From daemon@optimus.ietf.org Sat Mar 2 11:20:03 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA02127 for ; Sat, 2 Mar 2002 11:20:03 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id LAA22250 for rmt-archive@odin.ietf.org; Sat, 2 Mar 2002 11:20:03 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA21692; Sat, 2 Mar 2002 11:01:20 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA21663 for ; Sat, 2 Mar 2002 11:01:18 -0500 (EST) Received: from postal1.lbl.gov (postal1.lbl.gov [128.3.7.82]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA01731 for ; Sat, 2 Mar 2002 11:01:15 -0500 (EST) Received: from SpamWall.lbl.gov (localhost [127.0.0.1]) by postal1.lbl.gov (8.11.2/8.11.2) with ESMTP id g22G1Ha18638 for ; Sat, 2 Mar 2002 08:01:17 -0800 (PST) Received: from hanmir.com ([211.208.94.146]) by SpamWall.lbl.gov (8.11.2/8.11.2) with SMTP id g22G1F818624 for ; Sat, 2 Mar 2002 08:01:16 -0800 (PST) Message-Id: <200203021601.g22G1F818624@SpamWall.lbl.gov> Reply-To: sua95@hanmir.com From: K보트닷컴 To: Mime-Version: 1.0 Content-Type: text/html; charset="ks_c_5601-1987" Date: Sun, 3 Mar 2002 01:00:35 +0900 Subject: [Rmt] [광고] 고무보트&엔진 팝니다!! Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org 광고 메일
본 메일은 발신 전용 광고 메일 입니다.
수신거부를 원 하시면 여기를 클릭하여 주세요!
 
고무보트 / 아웃보드 엔진 전문 싸이버 전시장!! "케이보트닷컴"
_______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Sat Mar 2 13:51:35 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA03259 for ; Sat, 2 Mar 2002 13:51:35 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id NAA28223 for rmt-archive@odin.ietf.org; Sat, 2 Mar 2002 13:51:37 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA27839; Sat, 2 Mar 2002 13:36:35 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA27810 for ; Sat, 2 Mar 2002 13:36:34 -0500 (EST) Received: from postal2.lbl.gov (postal2.lbl.gov [131.243.248.26]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA03136 for ; Sat, 2 Mar 2002 13:36:30 -0500 (EST) Received: from SpamWall.lbl.gov (localhost [127.0.0.1]) by postal2.lbl.gov (8.11.2/8.11.2) with ESMTP id g22IaTv17157 for ; Sat, 2 Mar 2002 10:36:29 -0800 (PST) Received: from simmani.com ([211.200.172.156]) by SpamWall.lbl.gov (8.11.2/8.11.2) with SMTP id g22IZvP17149 for ; Sat, 2 Mar 2002 10:35:58 -0800 (PST) Message-Id: <200203021835.g22IZvP17149@SpamWall.lbl.gov> Reply-To: woo910@simmani.com From: 미화 To: Mime-Version: 1.0 Content-Type: text/html; charset="ks_c_5601-1987" Date: Sun, 3 Mar 2002 03:37:09 +0900 Subject: [Rmt] [광-고] "플라워도사" 2002년 꽃점으로 운세를... Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org ♧♡★ "플라워도사" 가라사대...
   

"그대 나를 모른다면,  운세를 논하지 말라!!"

                           싱그러운 꽃과 로맨스 향기가 넘치는  동양마을!
                           최고의 "꽃 + 로맨스 + 동양철학" 정보,
                           그동안 여러분의 사랑속에 꾸준히 성장해온 <신비의 꽃점>이
                          【플라워도사】이름으로  드디어 인터넷세계에 새롭게  탄생!!
                           2002년,  향기로운 꽃내음과 함께 찾아온 "FlowerDosa"!

                           ----------------------------------------------------------
                           ♧  꽃자리로 알아보는 나의 운명!  ⇒  "탄생화-운명의 꽃자리"
                           ▶  "신비의 주역꽃점"에서  내일의 바람을 점쳐볼까요?
                           ★  과연 당신의 2002년 애정운과 금전운은?  ⇒  "띠&꽃 운세"로!

                           ♡  착한 남자/여자도 바람핀다?^^  ⇒  과연 나의 바람끼 점수는?
                           ▶  지피지기면 백전백승! "당신의 혈액형 특성"을 꼭 알아두세요!
                           ♡  사귀고픈 이성에게 대쉬!  ⇒  효과만점  "로맨스프로포즈노하우"!

                           ★  동양의 섹스경전 "소녀경(素女經)"!  ⇒  "흥미진진 동양철학"
                           ▶  탑스타들의 화려한 꽃자리 운명은?  ⇒  "스타운명 따라잡기!"
                           ♡  "스타 꽃궁합"에서  탑스타들과 궁합을 맞춰볼까요?

                           [신비의 꽃점]은 아름다운 꽃을 통해서 운명의 바람을 알려드립니다.
                            "하늘에 찬란한 별이 빛나고  땅에는 아름다운 꽃이 피어나듯이,
                      사람에게는  따스한 사랑이 있어야 한다."
                       www.FlowerDosa.com
                                                                            ♧  싱그러운 낭만의 향기가 철철!
                                                                                                       플/라/워/도/사
                                             사전 양해없이 메일을 보내서 죄송합니다.
                                             본 메일은 인터넷상에 올라온 메일주소를 발췌하여 발송하였습니다.
                                             본 메일은 정보 통신망 이용 촉진 및 정보보호 등에 관한
                                             법률 제 50조에 의거한 [광고] 메일입니다.
                                             원치 않으시면 삭제하시거나, [수신거부]를 눌러주세요!

Copyright ⓒ 2001-2002 .J&Y 조나단 정보통신. All Rights Reserved.
_______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Sun Mar 3 14:35:50 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28654 for ; Sun, 3 Mar 2002 14:35:50 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id OAA27754 for rmt-archive@odin.ietf.org; Sun, 3 Mar 2002 14:35:52 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA27117; Sun, 3 Mar 2002 14:27:53 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA27089 for ; Sun, 3 Mar 2002 14:27:51 -0500 (EST) Received: from e1.ny.us.ibm.com (e1.ny.us.ibm.com [32.97.182.101]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28479 for ; Sun, 3 Mar 2002 14:27:48 -0500 (EST) Received: from northrelay02.pok.ibm.com (northrelay02.pok.us.ibm.com [9.117.200.22]) by e1.ny.us.ibm.com (8.9.3/8.9.3) with ESMTP id OAA244114 for ; Sun, 3 Mar 2002 14:24:04 -0500 Received: from d27mc103.RCHLAND.IBM.COM (d27mc103.rchland.ibm.com [9.10.226.35]) by northrelay02.pok.ibm.com (8.11.1m3/NCO/VER6.00) with ESMTP id g23JRH8129624 for ; Sun, 3 Mar 2002 14:27:17 -0500 To: rmt@ietf.org X-Mailer: Lotus Notes Release 5.0.5 September 22, 2000 Message-ID: From: "Edward Boden" Date: Sun, 3 Mar 2002 14:27:16 -0500 X-MIMETrack: Serialize by Router on D27mc103/27/M/IBM(Release 5.0.9a |January 7, 2002) at 03/03/2002 01:27:18 PM MIME-Version: 1.0 Content-type: text/plain; charset=us-ascii Subject: [Rmt] spam on rmt Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org I agree with Ross's comments -- lets stop the spam. _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Sun Mar 3 14:36:04 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28670 for ; Sun, 3 Mar 2002 14:36:04 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id OAA27768 for rmt-archive@odin.ietf.org; Sun, 3 Mar 2002 14:36:06 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA26475; Sun, 3 Mar 2002 14:10:19 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA26419 for ; Sun, 3 Mar 2002 14:10:17 -0500 (EST) Received: from ns.live.com (ns.live.com [66.80.62.34]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28219; Sun, 3 Mar 2002 14:10:13 -0500 (EST) Received: (from rsf@localhost) by ns.live.com (8.9.3/8.9.3) id LAA09041; Sun, 3 Mar 2002 11:09:37 -0800 (PST) (envelope-from rsf) Message-Id: <4.3.1.1.20020303110601.00b7af00@localhost> X-Sender: rsf@localhost X-Mailer: QUALCOMM Windows Eudora Version 4.3.1 Date: Sun, 03 Mar 2002 11:09:15 -0800 To: tsvwg@ietf.org, rmt@ietf.org From: Ross Finlayson Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Subject: [Rmt] Enough with the spam on these mailing lists! Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org I'm sure I'm not the only person who's getting really tired of the spam that's being sent to these mailing lists ("tsvwg" and "rmt"). Could the organizer(s) of these mailing list please change them so that only subscribers can post to them? (And please, no more comments about this not being the traditional policy for IETF mailing lists. Times have changed.) Ross. _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Sun Mar 3 15:22:51 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA29204 for ; Sun, 3 Mar 2002 15:22:51 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id PAA29453 for rmt-archive@odin.ietf.org; Sun, 3 Mar 2002 15:22:52 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA28833; Sun, 3 Mar 2002 15:04:26 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA28775 for ; Sun, 3 Mar 2002 15:04:23 -0500 (EST) Received: from mx.webfountain.com (mx.webfountain.com [63.161.54.39]) by ietf.org (8.9.1a/8.9.1a) with SMTP id PAA29062 for ; Sun, 3 Mar 2002 15:04:19 -0500 (EST) Received: (qmail 5134 invoked from network); 3 Mar 2002 20:03:50 -0000 Received: from mail.intranet (10.1.1.37) by mx.webfountain.com with SMTP; 3 Mar 2002 20:03:50 -0000 Received: by mail.intranet (8.12.1/8.12.1/Debian -5) with SMTP id g23K43lk026130; Sun, 3 Mar 2002 12:04:03 -0800 From: "Michael Luby" To: "Ross Finlayson" , , Subject: RE: [Rmt] Enough with the spam on these mailing lists! Date: Sun, 3 Mar 2002 12:04:40 -0800 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4807.1700 In-Reply-To: <4.3.1.1.20020303110601.00b7af00@localhost> Content-Transfer-Encoding: 7bit Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org Content-Transfer-Encoding: 7bit I agree with this. A serious portion of my junk email is now from this list, and I would unsubscribe if I could afford to do this. Most of my other junk email is filtered out, but I can't afford to filter out the RMT alias, so there should be tighter controls on sending to this list. Mike -----Original Message----- From: rmt-admin@ietf.org [mailto:rmt-admin@ietf.org]On Behalf Of Ross Finlayson Sent: Sunday, March 03, 2002 11:09 AM To: