for ; Tue, 26 Oct 2010 18:58:14 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.248 X-Spam-Level: X-Spam-Status: No, score=-6.248 tagged_above=-999 required=5 tests=[AWL=-0.000, BAYES_00=-2.599, HELO_EQ_DE=0.35, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id GgQ-pob26drg for ; Tue, 26 Oct 2010 18:58:13 -0700 (PDT) Received: from mx06.uni-tuebingen.de (mx06.uni-tuebingen.de [134.2.3.3]) by core3.amsl.com (Postfix) with ESMTP id AEC7A3A6825 for ; Tue, 26 Oct 2010 18:58:12 -0700 (PDT) Received: from hoeneT60 (dslb-094-216-253-211.pools.arcor-ip.net [94.216.253.211]) (authenticated bits=0) by mx06.uni-tuebingen.de (8.13.6/8.13.6) with ESMTP id o9R1xqWE016501 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Wed, 27 Oct 2010 03:59:58 +0200 From: "Christian Hoene" To: "'Stephen Botzko'" , "'Michael Knappe'" References: <05542EC42316164383B5180707A489EE1D6D8FBC90@EMBX02-HQ.jnpr.net> In-Reply-To: Date: Wed, 27 Oct 2010 03:59:53 +0200 Organization: =?UTF-8?Q?Universit=C3=A4t_T=C3=BCbingen?= Message-ID: <000301cb757a$a9916890$fcb439b0$@de> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0004_01CB758B.6D1A3890" X-Mailer: Microsoft Office Outlook 12.0 Thread-Index: Act1QrLn2GkY6POESEmVzOaT1fNS3AAN+fUw Content-Language: de X-AntiVirus: NOT checked by Avira MailGate (version: 3.0.0-4; host: mx06) Cc: codec@ietf.org Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Oct 2010 01:58:14 -0000 This is a multi-part message in MIME format. ------=_NextPart_000_0004_01CB758B.6D1A3890 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Me, too. =20 --------------------------------------------------------------- Dr.-Ing. Christian Hoene Interactive Communication Systems (ICS), University of T=C3=BCbingen=20 Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532=20 http://www.net.uni-tuebingen.de/ =20 From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf = Of Stephen Botzko Sent: Tuesday, October 26, 2010 9:19 PM To: Michael Knappe Cc: codec@ietf.org Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing =20 I will be there, and would be happy to join you on Monday evening. Stephen Botzko On Tue, Oct 26, 2010 at 3:12 PM, Michael Knappe = wrote: Stephan, This brings up a good point overall - who is planning to be at IETF79? Mike ----- Original Message ----- From: codec-bounces@ietf.org To: codec@ietf.org Sent: Tue Oct 26 14:57:35 2010 Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing I'll not be in China, but I wish you good luck and good progress for the meeting. Stephan On 10.26.2010 10:44 , "Michael Knappe" wrote: >Hi everyone! I would like to invite everyone as an informal activity to >join me after the Monday technical plenary at a nearby restaurant for >dinner and drinks. Email back if you=C2=B9re interested so I can tally = a rough >number for reservations. Suggestions for the restaurant are welcome, = and >I=C2=B9ll send out logistics for meeting up after the plenary once the >restaurant is decided. > >It=C2=B9s a great opportunity to meet and chat with fellow codec WG >participants in an informal setting, hope to see you there! > >Cheers, > >Mike > > > >_______________________________________________ >codec mailing list >codec@ietf.org >https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec =20 ------=_NextPart_000_0004_01CB758B.6D1A3890 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable

Me, too.

 

----------------------------------------------------------= -----

Dr.-Ing. Christian Hoene

Interactive Communication Systems (ICS), University of = T=C3=BCbingen

Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 = 2970532

http://www.net.uni-tuebingen.de= /

 

From:= codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of = Stephen Botzko
Sent: Tuesday, October 26, 2010 9:19 PM
To: Michael Knappe
Cc: codec@ietf.org
Subject: Re: [codec] Monday codec WG dinner and drinks in = Beijing

 

I will be there, and = would be happy to join you on Monday evening.

Stephen Botzko

On Tue, Oct 26, 2010 at 3:12 PM, Michael Knappe = <mknappe@juniper.net> = wrote:

Stephan,

This brings up a good point overall - who is planning to be at = IETF79?

Mike


----- Original Message -----
From: codec-bounces@ietf.org <codec-bounces@ietf.org>
= To: codec@ietf.org <codec@ietf.org>
Sent: Tue Oct 26 14:57:35 2010
Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing

I'll not be in China, but I wish you good luck and good progress for = the
meeting.
Stephan


On 10.26.2010 10:44 , "Michael Knappe" <mknappe@juniper.net> = wrote:

>Hi everyone! I would like to invite everyone as an informal activity = to
>join me after the Monday technical plenary at a nearby restaurant = for
>dinner and drinks. Email back if you=C2=B9re interested so I can = tally a rough
>number for reservations. Suggestions for the restaurant are welcome, = and
>I=C2=B9ll send out logistics for meeting up after the plenary once = the
>restaurant is decided.
>
>It=C2=B9s a great opportunity to meet and chat with fellow codec = WG
>participants in an informal setting, hope to see you there!
>
>Cheers,
>
>Mike
>
>
>
>_______________________________________________
>codec mailing list
>codec@ietf.org
>https://www.ietf.org/mailman/listinfo/codec


_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/mailman/listinfo/codec
_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/mailman/listinfo/codec

 

------=_NextPart_000_0004_01CB758B.6D1A3890-- From koen.vos@skype.net Tue Oct 26 20:15:18 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id E4D573A697D for ; Tue, 26 Oct 2010 20:15:17 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.394 X-Spam-Level: X-Spam-Status: No, score=-2.394 tagged_above=-999 required=5 tests=[AWL=0.204, BAYES_00=-2.599, HTML_MESSAGE=0.001] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id O4m9eYbpT7ge for ; Tue, 26 Oct 2010 20:15:15 -0700 (PDT) Received: from mx.skype.net (mx.skype.net [78.141.177.88]) by core3.amsl.com (Postfix) with ESMTP id BAAB93A6958 for ; Tue, 26 Oct 2010 20:15:14 -0700 (PDT) Received: from mx.skype.net (localhost [127.0.0.1]) by mx.skype.net (Postfix) with ESMTP id 1B0AF170B; Wed, 27 Oct 2010 05:17:02 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=skype.net; h=date:from:to :cc:message-id:in-reply-to:subject:mime-version:content-type; s= mx; bh=1HLKdLtewl4jfEZ9yyu8+ddGHCk=; b=IdOl0himhsluwffqexYE27zlJ 1utbWf26dlEWN8Xg3RfS1EVtMw6tUUIUuvYzXVENVQAPcay0yLX1VHeopEFFg7ig vxtdJUFTQ3IQKXuATEDUXiLys+FyDO8REi7xVzuOd5Mj5gmIgjGGJLAlOKMo/I6B jzSUSd4un6RJ2nN400= DomainKey-Signature: a=rsa-sha1; c=nofws; d=skype.net; h=date:from:to:cc :message-id:in-reply-to:subject:mime-version:content-type; q=dns ; s=mx; b=uGHWbFojHxGlFAse7+X6qzM6VqdBo8CaLZ1GS7D2pmeVwM1enp4Jjv gqtAp+Ijxr9FlEukNn+eUNPlb2cIVy4x9NB6G/UH9eVfYipfze1i4T5L45EVs3U5 wongFKOOr7EWvDukQHkW7kpEgeM7NLVMyD8yKapIIYI4Dl23lsXxw= Received: from zimbra.skype.net (zimbra.skype.net [78.141.177.82]) by mx.skype.net (Postfix) with ESMTP id 1953A16FF; Wed, 27 Oct 2010 05:17:02 +0200 (CEST) Received: from localhost (localhost [127.0.0.1]) by zimbra.skype.net (Postfix) with ESMTP id F3B8E3507679; Wed, 27 Oct 2010 05:17:01 +0200 (CEST) X-Virus-Scanned: amavisd-new at lu2-zimbra.skype.net Received: from zimbra.skype.net ([127.0.0.1]) by localhost (zimbra.skype.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ol7DlV+tbwrA; Wed, 27 Oct 2010 05:16:49 +0200 (CEST) Received: from zimbra.skype.net (lu2-zimbra.skype.net [78.141.177.82]) by zimbra.skype.net (Postfix) with ESMTP id 63DAF3507674; Wed, 27 Oct 2010 05:16:49 +0200 (CEST) Date: Wed, 27 Oct 2010 05:16:49 +0200 (CEST) From: Koen Vos To: Christian Hoene Message-ID: <247946901.1136671.1288149409289.JavaMail.root@lu2-zimbra> In-Reply-To: <000301cb757a$a9916890$fcb439b0$@de> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_1136670_559150139.1288149409287" X-Originating-IP: [69.181.192.115] X-Mailer: Zimbra 6.0.6_GA_2330.UBUNTU8_64 (ZimbraWebClient - FF3.0 (Win)/6.0.6_GA_2330.UBUNTU8_64) Cc: codec@ietf.org Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Oct 2010 03:15:18 -0000 ------=_Part_1136670_559150139.1288149409287 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable +1=20 koen=20 From: "Christian Hoene" =20 To: "Stephen Botzko" , "Michael Knappe" =20 Cc: codec@ietf.org=20 Sent: Tuesday, October 26, 2010 6:59:53 PM=20 Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing=20 Me, too.=20 ---------------------------------------------------------------=20 Dr.-Ing. Christian Hoene=20 Interactive Communication Systems (ICS), University of T=C3=BCbingen=20 Sand 13, 72076 T=C3=BCbingen, Germany, Phone +49 7071 2970532=20 http://www.net.uni-tuebingen.de/=20 From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of S= tephen Botzko=20 Sent: Tuesday, October 26, 2010 9:19 PM=20 To: Michael Knappe=20 Cc: codec@ietf.org=20 Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing=20 I will be there, and would be happy to join you on Monday evening.=20 Stephen Botzko=20 On Tue, Oct 26, 2010 at 3:12 PM, Michael Knappe < mknappe@juniper.net > wro= te:=20 Stephan,=20 This brings up a good point overall - who is planning to be at IETF79?=20 Mike=20 ----- Original Message -----=20 From: codec-bounces@ietf.org < codec-bounces@ietf.org >=20 To: codec@ietf.org < codec@ietf.org >=20 Sent: Tue Oct 26 14:57:35 2010=20 Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing=20 I'll not be in China, but I wish you good luck and good progress for the=20 meeting.=20 Stephan=20 On 10.26.2010 10:44 , "Michael Knappe" < mknappe@juniper.net > wrote:=20 >Hi everyone! I would like to invite everyone as an informal activity to=20 >join me after the Monday technical plenary at a nearby restaurant for=20 >dinner and drinks. Email back if you=C2=B9re interested so I can tally a r= ough=20 >number for reservations. Suggestions for the restaurant are welcome, and= =20 >I=C2=B9ll send out logistics for meeting up after the plenary once the=20 >restaurant is decided.=20 >=20 >It=C2=B9s a great opportunity to meet and chat with fellow codec WG=20 >participants in an informal setting, hope to see you there!=20 >=20 >Cheers,=20 >=20 >Mike=20 >=20 >=20 >=20 >_______________________________________________=20 >codec mailing list=20 > codec@ietf.org=20 > https://www.ietf.org/mailman/listinfo/codec=20 _______________________________________________=20 codec mailing list=20 codec@ietf.org=20 https://www.ietf.org/mailman/listinfo/codec=20 _______________________________________________=20 codec mailing list=20 codec@ietf.org=20 https://www.ietf.org/mailman/listinfo/codec=20 _______________________________________________=20 codec mailing list=20 codec@ietf.org=20 https://www.ietf.org/mailman/listinfo/codec=20 ------=_Part_1136670_559150139.1288149409287 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable <= div style=3D'font-family: Times New Roman; font-size: 12pt; color: #000000'= >+1
koen

From: "Christian Hoene" <hoene@uni-tuebin= gen.de>
To: "Stephen Botzko" <stephen.botzko@gmail.com>,= "Michael Knappe" <mknappe@juniper.net>
Cc: codec@ietf.org<= br>Sent: Tuesday, October 26, 2010 6:59:53 PM
Subject: Re:= [codec] Monday codec WG dinner and drinks in Beijing

Me, too.

 

-----------------------------= ----------------------------------

Dr.-Ing. Christian Hoene

Interactive Communication Sys= tems (ICS), University of T=C3=BCbingen

Sand 13, 72076 T=C3=BCbingen,= Germany, Phone +49 7071 2970532

http://www.net.uni-tuebingen.de/

 

From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of = Stephen Botzko
Sent: Tuesday, October 26, 2010 9:19 PM
To: Michael Knappe
Cc: codec@ietf.org
Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing

 

I will be there, and = would be happy to join you on Monday evening.

Stephen Botzko

On Tue, Oct 26, 2010 at 3:12 PM, Michael Knappe <= mknappe@juniper.ne= t> wrote:

Stephan,

This brings up a good point overall - who is planning to be at IETF79?

Mike


----- Original Message -----
From: codec-bou= nces@ietf.org <codec-bounces@ietf.org>
To: codec@ietf.org = <codec@ietf.org&= gt;
Sent: Tue Oct 26 14:57:35 2010
Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing

I'll not be in China, but I wish you good luck and good progress for the meeting.
Stephan


On 10.26.2010 10:44 , "Michael Knappe" <mknappe@juniper.net> wrote:

>Hi everyone! I would like to invite everyone as an informal activity to=
>join me after the Monday technical plenary at a nearby restaurant for >dinner and drinks. Email back if you=C2=B9re interested so I can tally = a rough
>number for reservations. Suggestions for the restaurant are welcome, an= d
>I=C2=B9ll send out logistics for meeting up after the plenary once the<= br> >restaurant is decided.
>
>It=C2=B9s a great opportunity to meet and chat with fellow codec WG
>participants in an informal setting, hope to see you there!
>
>Cheers,
>
>Mike
>
>
>
>_______________________________________________
>codec mailing list
>codec@ietf.org<= br> >https://www.ietf.org/mailman/listinfo/codec


_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec
_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec

 


_______________________________________________
codec mailing listcodec@ietf.org
https://www.ietf.org/mailman/listinfo/codec
------=_Part_1136670_559150139.1288149409287-- From rchen@broadcom.com Tue Oct 26 23:18:15 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id E20C13A69A6 for ; Tue, 26 Oct 2010 23:18:10 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.382 X-Spam-Level: X-Spam-Status: No, score=-2.382 tagged_above=-999 required=5 tests=[AWL=0.217, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id JbIjEpuKc1bg for ; Tue, 26 Oct 2010 23:17:55 -0700 (PDT) Received: from mms1.broadcom.com (mms1.broadcom.com [216.31.210.17]) by core3.amsl.com (Postfix) with ESMTP id 79D8C3A69B7 for ; Tue, 26 Oct 2010 23:16:31 -0700 (PDT) Received: from [10.9.200.133] by mms1.broadcom.com with ESMTP (Broadcom SMTP Relay (Email Firewall v6.3.2)); Tue, 26 Oct 2010 23:18:01 -0700 X-Server-Uuid: 02CED230-5797-4B57-9875-D5D2FEE4708A Received: from IRVEXCHCCR01.corp.ad.broadcom.com ([10.252.49.30]) by IRVEXCHHUB02.corp.ad.broadcom.com ([10.9.200.133]) with mapi; Tue, 26 Oct 2010 23:19:19 -0700 From: "Raymond (Juin-Hwey) Chen" To: "Jean-Marc Valin" Importance: low X-Priority: 5 Date: Tue, 26 Oct 2010 23:17:59 -0700 Thread-Topic: [codec] A pitch filtering method to improve the mode 1 CELT codec Thread-Index: Act1PgknOmY06GMaSIiB4lQYomuRUgAXlP9w Message-ID: References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC37F42.2040103@usherbrooke.ca> <4CC71C7A.3090002@octasic.com> In-Reply-To: <4CC71C7A.3090002@octasic.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US MIME-Version: 1.0 X-WSS-ID: 60D91B9347838039595-01-01 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Cc: "'codec@ietf.org'" , Jean-Marc Valin Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Oct 2010 06:18:19 -0000 Hi Jean-Marc, Are you saying the problem is in the pitch estimation for that test file ra= ther than the pitch filtering? =20 I will take a look at that file. Is that a raw file? I am not familiar wi= th the ".sw" file extension. I could hear audible improvement when I tried the method on the other solo = instrument files that I am working on, although the degree of improvement d= epends on which file it is, as expected. I am starting to wonder how many people in the codec WG email list still wa= nt to continue to see our low-level technical discussions about the algorit= hm development in the reflector. I wonder if some people in the codec WG e= mail list would prefer not to be bombarded with such emails. Should we con= sider moving such detailed codec algorithm development discussions off the = WG mailing list and just between those of us who are interested or are dire= ctly involved? Or is this how IETF supposed to work (discussing all technical development = in emails)? Thanks. Raymond -----Original Message----- From: Jean-Marc Valin [mailto:jean-marc.valin@octasic.com]=20 Sent: Tuesday, October 26, 2010 11:23 AM To: Raymond (Juin-Hwey) Chen Cc: Jean-Marc Valin; 'codec@ietf.org' Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT co= dec Hi Raymond, I've been able to reproduce results that are mostly similar to what you got= =20 earlier on the trumpet file. Unfortunately, it doesn't work as well on=20 other files I'm trying. For example on this one=20 (http://jmvalin.ca/misc_stuff/s02a_102.sw) there is simply too much pitch=20 doubling and I'm not actually sure it's possible to do a good job. Could=20 you try with your setup and see if you can improve things. I think one of=20 the tricky thing with the postfilter will be to make it work in all=20 circumstances and -- most importantly -- prevent it from causing harm in=20 cases where it cannot help. Regarding the periods we'd need to support, I think the important question= =20 is not that much "on what range does it help", but "on what range is it=20 necessary". From what I observed, CELT copes well with regular male and=20 even female speech and even the vega sample tends to sound OK. It's only on= =20 strongly tonal samples with higher pitches that problems occur. The main=20 reason I want to limit the maximum period allowed is that there's a memory= =20 footprint issue that's directly related. In any case, we'll need to=20 experiment as with anything else. Cheers, Jean-Marc On 10-10-26 02:40 AM, Raymond (Juin-Hwey) Chen wrote: > Hi Jean-Marc, > > See my in-line comments below. > > You wrote: >> My concern wasn't that much about the period, but the gain. I'll >> probably have to try both approaches, but I was mainly concerned that >> a non-harmonic signal with one strong sinusoid could cause the comb >> filter to be turned on with a high gain. > > [Raymond]: I see your concern with introducing artificial harmonics for > a single sinusoid. On the other hand, my concern is that if you derive > the gain (or taps) based on partially whitened input signal, the gain or > taps may be too small for legitimate periodic music or speech signals. > To address both your and my concerns, I think it is probably best to use > a 3-tap filter where the taps are derived directly from the input signal > without whitening. Since a 3-tap filter has the ability to taper off > the degree of comb filtering with increasing frequency, a set of 3 > filter taps that is chosen to match the input spectrum can probably have > a reasonable spectral match for either a single tone or music/speech > with a harmonic structure. The BroadVoice codecs use a 3-tap pitch > predictor. The way those 3 taps are jointly searched and vector > quantized might provide a way to quantize the 3 taps and do a spectral > match at the same time for our current pitch prefilter/postfilter > approa ch. We need to experiment to find out. The BroadVoice codebooks > for the 3 taps were designed using mainly speech. If we adopt this > approach, we need to redesign the codebook with sufficient music signals > in the training set. If you are particularly concerned with a single > sinusoid, we may want to train one or a few 3-tap codevectors using just > sinusoids. If this doesn't work well, in the worst case, we could > potentially design a detector that detect (nearly) periodic time-domain > waveform that does not have a spectrum with a harmonic structure (like a > single sinusoid) and turn off the prefilter/postfilter accordingly. > >> CELT currently has a pitch predictor used for PLC, but I doubt it's >> very good for this as it is now. OTOH, it's fairly low complexity too. >> It's part of my plans to compare it with the one in BV16/32. I assume >> that Broadcom would be willing to relicense it to the simplified BSD >> license (it's currently LGPL) if we wanted to use it instead of the >> current CELT estimator. > > [Raymond]: Broadcom's legal counsel will formally answer that question > soon. My assumption is that if we were previously willing to allow > royalty-free use of BroadVoice in the IETF codec, then allowing certain > components of it to be used in the IETF codec under IETF's licensing > terms is probably not a problem, but this is just my personal thought > and not an official Broadcom position yet. I will talk to our lawyer > and let him give the formal answer. > >> The pitch range is another thing that needs experimenting with, but >> my intuition here is that it's probably not necessary to go below 100 >> Hz because for these very low pitches, there is enough masking between >> the harmonics to make the coding noise inaudible even without a >> post-filter. Again worth testing. > > [Raymond]: I agree that lower pitch frequency generally has better noise > masking, so time-domain predictive codecs often sound noisier for female > speech than for male speech. However, my previous work with pitch > postfiltering during my Ph.D. study showed that pitch postfiltering > alone could still give audible quality improvement to low-bit-rate-coded > male speech. Does it improve quality when the pitch frequency is below > 100 Hz? I don't remember the pitch frequencies of my pitch postfilter > tests. I agree it is worth testing. > > > _______________________________________________ codec mailing list > codec@ietf.org https://www.ietf.org/mailman/listinfo/codec From erik.norvell@ericsson.com Tue Oct 26 23:37:24 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 027033A681F for ; Tue, 26 Oct 2010 23:37:24 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -5.622 X-Spam-Level: X-Spam-Status: No, score=-5.622 tagged_above=-999 required=5 tests=[AWL=0.976, BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 2QVw+PrtnhVM for ; Tue, 26 Oct 2010 23:37:21 -0700 (PDT) Received: from mailgw9.se.ericsson.net (mailgw9.se.ericsson.net [193.180.251.57]) by core3.amsl.com (Postfix) with ESMTP id 330FD3A69AD for ; Tue, 26 Oct 2010 23:37:21 -0700 (PDT) X-AuditID: c1b4fb39-b7b54ae000003464-32-4cc7c90c6c31 Received: from esessmw0256.eemea.ericsson.se (Unknown_Domain [153.88.253.125]) by mailgw9.se.ericsson.net (Symantec Mail Security) with SMTP id BB.63.13412.C09C7CC4; Wed, 27 Oct 2010 08:39:09 +0200 (CEST) Received: from ESESSCMS0351.eemea.ericsson.se ([169.254.1.175]) by esessmw0256.eemea.ericsson.se ([10.2.3.125]) with mapi; Wed, 27 Oct 2010 08:39:07 +0200 From: Erik Norvell To: Koen Vos , Christian Hoene Date: Wed, 27 Oct 2010 08:39:06 +0200 Thread-Topic: [codec] Monday codec WG dinner and drinks in Beijing Thread-Index: Act1hXJoxy5qnj0BReWNe8tThZHQsQAHAREg Message-ID: <027A93CE4A670242BD91A44E37105AEF0CACAE8C03@ESESSCMS0351.eemea.ericsson.se> References: <000301cb757a$a9916890$fcb439b0$@de> <247946901.1136671.1288149409289.JavaMail.root@lu2-zimbra> In-Reply-To: <247946901.1136671.1288149409289.JavaMail.root@lu2-zimbra> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US Content-Type: multipart/alternative; boundary="_000_027A93CE4A670242BD91A44E37105AEF0CACAE8C03ESESSCMS0351e_" MIME-Version: 1.0 X-Brightmail-Tracker: AAAAAA== Cc: "codec@ietf.org" Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Oct 2010 06:37:24 -0000 --_000_027A93CE4A670242BD91A44E37105AEF0CACAE8C03ESESSCMS0351e_ Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable I am also coming to Beijing and I'll be happy to join you for dinner. BR, Erik ________________________________ From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of K= oen Vos Sent: den 27 oktober 2010 05:17 To: Christian Hoene Cc: codec@ietf.org Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing +1 koen ________________________________ From: "Christian Hoene" To: "Stephen Botzko" , "Michael Knappe" Cc: codec@ietf.org Sent: Tuesday, October 26, 2010 6:59:53 PM Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing Me, too. --------------------------------------------------------------- Dr.-Ing. Christian Hoene Interactive Communication Systems (ICS), University of T=FCbingen Sand 13, 72076 T=FCbingen, Germany, Phone +49 7071 2970532 http://www.net.uni-tuebingen.de/ From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of S= tephen Botzko Sent: Tuesday, October 26, 2010 9:19 PM To: Michael Knappe Cc: codec@ietf.org Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing I will be there, and would be happy to join you on Monday evening. Stephen Botzko On Tue, Oct 26, 2010 at 3:12 PM, Michael Knappe > wrote: Stephan, This brings up a good point overall - who is planning to be at IETF79? Mike ----- Original Message ----- From: codec-bounces@ietf.org > To: codec@ietf.org > Sent: Tue Oct 26 14:57:35 2010 Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing I'll not be in China, but I wish you good luck and good progress for the meeting. Stephan On 10.26.2010 10:44 , "Michael Knappe" > wrote: >Hi everyone! I would like to invite everyone as an informal activity to >join me after the Monday technical plenary at a nearby restaurant for >dinner and drinks. Email back if you=B9re interested so I can tally a roug= h >number for reservations. Suggestions for the restaurant are welcome, and >I=B9ll send out logistics for meeting up after the plenary once the >restaurant is decided. > >It=B9s a great opportunity to meet and chat with fellow codec WG >participants in an informal setting, hope to see you there! > >Cheers, > >Mike > > > >_______________________________________________ >codec mailing list >codec@ietf.org >https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec --_000_027A93CE4A670242BD91A44E37105AEF0CACAE8C03ESESSCMS0351e_ Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
I am also coming to Beijing and I'll be happy to j= oin you=20 for dinner.
 
BR,
Erik

From: codec-bounces@ietf.org=20 [mailto:codec-bounces@ietf.org] On Behalf Of Koen Vos
Sent: den=20 27 oktober 2010 05:17
To: Christian Hoene
Cc:=20 codec@ietf.org
Subject: Re: [codec] Monday codec WG dinner and dr= inks=20 in Beijing

To:=20 "Stephen Botzko" <stephen.botzko@gmail.com>, "Michael Knappe"=20 <mknappe@juniper.net>
Cc: codec@ietf.org
Sent:=20 Tuesday, October 26, 2010 6:59:53 PM
Subject: Re: [codec] Mon= day=20 codec WG dinner and drinks in Beijing

Me,=20 too.

 

-= --------------------------------------------------------------

D= r.-Ing.=20 Christian Hoene

I= nteractive=20 Communication Systems (ICS), University of T=FCbingen

S= and 13,=20 72076 T=FCbingen, Germany, Phone +49 7071 2970532

<= A=20 href=3D"http://www.net.uni-tuebingen.de/"=20 target=3D_blank>http://www.net.uni-tuebingen.de/

 

From:<= /B>=20 codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of=20 Stephen Botzko
Sent: Tuesday, October 26, 2010 9:19=20 PM
To: Michael Knappe
Cc: codec@ietf.org
Subject:= =20 Re: [codec] Monday codec WG dinner and drinks in Beijing

 

I will be there, and wou= ld be=20 happy to join you on Monday evening.

Stephen Botzko

On Tue, Oct 26, 2010 at 3:12 PM, Michael Knappe <mknappe@juniper.net= >=20 wrote:

Stephan,

This brings up a good point overall - = who is=20 planning to be at IETF79?

Mike


----- Original Message -----
From: codec-bounces@ietf.o= rg=20 <codec-bounces@ietf.org>
To: codec@ietf.org <codec@ietf.org>
Se= nt: Tue=20 Oct 26 14:57:35 2010
Subject: Re: [codec] Monday codec WG dinner and dri= nks=20 in Beijing

I'll not be in China, but I wish you good luck and good=20 progress for the
meeting.
Stephan


On 10.26.2010 10:44 ,=20 "Michael Knappe" <mknappe@juniper.net> wrote:

>Hi everyone! = I would=20 like to invite everyone as an informal activity to
>join me after the= =20 Monday technical plenary at a nearby restaurant for
>dinner and drink= s.=20 Email back if you=B9re interested so I can tally a rough
>number for= =20 reservations. Suggestions for the restaurant are welcome, and
>I=B9ll= send=20 out logistics for meeting up after the plenary once the
>restaurant i= s=20 decided.
>
>It=B9s a great opportunity to meet and chat with fe= llow=20 codec WG
>participants in an informal setting, hope to see you=20 there!
>
>Cheers,
>
>Mike
>
>
>>_______________________________________________
>codec=20 mailing list
>codec@ietf.org
>https://www.ietf.org/mailman/listinfo/codec


= _______________________________________________
codec=20 mailing list
codec@ietf.org
https://www.ietf.org/mailman/listinfo/codec
________= _______________________________________
codec=20 mailing list
codec@ietf.org
https://www.ietf.org/mailman/listinfo/codec

 


_____________________________________= __________
codec=20 mailing=20 list
codec@ietf.org
https://www.ietf.org/mailman/listinfo/codec
--_000_027A93CE4A670242BD91A44E37105AEF0CACAE8C03ESESSCMS0351e_-- From stephen.botzko@gmail.com Wed Oct 27 03:14:24 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id AF2D73A6A09 for ; Wed, 27 Oct 2010 03:14:16 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.681 X-Spam-Level: X-Spam-Status: No, score=-2.681 tagged_above=-999 required=5 tests=[AWL=-0.083, BAYES_00=-2.599, HTML_MESSAGE=0.001] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id NseQj9-37R-c for ; Wed, 27 Oct 2010 03:14:03 -0700 (PDT) Received: from mail-bw0-f44.google.com (mail-bw0-f44.google.com [209.85.214.44]) by core3.amsl.com (Postfix) with ESMTP id C731B3A6905 for ; Wed, 27 Oct 2010 03:12:47 -0700 (PDT) Received: by bwz12 with SMTP id 12so401159bwz.31 for ; Wed, 27 Oct 2010 03:14:28 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type; bh=va09Gh8KnjziAVTdR3RJFsJ7OMQeDw6FnCAPPobbRdo=; b=idszZj7H4z8VYfHo9p+GxAHGWvZBYzTB2yUsAFowt6ijDO6wO5AtSL2Odnsl1s6JwP g37p8zoT8jtD4O3tEYIZnGSYe39tLkBSTL4xh6SP6C9xbzWdfncKE7pbzYY6YB7k5Crh Avb3f05TC6ovDfN7Mis2J/DqN81iLKZFo74hk= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=JlmYdmbXotH1og/CSO9E3BHGl3wrRr83TMkQMYocK8JgOCeQzSHQssZWSeqLT3swNW 6ngFnlhQgpRTdnyHairGB2Vzmrs2BLKL0D7y5+iO26UwbtJ+Q73fDSWOScgPrIsTMkEu h0gpKQkvl2FAm8jgkP+QjbS8oFPyFYEIQGRR8= MIME-Version: 1.0 Received: by 10.204.52.193 with SMTP id j1mr7079044bkg.52.1288174467476; Wed, 27 Oct 2010 03:14:27 -0700 (PDT) Received: by 10.204.80.213 with HTTP; Wed, 27 Oct 2010 03:14:26 -0700 (PDT) In-Reply-To: References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC37F42.2040103@usherbrooke.ca> <4CC71C7A.3090002@octasic.com> Date: Wed, 27 Oct 2010 06:14:26 -0400 Message-ID: From: Stephen Botzko To: "Raymond (Juin-Hwey) Chen" Content-Type: multipart/alternative; boundary=001636c5991b8aeba10493967d19 Cc: "codec@ietf.org" , Jean-Marc Valin Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Oct 2010 10:14:29 -0000 --001636c5991b8aeba10493967d19 Content-Type: text/plain; charset=ISO-8859-1 I think it'd be good to have development discussions like this on the list. Stephen Botzko On Wed, Oct 27, 2010 at 2:17 AM, Raymond (Juin-Hwey) Chen < rchen@broadcom.com> wrote: > Hi Jean-Marc, > > Are you saying the problem is in the pitch estimation for that test file > rather than the pitch filtering? > > I will take a look at that file. Is that a raw file? I am not familiar > with the ".sw" file extension. > > I could hear audible improvement when I tried the method on the other solo > instrument files that I am working on, although the degree of improvement > depends on which file it is, as expected. > > I am starting to wonder how many people in the codec WG email list still > want to continue to see our low-level technical discussions about the > algorithm development in the reflector. I wonder if some people in the > codec WG email list would prefer not to be bombarded with such emails. > Should we consider moving such detailed codec algorithm development > discussions off the WG mailing list and just between those of us who are > interested or are directly involved? > > Or is this how IETF supposed to work (discussing all technical development > in emails)? > > Thanks. > > Raymond > > -----Original Message----- > From: Jean-Marc Valin [mailto:jean-marc.valin@octasic.com] > Sent: Tuesday, October 26, 2010 11:23 AM > To: Raymond (Juin-Hwey) Chen > Cc: Jean-Marc Valin; 'codec@ietf.org' > Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT > codec > > Hi Raymond, > > I've been able to reproduce results that are mostly similar to what you got > earlier on the trumpet file. Unfortunately, it doesn't work as well on > other files I'm trying. For example on this one > (http://jmvalin.ca/misc_stuff/s02a_102.sw) there is simply too much pitch > doubling and I'm not actually sure it's possible to do a good job. Could > you try with your setup and see if you can improve things. I think one of > the tricky thing with the postfilter will be to make it work in all > circumstances and -- most importantly -- prevent it from causing harm in > cases where it cannot help. > > Regarding the periods we'd need to support, I think the important question > is not that much "on what range does it help", but "on what range is it > necessary". From what I observed, CELT copes well with regular male and > even female speech and even the vega sample tends to sound OK. It's only on > strongly tonal samples with higher pitches that problems occur. The main > reason I want to limit the maximum period allowed is that there's a memory > footprint issue that's directly related. In any case, we'll need to > experiment as with anything else. > > Cheers, > > Jean-Marc > > On 10-10-26 02:40 AM, Raymond (Juin-Hwey) Chen wrote: > > Hi Jean-Marc, > > > > See my in-line comments below. > > > > You wrote: > >> My concern wasn't that much about the period, but the gain. I'll > >> probably have to try both approaches, but I was mainly concerned that > >> a non-harmonic signal with one strong sinusoid could cause the comb > >> filter to be turned on with a high gain. > > > > [Raymond]: I see your concern with introducing artificial harmonics for > > a single sinusoid. On the other hand, my concern is that if you derive > > the gain (or taps) based on partially whitened input signal, the gain or > > taps may be too small for legitimate periodic music or speech signals. > > To address both your and my concerns, I think it is probably best to use > > a 3-tap filter where the taps are derived directly from the input signal > > without whitening. Since a 3-tap filter has the ability to taper off > > the degree of comb filtering with increasing frequency, a set of 3 > > filter taps that is chosen to match the input spectrum can probably have > > a reasonable spectral match for either a single tone or music/speech > > with a harmonic structure. The BroadVoice codecs use a 3-tap pitch > > predictor. The way those 3 taps are jointly searched and vector > > quantized might provide a way to quantize the 3 taps and do a spectral > > match at the same time for our current pitch prefilter/postfilter > > approa ch. We need to experiment to find out. The BroadVoice codebooks > > for the 3 taps were designed using mainly speech. If we adopt this > > approach, we need to redesign the codebook with sufficient music signals > > in the training set. If you are particularly concerned with a single > > sinusoid, we may want to train one or a few 3-tap codevectors using just > > sinusoids. If this doesn't work well, in the worst case, we could > > potentially design a detector that detect (nearly) periodic time-domain > > waveform that does not have a spectrum with a harmonic structure (like a > > single sinusoid) and turn off the prefilter/postfilter accordingly. > > > >> CELT currently has a pitch predictor used for PLC, but I doubt it's > >> very good for this as it is now. OTOH, it's fairly low complexity too. > >> It's part of my plans to compare it with the one in BV16/32. I assume > >> that Broadcom would be willing to relicense it to the simplified BSD > >> license (it's currently LGPL) if we wanted to use it instead of the > >> current CELT estimator. > > > > [Raymond]: Broadcom's legal counsel will formally answer that question > > soon. My assumption is that if we were previously willing to allow > > royalty-free use of BroadVoice in the IETF codec, then allowing certain > > components of it to be used in the IETF codec under IETF's licensing > > terms is probably not a problem, but this is just my personal thought > > and not an official Broadcom position yet. I will talk to our lawyer > > and let him give the formal answer. > > > >> The pitch range is another thing that needs experimenting with, but > >> my intuition here is that it's probably not necessary to go below 100 > >> Hz because for these very low pitches, there is enough masking between > >> the harmonics to make the coding noise inaudible even without a > >> post-filter. Again worth testing. > > > > [Raymond]: I agree that lower pitch frequency generally has better noise > > masking, so time-domain predictive codecs often sound noisier for female > > speech than for male speech. However, my previous work with pitch > > postfiltering during my Ph.D. study showed that pitch postfiltering > > alone could still give audible quality improvement to low-bit-rate-coded > > male speech. Does it improve quality when the pitch frequency is below > > 100 Hz? I don't remember the pitch frequencies of my pitch postfilter > > tests. I agree it is worth testing. > > > > > > _______________________________________________ codec mailing list > > codec@ietf.org https://www.ietf.org/mailman/listinfo/codec > > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > --001636c5991b8aeba10493967d19 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable I think it'd be good to have development discussions like this on the l= ist.