tsvwg@ietf.org; rmt@ietf.org Subject: [Rmt] Enough with the spam on these mailing lists! I'm sure I'm not the only person who's getting really tired of the spam that's being sent to these mailing lists ("tsvwg" and "rmt"). Could the organizer(s) of these mailing list please change them so that only subscribers can post to them? (And please, no more comments about this not being the traditional policy for IETF mailing lists. Times have changed.) Ross. _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Sun Mar 3 23:12:15 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA05310 for ; Sun, 3 Mar 2002 23:12:15 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id XAA18185 for rmt-archive@odin.ietf.org; Sun, 3 Mar 2002 23:12:17 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id XAA18030; Sun, 3 Mar 2002 23:05:50 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id XAA17971 for ; Sun, 3 Mar 2002 23:05:47 -0500 (EST) Received: from minotaur.nge.isi.edu (minotaur.nge.isi.edu [65.114.169.202]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA05228; Sun, 3 Mar 2002 23:05:44 -0500 (EST) Received: from minotaur (mankin@localhost) by minotaur.nge.isi.edu (8.11.6/8.11.6) with ESMTP id g2445Wx28342; Sun, 3 Mar 2002 23:05:33 -0500 Message-Id: <200203040405.g2445Wx28342@minotaur.nge.isi.edu> To: "Michael Luby" cc: "Ross Finlayson" , tsvwg@ietf.org, rmt@ietf.org Reply-To: mankin@isi.edu Subject: Re: [Tsvwg] RE: [Rmt] Enough with the spam on these mailing lists! In-reply-to: Your message of Sun, 03 Mar 2002 12:04:40 -0800. Mime-Version: 1.0 (generated by tm-edit 1.7) Content-Type: text/plain; charset=US-ASCII Date: Sun, 03 Mar 2002 23:05:32 -0500 From: Allison Mankin Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org All, I don't have the admin passwords on either list, or I would set up moderation of non-subscriber posts right now. That is becoming the IETF list norm, actually, and it is *well* worth doing. Lorenzo, Roger, Scott...can you do this asap for rmt and tsvwg respectively? Allison _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Mon Mar 4 00:26:49 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA06783 for ; Mon, 4 Mar 2002 00:26:48 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id AAA21919 for rmt-archive@odin.ietf.org; Mon, 4 Mar 2002 00:26:49 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id AAA21423; Mon, 4 Mar 2002 00:16:00 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id AAA20172 for ; Mon, 4 Mar 2002 00:00:18 -0500 (EST) Received: from motgate.mot.com (motgate.mot.com [129.188.136.100]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA06508; Mon, 4 Mar 2002 00:00:16 -0500 (EST) Received: [from mothost.mot.com (mothost.mot.com [129.188.137.101]) by motgate.mot.com (motgate 2.1) with ESMTP id WAA25935; Sun, 3 Mar 2002 22:00:14 -0700 (MST)] Received: [from homer.arc.corp.mot.com (homer.arc.corp.mot.com [10.238.80.38]) by mothost.mot.com (MOT-mothost 2.0) with ESMTP id WAA14763; Sun, 3 Mar 2002 22:00:11 -0700 (MST)] Received: from arc.corp.mot.com (rogerk.arc.corp.mot.com [10.238.80.74]) by homer.arc.corp.mot.com (8.12.2/8.12.2) with ESMTP id g24509rA007459; Mon, 4 Mar 2002 16:00:09 +1100 (EST) Message-ID: <3C82FF59.F027305B@arc.corp.mot.com> Date: Mon, 04 Mar 2002 16:00:09 +1100 From: Roger Kermode Organization: Motorola X-Mailer: Mozilla 4.79 [en] (Windows NT 5.0; U) X-Accept-Language: en MIME-Version: 1.0 To: mankin@isi.edu CC: Michael Luby , Ross Finlayson , tsvwg@ietf.org, rmt@ietf.org Subject: Re: [Tsvwg] RE: [Rmt] Enough with the spam on these mailing lists! References: <200203040405.g2445Wx28342@minotaur.nge.isi.edu> Content-Type: multipart/mixed; boundary="------------A64546D5293753EFE9B38510" Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org This is a multi-part message in MIME format. --------------A64546D5293753EFE9B38510 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Ugh...sorry about the delay in taking care of this. I finally found the appropriate web page to turn the filter on. Only members to the list are now allowed t o post to it, cheers, Roger Allison Mankin wrote: > All, > > I don't have the admin passwords on either list, or > I would set up moderation of non-subscriber posts right now. > That is becoming the IETF list norm, actually, and it is > *well* worth doing. > > Lorenzo, Roger, Scott...can you do this asap for rmt > and tsvwg respectively? > > Allison > > _______________________________________________ > Rmt mailing list > Rmt@ietf.org > https://www1.ietf.org/mailman/listinfo/rmt --------------A64546D5293753EFE9B38510 Content-Type: text/x-vcard; charset=us-ascii; name="rkermode.vcf" Content-Description: Card for Roger Kermode Content-Disposition: attachment; filename="rkermode.vcf" Content-Transfer-Encoding: 7bit begin:vcard n:Kermode;Roger tel;cell:+61-408-212-284 tel;fax:+61-2-9666-0501 tel;work:+61-2-9666-0558 x-mozilla-html:FALSE org:Motorola Labs, Motorola Australia Research Centre;Sydney Networks Research Laboratory adr:;;Locked Bag 5028;Botany;NSW;1455;Australia version:2.1 email;internet:Roger.Kermode@motorola.com title:Prinicpal Research Engineer / Lab Manager fn:Roger Kermode end:vcard --------------A64546D5293753EFE9B38510-- _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Mon Mar 4 08:28:19 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA23494 for ; Mon, 4 Mar 2002 08:28:18 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id IAA28451 for rmt-archive@odin.ietf.org; Mon, 4 Mar 2002 08:28:22 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA27920; Mon, 4 Mar 2002 08:23:06 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id HAA23973 for ; Mon, 4 Mar 2002 07:12:26 -0500 (EST) Received: from prue.eim.surrey.ac.uk (IDENT:exim@prue.eim.surrey.ac.uk [131.227.76.5]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA21152; Mon, 4 Mar 2002 07:12:22 -0500 (EST) Received: from phaestos.ee.surrey.ac.uk ([131.227.88.14] ident=eep1lw) by prue.eim.surrey.ac.uk with esmtp (Exim 3.33 #4) id 16hrJU-0002mX-00; Mon, 04 Mar 2002 12:11:32 +0000 Date: Mon, 4 Mar 2002 12:11:29 +0000 (GMT) From: Lloyd Wood X-X-Sender: eep1lw@phaestos.ee.surrey.ac.uk Reply-To: Lloyd Wood To: Roger Kermode cc: mankin@isi.edu, Michael Luby , Ross Finlayson , , Subject: Re: [Tsvwg] RE: [Rmt] Enough with the spam on these mailing lists! In-Reply-To: <3C82FF59.F027305B@arc.corp.mot.com> Message-ID: Organization: speaking for none X-url: http://www.ee.surrey.ac.uk/Personal/L.Wood/ X-no-archive: yes MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Scanner: exiscan *16hrJU-0002mX-00*UkuiU6fkkCs* (SECM, UniS) Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org On Mon, 4 Mar 2002, Roger Kermode wrote: > Only members to the list are now allowed > t o post to it, moderation of non-subscriber posts is the norm. that's a slightly different thing, as it's the only way to cope with variations in email address. L. > Allison Mankin wrote: > > > All, > > > > I don't have the admin passwords on either list, or > > I would set up moderation of non-subscriber posts right now. > > That is becoming the IETF list norm, actually, and it is > > *well* worth doing. > > > > Lorenzo, Roger, Scott...can you do this asap for rmt > > and tsvwg respectively? > > > > Allison > > > > _______________________________________________ > > Rmt mailing list > > Rmt@ietf.org > > https://www1.ietf.org/mailman/listinfo/rmt > PGP _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Mon Mar 4 08:45:47 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA24118 for ; Mon, 4 Mar 2002 08:45:45 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id IAA29678 for rmt-archive@odin.ietf.org; Mon, 4 Mar 2002 08:45:46 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA27861; Mon, 4 Mar 2002 08:22:47 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id BAA06431 for ; Mon, 4 Mar 2002 01:53:14 -0500 (EST) Received: from mx.webfountain.com (mx.webfountain.com [63.161.54.39]) by ietf.org (8.9.1a/8.9.1a) with SMTP id BAA07795 for ; Mon, 4 Mar 2002 01:53:08 -0500 (EST) Received: (qmail 7008 invoked from network); 4 Mar 2002 06:52:39 -0000 Received: from mail.intranet (10.1.1.37) by mx.webfountain.com with SMTP; 4 Mar 2002 06:52:39 -0000 Received: by mail.intranet (8.12.1/8.12.1/Debian -5) with SMTP id g246qclk003115; Sun, 3 Mar 2002 22:52:38 -0800 From: "Michael Luby" To: "Rmt@ietf. org" Cc: "Michael Luby" Date: Sun, 3 Mar 2002 22:53:15 -0800 Message-ID: MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_NextPart_000_0000_01C1C306.34596B20" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4807.1700 Subject: [Rmt] WEBRC 01 Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org This is a multi-part message in MIME format. ------=_NextPart_000_0000_01C1C306.34596B20 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Attached is the 01 revision of the WEBRC building block document in both .txt and .ps formats, together with a list of changes from the WEBRC 00 draft. I inadvertently missed the submission deadline by 3 hours on Friday, much to my dismay. This will be officially submitted on March 12, and will appear officially as a draft after the upcoming IETF meeting. Instead of talking about updates to the draft at the meeting, I propose instead to like to see what the working group thinks are the outstanding issues with the WEBRC congestion control and how they can be addressed (for example, amount of IGMP join/leave traffic generated by the protocol, and the IGMP and PIM join latencies). I would also like to suggest that future deadlines for submission not be biased against the West Coast by having the cutoff time at 5 PM EST, how about something like midnight PST instead in the future so that there is no bias in this? Mike ------=_NextPart_000_0000_01C1C306.34596B20 Content-Type: application/postscript; name="draft-ietf-rmt-bb-webrc-01.ps" Content-Disposition: attachment; filename="draft-ietf-rmt-bb-webrc-01.ps" Content-Transfer-Encoding: quoted-printable %!PS-Adobe-3.0=0A= %%Creator: groff version 1.15=0A= %%CreationDate: Fri Mar 1 17:41:48 2002=0A= %%DocumentNeededResources: font Courier-Bold=0A= %%+ font Times-Bold=0A= %%+ font Times-Roman=0A= %%+ font Courier=0A= %%DocumentSuppliedResources: procset grops 1.15 3=0A= %%Pages: 21=0A= %%PageOrder: Ascend=0A= %%Orientation: Portrait=0A= %%EndComments=0A= %%BeginProlog=0A= %%BeginResource: procset grops 1.15 3=0A= /setpacking where{=0A= pop=0A= currentpacking=0A= true setpacking=0A= }if=0A= /grops 120 dict dup begin=0A= /SC 32 def=0A= /A/show load def=0A= /B{0 SC 3 -1 roll widthshow}bind def=0A= /C{0 exch ashow}bind def=0A= /D{0 exch 0 SC 5 2 roll awidthshow}bind def=0A= /E{0 rmoveto show}bind def=0A= /F{0 rmoveto 0 SC 3 -1 roll widthshow}bind def=0A= /G{0 rmoveto 0 exch ashow}bind def=0A= /H{0 rmoveto 0 exch 0 SC 5 2 roll awidthshow}bind def=0A= /I{0 exch rmoveto show}bind def=0A= /J{0 exch rmoveto 0 SC 3 -1 roll widthshow}bind def=0A= /K{0 exch rmoveto 0 exch ashow}bind def=0A= /L{0 exch rmoveto 0 exch 0 SC 5 2 roll awidthshow}bind def=0A= /M{rmoveto show}bind def=0A= /N{rmoveto 0 SC 3 -1 roll widthshow}bind def=0A= /O{rmoveto 0 exch ashow}bind def=0A= /P{rmoveto 0 exch 0 SC 5 2 roll awidthshow}bind def=0A= /Q{moveto show}bind def=0A= /R{moveto 0 SC 3 -1 roll widthshow}bind def=0A= /S{moveto 0 exch ashow}bind def=0A= /T{moveto 0 exch 0 SC 5 2 roll awidthshow}bind def=0A= /SF{=0A= findfont exch=0A= [exch dup 0 exch 0 exch neg 0 0]makefont=0A= dup setfont=0A= [exch/setfont cvx]cvx bind def=0A= }bind def=0A= /MF{=0A= findfont=0A= [5 2 roll=0A= 0 3 1 roll=0A= neg 0 0]makefont=0A= dup setfont=0A= [exch/setfont cvx]cvx bind def=0A= }bind def=0A= /level0 0 def=0A= /RES 0 def=0A= /PL 0 def=0A= /LS 0 def=0A= /MANUAL{=0A= statusdict begin/manualfeed true store end=0A= }bind def=0A= /PLG{=0A= gsave newpath clippath pathbbox grestore=0A= exch pop add exch pop=0A= }bind def=0A= /BP{=0A= /level0 save def=0A= 1 setlinecap=0A= 1 setlinejoin=0A= 72 RES div dup scale=0A= LS{=0A= 90 rotate=0A= }{=0A= 0 PL translate=0A= }ifelse=0A= 1 -1 scale=0A= }bind def=0A= /EP{=0A= level0 restore=0A= showpage=0A= }bind def=0A= /DA{=0A= newpath arcn stroke=0A= }bind def=0A= /SN{=0A= transform=0A= .25 sub exch .25 sub exch=0A= round .25 add exch round .25 add exch=0A= itransform=0A= }bind def=0A= /DL{=0A= SN=0A= moveto=0A= SN=0A= lineto stroke=0A= }bind def=0A= /DC{=0A= newpath 0 360 arc closepath=0A= }bind def=0A= /TM matrix def=0A= /DE{=0A= TM currentmatrix pop=0A= translate scale newpath 0 0 .5 0 360 arc closepath=0A= TM setmatrix=0A= }bind def=0A= /RC/rcurveto load def=0A= /RL/rlineto load def=0A= /ST/stroke load def=0A= /MT/moveto load def=0A= /CL/closepath load def=0A= /FL{=0A= currentgray exch setgray fill setgray=0A= }bind def=0A= /BL/fill load def=0A= /LW/setlinewidth load def=0A= /RE{=0A= findfont=0A= dup maxlength 1 index/FontName known not{1 add}if dict begin=0A= {=0A= 1 index/FID ne{def}{pop pop}ifelse=0A= }forall=0A= /Encoding exch def=0A= dup/FontName exch def=0A= currentdict end definefont pop=0A= }bind def=0A= /DEFS 0 def=0A= /EBEGIN{=0A= moveto=0A= DEFS begin=0A= }bind def=0A= /EEND/end load def=0A= /CNT 0 def=0A= /level1 0 def=0A= /PBEGIN{=0A= /level1 save def=0A= translate=0A= div 3 1 roll div exch scale=0A= neg exch neg exch translate=0A= 0 setgray=0A= 0 setlinecap=0A= 1 setlinewidth=0A= 0 setlinejoin=0A= 10 setmiterlimit=0A= []0 setdash=0A= /setstrokeadjust where{=0A= pop=0A= false setstrokeadjust=0A= }if=0A= /setoverprint where{=0A= pop=0A= false setoverprint=0A= }if=0A= newpath=0A= /CNT countdictstack def=0A= userdict begin=0A= /showpage{}def=0A= }bind def=0A= /PEND{=0A= clear=0A= countdictstack CNT sub{end}repeat=0A= level1 restore=0A= }bind def=0A= end def=0A= /setpacking where{=0A= pop=0A= setpacking=0A= }if=0A= %%EndResource=0A= %%IncludeResource: font Courier-Bold=0A= %%IncludeResource: font Times-Bold=0A= %%IncludeResource: font Times-Roman=0A= %%IncludeResource: font Courier=0A= grops begin/DEFS 1 dict def DEFS begin/u{.001 mul}bind def end/RES 72=0A= def/PL 792 def/LS false def/ENC0[/asciicircum/asciitilde/Scaron/Zcaron=0A= /scaron/zcaron/Ydieresis/trademark/quotesingle/.notdef/.notdef/.notdef=0A= /.notdef/.notdef/.notdef/.notdef/.notdef/.notdef/.notdef/.notdef/.notdef=0A= /.notdef/.notdef/.notdef/.notdef/.notdef/.notdef/.notdef/.notdef/.notdef=0A= /.notdef/.notdef/space/exclam/quotedbl/numbersign/dollar/percent=0A= /ampersand/quoteright/parenleft/parenright/asterisk/plus/comma/hyphen=0A= /period/slash/zero/one/two/three/four/five/six/seven/eight/nine/colon=0A= /semicolon/less/equal/greater/question/at/A/B/C/D/E/F/G/H/I/J/K/L/M/N/O=0A= /P/Q/R/S/T/U/V/W/X/Y/Z/bracketleft/backslash/bracketright/circumflex=0A= /underscore/quoteleft/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y=0A= /z/braceleft/bar/braceright/tilde/.notdef/quotesinglbase/guillemotleft=0A= /guillemotright/bullet/florin/fraction/perthousand/dagger/daggerdbl=0A= /endash/emdash/ff/fi/fl/ffi/ffl/dotlessi/dotlessj/grave/hungarumlaut=0A= /dotaccent/breve/caron/ring/ogonek/quotedblleft/quotedblright/oe/lslash=0A= /quotedblbase/OE/Lslash/.notdef/exclamdown/cent/sterling/currency/yen=0A= /brokenbar/section/dieresis/copyright/ordfeminine/guilsinglleft=0A= /logicalnot/minus/registered/macron/degree/plusminus/twosuperior=0A= /threesuperior/acute/mu/paragraph/periodcentered/cedilla/onesuperior=0A= /ordmasculine/guilsinglright/onequarter/onehalf/threequarters=0A= /questiondown/Agrave/Aacute/Acircumflex/Atilde/Adieresis/Aring/AE=0A= /Ccedilla/Egrave/Eacute/Ecircumflex/Edieresis/Igrave/Iacute/Icircumflex=0A= /Idieresis/Eth/Ntilde/Ograve/Oacute/Ocircumflex/Otilde/Odieresis=0A= /multiply/Oslash/Ugrave/Uacute/Ucircumflex/Udieresis/Yacute/Thorn=0A= /germandbls/agrave/aacute/acircumflex/atilde/adieresis/aring/ae/ccedilla=0A= /egrave/eacute/ecircumflex/edieresis/igrave/iacute/icircumflex/idieresis=0A= /eth/ntilde/ograve/oacute/ocircumflex/otilde/odieresis/divide/oslash=0A= /ugrave/uacute/ucircumflex/udieresis/yacute/thorn/ydieresis]def=0A= /Courier@0 ENC0/Courier RE/Times-Roman@0 ENC0/Times-Roman RE=0A= /Times-Bold@0 ENC0/Times-Bold RE/Courier-Bold@0 ENC0/Courier-Bold RE=0A= %%EndProlog=0A= %%Page: 1 1=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 10/Courier-Bold@0 SF(Internet Engineering Task Force)72 85 Q(RMT WG)=0A= 210 E 198(INTERNET-DRAFT M.)72 98 R(Luby/Digital Fountain)6 E 102=0A= (draft-ietf-rmt-bb-webrc-01.ps V.)72 111 R(Goyal/Digital Fountain)6 E 6=0A= (1M)432 124 S(arch 2002)-6 E(Expires: September 2002)366 137 Q/F1 14=0A= /Times-Bold@0 SF -2.03 -.91(Wa v)124.242 162 T 3.5(ea).77 G=0A= (nd Equation Based Rate Contr)-3.5 E(ol b)-.252 E(uilding block)-.28 E=0A= /F2 11/Times-Bold@0 SF(Status of this Document)72 207 Q/F3 11=0A= /Times-Roman@0 SF(This document is an Internet-Draft and is in full con\=0A= formance with all pro)72 223.6 Q(visions of Section 10 of)-.165 E=0A= (RFC2026.)72 236.6 Q(Internet-Drafts are w)72 262.6 Q=0A= (orking documents of the Internet Engineering T)-.11 E(ask F)-.88 E=0A= (orce \(IETF\), its areas,)-.165 E(and its w)72 275.6 Q(orking groups.)=0A= -.11 E(Note that other groups may also distrib)5.5 E(ute w)-.22 E=0A= (orking documents as)-.11 E(Internet-Drafts.)72 288.6 Q=0A= (Internet-Drafts are v)72 314.6 Q=0A= (alid for a maximum of six months and may be updated, replaced, or)-.275=0A= E(obsoleted by other documents at an)72 327.6 Q 2.75(yt)-.165 G 2.75=0A= (ime. It)-2.75 F(is inappropriate to use Internet-Drafts as reference)=0A= 2.75 E(material or to cite them other than as a "w)72 340.6 Q=0A= (ork in progress".)-.11 E=0A= (The list of current Internet-Drafts can be accessed at http://www)72=0A= 366.6 Q(.ietf.or)-.715 E(g/ietf/1id-abstracts.txt)-.198 E 1.76 -.88=0A= (To v)72 392.6 T(ie).88 E 2.75(wt)-.275 G(he list Internet-Draft Shado)=0A= -2.75 E 2.75(wD)-.275 G(irectories, see http://www)-2.75 E(.ietf.or)=0A= -.715 E(g/shado)-.198 E -.715(w.)-.275 G(html.).715 E=0A= (This document is a product of the IETF RMT WG.)72 418.6 Q=0A= (Comments should be addressed to the)5.5 E(authors, or the WG')72 431.6=0A= Q 2.75(sm)-.605 G(ailing list at rmt@lbl.go)-2.75 E -.715(v.)-.165 G F2=0A= (Abstract)267.534 463.6 Q F3 -1.21 -.88(Wa v)97 486.2 T 2.75(ea).715 G=0A= (nd Equation Based Rate Control \(WEBRC\) pro)-2.75 E=0A= (vides rate and congestion)-.165 E(control for data deli)97 499.2 Q=0A= -.165(ve)-.275 G(ry).165 E 5.5(.W)-.715 G=0A= (EBRC is speci\214cally designed to support protocols using)-5.5 E=0A= (IP multicast, b)97 512.2 Q(ut could also be used to pro)-.22 E=0A= (vide support to protocols that use unicast.)