Stephen Botzko

On Wed, Oct 27= , 2010 at 2:17 AM, Raymond (Juin-Hwey) Chen <rchen@broadcom.com> wrote:
Hi Jean-Marc,

Are you saying the problem is in the pitch estimation for that test file ra= ther than the pitch filtering?

I will take a look at that file. =A0Is that a raw file? =A0I am not familia= r with the ".sw" file extension.

I could hear audible improvement when I tried the method on the other solo = instrument files that I am working on, although the degree of improvement d= epends on which file it is, as expected.

I am starting to wonder how many people in the codec WG email list still wa= nt to continue to see our low-level technical discussions about the algorit= hm development in the reflector. =A0I wonder if some people in the codec WG= email list would prefer not to be bombarded with such emails. =A0Should we= consider moving such detailed codec algorithm development discussions off = the WG mailing list and just between those of us who are interested or are = directly involved?

Or is this how IETF supposed to work (discussing all technical development = in emails)?

Thanks.

Raymond

-----Original Message-----
From: Jean-Marc Valin [mailto:jean-marc.valin@octasic.com]
Sent: Tuesday, October 26, 2010 11:23 AM
To: Raymond (Juin-Hwey) Chen
Cc: Jean-Marc Valin; 'codec@ietf.org'
Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT co= dec

Hi Raymond,

I've been able to reproduce results that are mostly similar to what you= got
earlier on the trumpet file. Unfortunately, it doesn't work as well on<= br> other files I'm trying. For example on this one
(htt= p://jmvalin.ca/misc_stuff/s02a_102.sw) there is simply too much pitch doubling and I'm not actually sure it's possible to do a good job. = Could
you try with your setup and see if you can improve things. I think one of the tricky thing with the postfilter will be to make it work in all
circumstances and -- most importantly -- prevent it from causing harm in cases where it cannot help.

Regarding the periods we'd need to support, I think the important quest= ion
is not that much "on what range does it help", but "on what = range is it
necessary". From what I observed, CELT copes well with regular male an= d
even female speech and even the vega sample tends to sound OK. It's onl= y on
strongly tonal samples with higher pitches that problems occur. The main reason I want to limit the maximum period allowed is that there's a mem= ory
footprint issue that's directly related. In any case, we'll need to=
experiment as with anything else.

Cheers,

=A0 =A0 =A0 =A0Jean-Marc

On 10-10-26 02:40 AM, Raymond (Juin-Hwey) Chen wrote:
> Hi Jean-Marc,
>
> See my in-line comments below.
>
> You wrote:
>> My concern wasn't that much about the period, but the gain. I&= #39;ll
>> probably have to try both approaches, but I was mainly concerned t= hat
>> a non-harmonic signal with one strong sinusoid could cause the com= b
>> filter to be turned on with a high gain.
>
> [Raymond]: I see your concern with introducing artificial harmonics fo= r
> a single sinusoid. =A0On the other hand, my concern is that if you der= ive
> the gain (or taps) based on partially whitened input signal, the gain = or
> taps may be too small for legitimate periodic music or speech signals.=
> To address both your and my concerns, I think it is probably best to u= se
> a 3-tap filter where the taps are derived directly from the input sign= al
> without whitening. =A0Since a 3-tap filter has the ability to taper of= f
> the degree of comb filtering with increasing frequency, a set of 3
> filter taps that is chosen to match the input spectrum can probably ha= ve
> a reasonable spectral match for either a single tone or music/speech > with a harmonic structure. =A0The BroadVoice codecs use a 3-tap pitch<= br> > predictor. =A0The way those 3 taps are jointly searched and vector
> quantized might provide a way to quantize the 3 taps and do a spectral=
> match at the same time for our current pitch prefilter/postfilter
> approa ch. =A0We need to experiment to find out. =A0The BroadVoice cod= ebooks
> for the 3 taps were designed using mainly speech. =A0If we adopt this<= br> > approach, we need to redesign the codebook with sufficient music signa= ls
> in the training set. =A0If you are particularly concerned with a singl= e
> sinusoid, we may want to train one or a few 3-tap codevectors using ju= st
> sinusoids. =A0If this doesn't work well, in the worst case, we cou= ld
> potentially design a detector that detect (nearly) periodic time-domai= n
> waveform that does not have a spectrum with a harmonic structure (like= a
> single sinusoid) and turn off the prefilter/postfilter accordingly. >
>> CELT currently has a pitch predictor used for PLC, but I doubt it&= #39;s
>> very good for this as it is now. OTOH, it's fairly low complex= ity too.
>> It's part of my plans to compare it with the one in BV16/32. I= assume
>> that Broadcom would be willing to relicense it to the simplified B= SD
>> license (it's currently LGPL) if we wanted to use it instead o= f the
>> current CELT estimator.
>
> [Raymond]: Broadcom's legal counsel will formally answer that ques= tion
> soon. =A0My assumption is that if we were previously willing to allow<= br> > royalty-free use of BroadVoice in the IETF codec, then allowing certai= n
> components of it to be used in the IETF codec under IETF's licensi= ng
> terms is probably not a problem, but this is just my personal thought<= br> > and not an official Broadcom position yet. =A0I will talk to our lawye= r
> and let him give the formal answer.
>
>> The pitch range is another thing that needs experimenting with, bu= t
>> my intuition here is that it's probably not necessary to go be= low 100
>> Hz because for these very low pitches, there is enough masking bet= ween
>> the harmonics to make the coding noise inaudible even without a >> post-filter. Again worth testing.
>
> [Raymond]: I agree that lower pitch frequency generally has better noi= se
> masking, so time-domain predictive codecs often sound noisier for fema= le
> speech than for male speech. =A0However, my previous work with pitch > postfiltering during my Ph.D. study showed that pitch postfiltering > alone could still give audible quality improvement to low-bit-rate-cod= ed
> male speech. =A0Does it improve quality when the pitch frequency is be= low
> 100 Hz? I don't remember the pitch frequencies of my pitch postfil= ter
> tests. =A0I agree it is worth testing.
>
>
> _______________________________________________ codec mailing list
> codec@ietf.org https://www.ietf.o= rg/mailman/listinfo/codec



_______________________________________________
codec mailing list
codec@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/codec

--001636c5991b8aeba10493967d19-- From karsten.vandborg.sorensen@skype.net Wed Oct 27 04:30:25 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 27A653A6931 for ; Wed, 27 Oct 2010 04:30:25 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.149 X-Spam-Level: X-Spam-Status: No, score=-1.149 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, MSGID_MULTIPLE_AT=1.449] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id zhpioxhn8WB1 for ; Wed, 27 Oct 2010 04:30:23 -0700 (PDT) Received: from mx.skype.net (mx.skype.net [78.141.177.88]) by core3.amsl.com (Postfix) with ESMTP id A96B13A6803 for ; Wed, 27 Oct 2010 04:30:19 -0700 (PDT) Received: from mx.skype.net (localhost [127.0.0.1]) by mx.skype.net (Postfix) with ESMTP id F3C69170B for ; Wed, 27 Oct 2010 13:31:59 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=skype.net; h=from:to :references:in-reply-to:subject:date:message-id:mime-version: content-type; s=mx; bh=IUcPc6wSyvK4XNYuq3iZ2NmRxV0=; b=ddcKj0+sR 7m5UbJGYQITZg+uVAuWKR4EP+iyySzP/xROs9hVxmsYveZqNLM0k6sSyY8GpllY0 evGaRrVO7IVppcN7LpR/dHohNWEi6JC0GHofW+DyRFdbV8EAjUJpdXhkz2uuAeen Q+y46RvfLtTvTs792W0Oj1+uDuo2hCbbvg= DomainKey-Signature: a=rsa-sha1; c=nofws; d=skype.net; h=from:to:references :in-reply-to:subject:date:message-id:mime-version:content-type; q=dns; s=mx; b=i49Z8w+QrtQFZShe2n6/Sl/m49KmhTZrK2j8rXuD7ZZT0m7y 1SLNZ28IFK518ntPJ5IYhla4DG4EwdKFuTU91kLeCYLmoTfhq9yqX+B98SgWPa2t UEQFIjZv06aN3tO3QvvOBARGzTSMqgHzHUCupRhfYHULf9CoXLH9LOAAF+s= Received: from zimbra.skype.net (zimbra.skype.net [78.141.177.82]) by mx.skype.net (Postfix) with ESMTP id F233216F5 for ; Wed, 27 Oct 2010 13:31:59 +0200 (CEST) Received: from localhost (localhost [127.0.0.1]) by zimbra.skype.net (Postfix) with ESMTP id E05CB1672684 for ; Wed, 27 Oct 2010 13:31:59 +0200 (CEST) X-Virus-Scanned: amavisd-new at lu2-zimbra.skype.net Received: from zimbra.skype.net ([127.0.0.1]) by localhost (zimbra.skype.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id eKHTwprkjJlQ for ; Wed, 27 Oct 2010 13:31:58 +0200 (CEST) Received: from zimbra.skype.net (lu2-zimbra.skype.net [78.141.177.82]) by zimbra.skype.net (Postfix) with ESMTP id A7AFB3507E4F for ; Wed, 27 Oct 2010 13:31:58 +0200 (CEST) From: "Karsten Vandborg Sorensen" To: References: <000301cb757a$a9916890$fcb439b0$@de> <247946901.1136671.1288149409289.JavaMail.root@lu2-zimbra> <027A93CE4A670242BD91A44E37105AEF0CACAE8C03@ESESSCMS0351.eemea.ericsson.se> In-Reply-To: <027A93CE4A670242BD91A44E37105AEF0CACAE8C03@ESESSCMS0351.eemea.ericsson.se> Date: Wed, 27 Oct 2010 13:31:58 +0200 (CEST) Message-ID: <002d01cb75ca$6ba4d800$42ee8800$@vandborg.sorensen@skype.net> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_002E_01CB75DB.2F2DA800" X-Mailer: Microsoft Office Outlook 12.0 X-Mailer: Zimbra 6.0.6_GA_2330.UBUNTU8_64 (ZimbraConnectorForOutlook/6.0.6058.8) Thread-Index: Act1hXJoxy5qnj0BReWNe8tThZHQsQAHAREgAAou27A= Content-Language: sv X-Originating-IP: [84.55.127.131] Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Oct 2010 11:30:25 -0000 This is a multi-part message in MIME format. ------=_NextPart_000_002E_01CB75DB.2F2DA800 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable I will also be there, and would like to join for dinner and drinks. /Karsten From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Erik Norvell Sent: den 27 oktober 2010 08:39 To: Koen Vos; Christian Hoene Cc: codec@ietf.org Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing I am also coming to Beijing and I'll be happy to join you for dinner. BR, Erik _____ From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Koen Vos Sent: den 27 oktober 2010 05:17 To: Christian Hoene Cc: codec@ietf.org Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing +1 koen _____ From: "Christian Hoene" To: "Stephen Botzko" , "Michael Knappe" Cc: codec@ietf.org Sent: Tuesday, October 26, 2010 6:59:53 PM Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing Me, too. --------------------------------------------------------------- Dr.-Ing. Christian Hoene Interactive Communication Systems (ICS), University of T=FCbingen Sand 13, 72076 T=FCbingen, Germany, Phone +49 7071 2970532 http://www.net.uni-tuebingen.de/ From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Stephen Botzko Sent: Tuesday, October 26, 2010 9:19 PM To: Michael Knappe Cc: codec@ietf.org Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing I will be there, and would be happy to join you on Monday evening. Stephen Botzko On Tue, Oct 26, 2010 at 3:12 PM, Michael Knappe wrote: Stephan, This brings up a good point overall - who is planning to be at IETF79? Mike ----- Original Message ----- From: codec-bounces@ietf.org To: codec@ietf.org Sent: Tue Oct 26 14:57:35 2010 Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing I'll not be in China, but I wish you good luck and good progress for the meeting. Stephan On 10.26.2010 10:44 , "Michael Knappe" wrote: >Hi everyone! I would like to invite everyone as an informal activity to >join me after the Monday technical plenary at a nearby restaurant for >dinner and drinks. Email back if you=B9re interested so I can tally a ro= ugh >number for reservations. Suggestions for the restaurant are welcome, and >I=B9ll send out logistics for meeting up after the plenary once the >restaurant is decided. > >It=B9s a great opportunity to meet and chat with fellow codec WG >participants in an informal setting, hope to see you there! > >Cheers, > >Mike > > > >_______________________________________________ >codec mailing list >codec@ietf.org >https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec ------=_NextPart_000_002E_01CB75DB.2F2DA800 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

I will also be there, and would like to join for dinner = and drinks.

 

/Karsten

 

From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Erik Norvell
Sent: den 27 oktober 2010 08:39
To: Koen Vos; Christian Hoene
Cc: codec@ietf.org
Subject: Re: [codec] Monday codec WG dinner and drinks in = Beijing

 

I am also coming to Beijing and I'll be happy to join you = for dinner.

 

BR,

Erik

 

From:= codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of = Koen Vos
Sent: den 27 oktober 2010 05:17
To: Christian Hoene
Cc: codec@ietf.org
Subject: Re: [codec] Monday codec WG dinner and drinks in = Beijing

+1
koen

From: = "Christian Hoene" = <hoene@uni-tuebingen.de>
To: "Stephen Botzko" <stephen.botzko@gmail.com>, "Michael Knappe" <mknappe@juniper.net>
Cc: codec@ietf.org
Sent: Tuesday, October 26, 2010 6:59:53 PM
Subject: Re: [codec] Monday codec WG dinner and drinks in = Beijing


Me, too.

 

----------------------------------------------------------= -----

Dr.-Ing. Christian Hoene

Interactive Communication Systems (ICS), University of = T=FCbingen

Sand 13, 72076 T=FCbingen, Germany, Phone +49 7071 = 2970532

http://www.net.uni-tuebingen.de/

 

From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] = On Behalf Of Stephen Botzko
Sent: Tuesday, October 26, 2010 9:19 PM
To: Michael Knappe
Cc: codec@ietf.org
Subject: Re: [codec] Monday codec WG dinner and drinks in = Beijing

 

I will be there, and would be happy to join you on Monday evening.

Stephen Botzko

On Tue, Oct 26, 2010 at = 3:12 PM, Michael Knappe <mknappe@juniper.net> wrote:

Stephan,

This brings up a good point overall - who is planning to be at = IETF79?