-.165 E(WEBRC pro)97 525.2=0A= Q(vides multiple rate congestion controlled deli)-.165 E -.165(ve)-.275=0A= G(ry to recei).165 E -.165(ve)-.275 G(rs, i.e.,).165 E(dif)97 538.2 Q=0A= (ferent recei)-.275 E -.165(ve)-.275 G=0A= (rs joined to the same session may be recei).165 E(ving pack)-.275 E=0A= (ets at dif)-.11 E(ferent)-.275 E(rates depending on their indi)97 551.2=0A= Q(vidual bandwidth connection to the sender and on)-.275 E=0A= (competing traf)97 564.2 Q(\214c along that connection.)-.275 E=0A= (WEBRC requires no feedback from recei)5.5 E -.165(ve)-.275 G(rs).165 E=0A= (to the sender)97 577.2 Q 2.75(,i)-.44 G(.e., it is a completely recei)=0A= -2.75 E -.165(ve)-.275 G -.22(r-).165 G(dri).22 E -.165(ve)-.275 G 2.75=0A= (nc).165 G(ongestion control protocol.)-2.75 E(Thus,)5.5 E=0A= (WEBRC is designed to scale to potentially massi)97 590.2 Q .33 -.165=0A= (ve n)-.275 H(umbers of recei).165 E -.165(ve)-.275 G(rs attached to a)=0A= .165 E(session from a single sender)97 603.2 Q 5.5(.F)-.605 G=0A= (urthermore, because each indi)-5.5 E(vidual recei)-.275 E -.165(ve)=0A= -.275 G 2.75(ra).165 G(djusts to)-2.75 E(the a)97 616.2 Q -.275(va)-.22=0A= G(ilable bandwidth between the sender and that recei).275 E -.165(ve)=0A= -.275 G .88 -.44(r, t).165 H(here is the potential to).44 E(deli)97=0A= 629.2 Q -.165(ve)-.275 G 2.75(rd).165 G(ata to each indi)-2.75 E=0A= (vidual recei)-.275 E -.165(ve)-.275 G 2.75(ra).165 G 2.75(tt)-2.75 G=0A= (he f)-2.75 E(astest possible rate for that recei)-.11 E -.165(ve)-.275=0A= G -.44(r,).165 G -2.365 -.275(ev e)97 642.2 T 2.75(ni).275 G 2.75(nah)=0A= -2.75 G(ighly heterogeneous netw)-2.75 E=0A= (ork architecture, using a single sender)-.11 E(.)-.605 E(Luby/Go)72 769=0A= Q 339.402(yal [P)-.11 F(age 1])-.165 E EP=0A= %%Page: 2 2=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E/F1 13/Times-Bold@0 SF -1.196=0A= (Ta)239.126 85 S(ble of Contents)1.196 E/F2 10/Times-Roman@0 SF=0A= (1. Introduction)97 123 Q F0 11(......................)3.56 G F2(3)11.5=0A= E(2. Rationale)97 135 Q F0 11(.......................).92 G F2(4)11.5 E=0A= (3. Functionality)97 147 Q F0 11(.....................)13.41 G F2(4)11.5=0A= E(3.1. Sender Operation)107 159 Q F0 11(...................)6.49 G F2(6)=0A= 11.5 E(3.1.1. Sender inputs and initialization)117 171 Q F0 11=0A= (..............)3.28 G F2(6)11.5 E(3.1.2. Sending pack)117 183 Q=0A= (ets to the session)-.1 E F0 11(..............)6.44 G F2(7)11.5 E=0A= (3.2. Recei)107 195 Q -.15(ve)-.25 G 2.5(rO).15 G(peration)-2.5 E F0 11=0A= (..................)12.87 G F2(8)11.5 E(3.2.1. Recei)117 207 Q -.15(ve)=0A= -.25 G 2.5(ri).15 G(nputs and initialization)-2.5 E F0 11(.............)=0A= 9.66 G F2(8)11.5 E(3.2.2. Recei)117 219 Q -.15(ve)-.25 G 2.5(rm).15 G=0A= (easurements and calculations)-2.5 E F0 11(...........)7.19 G F2(9)11.5=0A= E(3.2.2.1. A)127 231 Q -.15(ve)-.74 G(rage loss probability).15 E F0 11=0A= (..............)13.07 G F2(9)11.5 E(3.2.2.2. A)127 243 Q -.15(ve)-.74 G=0A= (rage round-trip time).15 E F0 11(...............).44 G F2(9)11.5 E=0A= (3.2.2.3. Rate Equation)127 255 Q F0 11(.................)12.31 G F2(10)=0A= 6.5 E(3.2.2.4. Epochs)127 267 Q F0 11(...................)12.31 G F2(10)=0A= 6.5 E(3.2.2.5. A)127 279 Q -.15(ve)-.74 G(rage reception rate).15 E F0=0A= 11(...............)6.57 G F2(10)6.5 E(3.2.2.6. Slo)127 291 Q 2.5(ws)-.25=0A= G(tart)-2.5 E F0 11(...................)1.72 G F2(11)6.5 E(3.2.2.7. T)=0A= 127 303 Q(ar)-.8 E(get rate)-.18 E F0 11(..................)12.89 G F2=0A= (11)6.5 E(3.2.3. Recei)117 315 Q -.15(ve)-.25 G 2.5(re).15 G -.15(ve)=0A= -2.75 G(nts).15 E F0 11(..................)10.21 G F2(11)6.5 E=0A= (3.2.3.1. Epoch change)127 327 Q F0 11(.................)12.88 G F2(12)=0A= 6.5 E(3.2.3.2. T)127 339 Q(ime slot change)-.35 E F0 11=0A= (.................).72 G F2(12)6.5 E(3.2.3.3. Join a w)127 351 Q -2.25=0A= -.2(av e)-.1 H(channel)2.7 E F0 11(................)2.64 G F2(12)6.5 E=0A= (3.2.3.4. Loss e)127 363 Q -.15(ve)-.25 G(nt).15 E F0 11=0A= (..................)12.85 G F2(12)6.5 E(3.2.3.5. Exceptional timeouts)=0A= 127 375 Q F0 11(...............)12.03 G F2(12)6.5 E=0A= (4. Applicability Statement)97 387 Q F0 11(..................)12.72 G F2=0A= (13)6.5 E(4.1. En)107 399 Q(vironmental Requirements and Considerations)=0A= -.4 E F0 11(..........)3.97 G F2(13)6.5 E(5. P)97 411 Q(ack)-.15 E=0A= (et Header Fields)-.1 E F0 11(...................)9.52 G F2(14)6.5 E=0A= (5.1. Short F)107 423 Q(ormat Congestion Control Information)-.15 E F0=0A= 11(...........)2.45 G F2(15)6.5 E(5.2. Long F)107 435 Q=0A= (ormat Congestion Control Information)-.15 E F0 11(...........)3.01 G F2=0A= (15)6.5 E(6. Requirements From Other Building Blocks)97 447 Q F0 11=0A= (.............)3.97 G F2(16)6.5 E(7. Security Considerations)97 459 Q F0=0A= 11(..................)12.17 G F2(17)6.5 E(8. IAN)97 471 Q 2.5(AC)-.35 G=0A= (onsiderations)-2.5 E F0 11(...................)7.11 G F2(17)6.5 E=0A= (9. Intellectual Property Issues)97 483 Q F0 11(.................)12.88=0A= G F2(17)6.5 E(10. References)97 495 Q F0 11(......................)3.58=0A= G F2(17)6.5 E(11. Authors' Addresses)97 507 Q F0 11(...................)=0A= 10.1 G F2(19)6.5 E(12. Full Cop)97 519 Q(yright Statement)-.1 E F0 11=0A= (..................)1.42 G F2(20)6.5 E F0(Luby/Go)72 769 Q 339.402=0A= (yal [P)-.11 F(age 2])-.165 E EP=0A= %%Page: 3 3=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E/F1 11/Times-Bold@0 SF(1.)72=0A= 85 Q/F2 14/Times-Bold@0 SF(Intr)5.5 E(oduction)-.252 E F0=0A= (This document speci\214es W)72 101.6 Q -2.475 -.22(av e)-.88 H=0A= (and Equation Based Rate Control \(WEBRC\).)2.97 E(WEBRC is a)5.5 E=0A= (congestion control b)72 114.6 Q=0A= (uilding block that is designed to be massi)-.22 E -.165(ve)-.275 G=0A= (ly scalable when used with the IP).165 E(multicast netw)72 127.6 Q=0A= (ork service.)-.11 E=0A= (WEBRC is also suitable as the basis for unicast congestion control,)5.5=0A= E -.22(bu)72 140.6 S 2.75(tt).22 G=0A= (his is outside the scope of this document.)-2.75 E=0A= (WEBRC is designed to compete f)5.5 E(airly with TCP)-.11 E=0A= (and similar congestion-controlled sessions.)72 153.6 Q=0A= (WEBRC can be used as a congestion control protocol)5.5 E(for an)72=0A= 166.6 Q 2.75(yt)-.165 G(ype of data deli)-2.75 E -.165(ve)-.275 G(ry)=0A= .165 E 2.75(,i)-.715 G(ncluding reliable content deli)-2.75 E -.165(ve)=0A= -.275 G(ry and streaming deli).165 E -.165(ve)-.275 G(ry).165 E(.)-.715=0A= E(This document describes a b)72 192.6 Q=0A= (uilding block as de\214ned in RFC3048 [17] that conforms to RFC2357)=0A= -.22 E([14].)72 205.6 Q(WEBRC is a recei)72 235.2 Q -.165(ve)-.275 G=0A= -.22(r-).165 G(dri).22 E -.165(ve)-.275 G 2.75(nc).165 G=0A= (ongestion control protocol in the spirit of [16] and [3]. This means)=0A= -2.75 E(that all measurements and decisions to raise or lo)72 248.2 Q=0A= (wer the reception rate are made by each)-.275 E(indi)72 261.2 Q=0A= (vidual recei)-.275 E -.165(ve)-.275 G .88 -.44(r, a).165 H=0A= (nd these decisions are acted upon by sending join and lea).44 E .33=0A= -.165(ve m)-.22 H(essages for).165 E(channels to the netw)72 274.2 Q=0A= 2.75(ork. A)-.11 F(recei)2.75 E -.165(ve)-.275 G 2.75(ru).165 G=0A= (sing WEBRC adjusts its reception rate without re)-2.75 E -.055(ga)-.165=0A= G(rd for).055 E(other concerns such as reliability)72 287.2 Q 5.5(.T)=0A= -.715 G(his is dif)-5.5 E(ferent from TCP)-.275 E 2.75(,w)-1.221 G=0A= (here the congestion control)-2.75 E=0A= (protocol and the reliability protocol are intricately interw)72 300.2 Q=0A= -.165(ove)-.11 G 2.75(nw).165 G(ith one another)-2.75 E(.)-.605 E=0A= (WEBRC tak)72 326.2 Q=0A= (es the same basic equation-based approach as TFRC [5]. In particular)=0A= -.11 E 2.75(,e)-.44 G(ach WEBRC)-2.75 E(recei)72 339.2 Q -.165(ve)-.275=0A= G 2.75(rm).165 G(easures parameters that are plugged into a TCP-lik)=0A= -2.75 E 2.75(ee)-.11 G(quation to compute the recei)-2.75 E -.165(ve)=0A= -.275 G(r).165 E(tar)72 352.2 Q=0A= (get reception rate, and adjusts its reception rate up and do)-.198 E=0A= (wn to closely approximate the tar)-.275 E(get)-.198 E(reception rate.)=0A= 72 365.2 Q(The sender sends pack)5.5 E=0A= (ets to multiple channels called w)-.11 E -2.475 -.22(av e)-.11 H 2.75=0A= (channels. Each)2.97 F -.11(wa)2.75 G -.165(ve)-.11 G(channel follo)72=0A= 378.2 Q(ws the same pattern of pack)-.275 E=0A= (et rate transmission spread out o)-.11 E -.165(ve)-.165 G 2.75(re).165=0A= G(qually spaced)-2.75 E(interv)72 391.2 Q(als of time.)-.275 E=0A= (The pattern of each w)5.5 E -2.475 -.22(av e)-.11 H=0A= (is that it starts at a high rate that decreases gradually)2.97 E=0A= (and continually o)72 404.2 Q -.165(ve)-.165 G 2.75(ral).165 G=0A= (ong interv)-2.75 E(al of time.)-.275 E=0A= (\(Picture an in\214nite sequence of w)5.5 E -2.475 -.22(av e)-.11 H=0A= 2.75(s.\) The).22 F(recei)2.75 E -.165(ve)-.275 G(r).165 E=0A= (increases its reception rate by joining the ne)72 417.2 Q(xt w)-.165 E=0A= -2.475 -.22(av e)-.11 H(channel earlier in the descent of the w)2.97 E=0A= -2.475 -.22(av e)-.11 H(than it joined the pre)72 430.2 Q(vious w)-.275=0A= E -2.475 -.22(av e)-.11 H(channel, and the recei)2.97 E -.165(ve)-.275 G=0A= 2.75(rd).165 G(ecreases its reception rate by joining)-2.75 E(the ne)72=0A= 443.2 Q(xt w)-.165 E -2.475 -.22(av e)-.11 H=0A= (channel later in the descent of the w)2.97 E -2.475 -.22(av e)-.11 H=0A= (than it joined the pre)2.97 E(vious w)-.275 E -2.475 -.22(av e)-.11 H=0A= (channel.)2.97 E(WEBRC is designed for applications that use a \214x)72=0A= 469.2 Q(ed pack)-.165 E(et size, and v)-.11 E(ary their pack)-.275 E=0A= (et reception)-.11 E(rate in response to congestion.)72 482.2 Q=0A= (WEBRC is designed to be reasonably f)5.5 E(air when competing for)-.11=0A= E(bandwidth with TCP \215o)72 495.2 Q(ws, where a \215o)-.275 E 2.75(wi)=0A= -.275 G 2.75(s`)-2.75 G(`reasonably f)-3.564 E(air')-.11 E 2.75('i)-.814=0A= G 2.75(fi)-2.75 G(ts reception rate is generally)-2.75 E(within a f)72=0A= 508.2 Q(actor of tw)-.11 E 2.75(oo)-.11 G 2.75(ft)-2.75 G=0A= (he reception rate of a TCP \215o)-2.75 E 2.75(wu)-.275 G=0A= (nder the same conditions.)-2.75 E(Ho)5.5 E(we)-.275 E -.165(ve)-.275 G=0A= (r).165 E(WEBRC has a much lo)72 521.2 Q(wer v)-.275 E=0A= (ariation of throughput o)-.275 E -.165(ve)-.165 G 2.75(rt).165 G=0A= (ime compared to TCP)-2.75 E 2.75(,w)-1.221 G(hich mak)-2.75 E(es it)=0A= -.11 E(more suitable for applications such as telephon)72 534.2 Q 2.75=0A= (yo)-.165 G 2.75(rs)-2.75 G(treaming media where a relati)-2.75 E -.165=0A= (ve)-.275 G(ly smooth).165 E(rate is of importance.)72 547.2 Q=0A= (Furthermore, WEBRC is designed to be massi)5.5 E -.165(ve)-.275 G=0A= (ly scalable in the sense).165 E(that the sender is insensiti)72 560.2 Q=0A= .33 -.165(ve t)-.275 H 2.75(ot).165 G(he number of recei)-2.75 E -.165=0A= (ve)-.275 G(rs joined to a multicast session.).165 E(The penalty)5.5 E=0A= (of ha)72 573.2 Q(ving smoother throughput than TCP while competing f)=0A= -.22 E(airly for bandwidth is that WEBRC)-.11 E(responds slo)72 586.2 Q=0A= (wer than TCP to changes in a)-.275 E -.275(va)-.22 G(ilable bandwidth.)=0A= .275 E(The recei)72 612.2 Q -.165(ve)-.275 G 2.75(rm).165 G(easures and\=0A= performs the calculation of congestion control parameters \(e.g., the)=0A= -2.75 E -2.475 -.22(av e)72 625.2 T(rage loss probability\) and mak).22=0A= E(es decisions on ho)-.11 E 2.75(wt)-.275 G 2.75(oi)-2.75 G=0A= (ncrease or decrease its rate based on)-2.75 E(these parameters.)72=0A= 638.2 Q(The recei)5.5 E -.165(ve)-.275 G -.22(r-).165 G=0A= (based approach is well suited to an application where the sender).22 E=0A= (is handling man)72 651.2 Q 2.75(yc)-.165 G=0A= (oncurrent connections and therefore WEBRC is suitable as a b)-2.75 E=0A= (uilding block)-.22 E(for multicast congestion control.)72 664.2 Q=0A= (The k)72 690.2 Q .33 -.165(ey w)-.11 H(ords "MUST", "MUST NO).055 E=0A= (T", "REQ)-.44 E(UIRED", "SHALL", "SHALL NO)-.11 E(T",)-.44 E=0A= ("SHOULD", "SHOULD NO)72 703.2 Q(T", "RECOMMENDED", "MA)-.44 E=0A= (Y", and "OPTION)-1.155 E(AL" in this)-.385 E=0A= (document are to be interpreted as described in RFC2119.)72 716.2 Q=0A= (Luby/Go)72 769 Q 289.902(yal Section)-.11 F 2.75(1. [P)2.75 F(age 3])=0A= -.165 E EP=0A= %%Page: 4 4=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E/F1 11/Times-Bold@0 SF(2.)72=0A= 85 Q/F2 14/Times-Bold@0 SF(Rationale)5.5 E F0(WEBRC pro)72 101.6 Q=0A= (vides congestion control for massi)-.165 E -.165(ve)-.275 G=0A= (ly scalable protocols using the IP multicast).165 E(netw)72 114.6 Q=0A= (ork service.)-.11 E(The congestion control that WEBRC pro)5.5 E=0A= (vides is common to a v)-.165 E(ariety of)-.275 E=0A= (applications, including reliable content deli)72 127.6 Q -.165(ve)-.275=0A= G(ry and streaming applications.).165 E(WEBRC is designed to pro)72=0A= 153.6 Q(vide congestion control for all pack)-.165 E=0A= (ets that are sent to a session.)-.11 E(A)5.5 E(session comprises multi\=0A= ple channels originating at a single sender that are used for some peri\=0A= od)72 166.6 Q(of time to carry pack)72 179.6 Q(ets pertaining to the tr\=0A= ansmission of one or more objects that can be of)-.11 E=0A= (interest to recei)72 192.6 Q -.165(ve)-.275 G 2.75(rs. The).165 F(logi\=0A= c behind de\214ning a session as originating from a single sender is)=0A= 2.75 E(that this is the right granularity to re)72 205.6 Q(gulate pack)=0A= -.165 E(et traf)-.11 E(\214c via congestion control.)-.275 E=0A= (The rationale for)5.5 E(pro)72 218.6 Q(viding congestion control that \=0A= uses multiple channels within the same session is that this)-.165 E=0A= (allo)72 231.6 Q(ws the data on the channels to be layered, and a recei)=0A= -.275 E -.165(ve)-.275 G 2.75(rj).165 G(oins and lea)-2.75 E -.165(ve)=0A= -.22 G 2.75(sc).165 G(hannels in a layered)-2.75 E=0A= (order during its participation in the session.)72 244.6 Q=0A= (There are adv)5.5 E(antages to layered data for streaming,)-.275 E=0A= (where more data can be sent to the lo)72 257.6 Q=0A= (wer layers and incrementally important data to higher layers.)-.275 E=0A= -.165(Fo)72 270.6 S 2.75(rr).165 G(eliable content deli)-2.75 E -.165=0A= (ve)-.275 G(ry).165 E 2.75(,a)-.715 G 2.75(sd)-2.75 G=0A= (escribed in [12] an application can send in pack)-2.75 E=0A= (ets encoded data)-.11 E(generated from an object in such a w)72 283.6 Q=0A= (ay that the arri)-.11 E -.275(va)-.275 G 2.75(lo).275 G 2.75(fe)-2.75 G=0A= (nough pack)-2.75 E(ets by a recei)-.11 E -.165(ve)-.275 G 2.75(ri).165=0A= G(s)-2.75 E(suf)72 296.6 Q=0A= (\214cient to reliably reconstruct original object.)-.275 E 2.75(Ap)5.5=0A= G(rimary adv)-2.75 E(antage of WEBRC is that each)-.275 E(recei)72 309.6=0A= Q -.165(ve)-.275 G 2.75(rc).165 G=0A= (ontrols it reception rate independent of other recei)-2.75 E -.165(ve)=0A= -.275 G 2.75(rs. Thus,).165 F(for e)2.75 E(xample, a recei)-.165 E -.165=0A= (ve)-.275 G(r).165 E(with a slo)72 322.6 Q 2.75(wc)-.275 G=0A= (onnection to the sender does not slo)-2.75 E 2.75(wd)-.275 G -.275(ow)=0A= -2.75 G 2.75(nt).275 G(he recei)-2.75 E -.165(ve)-.275 G(rs with f).165=0A= E(aster connections.)-.11 E(There are coding techniques that pro)72=0A= 348.6 Q(vide massi)-.165 E -.165(ve)-.275 G=0A= (ly scalable reliability and asynchronous deli).165 E -.165(ve)-.275 G=0A= (ry).165 E(which are compatible with WEBRC.)72 361.6 Q=0A= (When combined the result is a massi)5.5 E -.165(ve)-.275 G=0A= (ly scalable, reliable,).165 E(asynchronous content deli)72 374.6 Q=0A= -.165(ve)-.275 G(ry protocol that is netw).165 E(ork friendly)-.11 E 5.5=0A= (.W)-.715 G(EBRC also pro)-5.5 E(vides)-.165 E=0A= (congestion control that suitable for streaming applications.)72 387.6 Q=0A= (WEBRC a)72 413.6 Q -.22(vo)-.22 G=0A= (ids using techniques that are not massi).22 E -.165(ve)-.275 G=0A= (ly scalable.).165 E -.165(Fo)5.5 G 2.75(re).165 G=0A= (xample, WEBRC does not)-2.915 E(pro)72 426.6 Q(vide an)-.165 E 2.75(ym)=0A= -.165 G(echanisms for sending information from recei)-2.75 E -.165(ve)=0A= -.275 G(rs to senders, although this does not).165 E(rule out protocols\=0A= that both use WEBRC and do require sending information from recei)72=0A= 439.6 Q -.165(ve)-.275 G(rs to).165 E(senders.)72 452.6 Q(WEBRC pro)72=0A= 478.6 Q(vides congestion control that can be tuned for dif)-.165 E=0A= (ferent applications that may ha)-.275 E -.165(ve)-.22 G(dif)72 491.6 Q=0A= (fering application requirements. F)-.275 E(or e)-.165 E=0A= (xample, a content deli)-.165 E -.165(ve)-.275 G=0A= (ry protocol may aggressi).165 E -.165(ve)-.275 G(ly).165 E(stri)72=0A= 504.6 Q .33 -.165(ve t)-.275 H 2.75(ou).165 G(se all a)-2.75 E -.275(va)=0A= -.22 G(ilable bandwidth between recei).275 E -.165(ve)-.275 G=0A= (rs and the sender).165 E 2.75(,a)-.44 G(nd thus it must drastically)=0A= -2.75 E(back of)72 517.6 Q 2.75(fi)-.275 G=0A= (ts rate when there is competing traf)-2.75 E 2.75(\214c. On)-.275 F=0A= (the other hand, a streaming deli)2.75 E -.165(ve)-.275 G(ry protocol)=0A= .165 E(may stri)72 530.6 Q .33 -.165(ve t)-.275 H 2.75(om).165 G=0A= (aintain a constant rate instead of trying to use all a)-2.75 E -.275=0A= (va)-.22 G(ilable bandwidth, and thus it).275 E(may not back of)72 543.6=0A= Q 2.75(fi)-.275 G(ts rate as f)-2.75 E(ast when there is competing traf)=0A= -.11 E(\214c.)-.275 E(WEBRC does not pro)72 569.6 Q(vide an)-.165 E 2.75=0A= (ys)-.165 G(upport be)-2.75 E=0A= (yond congestion control, and thus it must be combined)-.165 E=0A= (with other b)72 582.6 Q(uilding blocks to pro)-.22 E=0A= (vide a complete protocol instantiation.)-.165 E -.165(Fo)5.5 G 2.75(re)=0A= .165 G(xample, WEBRC)-2.915 E(does not pro)72 595.6 Q(vide an)-.165 E=0A= 2.75(ym)-.165 G=0A= (eans that can be used to identify which session each recei)-2.75 E=0A= -.165(ve)-.275 G 2.75(dp).165 G(ack)-2.75 E(et)-.11 E(belongs to.)72=0A= 608.6 Q(As another e)5.5 E(xample, WEBRC does not pro)-.165 E=0A= (vide support for identifying which object)-.165 E(each pack)72 621.6 Q=0A= (et is carrying information about.)