Mike


----- Original Message -----
From: codec-bounces@ietf.org <codec-bounces@ietf.org>
To: codec@ietf.org <codec@ietf.org>
Sent: Tue Oct 26 14:57:35 2010
Subject: Re: [codec] Monday codec WG dinner and drinks in Beijing

I'll not be in China, but I wish you good luck and good progress for = the
meeting.
Stephan


On 10.26.2010 10:44 , "Michael Knappe" <mknappe@juniper.net> wrote:

>Hi everyone! I would like to invite everyone as an informal activity = to
>join me after the Monday technical plenary at a nearby restaurant = for
>dinner and drinks. Email back if you=B9re interested so I can tally = a rough
>number for reservations. Suggestions for the restaurant are welcome, = and
>I=B9ll send out logistics for meeting up after the plenary once = the
>restaurant is decided.
>
>It=B9s a great opportunity to meet and chat with fellow codec WG
>participants in an informal setting, hope to see you there!
>
>Cheers,
>
>Mike
>
>
>
>_______________________________________________
>codec mailing list
>codec@ietf.org
>https://www.ietf.org/mailman/listinfo/codec


_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/mailman/listinfo/codec
_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/mailman/listinfo/codec

 


_______________________________________________
codec mailing list
codec@ietf.org
https://www.ietf.org/mailman/listinfo/codec