-.11 E F1(3.)72 664.2 Q F2=0A= (Functionality)5.5 E F0 2.75(AW)72 680.8 S(EBRC session comprises a log\=0A= ically related set of channels originating from a single sender)-2.75 E=0A= (that are used for some period of time to carry data pack)72 693.8 Q=0A= (ets with a header carrying WEBRC)-.11 E=0A= (Congestion Control Information.)72 706.8 Q(At the netw)5.5 E(ork layer)=0A= -.11 E 2.75(,ac)-.44 G(hannel can be uniquely identi\214ed by a)-2.75 E=0A= (\(sender IP address, multicast group address\) pair)72 719.8 Q(.)-.605=0A= E(Luby/Go)72 769 Q 289.902(yal Section)-.11 F 2.75(3. [P)2.75 F(age 4])=0A= -.165 E EP=0A= %%Page: 5 5=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E -.165(Fo)72 85 S 2.75(re)=0A= .165 G(ach WEBRC session, the channels within the session are mapped un\=0A= iquely to consecuti)-2.75 E -.165(ve)-.275 G(channel numbers.)72 98 Q=0A= (In each pack)5.5 E=0A= (et sent to a channel, the channel number that corresponds to the)-.11 E=0A= (channel is carried in the WEBRC Congestion Control Information.)72 111=0A= Q 2.75(AW)5.5 G(EBRC recei)-2.75 E -.165(ve)-.275 G 2.75(ru).165 G=0A= (ses the)-2.75 E=0A= (channel number to determine which channel within a session a pack)72=0A= 124 Q(et is recei)-.11 E -.165(ve)-.275 G 2.75(df).165 G 2.75(rom. When)=0A= -2.75 F(pack)72 137 Q(ets are recei)-.11 E -.165(ve)-.275 G(d, the).165=0A= E 2.75(ya)-.165 G(re \214rst check)-2.75 E(ed to see that the)-.11 E=0A= 2.75(yb)-.165 G(elong to the appropriate session)-2.75 E=0A= (before WEBRC is applied.)72 150 Q -.165(Fo)5.5 G 2.75(re).165 G=0A= (xample, if LCT [11] is being used with the session, then the)-2.915 E=0A= (sender IP address together with the T)72 163 Q=0A= (ransport Session Identi\214er supported by LCT w)-.385 E(ould be used)=0A= -.11 E(to determine which session a recei)72 176 Q -.165(ve)-.275 G 2.75=0A= (dp).165 G(ack)-2.75 E(et belongs to.)-.11 E=0A= (The particular details of ho)5.5 E 2.75(wt)-.275 G(his)-2.75 E=0A= (\214ltering is performed it outside the scope of this document.)72 189=0A= Q(In the remainder of this document,)5.5 E(channels will be referred to\=0A= by their channel number in the scope of the session.)72 202 Q=0A= (At the sender)72 228 Q 2.75(,t)-.44 G=0A= (ime is partitioned into time slots, each of duration TSD seconds.)-2.75=0A= E(There are a \214x)5.5 E(ed)-.165 E=0A= (number T of time slot indices associated with a session.)72 241 Q=0A= (As time progresses, the current time slot)5.5 E(inde)72 254 Q 2.75(xi)=0A= -.165 G(ncreases by one modulo T each TSD seconds.)-2.75 E=0A= (The current time slot inde)5.5 E 2.75(xC)-.165 G(TSI is carried)-2.75 E=0A= (in the WEBRC Congestion Control Information.)72 267 Q(This allo)5.5 E=0A= (ws recei)-.275 E -.165(ve)-.275 G(rs to perform v).165 E(ery coarse)=0A= -.165 E(grained synchronization within a session.)72 280 Q=0A= (WEBRC congestion control is achie)72 306 Q -.165(ve)-.275 G 2.75(db)=0A= .165 G 2.75(yh)-2.75 G -.22(av)-2.75 G(ing the sender send pack).22 E=0A= (ets associated with a)-.11 E(gi)72 319 Q -.165(ve)-.275 G 2.75(ns).165=0A= G(ession to se)-2.75 E -.165(ve)-.275 G(ral dif).165 E(ferent channels.)=0A= -.275 E(Indi)5.5 E(vidual recei)-.275 E -.165(ve)-.275 G=0A= (rs dynamically join and lea).165 E .33 -.165(ve t)-.22 H(hese).165 E=0A= (channels according to the netw)72 332 Q(ork congestion as e)-.11 E=0A= (xperienced by the recei)-.165 E -.165(ve)-.275 G 3.96 -.605(r. T).165 H=0A= (hese congestion).605 E(control adjustments are performed at each recei)=0A= 72 345 Q -.165(ve)-.275 G 2.75(ri).165 G=0A= (ndependently of all other recei)-2.75 E -.165(ve)-.275 G=0A= (rs, and in the).165 E(case of multicast without an)72 358 Q 2.75(yi)=0A= -.165 G(mpact on the sender)-2.75 E 5.5(.A)-.605 G(pack)-2.75 E=0A= (et sequence number is carried in the)-.11 E=0A= (WEBRC Congestion Control Information.)72 371 Q(The pack)5.5 E=0A= (et sequence numbers are consecuti)-.11 E -.165(ve)-.275 G(ly).165 E=0A= (numbered per channel and are used by recei)72 384 Q -.165(ve)-.275 G=0A= (rs to measure pack).165 E(et loss.)-.11 E(The channels associated with\=0A= a session consist of one base channel and T w)72 410 Q -2.475 -.22=0A= (av e)-.11 H 2.75(channels. The)2.97 F(pack)72 423 Q=0A= (et rate for each channel v)-.11 E(aries o)-.275 E -.165(ve)-.165 G 2.75=0A= (rt).165 G 2.75(ime. F)-2.75 F(or the base channel, pack)-.165 E=0A= (ets are sent to the)-.11 E(channel at a lo)72 436 Q 2.75(wr)-.275 G=0A= (ate BCR_P at the be)-2.75 E=0A= (ginning of a time slot and this rate gradually decreases to)-.165 E(P*\=0A= BCR_P at the end of the time slot, where P < 1 is a constant de\214ned \=0A= later)72 449 Q 5.5(.T)-.605 G(his pattern for the)-5.5 E=0A= (base channel repeats o)72 462 Q -.165(ve)-.165 G 2.75(re).165 G=0A= (ach time slot.)-2.75 E -.165(Fo)5.5 G 2.75(re).165 G(ach w)-2.75 E=0A= -2.475 -.22(av e)-.11 H(channel i, pack)2.97 E=0A= (ets are sent to channel i)-.11 E=0A= (starting at a high rate MWCR_P and decreasing o)72 475 Q -.165(ve)-.165=0A= G 2.75(rt).165 G(ime by a \214x)-2.75 E=0A= (ed fraction P per time slot until)-.165 E 2.75(ar)72 488 S=0A= (ate of BCR_P is reached at the end of time slot i.)-2.75 E=0A= (Then, for a period of time called the)5.5 E(quiescent period, no pack)=0A= 72 501 Q(ets are sent to w)-.11 E -2.475 -.22(av e)-.11 H=0A= (channel i, and thereafter the whole c)2.97 E(ycle repeats)-.165 E=0A= (itself, where the duration of the c)72 514 Q(ycle is T*TSD seconds.)=0A= -.165 E(Thus, the w)5.5 E -2.475 -.22(av e)-.11 H(channels are going)=0A= 2.97 E(through the same c)72 527 Q(yclic pattern of pack)-.165 E=0A= (et rate transmission spaced out e)-.11 E -.165(ve)-.275 G=0A= (nly by TSD seconds.).165 E(Before joining a session, the recei)72 553 Q=0A= -.165(ve)-.275 G=0A= (rs MUST obtain enough of the session description to start the).165 E=0A= 2.75(session. This)72 566 R(MUST include the rele)2.75 E -.275(va)-.275=0A= G(nt session parameters needed by a recei).275 E -.165(ve)-.275 G 2.75=0A= (rt).165 G 2.75(op)-2.75 G(articipate)-2.75 E=0A= (in the session and perform WEBRC congestion control.)72 579 Q=0A= (The session description is determined by)5.5 E=0A= (the sender and is typically communicated to the recei)72 592 Q -.165=0A= (ve)-.275 G(rs out of band.).165 E(Ho)5.5 E 2.75(wr)-.275 G(ecei)-2.75 E=0A= -.165(ve)-.275 G(rs obtain the).165 E=0A= (session description is outside the scope of this document.)72 605 Q=0A= (When a recei)72 631 Q -.165(ve)-.275 G 2.75(ri).165 G=0A= (nitiates a session, it \214rst joins the base channel.)-2.75 E=0A= (The pack)5.5 E(ets in the base channel)-.11 E(help the recei)72 644 Q=0A= -.165(ve)-.275 G 2.75(ro).165 G=0A= (rient itself in terms of what the current time slot inde)-2.75 E 2.75=0A= (xi)-.165 G(s, which in turn allo)-2.75 E(ws)-.275 E(the recei)72 657 Q=0A= -.165(ve)-.275 G 2.75(rt).165 G 2.75(ok)-2.75 G(no)-2.75 E 2.75(wt)-.275=0A= G(he relati)-2.75 E .33 -.165(ve r)-.275 H(ates on the w).165 E -2.475=0A= -.22(av e)-.11 H 2.75(channels. The)2.97 F(recei)2.75 E -.165(ve)-.275 G=0A= 2.75(rr).165 G(emains joined to the)-2.75 E=0A= (base channel for the duration of its participation in the session.)72=0A= 670 Q(After joining a session the recei)72 696 Q -.165(ve)-.275 G 2.75=0A= (ra).165 G(djusts its rate upw)-2.75 E(ards and do)-.11 E(wnw)-.275 E=0A= (ards by joining w)-.11 E -2.475 -.22(av e)-.11 H=0A= (channels in sequence, from the lo)72 709 Q(west rate w)-.275 E -2.475=0A= -.22(av e)-.11 H(channel and mo)2.97 E(ving to)-.165 E -.11(wa)-.275 G=0A= (rds the higher rate).11 E -.11(wa)72 722 S .33 -.165(ve c)-.11 H 2.75=0A= (hannels. Since).165 F(the relati)2.75 E .33 -.165(ve o)-.275 H=0A= (rdering among the channels with respect to their rate depends).165 E=0A= (Luby/Go)72 769 Q 289.902(yal Section)-.11 F 2.75(3. [P)2.75 F(age 5])=0A= -.165 E EP=0A= %%Page: 6 6=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E=0A= (on the current time slot inde)72 85 Q=0A= (x, it is important that the recei)-.165 E -.165(ve)-.275 G 2.75(rc).165=0A= G(ontinually monitor the current time)-2.75 E(slot inde)72 98 Q 2.75(xc)=0A= -.165 G(ontained in recei)-2.75 E -.165(ve)-.275 G 2.75(dp).165 G(ack)=0A= -2.75 E 2.75(ets. The)-.11 F(reception rate at the recei)2.75 E -.165=0A= (ve)-.275 G 2.75(ri).165 G 2.75(sd)-2.75 G(etermined by ho)-2.75 E(w)=0A= -.275 E(early each w)72 111 Q -2.475 -.22(av e)-.11 H=0A= (channel is joined by the recei)2.97 E -.165(ve)-.275 G=0A= (r: the earlier the recei).165 E -.165(ve)-.275 G 2.75(rj).165 G=0A= (oins a channel with)-2.75 E(respect to when its w)72 124 Q -2.475 -.22=0A= (av e)-.11 H(started, the higher the reception rate.)2.97 E=0A= (When the recei)5.5 E -.165(ve)-.275 G 2.75(rw).165 G(ants to)-2.86 E=0A= (decrease its rate, it joins the ne)72 137 Q(xt w)-.165 E -2.475 -.22=0A= (av e)-.11 H(channel at a later time relati)2.97 E .33 -.165(ve t)-.275=0A= H 2.75(ow).165 G(hen it joined the)-2.75 E(pre)72 150 Q(vious w)-.275 E=0A= -2.475 -.22(av e)-.11 H 5.5(.W).22 G(hen the recei)-5.5 E -.165(ve)-.275=0A= G 2.75(rw).165 G(ants to increase its rate, it joins the ne)-2.86 E=0A= (xt w)-.165 E -2.475 -.22(av e)-.11 H(channel at an)2.97 E=0A= (earlier time relati)72 163 Q .33 -.165(ve t)-.275 H 2.75(ow).165 G=0A= (hen it joined the pre)-2.75 E(vious w)-.275 E -2.475 -.22(av e)-.11 H=0A= (.).22 E(Once the recei)72 189 Q -.165(ve)-.275 G 2.75(ri).165 G 2.75=0A= (sj)-2.75 G(oined to a w)-2.75 E -2.475 -.22(av e)-.11 H=0A= (channel, the recei)2.97 E -.165(ve)-.275 G 2.75(rr).165 G=0A= (emains joined to the w)-2.75 E -2.475 -.22(av e)-.11 H(channel)2.97 E=0A= (until the channel goes quiescent, at which point the recei)72 202 Q=0A= -.165(ve)-.275 G 2.75(rM).165 G(UST lea)-2.75 E .33 -.165(ve t)-.22 H=0A= (he channel.).165 E(The w)72 228 Q(ay the recei)-.11 E -.165(ve)-.275 G=0A= 2.75(ra).165 G=0A= (djusts its reception rate is inspired by TFRC [5]. The recei)-2.75 E=0A= -.165(ve)-.275 G 2.75(ra).165 G 2.75(ta)-2.75 G(ll points)-2.75 E=0A= (in time maintains a tar)72 241 Q(get reception rate, and the recei)=0A= -.198 E -.165(ve)-.275 G 2.75(ri).165 G 2.75(sa)-2.75 G(llo)-2.75 E=0A= (wed to join the ne)-.275 E(xt w)-.165 E -2.475 -.22(av e)-.11 H=0A= (channel)2.97 E(if after joining its anticipated reception rate w)72 254=0A= Q(ould be at most its tar)-.11 E(get reception rate.)