------=_NextPart_000_002E_01CB75DB.2F2DA800-- From jean-marc.valin@octasic.com Wed Oct 27 06:13:41 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 51FFD3A6A3F for ; Wed, 27 Oct 2010 06:13:41 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.499 X-Spam-Level: X-Spam-Status: No, score=-2.499 tagged_above=-999 required=5 tests=[AWL=0.100, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id a4XxZdRWnLyG for ; Wed, 27 Oct 2010 06:13:37 -0700 (PDT) Received: from toroondcbmts05-srv.bellnexxia.net (toroondcbmts05.bellnexxia.net [207.236.237.39]) by core3.amsl.com (Postfix) with ESMTP id 9E1513A6A45 for ; Wed, 27 Oct 2010 06:11:46 -0700 (PDT) Received: from toip57-bus.srvr.bell.ca ([67.69.240.184]) by toroondcbmts05-srv.bellnexxia.net (InterMail vM.8.00.01.00 201-2244-105-20090324) with ESMTP id <20101027131330.HSYV28711.toroondcbmts05-srv.bellnexxia.net@toip57-bus.srvr.bell.ca>; Wed, 27 Oct 2010 09:13:30 -0400 Received: from toip40-bus.srvr.bell.ca ([67.69.240.41]) by toip57-bus.srvr.bell.ca with ESMTP; 27 Oct 2010 09:13:30 -0400 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: ApsEADzCx0xGNv5q/2dsb2JhbACiSrxfhUgEjVsG Received: from mail.octasic.com (HELO MAILEXCH.octasic.com) ([70.54.254.106]) by toip40-bus.srvr.bell.ca with ESMTP; 27 Oct 2010 09:13:29 -0400 Received: from [10.100.60.27] ([10.100.60.27]) by MAILEXCH.octasic.com with Microsoft SMTPSVC(6.0.3790.4675); Wed, 27 Oct 2010 09:13:07 -0400 Message-ID: <4CC82563.9070403@octasic.com> Date: Wed, 27 Oct 2010 09:13:07 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.14) Gecko/20101006 Lightning/1.0b1 Thunderbird/3.0.9 MIME-Version: 1.0 To: "Raymond (Juin-Hwey) Chen" References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC37F42.2040103@usherbrooke.ca> <4CC71C7A.3090002@octasic.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 27 Oct 2010 13:13:07.0653 (UTC) FILETIME=[B264D750:01CB75D8] Cc: "'codec@ietf.org'" , Jean-Marc Valin Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Oct 2010 13:13:41 -0000 Hi Raymond, On 10-10-27 02:17 AM, Raymond (Juin-Hwey) Chen wrote: > Are you saying the problem is in the pitch estimation for that test file > rather than the pitch filtering? The fundamental problem is definitely with the estimation. Reliable and consistent pitch estimation is hard. The question is thus whether it's possible to either have a reliable enough estimator *or* have a robust enough filtering scheme that it can tolerate occasional errors in the estimation. I'm still working on both. > I will take a look at that file. Is that a raw file? I am not familiar > with the ".sw" file extension. That's just a 48 kHz raw 16-bit PCM file (little endian). sw is just the sox program's extension for "signed word". > I could hear audible improvement when I tried the method on the other > solo instrument files that I am working on, although the degree of > improvement depends on which file it is, as expected. Yes, I'm also observing that. What I mainly want to improve is how it works on more complicated signals (like the one I linked to), or non-harmonic signals. As long as we're in a situation where it improves on a subset of the samples while not degrading the others, I'm happy. > I am starting to wonder how many people in the codec WG email list still > want to continue to see our low-level technical discussions about the > algorithm development in the reflector. I wonder if some people in the > codec WG email list would prefer not to be bombarded with such emails. > Should we consider moving such detailed codec algorithm development > discussions off the WG mailing list and just between those of us who are > interested or are directly involved? > > Or is this how IETF supposed to work (discussing all technical > development in emails)? I think we should keep having this discussion on the list. The IETF lists are normally meant for technical discussions so this just helps improve the SNR :-) Cheers, Jean-Marc From jdrosen@jdrosen.net Wed Oct 27 14:12:10 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 3D0493A6768 for ; Wed, 27 Oct 2010 14:12:10 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.371 X-Spam-Level: X-Spam-Status: No, score=-102.371 tagged_above=-999 required=5 tests=[AWL=0.228, BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id g5dfrZMbM0b2 for ; Wed, 27 Oct 2010 14:12:05 -0700 (PDT) Received: from ecbiz71.inmotionhosting.com (ecbiz71.inmotionhosting.com [173.205.126.243]) by core3.amsl.com (Postfix) with ESMTP id 043D43A6405 for ; Wed, 27 Oct 2010 14:12:04 -0700 (PDT) Received: from pool-98-109-139-97.nwrknj.fios.verizon.net ([98.109.139.97] helo=[192.168.1.8]) by ecbiz71.inmotionhosting.com with esmtpsa (TLSv1:AES256-SHA:256) (Exim 4.69) (envelope-from ) id 1PBDJg-00042e-G0 for codec@ietf.org; Wed, 27 Oct 2010 17:13:52 -0400 Message-ID: <4CC89610.70107@jdrosen.net> Date: Wed, 27 Oct 2010 17:13:52 -0400 From: Jonathan Rosenberg User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.2.11) Gecko/20101013 Thunderbird/3.1.5 MIME-Version: 1.0 To: codec@ietf.org References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC37F42.2040103@usherbrooke.ca> <4CC71C7A.3090002@octasic.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - ecbiz71.inmotionhosting.com X-AntiAbuse: Original Domain - ietf.org X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - jdrosen.net Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Oct 2010 21:12:10 -0000 Absolutely, these discussions should happen here. This working group is about producing this codec, and technical discussions about that deliverable belong here. -Jonathan R. (with WG chair hat on) On 10/27/2010 6:14 AM, Stephen Botzko wrote: > I think it'd be good to have development discussions like this on the list. > > Stephen Botzko > > On Wed, Oct 27, 2010 at 2:17 AM, Raymond (Juin-Hwey) Chen > > wrote: > > Hi Jean-Marc, > > Are you saying the problem is in the pitch estimation for that test > file rather than the pitch filtering? > > I will take a look at that file. Is that a raw file? I am not > familiar with the ".sw" file extension. > > I could hear audible improvement when I tried the method on the > other solo instrument files that I am working on, although the > degree of improvement depends on which file it is, as expected. > > I am starting to wonder how many people in the codec WG email list > still want to continue to see our low-level technical discussions > about the algorithm development in the reflector. I wonder if some > people in the codec WG email list would prefer not to be bombarded > with such emails. Should we consider moving such detailed codec > algorithm development discussions off the WG mailing list and just > between those of us who are interested or are directly involved? > > Or is this how IETF supposed to work (discussing all technical > development in emails)? > > Thanks. > > Raymond > > -----Original Message----- > From: Jean-Marc Valin [mailto:jean-marc.valin@octasic.com > ] > Sent: Tuesday, October 26, 2010 11:23 AM > To: Raymond (Juin-Hwey) Chen > Cc: Jean-Marc Valin; 'codec@ietf.org ' > Subject: Re: [codec] A pitch filtering method to improve the mode 1 > CELT codec > > Hi Raymond, > > I've been able to reproduce results that are mostly similar to what > you got > earlier on the trumpet file. Unfortunately, it doesn't work as well on > other files I'm trying. For example on this one > (http://jmvalin.ca/misc_stuff/s02a_102.sw) there is simply too much > pitch > doubling and I'm not actually sure it's possible to do a good job. Could > you try with your setup and see if you can improve things. I think > one of > the tricky thing with the postfilter will be to make it work in all > circumstances and -- most importantly -- prevent it from causing harm in > cases where it cannot help. > > Regarding the periods we'd need to support, I think the important > question > is not that much "on what range does it help", but "on what range is it > necessary". From what I observed, CELT copes well with regular male and > even female speech and even the vega sample tends to sound OK. It's > only on > strongly tonal samples with higher pitches that problems occur. The main > reason I want to limit the maximum period allowed is that there's a > memory > footprint issue that's directly related. In any case, we'll need to > experiment as with anything else. > > Cheers, > > Jean-Marc > > On 10-10-26 02:40 AM, Raymond (Juin-Hwey) Chen wrote: > > Hi Jean-Marc, > > > > See my in-line comments below. > > > > You wrote: > >> My concern wasn't that much about the period, but the gain. I'll > >> probably have to try both approaches, but I was mainly concerned > that > >> a non-harmonic signal with one strong sinusoid could cause the comb > >> filter to be turned on with a high gain. > > > > [Raymond]: I see your concern with introducing artificial > harmonics for > > a single sinusoid. On the other hand, my concern is that if you > derive > > the gain (or taps) based on partially whitened input signal, the > gain or > > taps may be too small for legitimate periodic music or speech > signals. > > To address both your and my concerns, I think it is probably best > to use > > a 3-tap filter where the taps are derived directly from the input > signal > > without whitening. Since a 3-tap filter has the ability to taper off > > the degree of comb filtering with increasing frequency, a set of 3 > > filter taps that is chosen to match the input spectrum can > probably have > > a reasonable spectral match for either a single tone or music/speech > > with a harmonic structure. The BroadVoice codecs use a 3-tap pitch > > predictor. The way those 3 taps are jointly searched and vector > > quantized might provide a way to quantize the 3 taps and do a > spectral > > match at the same time for our current pitch prefilter/postfilter > > approa ch. We need to experiment to find out. The BroadVoice > codebooks > > for the 3 taps were designed using mainly speech. If we adopt this > > approach, we need to redesign the codebook with sufficient music > signals > > in the training set. If you are particularly concerned with a single > > sinusoid, we may want to train one or a few 3-tap codevectors > using just > > sinusoids. If this doesn't work well, in the worst case, we could > > potentially design a detector that detect (nearly) periodic > time-domain > > waveform that does not have a spectrum with a harmonic structure > (like a > > single sinusoid) and turn off the prefilter/postfilter accordingly. > > > >> CELT currently has a pitch predictor used for PLC, but I doubt it's > >> very good for this as it is now. OTOH, it's fairly low > complexity too. > >> It's part of my plans to compare it with the one in BV16/32. I > assume > >> that Broadcom would be willing to relicense it to the simplified BSD > >> license (it's currently LGPL) if we wanted to use it instead of the > >> current CELT estimator. > > > > [Raymond]: Broadcom's legal counsel will formally answer that > question > > soon. My assumption is that if we were previously willing to allow > > royalty-free use of BroadVoice in the IETF codec, then allowing > certain > > components of it to be used in the IETF codec under IETF's licensing > > terms is probably not a problem, but this is just my personal thought > > and not an official Broadcom position yet. I will talk to our lawyer > > and let him give the formal answer. > > > >> The pitch range is another thing that needs experimenting with, but > >> my intuition here is that it's probably not necessary to go > below 100 > >> Hz because for these very low pitches, there is enough masking > between > >> the harmonics to make the coding noise inaudible even without a > >> post-filter. Again worth testing. > > > > [Raymond]: I agree that lower pitch frequency generally has > better noise > > masking, so time-domain predictive codecs often sound noisier for > female > > speech than for male speech. However, my previous work with pitch > > postfiltering during my Ph.D. study showed that pitch postfiltering > > alone could still give audible quality improvement to > low-bit-rate-coded > > male speech. Does it improve quality when the pitch frequency is > below > > 100 Hz? I don't remember the pitch frequencies of my pitch postfilter > > tests. I agree it is worth testing. > > > > > > _______________________________________________ codec mailing list > > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec -- Jonathan D. Rosenberg, Ph.D. SkypeID: jdrosen Chief Technology Strategist Mobile: +1 (732) 766-2496 Skype SkypeIn: +1 (408) 465-0361 jdrosen@skype.net http://www.skype.com jdrosen@jdrosen.net http://www.jdrosen.net From jdrosen@jdrosen.net Wed Oct 27 14:19:52 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id D6B0D3A67AF for ; Wed, 27 Oct 2010 14:19:52 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.399 X-Spam-Level: X-Spam-Status: No, score=-102.399 tagged_above=-999 required=5 tests=[AWL=0.200, BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id zVK5b-xug-Wh for ; Wed, 27 Oct 2010 14:19:51 -0700 (PDT) Received: from ecbiz71.inmotionhosting.com (ecbiz71.inmotionhosting.com [173.205.126.243]) by core3.amsl.com (Postfix) with ESMTP id C1B863A659A for ; Wed, 27 Oct 2010 14:19:51 -0700 (PDT) Received: from pool-98-109-139-97.nwrknj.fios.verizon.net ([98.109.139.97] helo=[192.168.1.8]) by ecbiz71.inmotionhosting.com with esmtpsa (TLSv1:AES256-SHA:256) (Exim 4.69) (envelope-from ) id 1PBDRF-00019K-Ah for codec@ietf.org; Wed, 27 Oct 2010 17:21:41 -0400 Message-ID: <4CC897E7.7070302@jdrosen.net> Date: Wed, 27 Oct 2010 17:21:43 -0400 From: Jonathan Rosenberg User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.2.11) Gecko/20101013 Thunderbird/3.1.5 MIME-Version: 1.0 To: codec@ietf.org References: <390831ED3DF58E41A3D2FB82591E2C3605A34C4D@MAILEXCH.octasic.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - ecbiz71.inmotionhosting.com X-AntiAbuse: Original Domain - ietf.org X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - jdrosen.net Subject: Re: [codec] Opus codec licensing X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Oct 2010 21:19:52 -0000 The Skype IPR declaration uses standard in the lower-case sense - any standards track RFC. It does not mean that it only applies to a full standard protocol. For those not familiar - IETF standards track documents can be in one of three categories - proposed standard, draft standard, and standard. A specification, in theory progresses through these three phases in that order, each with a different RFC assigned. In practice, the vast majority of standards-track RFCs are proposed standard, and never make it beyond that stage. Thanks, Jonathan R. On 10/24/2010 3:09 AM, James Cloos wrote: >>>>>> "JV" == Jean-Marc Valin writes: > > JV> Opus is available under the BSD license. As for patents, Skype has > JV> pledged to make them available royalty-free once the codec is > JV> accepted as an IETF standard. > > As a Standard? Or just published as an standards-track RFC? > > That is a /huge/ difference, time-wise. > > -JimC -- Jonathan D. Rosenberg, Ph.D. SkypeID: jdrosen Chief Technology Strategist Mobile: +1 (732) 766-2496 Skype SkypeIn: +1 (408) 465-0361 jdrosen@skype.net http://www.skype.com jdrosen@jdrosen.net http://www.jdrosen.net From jean-marc.valin@octasic.com Wed Oct 27 15:01:55 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 432DF3A6807 for ; Wed, 27 Oct 2010 15:01:55 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.506 X-Spam-Level: X-Spam-Status: No, score=-2.506 tagged_above=-999 required=5 tests=[AWL=0.093, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id LIdVps5lo8j5 for ; Wed, 27 Oct 2010 15:01:45 -0700 (PDT) Received: from toroondcbmts07-srv.bellnexxia.net (toroondcbmts07-srv.bellnexxia.net [207.236.237.41]) by core3.amsl.com (Postfix) with ESMTP id B80463A67B2 for ; Wed, 27 Oct 2010 15:01:37 -0700 (PDT) Received: from toip56-bus.srvr.bell.ca ([67.69.240.142]) by toroondcbmts07-srv.bellnexxia.net (InterMail vM.8.00.01.00 201-2244-105-20090324) with ESMTP id <20101027220326.NXHC24071.toroondcbmts07-srv.bellnexxia.net@toip56-bus.srvr.bell.ca>; Wed, 27 Oct 2010 18:03:26 -0400 Received: from toip41-bus.srvr.bell.ca ([67.69.240.42]) by toip56-bus.srvr.bell.ca with ESMTP; 27 Oct 2010 18:03:18 -0400 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: ApsEALM6yExGNv5q/2dsb2JhbACiNMB6gw2COwSNWwY Received: from mail.octasic.com (HELO MAILEXCH.octasic.com) ([70.54.254.106]) by toip41-bus.srvr.bell.ca with ESMTP; 27 Oct 2010 18:03:17 -0400 Received: from [10.100.60.27] ([10.100.60.27]) by MAILEXCH.octasic.com with Microsoft SMTPSVC(6.0.3790.4675); Wed, 27 Oct 2010 18:03:17 -0400 Message-ID: <4CC8A1A5.4070603@octasic.com> Date: Wed, 27 Oct 2010 18:03:17 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.14) Gecko/20101006 Lightning/1.0b1 Thunderbird/3.0.9 MIME-Version: 1.0 To: Jonathan Rosenberg References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC37F42.2040103@usherbrooke.ca> <4CC71C7A.3090002@octasic.com> <4CC89610.70107@jdrosen.net> In-Reply-To: <4CC89610.70107@jdrosen.net> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 27 Oct 2010 22:03:17.0279 (UTC) FILETIME=[C26892F0:01CB7622] Cc: codec@ietf.org Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Oct 2010 22:01:55 -0000 For those curious about what this is all about. This is the frequency response of the post-filter for a certain pitch value: http://jmvalin.ca/misc_stuff/postfilter.png Note that the roll-off is due to the fact that I've implemented a (fixed) 3-tap filter to prevent HF artefacts. I'll post some samples a bit later. Jean-Marc On 10-10-27 05:13 PM, Jonathan Rosenberg wrote: > Absolutely, these discussions should happen here. This working group is > about producing this codec, and technical discussions about that > deliverable belong here. > > -Jonathan R. > (with WG chair hat on) > > On 10/27/2010 6:14 AM, Stephen Botzko wrote: >> I think it'd be good to have development discussions like this on the >> list. >> >> Stephen Botzko >> >> On Wed, Oct 27, 2010 at 2:17 AM, Raymond (Juin-Hwey) Chen >> > wrote: >> >> Hi Jean-Marc, >> >> Are you saying the problem is in the pitch estimation for that test >> file rather than the pitch filtering? >> >> I will take a look at that file. Is that a raw file? I am not >> familiar with the ".sw" file extension. >> >> I could hear audible improvement when I tried the method on the >> other solo instrument files that I am working on, although the >> degree of improvement depends on which file it is, as expected. >> >> I am starting to wonder how many people in the codec WG email list >> still want to continue to see our low-level technical discussions >> about the algorithm development in the reflector. I wonder if some >> people in the codec WG email list would prefer not to be bombarded >> with such emails. Should we consider moving such detailed codec >> algorithm development discussions off the WG mailing list and just >> between those of us who are interested or are directly involved? >> >> Or is this how IETF supposed to work (discussing all technical >> development in emails)? >> >> Thanks. >> >> Raymond >> >> -----Original Message----- >> From: Jean-Marc Valin [mailto:jean-marc.valin@octasic.com >> ] >> Sent: Tuesday, October 26, 2010 11:23 AM >> To: Raymond (Juin-Hwey) Chen >> Cc: Jean-Marc Valin; 'codec@ietf.org ' >> Subject: Re: [codec] A pitch filtering method to improve the mode 1 >> CELT codec >> >> Hi Raymond, >> >> I've been able to reproduce results that are mostly similar to what >> you got >> earlier on the trumpet file. Unfortunately, it doesn't work as well on >> other files I'm trying. For example on this one >> (http://jmvalin.ca/misc_stuff/s02a_102.sw) there is simply too much >> pitch >> doubling and I'm not actually sure it's possible to do a good job. Could >> you try with your setup and see if you can improve things. I think >> one of >> the tricky thing with the postfilter will be to make it work in all >> circumstances and -- most importantly -- prevent it from causing harm in >> cases where it cannot help. >> >> Regarding the periods we'd need to support, I think the important >> question >> is not that much "on what range does it help", but "on what range is it >> necessary". From what I observed, CELT copes well with regular male and >> even female speech and even the vega sample tends to sound OK. It's >> only on >> strongly tonal samples with higher pitches that problems occur. The main >> reason I want to limit the maximum period allowed is that there's a >> memory >> footprint issue that's directly related. In any case, we'll need to >> experiment as with anything else. >> >> Cheers, >> >> Jean-Marc >> >> On 10-10-26 02:40 AM, Raymond (Juin-Hwey) Chen wrote: >> > Hi Jean-Marc, >> > >> > See my in-line comments below. >> > >> > You wrote: >> >> My concern wasn't that much about the period, but the gain. I'll >> >> probably have to try both approaches, but I was mainly concerned >> that >> >> a non-harmonic signal with one strong sinusoid could cause the comb >> >> filter to be turned on with a high gain. >> > >> > [Raymond]: I see your concern with introducing artificial >> harmonics for >> > a single sinusoid. On the other hand, my concern is that if you >> derive >> > the gain (or taps) based on partially whitened input signal, the >> gain or >> > taps may be too small for legitimate periodic music or speech >> signals. >> > To address both your and my concerns, I think it is probably best >> to use >> > a 3-tap filter where the taps are derived directly from the input >> signal >> > without whitening. Since a 3-tap filter has the ability to taper off >> > the degree of comb filtering with increasing frequency, a set of 3 >> > filter taps that is chosen to match the input spectrum can >> probably have >> > a reasonable spectral match for either a single tone or music/speech >> > with a harmonic structure. The BroadVoice codecs use a 3-tap pitch >> > predictor. The way those 3 taps are jointly searched and vector >> > quantized might provide a way to quantize the 3 taps and do a >> spectral >> > match at the same time for our current pitch prefilter/postfilter >> > approa ch. We need to experiment to find out. The BroadVoice >> codebooks >> > for the 3 taps were designed using mainly speech. If we adopt this >> > approach, we need to redesign the codebook with sufficient music >> signals >> > in the training set. If you are particularly concerned with a single >> > sinusoid, we may want to train one or a few 3-tap codevectors >> using just >> > sinusoids. If this doesn't work well, in the worst case, we could >> > potentially design a detector that detect (nearly) periodic >> time-domain >> > waveform that does not have a spectrum with a harmonic structure >> (like a >> > single sinusoid) and turn off the prefilter/postfilter accordingly. >> > >> >> CELT currently has a pitch predictor used for PLC, but I doubt it's >> >> very good for this as it is now. OTOH, it's fairly low >> complexity too. >> >> It's part of my plans to