-.198 E(The tar)5.5=0A= E(get)-.198 E=0A= (rate is continually updated based on a set of measured parameters.)72=0A= 267 Q(The primary parameters are)5.5 E(the a)72 280 Q -.165(ve)-.22 G=0A= (rage loss probability \(measured in much the same w).165 E=0A= (ay as described in TFRC\) and the)-.11 E -2.475 -.22(av e)72 293 T=0A= (rage round-trip time \(measured as described belo).22 E(w\).)-.275 E/F1=0A= 11/Times-Bold@0 SF(3.1.)72 335.6 Q/F2 13/Times-Bold@0 SF=0A= (Sender Operation)5.5 E F0=0A= (The sender operation is by design much simpler than the recei)72 352.2=0A= Q -.165(ve)-.275 G 2.75(ro).165 G(peration.)-2.75 E F1 2.75=0A= (3.1.1. Sender)72 394.8 R(inputs and initialization)2.75 E F0=0A= (The primary input to the sender for the session is MSR_b)72 411.4 Q 5.5=0A= (.M)-.44 G(SR_b is the maximum sender)-5.5 E=0A= (transmission rate in bits per second at an)72 424.4 Q 2.75(yp)-.165 G=0A= (oint in time in aggre)-2.75 E -.055(ga)-.165 G(te to all channels.).055=0A= E(This is also)5.5 E(the maximum rate in bits per second that an)72=0A= 437.4 Q 5.5(yr)-.165 G(ecei)-5.5 E -.165(ve)-.275 G 2.75(rc).165 G=0A= (ould recei)-2.75 E .33 -.165(ve d)-.275 H(ata from the session at an)=0A= .165 E(y)-.165 E(point in time.)72 450.4 Q=0A= (The secondary inputs to the sender are listed belo)72 476.4 Q 4.18=0A= -.715(w. T)-.275 H(hese are secondary inputs because in).715 E=0A= (general the v)72 489.4 Q(alues for these inputs will be \214x)-.275 E=0A= (ed to def)-.165 E(ault v)-.11 E(alues that will not change, or because)=0A= -.275 E(the)72 502.4 Q 2.75(ya)-.165 G=0A= (re set based on non-WEBRC considerations.)-2.75 E 11(oL)77.5 519 S=0A= (ENP_B is the length of pack)-11 E(ets in bytes sent to the session.)=0A= -.11 E(The v)5.5 E(alue of LENP_B depends)-.275 E=0A= (on the complete protocol, b)94 532 Q=0A= (ut in general this should be set to as high a v)-.22 E=0A= (alue as possible)-.275 E(without e)94 545 Q=0A= (xceeding the MTU size for the netw)-.165 E(ork that w)-.11 E=0A= (ould cause fragmentation.)-.11 E 11(oB)77.5 561.6 S=0A= (CR_P is the transmission rate on the base channel at the be)-11 E=0A= (ginning of a time slot in pack)-.165 E(ets)-.11 E(per second.)94 574.6=0A= Q(The def)5.5 E(ault v)-.11 E(alue for BCR_P is 1.)-.275 E 11(oT)77.5=0A= 591.2 S(SD is the time slot duration measured in seconds.)-11 E(The def)=0A= 5.5 E(ault v)-.11 E(alue for TSD is 10.)-.275 E 11(oQ)77.5 607.8 S 2.75=0A= (Di)-11 G 2.75(st)-2.75 G=0A= (he minimum quiescent period duration measured in seconds.)-2.75 E=0A= (The def)5.5 E(ault v)-.11 E(alue for)-.275 E(QD is 300.)94 620.8 Q 11=0A= (oP)77.5 637.4 S(is the multiplicati)-8.25 E .33 -.165(ve d)-.275 H=0A= (rop in e).165 E -.165(ve)-.275 G(ry channel rate o).165 E -.165(ve)=0A= -.165 G 2.75(re).165 G(ach time slot.)-2.75 E(The def)5.5 E(ault v)-.11=0A= E(alue for P)-.275 E(is 0.75.)94 650.4 Q(From these inputs the follo)72=0A= 667 Q(wing \214x)-.275 E(ed sender parameters can be deri)-.165 E -.165=0A= (ve)-.275 G 2.75(da).165 G 2.75(sf)-2.75 G(ollo)-2.75 E(ws.)-.275 E 11=0A= (oM)77.5 683.6 S(SR_P =3D MSR_b/\(8*LENP_B\) is the maximum sender = transm\=0A= ission rate in pack)-11 E(ets per)-.11 E(second.)94 696.6 Q(Luby/Go)72=0A= 769 Q 273.402(yal Section)-.11 F 2.75(3.1.1. [P)2.75 F(age 6])-.165 E EP=0A= %%Page: 7 7=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E 11(oB)77.5 85 S=0A= (CR_b =3D 8*LENP_B*BCR_P is the rate of the base channel at the be)-11 E=0A= (ginning of a time slot)-.165 E(in bits per second.)94 98 Q 11(oN)77.5=0A= 114.6 S 2.75<3d8d>-8.25 G(oor{log_{1/P}\(\(1+\(1-P\)/P^2\)*\(MSR_P/BCR_\=0A= P\)\)}-1 is the number of acti)-2.75 E .33 -.165(ve t)-.275 H(ime slots)=0A= .165 E(for a w)94 127.6 Q -2.475 -.22(av e)-.11 H 2.75(channel. A)2.97 F=0A= -.11(wa)2.75 G .33 -.165(ve c)-.11 H(hannel is acti).165 E .33 -.165=0A= (ve i)-.275 H 2.75(na).165 G .33 -.165(ny t)-2.75 H=0A= (ime slot if it is not quiescent.).165 E 2.75(Ni)5.5 G 2.75(sa)-2.75 G=0A= (lso)-2.75 E(the number of w)94 140.6 Q -2.475 -.22(av e)-.11 H=0A= (channels acti)2.97 E .33 -.165(ve i)-.275 H 2.75(ne).165 G -.165(ve)=0A= -3.025 G(ry time slot.).165 E 11(oQ)77.5 157.2 S 2.75(=3Dc)-8.25 G=0A= (eil\(QD/TSD\) is the number of quiescent time slots for a w)-2.75 E=0A= -2.475 -.22(av e)-.11 H(channel.)2.97 E 11(oT)77.5 173.8 S 2.75(=3DN+Qi)=0A= -8.25 G 2.75(st)-2.75 G(he total number of time slots in a c)-2.75 E=0A= 2.75(ycle. T)-.165 F(is also the total number of w)2.75 E -2.475 -.22=0A= (av e)-.11 H(channels.)94 186.8 Q 11(oF)77.5 203.4 S=0A= (or the base channel CN =3D T and for the w)-11.165 E -2.475 -.22(av e)=0A= -.11 H(channels CN =3D 0,1,...,T)2.97 E 2.75(-1. The)-1.012 F=0A= (sender has the)2.75 E(description of the channels assigned to the sess\=0A= ion and the mapping between the channels)94 216.4 Q(and the CNs.)94=0A= 229.4 Q 11(oC)77.5 246 S 2.75(=3DT)-8.25 G=0A= (SD*T is the total duration in seconds of a c)-2.75 E(ycle.)-.165 E/F1=0A= 11/Times-Bold@0 SF 2.75(3.1.2. Sending)72 288.6 R(pack)2.75 E=0A= (ets to the session)-.11 E F0(The sender k)72 305.2 Q=0A= (eeps track of the current time slot inde)-.11 E 2.75(xC)-.165 G 2.75=0A= (TSI. The)-2.75 F -.275(va)2.75 G(lue of CTSI is incremented by).275 E=0A= 2.75(1m)72 318.2 S(odulo T each TSD seconds.)-2.75 E(The v)5.5 E=0A= (alue of CTSI is placed into each pack)-.275 E(et in the format)-.11 E=0A= (described in Section 5. F)72 331.2 Q(or each pack)-.165 E=0A= (et sent to the session, the sender also places the channel)-.11 E=0A= (number CN of the channel into the pack)72 344.2 Q=0A= (ets in the format described in Section 5. Recall that CN =3D)-.11 E 2.75=0A= (Tf)72 357.2 S(or the base channel and CN =3D 0,1,...,T)-2.75 E=0A= (-1 for the w)-1.012 E -2.475 -.22(av e)-.11 H(channels.)2.97 E -.165=0A= (Fo)72 373.8 S 2.75(re).165 G(ach pack)-2.75 E=0A= (et sent to the session, the sender calculates and places a pack)-.11 E=0A= (et sequence number)-.11 E(PSN into the pack)72 386.8 Q 2.75(et. The)=0A= -.11 F -.275(va)2.75 G(lue of PSN is scoped by CN, and the v).275 E=0A= (alue of PSN is consecuti)-.275 E -.165(ve)-.275 G(ly).165 E=0A= (increasing within each channel.)72 399.8 Q(Furthermore, for each w)5.5=0A= E -2.475 -.22(av e)-.11 H(channel, the last pack)2.97 E(et sent before)=0A= -.11 E(the channel becomes quiescent must ha)72 412.8 Q .33 -.165(ve t)=0A= -.22 H(he maximum possible PSN v).165 E 2.75(alue. When)-.275 F=0A= (the short)2.75 E(format for Congestion Control Information is used \(s\=0A= ee Section 5.1\), this implies that for an)72 425.8 Q(y)-.165 E -.11(wa)=0A= 72 438.8 S .33 -.165(ve c)-.11 H(hannel the PSN v).165 E=0A= (alues are 65 535, 65 534, 65 533, ..., going backw)-.275 E=0A= (ard in time starting at)-.11 E(the last pack)72 451.8 Q=0A= (et sent to the channel just before the channel becomes quiescent.)-.11=0A= E(Similarly)5.5 E 2.75(,w)-.715 G(hen the)-2.75 E(long format for Conge\=0A= stion Control Information is used \(see Section 5.2\), the PSN for the \=0A= \214nal)72 464.8 Q(pack)72 477.8 Q(et of an)-.11 E 2.75(yw)-.165 G=0A= -2.475 -.22(av e)-2.86 H(is 4 294 967 295.)2.97 E=0A= (The PSN of the initial pack)5.5 E(et of a w)-.11 E -2.475 -.22(av e)=0A= -.11 H(thus depends on)2.97 E(TSD, P)72 490.8 Q 2.75(,B)-1.221 G=0A= (CR_P and MSR_P)-2.75 E 5.5(.T)-1.221 G=0A= (he format for the PSN within pack)-5.5 E=0A= (ets is described in Section 5.)-.11 E -.165(Pa)72 507.4 S(ck).165 E=0A= (ets are sent to the base channel at a rate of BCR_P pack)-.11 E=0A= (ets per second at the be)-.11 E(ginning of a)-.165 E=0A= (time slot decreasing at a constant relati)72 520.4 Q .33 -.165(ve r)=0A= -.275 H(ate till the rate reaches P*BCR_P at the end of a time).165 E=0A= 2.75(slot. T)72 533.4 R 2.75(oe)-.88 G(xplain the pack)-2.915 E=0A= (et rate for w)-.11 E -2.475 -.22(av e)-.11 H=0A= (channels, suppose for simplicity that MSR_P/BCR_P =3D)2.97 E = 2.75(1+\()72=0A= 546.4 S=0A= (1/P\) + \(1/P\)^2 + ... + \(1/P\)^N =3D \(\(1/P\)^{N+1}-1\)*P/\(1-P\).)