compare it with the one in BV16/32. I >> assume >> >> that Broadcom would be willing to relicense it to the simplified BSD >> >> license (it's currently LGPL) if we wanted to use it instead of the >> >> current CELT estimator. >> > >> > [Raymond]: Broadcom's legal counsel will formally answer that >> question >> > soon. My assumption is that if we were previously willing to allow >> > royalty-free use of BroadVoice in the IETF codec, then allowing >> certain >> > components of it to be used in the IETF codec under IETF's licensing >> > terms is probably not a problem, but this is just my personal thought >> > and not an official Broadcom position yet. I will talk to our lawyer >> > and let him give the formal answer. >> > >> >> The pitch range is another thing that needs experimenting with, but >> >> my intuition here is that it's probably not necessary to go >> below 100 >> >> Hz because for these very low pitches, there is enough masking >> between >> >> the harmonics to make the coding noise inaudible even without a >> >> post-filter. Again worth testing. >> > >> > [Raymond]: I agree that lower pitch frequency generally has >> better noise >> > masking, so time-domain predictive codecs often sound noisier for >> female >> > speech than for male speech. However, my previous work with pitch >> > postfiltering during my Ph.D. study showed that pitch postfiltering >> > alone could still give audible quality improvement to >> low-bit-rate-coded >> > male speech. Does it improve quality when the pitch frequency is >> below >> > 100 Hz? I don't remember the pitch frequencies of my pitch postfilter >> > tests. I agree it is worth testing. >> > >> > >> > _______________________________________________ codec mailing list >> > codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >> >> >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec >> >> >> >> >> _______________________________________________ >> codec mailing list >> codec@ietf.org >> https://www.ietf.org/mailman/listinfo/codec > > -- > Jonathan D. Rosenberg, Ph.D. SkypeID: jdrosen > Chief Technology Strategist Mobile: +1 (732) 766-2496 > Skype SkypeIn: +1 (408) 465-0361 > jdrosen@skype.net http://www.skype.com > jdrosen@jdrosen.net http://www.jdrosen.net > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec From koen.vos@skype.net Wed Oct 27 18:19:45 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id AE9E53A67AF for ; Wed, 27 Oct 2010 18:19:45 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.413 X-Spam-Level: X-Spam-Status: No, score=-2.413 tagged_above=-999 required=5 tests=[AWL=0.186, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ZuhVzzxLFXI8 for ; Wed, 27 Oct 2010 18:19:43 -0700 (PDT) Received: from mx.skype.net (mx.skype.net [78.141.177.88]) by core3.amsl.com (Postfix) with ESMTP id E51F93A67A6 for ; Wed, 27 Oct 2010 18:19:42 -0700 (PDT) Received: from mx.skype.net (localhost [127.0.0.1]) by mx.skype.net (Postfix) with ESMTP id D66E61711; Thu, 28 Oct 2010 03:21:32 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=skype.net; h=date:from:to :cc:message-id:in-reply-to:subject:mime-version:content-type: content-transfer-encoding; s=mx; bh=yLWzxR0Iv7orObEpVHkNJpRVbBA= ; b=H/c5VjCOyR04KvUrfKFqmkkhkF+gsT9dZtgBqXA3qNwXb9zLTI0aM9NPMtnE AtzWOZmHE9Xpfk0mABC9l/Og65R4cCXr0lK5awzNjo6Ktm+a1wTAPV7mxfJcEoIa B/W3D1CJmOupMXOuNC9Qgvxvqt1PMcl+5iouCvbGzL8HfjA= DomainKey-Signature: a=rsa-sha1; c=nofws; d=skype.net; h=date:from:to:cc :message-id:in-reply-to:subject:mime-version:content-type: content-transfer-encoding; q=dns; s=mx; b=UCOCywpKR1vvv+Z30Y06z+ qaEOG1He5Mg2UGLBFqb+6ybjHDn5/JVdQlrFCJpCY9L83t6nXASEU5ADQZpXoYhq 81AV123I/NcI1yqjnNnWtCDtg2WutBI+r0Gj/zJ+noA48JnhJD0dNYv+BltMlZRB et4Eq3EmkWVoTF2luCn18= Received: from zimbra.skype.net (zimbra.skype.net [78.141.177.82]) by mx.skype.net (Postfix) with ESMTP id CB4881708; Thu, 28 Oct 2010 03:21:32 +0200 (CEST) Received: from localhost (localhost [127.0.0.1]) by zimbra.skype.net (Postfix) with ESMTP id AAA913507698; Thu, 28 Oct 2010 03:21:32 +0200 (CEST) X-Virus-Scanned: amavisd-new at lu2-zimbra.skype.net Received: from zimbra.skype.net ([127.0.0.1]) by localhost (zimbra.skype.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 2LG7gyEf9RVe; Thu, 28 Oct 2010 03:21:31 +0200 (CEST) Received: from zimbra.skype.net (lu2-zimbra.skype.net [78.141.177.82]) by zimbra.skype.net (Postfix) with ESMTP id 4697B3507632; Thu, 28 Oct 2010 03:21:31 +0200 (CEST) Date: Thu, 28 Oct 2010 03:21:31 +0200 (CEST) From: Koen Vos To: Jean-Marc Valin Message-ID: <207365714.1182401.1288228891167.JavaMail.root@lu2-zimbra> In-Reply-To: <1795514592.1182399.1288228845673.JavaMail.root@lu2-zimbra> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [69.181.192.115] X-Mailer: Zimbra 6.0.6_GA_2330.UBUNTU8_64 (ZimbraWebClient - FF3.0 (Win)/6.0.6_GA_2330.UBUNTU8_64) Cc: "codec@ietf.org" , Jean-Marc Valin Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Oct 2010 01:19:45 -0000 The need for reliable and consistent pitch estimates is there for SILK as well: SILK uses pitch lags for both harmonic noise shaping and prediction, using the same kind of 3-tap LTP pre/post filter. However, the pitch estimator may have different estimation sensitivites for audio and speech signals. So it may be useful to play around with these settings: PE_FLP_SHORTLAG_BIAS /* bias towards shorter pitch lags */ PE_FLP_PREVLAG_BIAS /* bias towards the lag of the previous frame */ It would be trivial to make these part of the state, so that CELT and SILK can have different bias values. SILK's pitch estimator is pretty simple: - partial LPC prewhitening, where the LPC order (6...16) depends on the complexity setting. - 4x downsampled domain: full correlation search, keep a few lags with high correlation - 2x downsampled domain: fine search with bias towards the previous pitch lag and short lags, keep winner - input domain: fine search around winner, allowing gradual pitch variations over subframes. I'm sure it can be improved, which would probably benefit CELT and SILK. best, koen. ----- Original Message ----- From: "Jean-Marc Valin" To: "Raymond (Juin-Hwey) Chen" Cc: "codec@ietf.org" , "Jean-Marc Valin" Sent: Wednesday, October 27, 2010 6:13:07 AM Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec Hi Raymond, On 10-10-27 02:17 AM, Raymond (Juin-Hwey) Chen wrote: > Are you saying the problem is in the pitch estimation for that test > file rather than the pitch filtering? The fundamental problem is definitely with the estimation. Reliable and consistent pitch estimation is hard. The question is thus whether it's possible to either have a reliable enough estimator *or* have a robust enough filtering scheme that it can tolerate occasional errors in the estimation. I'm still working on both. > I will take a look at that file. Is that a raw file? I am not familiar > with the ".sw" file extension. That's just a 48 kHz raw 16-bit PCM file (little endian). sw is just the sox program's extension for "signed word". > I could hear audible improvement when I tried the method on the other > solo instrument files that I am working on, although the degree of > improvement depends on which file it is, as expected. Yes, I'm also observing that. What I mainly want to improve is how it works on more complicated signals (like the one I linked to), or non-harmonic signals. As long as we're in a situation where it improves on a subset of the samples while not degrading the others, I'm happy. > I am starting to wonder how many people in the codec WG email list > still want to continue to see our low-level technical discussions > about the > algorithm development in the reflector. I wonder if some people in the > codec WG email list would prefer not to be bombarded with such emails. > Should we consider moving such detailed codec algorithm development > discussions off the WG mailing list and just between those of us who > are interested or are directly involved? > > Or is this how IETF supposed to work (discussing all technical > development in emails)? I think we should keep having this discussion on the list. The IETF lists are normally meant for technical discussions so this just helps improve the SNR :-) Cheers, Jean-Marc _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec From rchen@broadcom.com Wed Oct 27 22:43:30 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id B3EA23A6807 for ; Wed, 27 Oct 2010 22:43:29 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.4 X-Spam-Level: X-Spam-Status: No, score=-2.4 tagged_above=-999 required=5 tests=[AWL=0.199, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id vqqqCnxcvUkt for ; Wed, 27 Oct 2010 22:43:24 -0700 (PDT) Received: from mms1.broadcom.com (mms1.broadcom.com [216.31.210.17]) by core3.amsl.com (Postfix) with ESMTP id 45E5A3A67D7 for ; Wed, 27 Oct 2010 22:42:06 -0700 (PDT) Received: from [10.9.200.133] by mms1.broadcom.com with ESMTP (Broadcom SMTP Relay (Email Firewall v6.3.2)); Wed, 27 Oct 2010 22:42:52 -0700 X-Server-Uuid: 02CED230-5797-4B57-9875-D5D2FEE4708A Received: from IRVEXCHCCR01.corp.ad.broadcom.com ([10.252.49.30]) by IRVEXCHHUB02.corp.ad.broadcom.com ([10.9.200.133]) with mapi; Wed, 27 Oct 2010 22:44:10 -0700 From: "Raymond (Juin-Hwey) Chen" To: "Jonathan Rosenberg" , "codec@ietf.org" Importance: low X-Priority: 5 Date: Wed, 27 Oct 2010 22:42:51 -0700 Thread-Topic: [codec] A pitch filtering method to improve the mode 1 CELT codec Thread-Index: Act2G+d2YEVXslPGSQ6MCORhAhuukQARvf+w Message-ID: References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC37F42.2040103@usherbrooke.ca> <4CC71C7A.3090002@octasic.com> <4CC89610.70107@jdrosen.net> In-Reply-To: <4CC89610.70107@jdrosen.net> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US MIME-Version: 1.0 X-WSS-ID: 60D7D2D647840374212-01-01 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Oct 2010 05:43:30 -0000 Hi Jonathan, Thanks for the instruction. Will do. Raymond -----Original Message----- From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of J= onathan Rosenberg Sent: Wednesday, October 27, 2010 2:14 PM To: codec@ietf.org Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT co= dec Absolutely, these discussions should happen here. This working group is=20 about producing this codec, and technical discussions about that=20 deliverable belong here. -Jonathan R. (with WG chair hat on) On 10/27/2010 6:14 AM, Stephen Botzko wrote: > I think it'd be good to have development discussions like this on the lis= t. > > Stephen Botzko > > On Wed, Oct 27, 2010 at 2:17 AM, Raymond (Juin-Hwey) Chen > > wrote: > > Hi Jean-Marc, > > Are you saying the problem is in the pitch estimation for that test > file rather than the pitch filtering? > > I will take a look at that file. Is that a raw file? I am not > familiar with the ".sw" file extension. > > I could hear audible improvement when I tried the method on the > other solo instrument files that I am working on, although the > degree of improvement depends on which file it is, as expected. > > I am starting to wonder how many people in the codec WG email list > still want to continue to see our low-level technical discussions > about the algorithm development in the reflector. I wonder if some > people in the codec WG email list would prefer not to be bombarded > with such emails. Should we consider moving such detailed codec > algorithm development discussions off the WG mailing list and just > between those of us who are interested or are directly involved? > > Or is this how IETF supposed to work (discussing all technical > development in emails)? > > Thanks. > > Raymond > > -----Original Message----- > From: Jean-Marc Valin [mailto:jean-marc.valin@octasic.com > ] > Sent: Tuesday, October 26, 2010 11:23 AM > To: Raymond (Juin-Hwey) Chen > Cc: Jean-Marc Valin; 'codec@ietf.org ' > Subject: Re: [codec] A pitch filtering method to improve the mode 1 > CELT codec > > Hi Raymond, > > I've been able to reproduce results that are mostly similar to what > you got > earlier on the trumpet file. Unfortunately, it doesn't work as well o= n > other files I'm trying. For example on this one > (http://jmvalin.ca/misc_stuff/s02a_102.sw) there is simply too much > pitch > doubling and I'm not actually sure it's possible to do a good job. Co= uld > you try with your setup and see if you can improve things. I think > one of > the tricky thing with the postfilter will be to make it work in all > circumstances and -- most importantly -- prevent it from causing harm= in > cases where it cannot help. > > Regarding the periods we'd need to support, I think the important > question > is not that much "on what range does it help", but "on what range is = it > necessary". From what I observed, CELT copes well with regular male a= nd > even female speech and even the vega sample tends to sound OK. It's > only on > strongly tonal samples with higher pitches that problems occur. The m= ain > reason I want to limit the maximum period allowed is that there's a > memory > footprint issue that's directly related. In any case, we'll need to > experiment as with anything else. > > Cheers, > > Jean-Marc > > On 10-10-26 02:40 AM, Raymond (Juin-Hwey) Chen wrote: > > Hi Jean-Marc, > > > > See my in-line comments below. > > > > You wrote: > >> My concern wasn't that much about the period, but the gain. I'll > >> probably have to try both approaches, but I was mainly concerned > that > >> a non-harmonic signal with one strong sinusoid could cause the co= mb > >> filter to be turned on with a high gain. > > > > [Raymond]: I see your concern with introducing artificial > harmonics for > > a single sinusoid. On the other hand, my concern is that if you > derive > > the gain (or taps) based on partially whitened input signal, the > gain or > > taps may be too small for legitimate periodic music or speech > signals. > > To address both your and my concerns, I think it is probably best > to use > > a 3-tap filter where the taps are derived directly from the input > signal > > without whitening. Since a 3-tap filter has the ability to taper = off > > the degree of comb filtering with increasing frequency, a set of 3 > > filter taps that is chosen to match the input spectrum can > probably have > > a reasonable spectral match for either a single tone or music/spee= ch > > with a harmonic structure. The BroadVoice codecs use a 3-tap pitc= h > > predictor. The way those 3 taps are jointly searched and vector > > quantized might provide a way to quantize the 3 taps and do a > spectral > > match at the same time for our current pitch prefilter/postfilter > > approa ch. We need to experiment to find out. The BroadVoice > codebooks > > for the 3 taps were designed using mainly speech. If we adopt thi= s > > approach, we need to redesign the codebook with sufficient music > signals > > in the training set. If you are particularly concerned with a sin= gle > > sinusoid, we may want to train one or a few 3-tap codevectors > using just > > sinusoids. If this doesn't work well, in the worst case, we could > > potentially design a detector that detect (nearly) periodic > time-domain > > waveform that does not have a spectrum with a harmonic structure > (like a > > single sinusoid) and turn off the prefilter/postfilter accordingly= . > > > >> CELT currently has a pitch predictor used for PLC, but I doubt it= 's > >> very good for this as it is now. OTOH, it's fairly low > complexity too. > >> It's part of my plans to compare it with the one in BV16/32. I > assume > >> that Broadcom would be willing to relicense it to the simplified = BSD > >> license (it's currently LGPL) if we wanted to use it instead of t= he > >> current CELT estimator. > > > > [Raymond]: Broadcom's legal counsel will formally answer that > question > > soon. My assumption is that if we were previously willing to allo= w > > royalty-free use of BroadVoice in the IETF codec, then allowing > certain > > components of it to be used in the IETF codec under IETF's licensi= ng > > terms is probably not a problem, but this is just my personal thou= ght > > and not an official Broadcom position yet. I will talk to our law= yer > > and let him give the formal answer. > > > >> The pitch range is another thing that needs experimenting with, b= ut > >> my intuition here is that it's probably not necessary to go > below 100 > >> Hz because for these very low pitches, there is enough masking > between > >> the harmonics to make the coding noise inaudible even without a > >> post-filter. Again worth testing. > > > > [Raymond]: I agree that lower pitch frequency generally has > better noise > > masking, so time-domain predictive codecs often sound noisier for > female > > speech than for male speech. However, my previous work with pitch > > postfiltering during my Ph.D. study showed that pitch postfilterin= g > > alone could still give audible quality improvement to > low-bit-rate-coded > > male speech. Does it improve quality when the pitch frequency is > below > > 100 Hz? I don't remember the pitch frequencies of my pitch postfil= ter > > tests. I agree it is worth testing. > > > > > > _______________________________________________ codec mailing list > > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec > > > > > _______________________________________________ > codec mailing list > codec@ietf.org > https://www.ietf.org/mailman/listinfo/codec --=20 Jonathan D. Rosenberg, Ph.D. SkypeID: jdrosen Chief Technology Strategist Mobile: +1 (732) 766-2496 Skype SkypeIn: +1 (408) 465-0361 jdrosen@skype.net http://www.skype.com jdrosen@jdrosen.net http://www.jdrosen.net _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec From rchen@broadcom.com Wed Oct 27 23:12:48 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 198223A683B for ; Wed, 27 Oct 2010 23:12:45 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.416 X-Spam-Level: X-Spam-Status: No, score=-2.416 tagged_above=-999 required=5 tests=[AWL=0.183, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9T+9i6FiWBHs for ; Wed, 27 Oct 2010 23:11:05 -0700 (PDT) Received: from mms1.broadcom.com (mms1.broadcom.com [216.31.210.17]) by core3.amsl.com (Postfix) with ESMTP id 604933A6864 for ; Wed, 27 Oct 2010 23:08:57 -0700 (PDT) Received: from [10.9.200.133] by mms1.broadcom.com with ESMTP (Broadcom SMTP Relay (Email Firewall v6.3.2)); Wed, 27 Oct 2010 23:10:24 -0700 X-Server-Uuid: 02CED230-5797-4B57-9875-D5D2FEE4708A Received: from IRVEXCHCCR01.corp.ad.broadcom.com ([10.252.49.30]) by IRVEXCHHUB02.corp.ad.broadcom.com ([10.9.200.133]) with mapi; Wed, 27 Oct 2010 23:11:42 -0700 From: "Raymond (Juin-Hwey) Chen" To: "Jean-Marc Valin" Importance: low X-Priority: 5 Date: Wed, 27 Oct 2010 23:10:16 -0700 Thread-Topic: [codec] A pitch filtering method to improve the mode 1 CELT codec Thread-Index: Act12MsMh306ogOTQNaRF6c9N9Sc3AAinEbg Message-ID: References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC37F42.2040103@usherbrooke.ca> <4CC71C7A.3090002@octasic.com> <4CC82563.9070403@octasic.com> In-Reply-To: <4CC82563.9070403@octasic.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-cr-hashedpuzzle: Ayhm C617 F6SU MQwW MVuo Neoc PJk2 QRUs QTXp R5Xp SSbF SzHU VhWH YluE YtjF aL+5; 3; YwBvAGQAZQBjAEAAaQBlAHQAZgAuAG8AcgBnADsAagBlAGEAbgAtAG0AYQByAGMALgB2AGEAbABpAG4AQABvAGMAdABhAHMAaQBjAC4AYwBvAG0AOwBqAGUAYQBuAC0AbQBhAHIAYwAuAHYAYQBsAGkAbgBAAHUAcwBoAGUAcgBiAHIAbwBvAGsAZQAuAGMAYQA=; Sosha1_v1; 7; {9A548B0C-51BF-490D-907B-76FF98E29722}; cgBjAGgAZQBuAEAAYgByAG8AYQBkAGMAbwBtAC4AYwBvAG0A; Thu, 28 Oct 2010 06:10:16 GMT; UgBFADoAIABbAGMAbwBkAGUAYwBdACAAQQAgAHAAaQB0AGMAaAAgAGYAaQBsAHQAZQByAGkAbgBnACAAbQBlAHQAaABvAGQAIAB0AG8AIABpAG0AcAByAG8AdgBlACAAdABoAGUAIABtAG8AZABlACAAMQAgAEMARQBMAFQAIABjAG8AZABlAGMA x-cr-puzzleid: {9A548B0C-51BF-490D-907B-76FF98E29722} acceptlanguage: en-US MIME-Version: 1.0 X-WSS-ID: 60D7CC5A47840423266-01-01 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Cc: "'codec@ietf.org'" , Jean-Marc Valin Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Oct 2010 06:12:49 -0000 Hi Jean-Marc, Thanks for your comments. If the problem is with the pitch estimation, the= good thing is that it does not affect the bit-stream format if we just use= the integer pitch period within a certain range. Even after the bit-strea= m format is frozen, the pitch estimation accuracy can still be improved lat= er as it only affects the encoder side. Therefore, even if we can't get a v= ery reliable pitch estimator for music before the IETF 79 meeting, it doesn= 't necessarily need to hold up the bit-stream definition. Hence, I think the pitch estimation accuracy is not the most urgent issue. = I think the most urgent issue is to design the pitch filter tap(s) codeboo= k entries so they all correspond to stable pitch filters and there are enou= gh variety of filter tap sets in the codebook so they allow us to shape the= envelope of the frequency response of the pitch filters in a wide range fr= om fully periodic (no envelope shaping) to something that effectively turns= off the pitch filter. To finalize the bit-stream, we do need to finalize = such a filter tap codebook. I remember I have seen some papers on pitch estimation techniques for music= signals published at IEEE ICASSP conferences, and from the pitch contour f= igures in the papers, they seem to work well. I think we can first try to = improve the CELT PLC pitch estimator, BroadVoice pitch estimator, or SILK p= itch estimator to see if we can get satisfactory pitch estimation performan= ce for both speech and music with low enough complexity. If not, we can ch= eck those music pitch estimation papers out to see what they have there. W= e can do such pitch estimator improvement after IETF 79 if we can't finish = it before then. Raymond P.S. It just so happens that this week I am completely booked solid during = the weekdays -- a 3.5-day off-site conference followed by a 1.5-day meeting= , so I don't have much time to work on this method this week. This is very= rare for me. Sorry about that. Next week I expect to have more time to w= ork on this. -----Original Message----- From: Jean-Marc Valin [mailto:jean-marc.valin@octasic.com]=20 Sent: Wednesday, October 27, 2010 6:13 AM To: Raymond (Juin-Hwey) Chen Cc: Jean-Marc Valin; 'codec@ietf.org' Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT co= dec Hi Raymond, On 10-10-27 02:17 AM, Raymond (Juin-Hwey) Chen wrote: > Are you saying the problem is in the pitch estimation for that test file > rather than the pitch filtering? The fundamental problem is definitely with the estimation. Reliable and=20 consistent pitch estimation is hard. The question is thus whether it's=20 possible to either have a reliable enough estimator *or* have a robust=20 enough filtering scheme that it can tolerate occasional errors in the=20 estimation. I'm still working on both. > I will take a look at that file. Is that a raw file? I am not familiar > with the ".sw" file extension. That's just a 48 kHz raw 16-bit PCM file (little endian). sw is just the=20 sox program's extension for "signed word". > I could hear audible improvement when I tried the method on the other > solo instrument files that I am working on, although the degree of > improvement depends on which file it is, as expected. Yes, I'm also observing that. What I mainly want to improve is how it works= =20 on more complicated signals (like the one I linked to), or non-harmonic=20 signals. As long as we're in a situation where it improves on a subset of=20 the samples while not degrading the others, I'm happy. > I am starting to wonder how many people in the codec WG email list still > want to continue to see our low-level technical discussions about the > algorithm development in the reflector. I wonder if some people in the > codec WG email list would prefer not to be bombarded with such emails. > Should we consider moving such detailed codec algorithm development > discussions off the WG mailing list and just between those of us who are > interested or are directly involved? > > Or is this how IETF supposed to work (discussing all technical > development in emails)? I think we should keep having this discussion on the list. The IETF lists=20 are normally meant for technical discussions so this just helps improve the= =20 SNR :-) Cheers, Jean-Marc From rchen@broadcom.com Wed Oct 27 23:24:39 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id B8C603A67EF for ; Wed, 27 Oct 2010 23:24:39 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.429 X-Spam-Level: X-Spam-Status: No, score=-2.429 tagged_above=-999 required=5 tests=[AWL=0.170, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id hyAO0I-jOaIw for ; Wed, 27 Oct 2010 23:24:38 -0700 (PDT) Received: from mms1.broadcom.com (mms1.broadcom.com [216.31.210.17]) by core3.amsl.com (Postfix) with ESMTP id 7CFBF3A67D3 for ; Wed, 27 Oct 2010 23:24:38 -0700 (PDT) Received: from [10.9.200.133] by mms1.broadcom.com with ESMTP (Broadcom SMTP Relay (Email Firewall v6.3.2)); Wed, 27 Oct 2010 23:26:21 -0700 X-Server-Uuid: 02CED230-5797-4B57-9875-D5D2FEE4708A Received: from IRVEXCHCCR01.corp.ad.broadcom.com ([10.252.49.30]) by IRVEXCHHUB02.corp.ad.broadcom.com ([10.9.200.133]) with mapi; Wed, 27 Oct 2010 23:27:39 -0700 From: "Raymond (Juin-Hwey) Chen" To: "Jean-Marc Valin" Importance: low X-Priority: 5 Date: Wed, 27 Oct 2010 23:26:12 -0700 Thread-Topic: [codec] A pitch filtering method to improve the mode 1 CELT codec Thread-Index: Act12MsMh306ogOTQNaRF6c9N9Sc3AAinEbgAAEuLMA= Message-ID: References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC37F42.2040103@usherbrooke.ca> <4CC71C7A.3090002@octasic.com> <4CC82563.9070403@octasic.com> In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-cr-hashedpuzzle: DVWI Fj3E HFO+ H2UJ IHkE JzND QlOY Sjs5 XrJ1 XvHE YFk5 YWvt YYfg crOt c0+j dBRy; 3; YwBvAGQAZQBjAEAAaQBlAHQAZgAuAG8AcgBnADsAagBlAGEAbgAtAG0AYQByAGMALgB2AGEAbABpAG4AQABvAGMAdABhAHMAaQBjAC4AYwBvAG0AOwBqAGUAYQBuAC0AbQBhAHIAYwAuAHYAYQBsAGkAbgBAAHUAcwBoAGUAcgBiAHIAbwBvAGsAZQAuAGMAYQA=; Sosha1_v1; 7; {B61804E7-DC49-4141-B752-40E5E5ED1957}; cgBjAGgAZQBuAEAAYgByAG8AYQBkAGMAbwBtAC4AYwBvAG0A; Thu, 28 Oct 2010 06:26:12 GMT; UgBFADoAIABbAGMAbwBkAGUAYwBdACAAQQAgAHAAaQB0AGMAaAAgAGYAaQBsAHQAZQByAGkAbgBnACAAbQBlAHQAaABvAGQAIAB0AG8AIABpAG0AcAByAG8AdgBlACAAdABoAGUAIABtAG8AZABlACAAMQAgAEMARQBMAFQAIABjAG8AZABlAGMA x-cr-puzzleid: {B61804E7-DC49-4141-B752-40E5E5ED1957} acceptlanguage: en-US MIME-Version: 1.0 X-WSS-ID: 60D7C80747840452790-01-01 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Cc: "'codec@ietf.org'" , Jean-Marc Valin Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Oct 2010 06:24:39 -0000 Jean-Marc, I forgot to comment:=20 As long as the filter tap codebook adequately covers the full possible rang= e from purely periodic (no envelope shaping) to effectively turning off the= pitch filter, then even if we can't quite get a perfect pitch estimator or= pitch filter tap determination method that we are completely happy with be= fore IETF 79 meeting, we can improve them later even after the possible pit= ch periods and possible pitch filter tap sets are determined. The improved= pitch estimator and pitch filter tap determination method developed later = will improve the audio quality without changing the bit-stream format. The= y may increase the complexity of the encoder side, but the decoder side rem= ains low complexity and unchanged. And who knows, maybe with sufficient res= earch, it is possible that the pitch period and pitch filter tap determinat= ion can both be improved while reducing the encoder side complexity at the = same time. Raymond -----Original Message----- From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of R= aymond (Juin-Hwey) Chen Sent: Wednesday, October 27, 2010 11:10 PM To: Jean-Marc Valin Cc: 'codec@ietf.org'; Jean-Marc Valin Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT co= dec Importance: Low Hi Jean-Marc, Thanks for your comments. If the problem is with the pitch estimation, the= good thing is that it does not affect the bit-stream format if we just use= the integer pitch period within a certain range. Even after the bit-strea= m format is frozen, the pitch estimation accuracy can still be improved lat= er as it only affects the encoder side. Therefore, even if we can't get a v= ery reliable pitch estimator for music before the IETF 79 meeting, it doesn= 't necessarily need to hold up the bit-stream definition. Hence, I think the pitch estimation accuracy is not the most urgent issue. = I think the most urgent issue is to design the pitch filter tap(s) codeboo= k entries so they all correspond to stable pitch filters and there are enou= gh variety of filter tap sets in the codebook so they allow us to shape the= envelope of the frequency response of the pitch filters in a wide range fr= om fully periodic (no envelope shaping) to something that effectively turns= off the pitch filter. To finalize the bit-stream, we do need to finalize = such a filter tap codebook. I remember I have seen some papers on pitch estimation techniques for music= signals published at IEEE ICASSP conferences, and from the pitch contour f= igures in the papers, they seem to work well. I think we can first try to = improve the CELT PLC pitch estimator, BroadVoice pitch estimator, or SILK p= itch estimator to see if we can get satisfactory pitch estimation performan= ce for both speech and music with low enough complexity. If not, we can ch= eck those music pitch estimation papers out to see what they have there. W= e can do such pitch estimator improvement after IETF 79 if we can't finish = it before then. Raymond P.S. It just so happens that this week I am completely booked solid during = the weekdays -- a 3.5-day off-site conference followed by a 1.5-day meeting= , so I don't have much time to work on this method this week. This is very= rare for me. Sorry about that. Next week I expect to have more time to w= ork on this. -----Original Message----- From: Jean-Marc Valin [mailto:jean-marc.valin@octasic.com]=20 Sent: Wednesday, October 27, 2010 6:13 AM To: Raymond (Juin-Hwey) Chen Cc: Jean-Marc Valin; 'codec@ietf.org' Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT co= dec Hi Raymond, On 10-10-27 02:17 AM, Raymond (Juin-Hwey) Chen wrote: > Are you saying the problem is in the pitch estimation for that test file > rather than the pitch filtering? The fundamental problem is definitely with the estimation. Reliable and=20 consistent pitch estimation is hard. The question is thus whether it's=20 possible to either have a reliable enough estimator *or* have a robust=20 enough filtering scheme that it can tolerate occasional errors in the=20 estimation. I'm still working on both. > I will take a look at that file. Is that a raw file? I am not familiar > with the ".sw" file extension. That's just a 48 kHz raw 16-bit PCM file (little endian). sw is just the=20 sox program's extension for "signed word". > I could hear audible improvement when I tried the method on the other > solo instrument files that I am working on, although the degree of > improvement depends on which file it is, as expected. Yes, I'm also observing that. What I mainly want to improve is how it works= =20 on more complicated signals (like the one I linked to), or non-harmonic=20 signals. As long as we're in a situation where it improves on a subset of=20 the samples while not degrading the others, I'm happy. > I am starting to wonder how many people in the codec WG email list still > want to continue to see our low-level technical discussions about the > algorithm development in the reflector. I wonder if some people in the > codec WG email list would prefer not to be bombarded with such emails. > Should we consider moving such detailed codec algorithm development > discussions off the WG mailing list and just between those of us who are > interested or are directly involved? > > Or is this how IETF supposed to work (discussing all technical > development in emails)? I think we should keep having this discussion on the list. The IETF lists=20 are normally meant for technical discussions so this just helps improve the= =20 SNR :-) Cheers, Jean-Marc _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec From koen.vos@skype.net Thu Oct 28 01:51:30 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id F2BBF3A677E for ; Thu, 28 Oct 2010 01:51:29 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.428 X-Spam-Level: X-Spam-Status: No, score=-2.428 tagged_above=-999 required=5 tests=[AWL=0.171, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id AVHi+K7PkzG3 for ; Thu, 28 Oct 2010 01:51:22 -0700 (PDT) Received: from mx.skype.net (mx.skype.net [78.141.177.88]) by core3.amsl.com (Postfix) with ESMTP id 0BEC13A6831 for ; Thu, 28 Oct 2010 01:51:21 -0700 (PDT) Received: from mx.skype.net (localhost [127.0.0.1]) by mx.skype.net (Postfix) with ESMTP id 6F9B91711; Thu, 28 Oct 2010 10:53:12 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=skype.net; h=date:from:to :cc:message-id:in-reply-to:subject:mime-version:content-type: content-transfer-encoding; s=mx; bh=LOKHzXliSiLKoFP14kZiTB1CgeE= ; b=bpOC6qcye+/RXLiK9kDZ1+yAXTUmiZ4c+srA9Rh3OGmgQ6+rSerxb75QOiCd YHKyAE7nAIJ0MpqauFk/Tw5mmD8LphtC/2xemA7NdKXY14PL85biTrC7YmMdfrCz M9zCiC6kKYLOJrp6bTURATJWyP93sFFyfIq5vNt59nPSjZ8= DomainKey-Signature: a=rsa-sha1; c=nofws; d=skype.net; h=date:from:to:cc :message-id:in-reply-to:subject:mime-version:content-type: content-transfer-encoding; q=dns; s=mx; b=CB95VApW42lQvXFQfw3BL9 Wwko4yf+s4VcgYdX4jXKCOQTW/57cLYO2LPYQE5l8/MOSbjKLXUuklQL6krO4xnW uQcBmTF6J6SB+oOJiG5ccb5bPRjGOEhpsKSLFTnWYo/KPzeJRkLnHRW1buECrlLJ eACSDXiFWNcMU6RCMbQhM= Received: from zimbra.skype.net (zimbra.skype.net [78.141.177.82]) by mx.skype.net (Postfix) with ESMTP id 6DC5B16F3; Thu, 28 Oct 2010 10:53:12 +0200 (CEST) Received: from localhost (localhost [127.0.0.1]) by zimbra.skype.net (Postfix) with ESMTP id 4D4A91672687; Thu, 28 Oct 2010 10:53:12 +0200 (CEST) X-Virus-Scanned: amavisd-new at lu2-zimbra.skype.net Received: from zimbra.skype.net ([127.0.0.1]) by localhost (zimbra.skype.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id DE8KxPg5DAt0; Thu, 28 Oct 2010 10:53:11 +0200 (CEST) Received: from zimbra.skype.net (lu2-zimbra.skype.net [78.141.177.82]) by zimbra.skype.net (Postfix) with ESMTP id 19CE41672684; Thu, 28 Oct 2010 10:53:11 +0200 (CEST) Date: Thu, 28 Oct 2010 10:53:10 +0200 (CEST) From: Koen Vos To: "Raymond (Juin-Hwey) Chen" Message-ID: <816830837.1188990.1288255990888.JavaMail.root@lu2-zimbra> In-Reply-To: <1629993190.1188939.1288255860624.JavaMail.root@lu2-zimbra> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [69.181.192.115] X-Mailer: Zimbra 6.0.6_GA_2330.UBUNTU8_64 (ZimbraWebClient - FF3.0 (Win)/6.0.6_GA_2330.UBUNTU8_64) Cc: "codec@ietf.org" , Jean-Marc Valin Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Oct 2010 08:51:30 -0000 I guess pitch filtering hasn't been used for music before because very little music is actually anywhere near harmonic. Most tonal music contains chords: multiple tones with different fundamental frequencies. For these, pitch filtering easily causes more harm than anything else. But it is true that harmonic sounds like the harpsichord are among the trickiest to code. An interesting aspect of string-generated tones is that their harmonic frequencies are not exactly integer multiples, because of the stiffness of the string: http://en.wikipedia.org/wiki/Inharmonicity I remember from previous experiments with pitch filters that for signals like the harpsichord this effect is quite strong. Fortunately it's possible to create a pitch filter with similarly "inharmonic" harmonics: just add an allpass filter to the delay line. The allpass filter creates a shorter delay for higher frequencies, which moves the higher harmonics up a bit. The allpass coefficient can be fairly easily estimated with for instance a gradient descent algorithm, or even by just trying out a few values and chosing the one with the highest prediction gain. Might be worth an experiment. The downside is of course that the inharmonicity coefficient will need to be coded and transmitted. Another idea is to use a slighlty lower gain in the pitch prefilter than in the postfilter. This will make the decoder emphasize the harmonics a bit, which creates an effect similar to a traditional harmonic postfilter - for free. This may work especially well for speech coded at below-transparent quality. koen. ----- Original Message ----- From: "Raymond (Juin-Hwey) Chen" To: "Jean-Marc Valin" Cc: "codec@ietf.org" , "Jean-Marc Valin" Sent: Wednesday, October 27, 2010 11:26:12 PM Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec Jean-Marc, I forgot to comment: As long as the filter tap codebook adequately covers the full possible range from purely periodic (no envelope shaping) to effectively turning off the pitch filter, then even if we can't quite get a perfect pitch estimator or pitch filter tap determination method that we are completely happy with before IETF 79 meeting, we can improve them later even after the possible pitch periods and possible pitch filter tap sets are determined. The improved pitch estimator and pitch filter tap determination method developed later will improve the audio quality without changing the bit-stream format. They may increase the complexity of the encoder side, but the decoder side remains low complexity and unchanged. And who knows, maybe with sufficient research, it is possible that the pitch period and pitch filter tap determination can both be improved while reducing the encoder side complexity at the same time. Raymond -----Original Message----- From: codec-bounces@ietf.org [mailto:codec-bounces@ietf.org] On Behalf Of Raymond (Juin-Hwey) Chen Sent: Wednesday, October 27, 2010 11:10 PM To: Jean-Marc Valin Cc: 'codec@ietf.org'; Jean-Marc Valin Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec Importance: Low Hi Jean-Marc, Thanks for your comments. If the problem is with the pitch estimation, the good thing is that it does not affect the bit-stream format if we just use the integer pitch period within a certain range. Even after the bit-stream format is frozen, the pitch estimation accuracy can still be improved later as it only affects the encoder side. Therefore, even if we can't get a very reliable pitch estimator for music before the IETF 79 meeting, it doesn't necessarily need to hold up the bit-stream definition. Hence, I think the pitch estimation accuracy is not the most urgent issue. I think the most urgent issue is to design the pitch filter tap(s) codebook entries so they all correspond to stable pitch filters and there are enough variety of filter tap sets in the codebook so they allow us to shape the envelope of the frequency response of the pitch filters in a wide range from fully periodic (no envelope shaping) to something that effectively turns off the pitch filter. To finalize the bit-stream, we do need to finalize such a filter tap codebook. I remember I have seen some papers on pitch estimation techniques for music signals published at IEEE ICASSP conferences, and from the pitch contour figures in the papers, they seem to work well. I think we can first try to improve the CELT PLC pitch estimator, BroadVoice pitch estimator, or SILK pitch estimator to see if we can get satisfactory pitch estimation performance for both speech and music with low enough complexity. If not, we can check those music pitch estimation papers out to see what they have there. We can do such pitch estimator improvement after IETF 79 if we can't finish it before then. Raymond P.S. It just so happens that this week I am completely booked solid during the weekdays -- a 3.5-day off-site conference followed by a 1.5-day meeting, so I don't have much time to work on this method this week. This is very rare for me. Sorry about that. Next week I expect to have more time to work on this. -----Original Message----- From: Jean-Marc Valin [mailto:jean-marc.valin@octasic.com] Sent: Wednesday, October 27, 2010 6:13 AM To: Raymond (Juin-Hwey) Chen Cc: Jean-Marc Valin; 'codec@ietf.org' Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec Hi Raymond, On 10-10-27 02:17 AM, Raymond (Juin-Hwey) Chen wrote: > Are you saying the problem is in the pitch estimation for that test > file rather than the pitch filtering? The fundamental problem is definitely with the estimation. Reliable and consistent pitch estimation is hard. The question is thus whether it's possible to either have a reliable enough estimator *or* have a robust enough filtering scheme that it can tolerate occasional errors in the estimation. I'm still working on both. > I will take a look at that file. Is that a raw file? I am not familiar > with the ".sw" file extension. That's just a 48 kHz raw 16-bit PCM file (little endian). sw is just the sox program's extension for "signed word". > I could hear audible improvement when I tried the method on the other > solo instrument files that I am working on, although the degree of > improvement depends on which file it is, as expected. Yes, I'm also observing that. What I mainly want to improve is how it works on more complicated signals (like the one I linked to), or non-harmonic signals. As long as we're in a situation where it improves on a subset of the samples while not degrading the others, I'm happy. > I am starting to wonder how many people in the codec WG email list > still want to continue to see our low-level technical discussions > about the > algorithm development in the reflector. I wonder if some people in the > codec WG email list would prefer not to be bombarded with such emails. > Should we consider moving such detailed codec algorithm development > discussions off the WG mailing list and just between those of us who > are interested or are directly involved? > > Or is this how IETF supposed to work (discussing all technical > development in emails)? I think we should keep having this discussion on the list. The IETF lists are normally meant for technical discussions so this just helps improve the SNR :-) Cheers, Jean-Marc _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec _______________________________________________ codec mailing list codec@ietf.org https://www.ietf.org/mailman/listinfo/codec From xiphmont@gmail.com Thu Oct 28 02:45:19 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 1F50F3A689F for ; Thu, 28 Oct 2010 02:45:19 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.599 X-Spam-Level: X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id eEO9aSm5y6Mt for ; Thu, 28 Oct 2010 02:45:18 -0700 (PDT) Received: from mail-wy0-f172.google.com (mail-wy0-f172.google.com [74.125.82.172]) by core3.amsl.com (Postfix) with ESMTP id E3ED43A689D for ; Thu, 28 Oct 2010 02:45:17 -0700 (PDT) Received: by wyb28 with SMTP id 28so1666607wyb.31 for ; Thu, 28 Oct 2010 02:47:09 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type; bh=W0sbEC+3Ih27igHGSqRhns7bmyHICPD3ALC0v5fAMvc=; b=vOChxQaOOpAdGcJ192lSYAIye2dcvOw2Ip030BgAz1hCSgE5wV4ev5ZEXmrhaBCqZ8 Ss9wi/SzhBG0orbU2cCyvXRKYdTOXRLkVsBtck7pWFuo1PHkvj6jhu1ObimML/A69E63 7P/ykZvk6q9yf/I4LA/h58Uacp6b4eW5fzDsM= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=qxK4yzZgpBAS4xoRe4CqeyZMZjJ+5JNj+s9KkRV78how+1CX/PZOmmj938HoR52EVa Yp8m6Gal4ClkTCZjFCLJVVCc7rwM6FthyP/wlzP3k1oEFhtX5gJAXMidWLNMGuG5QbFx md7L1W7LNUkpMBmAQUtz5A+EgmAVCH0o8Xz60= MIME-Version: 1.0 Received: by 10.227.156.69 with SMTP id v5mr9964530wbw.189.1288259228888; Thu, 28 Oct 2010 02:47:08 -0700 (PDT) Received: by 10.216.86.142 with HTTP; Thu, 28 Oct 2010 02:47:08 -0700 (PDT) In-Reply-To: <207365714.1182401.1288228891167.JavaMail.root@lu2-zimbra> References: <1795514592.1182399.1288228845673.JavaMail.root@lu2-zimbra> <207365714.1182401.1288228891167.JavaMail.root@lu2-zimbra> Date: Thu, 28 Oct 2010 05:47:08 -0400 Message-ID: From: Monty Montgomery To: Koen Vos Content-Type: text/plain; charset=ISO-8859-1 Cc: "codec@ietf.org" , Jean-Marc Valin Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Oct 2010 09:45:19 -0000 On Wed, Oct 27, 2010 at 9:21 PM, Koen Vos wrote: > The need for reliable and consistent pitch estimates is there for SILK as well: Indeed, this problem, and the problem of a harmonic filter that does not cause harn for non-harmonic (chord) spectra is central to the needs of Ghost development as well. Monty From jean-marc.valin@usherbrooke.ca Thu Oct 28 04:06:40 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 87EFB3A67E1 for ; Thu, 28 Oct 2010 04:06:40 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.228 X-Spam-Level: X-Spam-Status: No, score=-2.228 tagged_above=-999 required=5 tests=[AWL=0.372, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4us1I9TxpbOS for ; Thu, 28 Oct 2010 04:06:38 -0700 (PDT) Received: from relais.videotron.ca (relais.videotron.ca [24.201.245.36]) by core3.amsl.com (Postfix) with ESMTP id C40363A683C for ; Thu, 28 Oct 2010 04:06:37 -0700 (PDT) MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; charset=ISO-8859-1 Received: from [192.168.1.14] ([70.81.109.112]) by vl-mh-mrz21.ip.videotron.ca (Sun Java(tm) System Messaging Server 6.3-8.01 (built Dec 16 2008; 32bit)) with ESMTP id <0LAZ00276YXUZGA0@vl-mh-mrz21.ip.videotron.ca> for codec@ietf.org; Thu, 28 Oct 2010 07:08:19 -0400 (EDT) Message-id: <4CC959D2.3060607@usherbrooke.ca> Date: Thu, 28 Oct 2010 07:09:06 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.14) Gecko/20101006 Thunderbird/3.0.9 To: "Raymond (Juin-Hwey) Chen" References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC37F42.2040103@usherbrooke.ca> <4CC71C7A.3090002@octasic.com> <4CC82563.9070403@octasic.com> In-reply-to: X-Enigmail-Version: 1.0.1 Cc: "'codec@ietf.org'" Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Oct 2010 11:06:40 -0000 Hi Raymond and others, See comments below. On 10-10-28 02:10 AM, Raymond (Juin-Hwey) Chen wrote: > Thanks for your comments. If the problem is with the pitch > estimation, the good thing is that it does not affect the bit-stream > format if we just use the integer pitch period within a certain > range. Even after the bit-stream format is frozen, the pitch > estimation accuracy can still be improved later as it only affects > the encoder side. Therefore, even if we can't get a very reliable > pitch estimator for music before the IETF 79 meeting, it doesn't > necessarily need to hold up the bit-stream definition. Actually, the good news is that I think I now have a pitch estimator that is good enough to show that the prefilter/postfilter is useful. It's not perfect, but the error rate no longer causes too many artefacts. Most importantly, we have run this on many non-harmonic signals and so far we have not been able to find samples that get noticeably degraded. That was my main fear with this idea and I think it's gone now. > Hence, I think the pitch estimation accuracy is not the most urgent > issue. I think the most urgent issue is to design the pitch filter > tap(s) codebook entries so they all correspond to stable pitch > filters and there are enough variety of filter tap sets in the > codebook so they allow us to shape the envelope of the frequency > response of the pitch filters in a wide range from fully periodic (no > envelope shaping) to something that effectively turns off the pitch > filter. To finalize the bit-stream, we do need to finalize such a > filter tap codebook. I'm currently using a fixed 3-tap structure with coefficients [1, sqrt(3), 1]/(2+sqrt(3) and the frequency response for a gain of 0.5 is the link i sent earlier: http://jmvalin.ca/misc_stuff/postfilter.png One issue we found is that using a single-tap filter would create too HF artefacts -- especially in signals that have a periodic component (i.e trigger the pitch predictor), but would not otherwise need the filtering. So unless we can clearly find a case that doesn't work with this single 3-tap predictor, I'm tempted to only have one set of taps. In general, I think that pitch is less important as frequency goes up for the following reasons: 1) In-harmonicity as Koen pointed out 2) Critical bands are wide enough that there is more masking (so less need for a postfilter) 3) Because of the high SNR in higher bands, there is more danger of changing the total energy of the band, which would be perceptually bad That being said, I'll revise my opinion if we can find classes of signals that don't work well with this fixed predictor and give good results with another predictor that we can't use for "normal" samples. > I remember I have seen some papers on pitch estimation techniques for > music signals published at IEEE ICASSP conferences, and from the > pitch contour figures in the papers, they seem to work well. I think > we can first try to improve the CELT PLC pitch estimator, BroadVoice > pitch estimator, or SILK pitch estimator to see if we can get > satisfactory pitch estimation performance for both speech and music > with low enough complexity. If not, we can check those music pitch > estimation papers out to see what they have there. We can do such > pitch estimator improvement after IETF 79 if we can't finish it > before then. As I sad, I'm no longer too worried about the pitch predictor. I'm currently using the CELT PLC predictor with a "de-doubling" step at the end, so the good news is that the first stages could probably use any predictor. Of course, it would be nice to merge code and end up with a single predictor and we can incorporate ideas from multiple places. And as you say, since this does not affect the bit-stream, it is less urgent indeed. As a note, this is what I'm currently getting with CELT and SILK for the pitch of the tricky sample i linked to earlier: http://jmvalin.ca/misc_stuff/pitch_silk_celt.png Note that this isn't meant to bash the SILK pitch estimator, since although it makes more error (even after I set the biases Koen mentioned) it's probably much better than what I have on speech -- which is what matters most for SILK. Cheers, Jean-Marc > Raymond > > P.S. It just so happens that this week I am completely booked solid > during the weekdays -- a 3.5-day off-site conference followed by a > 1.5-day meeting, so I don't have much time to work on this method > this week. This is very rare for me. Sorry about that. Next week I > expect to have more time to work on this. > > -----Original Message----- From: Jean-Marc Valin > [mailto:jean-marc.valin@octasic.com] Sent: Wednesday, October 27, > 2010 6:13 AM To: Raymond (Juin-Hwey) Chen Cc: Jean-Marc Valin; > 'codec@ietf.org' Subject: Re: [codec] A pitch filtering method to > improve the mode 1 CELT codec > > Hi Raymond, > > On 10-10-27 02:17 AM, Raymond (Juin-Hwey) Chen wrote: >> Are you saying the problem is in the pitch estimation for that test >> file rather than the pitch filtering? > > The fundamental problem is definitely with the estimation. Reliable > and consistent pitch estimation is hard. The question is thus whether > it's possible to either have a reliable enough estimator *or* have a > robust enough filtering scheme that it can tolerate occasional errors > in the estimation. I'm still working on both. > >> I will take a look at that file. Is that a raw file? I am not >> familiar with the ".sw" file extension. > > That's just a 48 kHz raw 16-bit PCM file (little endian). sw is just > the sox program's extension for "signed word". > >> I could hear audible improvement when I tried the method on the >> other solo instrument files that I am working on, although the >> degree of improvement depends on which file it is, as expected. > > Yes, I'm also observing that. What I mainly want to improve is how it > works on more complicated signals (like the one I linked to), or > non-harmonic signals. As long as we're in a situation where it > improves on a subset of the samples while not degrading the others, > I'm happy. > >> I am starting to wonder how many people in the codec WG email list >> still want to continue to see our low-level technical discussions >> about the algorithm development in the reflector. I wonder if some >> people in the codec WG email list would prefer not to be bombarded >> with such emails. Should we consider moving such detailed codec >> algorithm development discussions off the WG mailing list and just >> between those of us who are interested or are directly involved? >> >> Or is this how IETF supposed to work (discussing all technical >> development in emails)? > > I think we should keep having this discussion on the list. The IETF > lists are normally meant for technical discussions so this just helps > improve the SNR :-) > > Cheers, > > Jean-Marc > > > > > From rchen@broadcom.com Thu Oct 28 23:44:43 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 2B0033A6A06 for ; Thu, 28 Oct 2010 23:44:43 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.44 X-Spam-Level: X-Spam-Status: No, score=-2.44 tagged_above=-999 required=5 tests=[AWL=0.159, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id aGyzsFy-OOXu for ; Thu, 28 Oct 2010 23:44:37 -0700 (PDT) Received: from mms1.broadcom.com (mms1.broadcom.com [216.31.210.17]) by core3.amsl.com (Postfix) with ESMTP id B810B3A6A0D for ; Thu, 28 Oct 2010 23:44:36 -0700 (PDT) Received: from [10.9.200.131] by mms1.broadcom.com with ESMTP (Broadcom SMTP Relay (Email Firewall v6.3.2)); Thu, 28 Oct 2010 23:46:18 -0700 X-Server-Uuid: 02CED230-5797-4B57-9875-D5D2FEE4708A Received: from IRVEXCHCCR01.corp.ad.broadcom.com ([10.252.49.30]) by IRVEXCHHUB01.corp.ad.broadcom.com ([10.9.200.131]) with mapi; Thu, 28 Oct 2010 23:46:18 -0700 From: "Raymond (Juin-Hwey) Chen" To: "Jean-Marc Valin" Importance: low X-Priority: 5 Date: Thu, 28 Oct 2010 23:46:10 -0700 Thread-Topic: [codec] A pitch filtering method to improve the mode 1 CELT codec Thread-Index: Act2kHnOUYNKD/BbQpGmg0f6ZbTcCgAnvmqQ Message-ID: References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC37F42.2040103@usherbrooke.ca> <4CC71C7A.3090002@octasic.com> <4CC82563.9070403@octasic.com> <4CC959D2.3060607@usherbrooke.ca> In-Reply-To: <4CC959D2.3060607@usherbrooke.ca> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-cr-hashedpuzzle: UMo= E0/T FGII FxG6 HaV8 H11K JrVj MD7k No6a O94/ PA15 Wrcu XCYS ZrOP b7/e hYKM; 3; YwBvAGQAZQBjAEAAaQBlAHQAZgAuAG8AcgBnADsAagBlAGEAbgAtAG0AYQByAGMALgB2AGEAbABpAG4AQABvAGMAdABhAHMAaQBjAC4AYwBvAG0AOwBqAGUAYQBuAC0AbQBhAHIAYwAuAHYAYQBsAGkAbgBAAHUAcwBoAGUAcgBiAHIAbwBvAGsAZQAuAGMAYQA=; Sosha1_v1; 7; {64F5E350-0960-47B7-BAB0-48467FF92B4C}; cgBjAGgAZQBuAEAAYgByAG8AYQBkAGMAbwBtAC4AYwBvAG0A; Fri, 29 Oct 2010 06:46:10 GMT; UgBFADoAIABbAGMAbwBkAGUAYwBdACAAQQAgAHAAaQB0AGMAaAAgAGYAaQBsAHQAZQByAGkAbgBnACAAbQBlAHQAaABvAGQAIAB0AG8AIABpAG0AcAByAG8AdgBlACAAdABoAGUAIABtAG8AZABlACAAMQAgAEMARQBMAFQAIABjAG8AZABlAGMA x-cr-puzzleid: {64F5E350-0960-47B7-BAB0-48467FF92B4C} acceptlanguage: en-US MIME-Version: 1.0 X-WSS-ID: 60D4B23047842630012-01-01 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Cc: "'codec@ietf.org'" Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Oct 2010 06:44:43 -0000 Hi Jean-Marc, Glad to hear that you now have a pitch estimator that's satisfactory to you= . That a very good news. Regarding your comments about single-tap filter, yes, I mentioned in my ori= ginal email that for some signals it can produce more periodicity in the hi= gh frequencies than in the original signal. However, if you send a mode/fl= ag bit to turn the filter on or off (or make one of the candidate tap weigh= ts equal to zero) and if there is a evaluation process in the encoder side = to determine whether to turn on the pitch filter, then if we decide that a = single-tap pitch filter will degrade quality, we can always turn it off. (T= he evaluation can even be a closed-loop determination by trying both, altho= ugh this will increase the encoding complexity significantly.) I am not ad= vocating using a single-tap pitch filter. I am just saying that if for som= e reason a single-tap pitch filter is used for its simplicity, we don't nec= essarily have to suffer quality degradation because we can always turn it o= ff dynamically for those frames where it can degrade quality. Regarding your comments about a single fixed set of 3 filter taps, are you = planning to send different scaling factor to scale the three taps so as to = control the degree of comb filtering? I agree that generally it makes more= sense to use a 3-tap pitch filter than a single-tap pitch filter as you di= scussed. Having a single fixed set of 3 filter taps (and thus a single kin= d of envelope) may be fine, but don't you think we should have different sc= aling factors to scale these three taps so we can control the degree of com= b filtering to fit different kinds of audio signals? Also, although you can't find any audio signal that this fixed set doesn't = work, what if we later find some signals that this fixed set of taps doesn'= t work well? In a more conservative design, wouldn't you want at least a f= ew different envelope shapes for the frequency response of the pitch filter= s? I am not necessarily proposing some new development direction. I am just t= hrowing some questions for us to think about. Raymond -----Original Message----- From: Jean-Marc Valin [mailto:jean-marc.valin@usherbrooke.ca]=20 Sent: Thursday, October 28, 2010 4:09 AM To: Raymond (Juin-Hwey) Chen Cc: Jean-Marc Valin; 'codec@ietf.org' Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT co= dec Hi Raymond and others, See comments below. On 10-10-28 02:10 AM, Raymond (Juin-Hwey) Chen wrote: > Thanks for your comments. If the problem is with the pitch > estimation, the good thing is that it does not affect the bit-stream > format if we just use the integer pitch period within a certain > range. Even after the bit-stream format is frozen, the pitch > estimation accuracy can still be improved later as it only affects > the encoder side. Therefore, even if we can't get a very reliable > pitch estimator for music before the IETF 79 meeting, it doesn't > necessarily need to hold up the bit-stream definition. Actually, the good news is that I think I now have a pitch estimator that is good enough to show that the prefilter/postfilter is useful. It's not perfect, but the error rate no longer causes too many artefacts. Most importantly, we have run this on many non-harmonic signals and so far we have not been able to find samples that get noticeably degraded. That was my main fear with this idea and I think it's gone now. > Hence, I think the pitch estimation accuracy is not the most urgent > issue. I think the most urgent issue is to design the pitch filter > tap(s) codebook entries so they all correspond to stable pitch > filters and there are enough variety of filter tap sets in the > codebook so they allow us to shape the envelope of the frequency > response of the pitch filters in a wide range from fully periodic (no > envelope shaping) to something that effectively turns off the pitch > filter. To finalize the bit-stream, we do need to finalize such a > filter tap codebook. I'm currently using a fixed 3-tap structure with coefficients [1, sqrt(3), 1]/(2+sqrt(3) and the frequency response for a gain of 0.5 is the link i sent earlier: http://jmvalin.ca/misc_stuff/postfilter.png One issue we found is that using a single-tap filter would create too HF artefacts -- especially in signals that have a periodic component (i.e trigger the pitch predictor), but would not otherwise need the filtering. So unless we can clearly find a case that doesn't work with this single 3-tap predictor, I'm tempted to only have one set of taps. In general, I think that pitch is less important as frequency goes up for the following reasons: 1) In-harmonicity as Koen pointed out 2) Critical bands are wide enough that there is more masking (so less need for a postfilter) 3) Because of the high SNR in higher bands, there is more danger of changing the total energy of the band, which would be perceptually bad That being said, I'll revise my opinion if we can find classes of signals that don't work well with this fixed predictor and give good results with another predictor that we can't use for "normal" samples. > I remember I have seen some papers on pitch estimation techniques for > music signals published at IEEE ICASSP conferences, and from the > pitch contour figures in the papers, they seem to work well. I think > we can first try to improve the CELT PLC pitch estimator, BroadVoice > pitch estimator, or SILK pitch estimator to see if we can get > satisfactory pitch estimation performance for both speech and music > with low enough complexity. If not, we can check those music pitch > estimation papers out to see what they have there. We can do such > pitch estimator improvement after IETF 79 if we can't finish it > before then. As I sad, I'm no longer too worried about the pitch predictor. I'm currently using the CELT PLC predictor with a "de-doubling" step at the end, so the good news is that the first stages could probably use any predictor. Of course, it would be nice to merge code and end up with a single predictor and we can incorporate ideas from multiple places. And as you say, since this does not affect the bit-stream, it is less urgent indeed. As a note, this is what I'm currently getting with CELT and SILK for the pitch of the tricky sample i linked to earlier: http://jmvalin.ca/misc_stuff/pitch_silk_celt.png Note that this isn't meant to bash the SILK pitch estimator, since although it makes more error (even after I set the biases Koen mentioned) it's probably much better than what I have on speech -- which is what matters most for SILK. Cheers, Jean-Marc > Raymond >=20 > P.S. It just so happens that this week I am completely booked solid > during the weekdays -- a 3.5-day off-site conference followed by a > 1.5-day meeting, so I don't have much time to work on this method > this week. This is very rare for me. Sorry about that. Next week I > expect to have more time to work on this. >=20 > -----Original Message----- From: Jean-Marc Valin > [mailto:jean-marc.valin@octasic.com] Sent: Wednesday, October 27, > 2010 6:13 AM To: Raymond (Juin-Hwey) Chen Cc: Jean-Marc Valin; > 'codec@ietf.org' Subject: Re: [codec] A pitch filtering method to > improve the mode 1 CELT codec >=20 > Hi Raymond, >=20 > On 10-10-27 02:17 AM, Raymond (Juin-Hwey) Chen wrote: >> Are you saying the problem is in the pitch estimation for that test >> file rather than the pitch filtering? >=20 > The fundamental problem is definitely with the estimation. Reliable > and consistent pitch estimation is hard. The question is thus whether > it's possible to either have a reliable enough estimator *or* have a > robust enough filtering scheme that it can tolerate occasional errors > in the estimation. I'm still working on both. >=20 >> I will take a look at that file. Is that a raw file? I am not >> familiar with the ".sw" file extension. >=20 > That's just a 48 kHz raw 16-bit PCM file (little endian). sw is just > the sox program's extension for "signed word". >=20 >> I could hear audible improvement when I tried the method on the >> other solo instrument files that I am working on, although the >> degree of improvement depends on which file it is, as expected. >=20 > Yes, I'm also observing that. What I mainly want to improve is how it > works on more complicated signals (like the one I linked to), or > non-harmonic signals. As long as we're in a situation where it > improves on a subset of the samples while not degrading the others, > I'm happy. >=20 >> I am starting to wonder how many people in the codec WG email list >> still want to continue to see our low-level technical discussions >> about the algorithm development in the reflector. I wonder if some >> people in the codec WG email list would prefer not to be bombarded >> with such emails. Should we consider moving such detailed codec >> algorithm development discussions off the WG mailing list and just >> between those of us who are interested or are directly involved? >>=20 >> Or is this how IETF supposed to work (discussing all technical=20 >> development in emails)? >=20 > I think we should keep having this discussion on the list. The IETF > lists are normally meant for technical discussions so this just helps > improve the SNR :-) >=20 > Cheers, >=20 > Jean-Marc >=20 >=20 >=20 >=20 >=20 From jean-marc.valin@usherbrooke.ca Fri Oct 29 03:35:41 2010 Return-Path: X-Original-To: codec@core3.amsl.com Delivered-To: codec@core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id CB45E3A6940 for ; Fri, 29 Oct 2010 03:35:41 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.245 X-Spam-Level: X-Spam-Status: No, score=-2.245 tagged_above=-999 required=5 tests=[AWL=0.354, BAYES_00=-2.599] Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id GrX57iqdivAx for ; Fri, 29 Oct 2010 03:35:40 -0700 (PDT) Received: from relais.videotron.ca (relais.videotron.ca [24.201.245.36]) by core3.amsl.com (Postfix) with ESMTP id 03E713A67A2 for ; Fri, 29 Oct 2010 03:35:39 -0700 (PDT) MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; charset=ISO-8859-1 Received: from [192.168.1.14] ([70.81.109.112]) by VL-MO-MR002.ip.videotron.ca (Sun Java(tm) System Messaging Server 6.3-8.01 (built Dec 16 2008; 32bit)) with ESMTP id <0LB1001K1S6LBS70@VL-MO-MR002.ip.videotron.ca> for codec@ietf.org; Fri, 29 Oct 2010 06:37:33 -0400 (EDT) Message-id: <4CCAA417.9040707@usherbrooke.ca> Date: Fri, 29 Oct 2010 06:38:15 -0400 From: Jean-Marc Valin User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.14) Gecko/20101006 Thunderbird/3.0.9 To: "Raymond (Juin-Hwey) Chen" References: <4CC2DA7E.1060705@usherbrooke.ca> <4CC37F42.2040103@usherbrooke.ca> <4CC71C7A.3090002@octasic.com> <4CC82563.9070403@octasic.com> <4CC959D2.3060607@usherbrooke.ca> In-reply-to: X-Enigmail-Version: 1.0.1 Cc: "'codec@ietf.org'" Subject: Re: [codec] A pitch filtering method to improve the mode 1 CELT codec X-BeenThere: codec@ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Codec WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Oct 2010 10:35:42 -0000 Hi Raymond, On 10-10-29 02:46 AM, Raymond (Juin-Hwey) Chen wrote: > Regarding your comments about single-tap filter, yes, I mentioned in > my original email that for some signals it can produce more > periodicity in the high frequencies than in the original signal. > However, if you send a mode/flag bit to turn the filter on or off (or > make one of the candidate tap weights equal to zero) and if there is > a evaluation process in the encoder side to determine whether to turn > on the pitch filter, then if we decide that a single-tap pitch filter > will degrade quality, we can always turn it off. (The evaluation can > even be a closed-loop determination by trying both, although this > will increase the encoding complexity significantly.) I am not > advocating using a single-tap pitch filter. I am just saying that if > for some reason a single-tap pitch filter is used for its simplicity, > we don't necessarily have to suffer quality degradation because we > can always turn it off dynamically for those frames where it can > degrade quality. At this point, I really don't see what kind of audio signal would benefit from having a one-tap filter and I don't see how we could automatically detect this either in the encoder. If anything I think a 5-tap filter would be more useful, though I'm not sure we really want to go there either. > Regarding your comments about a single fixed set of 3 filter taps, > are you planning to send different scaling factor to scale the three > taps so as to control the degree of comb filtering? I agree that > generally it makes more sense to use a 3-tap pitch filter than a > single-tap pitch filter as you discussed. Having a single fixed set > of 3 filter taps (and thus a single kind of envelope) may be fine, > but don't you think we should have different scaling factors to scale > these three taps so we can control the degree of comb filtering to > fit different kinds of audio signals? Oh, of course. The current code I have has four levels for the "global gain" of that filter (in increments of 0.125), with 0.5 being the largest value value encodable. This is mostly probably not optimal and we would probably want to allow slightly larger gains even if we don't make use of them yet. Feel free to make suggestions here. > Also, although you can't find any audio signal that this fixed set > doesn't work, what if we later find some signals that this fixed set > of taps doesn't work well? In a more conservative design, wouldn't > you want at least a few different envelope shapes for the frequency > response of the pitch filters? I think we're more likely to come across audio that requires a different 3-tap filter than audio that requires a single tap. That being said, I still don't really see what that would look like and there's many possibilities. One issue here is that for small frame sizes, coding these parameters can end up being expensive, so before adding anything we need a reasonable expectation that it will be useful. > I am not necessarily proposing some new development direction. I am > just throwing some questions for us to think about. Sure. I think in the end we're going to just try a variety of audio samples and see if anything needs more than what we're currently doing. I think we will, but I currently have no idea yet in what direction it'll send us. Jean-Marc