=0A= -2.75 E -.165(Fo)5.5 G 2.75(ra).165 G(ll i =3D 0,1,...,T)-2.75 E=0A= (-1, the rate for)-1.012 E -.11(wa)72 559.4 S .33 -.165(ve c)-.11 H=0A= (hannel i beha).165 E -.165(ve)-.22 G 2.75(sa).165 G 2.75(sf)-2.75 G=0A= (ollo)-2.75 E 2.75(ws. At)-.275 F(the be)2.75 E=0A= (ginning of time slot i+Q+1 modulo T)-.165 E 2.75(,p)-.814 G(ack)-2.75 E=0A= (ets are)-.11 E(sent to w)72 572.4 Q -2.475 -.22(av e)-.11 H=0A= (channel i at the rate of BCR_P*\(1/P\)^N.)2.97 E(The rate of pack)5.5 E=0A= (ets sent to w)-.11 E -2.475 -.22(av e)-.11 H(channel i)2.97 E=0A= (steadily decreases geometrically by a f)72 585.4 Q=0A= (actor of P during each of the ne)-.11 E(xt N time slots until the end)=0A= -.165 E(of time slot i when the rate reaches BCR_P)72 598.4 Q 5.5(.D)=0A= -1.221 G(uring the Q time slots i+1 modulo T)-5.5 E 2.75(,i)-.814 G=0A= (+2 modulo T)-2.75 E(,)-.814 E(..., i+Q modulo T)72 611.4 Q 2.75(,w)=0A= -.814 G -2.475 -.22(av e)-2.86 H(channel i is quiescent, i.e., no pack)=0A= 2.97 E(ets are sent to the channel.)-.11 E F1(3.2.)72 654 Q/F2 13=0A= /Times-Bold@0 SF(Recei)5.5 E -.13(ve)-.13 G 3.25(rO).13 G(peration)-3.25=0A= E F0(The b)72 670.6 Q(ulk of the comple)-.22 E=0A= (xity in WEBRC is in the recei)-.165 E -.165(ve)-.275 G 2.75(ro).165 G=0A= 2.75(peration. F)-2.75 F(or ease of e)-.165 E(xplanation,)-.165 E=0A= (suppose for the moment that during the reception there is no pack)72=0A= 683.6 Q(et loss and pack)-.11 E(ets are arri)-.11 E(ving)-.275 E(at e)72=0A= 696.6 Q(xactly the rate at which the)-.165 E 2.75(yw)-.165 G(ere sent.)=0A= -2.75 E(The sender transmission rate to the channels is)5.5 E=0A= (designed so that the recei)72 709.6 Q -.165(ve)-.275 G 2.75(rr).165 G=0A= (eception rate beha)-2.75 E -.165(ve)-.22 G 2.75(sa).165 G 2.75(sf)-2.75=0A= G(ollo)-2.75 E(ws.)-.275 E(Luby/Go)72 769 Q 281.652(yal Section)-.11 F=0A= 2.75(3.2. [P)2.75 F(age 7])-.165 E EP=0A= %%Page: 8 8=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E=0A= (Upon entering a session, the recei)72 85 Q -.165(ve)-.275 G 2.75(ri)=0A= .165 G(mmediately joins the base channel.)-2.75 E(When the recei)5.5 E=0A= -.165(ve)-.275 G(r).165 E -.11(wa)72 98 S=0A= (nts to increase its rate, it consecuti).11 E -.165(ve)-.275 G=0A= (ly joins w).165 E -2.475 -.22(av e)-.11 H(channels from the lo)2.97 E=0A= (west rate channel to the)-.275 E(highest rate channel.)72 111 Q=0A= (\(Recall that the designations of lo)5.5 E=0A= (west to highest change at time slot)-.275 E 2.75(boundaries.\) When)72=0A= 124 R(the recei)2.75 E -.165(ve)-.275 G 2.75(rw).165 G=0A= (ants to maintain its current reception rate and it is already)-2.86 E=0A= (joined to NWC w)72 137 Q -2.475 -.22(av e)-.11 H=0A= (channels, if the recei)2.97 E -.165(ve)-.275 G 2.75(rj).165 G=0A= (oins channel i-1+NWC modulo T sometime during)-2.75 E=0A= (time slot i then the recei)72 150 Q -.165(ve)-.275 G 2.75(rj).165 G=0A= (oins channel i+NWC modulo T TSD seconds later in time slot i+1.)-2.75 E=0A= (As each w)72 163 Q -2.475 -.22(av e)-.11 H=0A= (channel becomes quiescent the recei)2.97 E -.165(ve)-.275 G 2.75(rl)=0A= .165 G(ea)-2.75 E -.165(ve)-.22 G 2.75(st).165 G(he channel.)-2.75 E=0A= (Suppose the recei)72 189 Q -.165(ve)-.275 G 2.75(rw).165 G=0A= (ants to decrease its rate till it is joined to just the base channel.)=0A= -2.86 E(Assume)5.5 E(that a recei)72 202 Q -.165(ve)-.275 G 2.75(ri).165=0A= G 2.75(sj)-2.75 G(oined to the NWC w)-2.75 E -2.475 -.22(av e)-.11 H=0A= (channels i, i+1 modulo T)2.97 E 2.75(,.)-.814 G=0A= (.., i+NWC-1 modulo T at the)-2.75 E(be)72 215 Q=0A= (ginning of time slot i.)-.165 E(Then, the aggre)5.5 E -.055(ga)-.165 G=0A= (te pack).055 E(et reception rate of the recei)-.11 E -.165(ve)-.275 G=0A= 2.75(ro).165 G -.165(ve)-2.915 G 2.75(rt).165 G(he ne)-2.75 E(xt)-.165 E=0A= (NWC time slots will beha)72 228 Q .33 -.165(ve a)-.22 H 2.75(sf).165 G=0A= (ollo)-2.75 E 2.75(ws. At)-.275 F(the be)2.75 E=0A= (ginning of time slot i the recei)-.165 E -.165(ve)-.275 G 2.75(rr).165=0A= G(eception rate)-2.75 E=0A= (is BCR_P*\(1 + \(1/P\) + \(1/P\)^2 + ... + \(1/P\)^NWC\).)72 241 Q=0A= (Then the recei)5.5 E -.165(ve)-.275 G 2.75(rr).165 G=0A= (eception rate decreases by)-2.75 E 2.75(af)72 254 S(actor of P o)-2.86=0A= E -.165(ve)-.165 G 2.75(rt).165 G(he duration of each time slot, and at\=0A= the end of each time slot the reception rate)-2.75 E=0A= (decreases by an additi)72 267 Q .33 -.165(ve a)-.275 H=0A= (mount of P*BCR_P).165 E 5.5(.A)-1.221 G 2.75(tt)-5.5 G=0A= (he end of time slot i+NWC-1 mod T)-2.75 E 2.75(,t)-.814 G(he)-2.75 E=0A= (recei)72 280 Q -.165(ve)-.275 G 2.75(rr).165 G=0A= (eception rate is BCR_P*\(1+P\), and at the be)-2.75 E=0A= (ginning of time slot i+NWC mod T the)-.165 E(recei)72 293 Q -.165(ve)=0A= -.275 G 2.75(ri).165 G 2.75(sj)-2.75 G=0A= (oined only to the base channel and its reception rate is BCR_P)-2.75 E=0A= (.)-1.221 E/F1 11/Times-Bold@0 SF 2.75(3.2.1. Recei)72 335.6 R -.11(ve)=0A= -.11 G 2.75(ri).11 G(nputs and initialization)-2.75 E F0=0A= (Before joining a session the recei)72 352.2 Q -.165(ve)-.275 G 2.75(rM)=0A= .165 G(UST kno)-2.75 E 2.75(wt)-.275 G=0A= (he mapping between the CNs and the)-2.75 E 2.75(channels. Upon)72 365.2=0A= R(joining the session, it should ha)2.75 E .33 -.165(ve t)-.22 H(he v)=0A= .165 E(alues of LENP_B, BCR_P)-.275 E 2.75(,T)-1.221 G(SD, P)-2.75 E=0A= 2.75(,N)-1.221 G 2.75(,Q)-2.75 G(and T)72 378.2 Q 5.5(.S)-.814 G=0A= (ome of these v)-5.5 E(alues may be obtained or measured once the recei)=0A= -.275 E -.165(ve)-.275 G 2.75(rh).165 G(as joined the)-2.75 E 2.75=0A= (session. F)72 391.2 R(or e)-.165 E(xample, the recei)-.165 E -.165(ve)=0A= -.275 G 2.75(rM).165 G 2.31 -1.155(AY o)-2.75 H=0A= (btain LENP_B and T from the \214rst pack)1.155 E(et recei)-.11 E -.165=0A= (ve)-.275 G(d).165 E(from the base channel, and the recei)72 404.2 Q=0A= -.165(ve)-.275 G 2.75(rM).165 G 2.31 -1.155(AY m)-2.75 H=0A= (easure BCR_P once it is joined to the base)1.155 E 2.75(channel. The)72=0A= 417.2 R -.275(va)2.75 G(lues of P).275 E 2.75(,Qa)-1.221 G(nd TSD MA)=0A= -2.75 E 2.75(Yb)-1.155 G 2.75<658c>-2.75 G -.165(xe)-2.75 G 2.75(dt).165=0A= G 2.75(od)-2.75 G(ef)-2.75 E(ault v)-.11 E(alues b)-.275 E=0A= (uilt into the recei)-.22 E -.165(ve)-.275 G 2.75(rt).165 G(hat)-2.75 E=0A= (do not change from session to session, and the v)72 430.2 Q=0A= (alue of N MA)-.275 E 2.75(Yb)-1.155 G 2.75(ec)-2.75 G(omputed as T)=0A= -2.75 E(-Q.)-1.012 E(When a recei)72 456.2 Q -.165(ve)-.275 G 2.75<728c>=0A= .165 G(rst joins a session, it MUST \214rst join just the base channel \=0A= and start recei)-2.75 E(ving)-.275 E(pack)72 469.2 Q=0A= (ets to determine the current time slot inde)-.11 E 2.75(x. If)-.165 F=0A= (during the course of the session the recei)2.75 E -.165(ve)-.275 G(r)=0A= .165 E(continually loses a high fraction of the pack)72 482.2 Q=0A= (ets from the base channel e)-.11 E -.165(ve)-.275 G 2.75(nw).165 G=0A= (hen the recei)-2.75 E -.165(ve)-.275 G 2.75(ri).165 G(s)-2.75 E=0A= (only joined to the base channel, the recei)72 495.2 Q -.165(ve)-.275 G=0A= 2.75(rS).165 G(HOULD lea)-2.75 E .33 -.165(ve t)-.22 H(he session.).165=0A= E(The recei)72 521.2 Q -.165(ve)-.275 G 2.75(rM).165 G 2.31 -1.155(AY a)=0A= -2.75 H(lso ha)1.155 E .33 -.165(ve o)-.22 H(ther indi).165 E=0A= (vidually set parameters that may be used to determine its)-.275 E(beha)=0A= 72 534.2 Q(vior)-.22 E 5.5(.T)-.605 G .22 -.11(wo s)-6.38 H=0A= (uch parameters are:).11 E 11(oM)77.5 550.8 S(RR_b is the maximum recei)=0A= -11 E -.165(ve)-.275 G 2.75(rr).165 G(eception rate in bits per second.)=0A= -2.75 E(This may be used to)5.5 E=0A= (determine the maximum reception rate this recei)94 563.8 Q -.165(ve)=0A= -.275 G 2.75(ri).165 G 2.75(sw)-2.75 G(illing to reach.)-2.75 E=0A= (Thus, the maximum)5.5 E(reception rate that the recei)94 576.8 Q -.165=0A= (ve)-.275 G 2.75(rc).165 G(an possibly achie)-2.75 E .33 -.165(ve i)=0A= -.275 H 2.75(nt).165 G(he session is the minimum of MSR_b)-2.75 E=0A= (and MRR_b)94 589.8 Q 5.5(.A)-.44 G(recommended v)-2.75 E=0A= (alue of MRR_b for a recei)-.275 E -.165(ve)-.275 G 2.75(ri).165 G 2.75=0A= (st)-2.75 G(he bandwidth capacity of)-2.75 E(the last link to the recei)=0A= 94 602.8 Q -.165(ve)-.275 G 3.96 -.605(r. M).165 H=0A= (RR_P is the maximum recei).605 E -.165(ve)-.275 G 2.75(rr).165 G=0A= (eception rate in pack)-2.75 E(ets per)-.11 E=0A= (second, i.e., MRR_P =3D MRR_b/\(8*LENP_B\).)94 615.8 Q 11(oC)77.5 632.4 = S=0A= (ONNEC is the number of virtual connections that the recei)-11 E -.165=0A= (ve)-.275 G 2.75(ri).165 G 2.75(sm)-2.75 G(aking to the sender to)-2.75=0A= E(recei)94 645.4 Q .33 -.165(ve d)-.275 H(ata from the session.).165 E=0A= (This af)5.5 E(fects ho)-.275 E 2.75(wc)-.275 G(ompetiti)-2.75 E .33=0A= -.165(ve t)-.275 H(he recei).165 E -.165(ve)-.275 G 2.75(rw).165 G=0A= (ill be compared to)-2.75 E(other recei)94 658.4 Q -.165(ve)-.275 G=0A= (rs \(both TCP and WEBRC recei).165 E -.165(ve)-.275 G=0A= (rs\) in competing for bandwidth o).165 E -.165(ve)-.165 G(r).165 E=0A= (bottleneck links.)94 671.4 Q(The standard def)5.5 E(ault v)-.11 E=0A= (alue for CONNEC is 1.)-.275 E(Luby/Go)72 769 Q 273.402(yal Section)-.11=0A= F 2.75(3.2.1. [P)2.75 F(age 8])-.165 E EP=0A= %%Page: 9 9=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E/F1 11/Times-Bold@0 SF 2.75=0A= (3.2.2. Recei)72 85 R -.11(ve)-.11 G 2.75(rm).11 G(easur)-2.75 E=0A= (ements and calculations)-.198 E F0=0A= (As outlined in the introduction, the w)72 101.6 Q(ay a recei)-.11 E=0A= -.165(ve)-.275 G 2.75(ra).165 G=0A= (djusts its reception rate is inspired by TFRC)-2.75 E([5]. The recei)72=0A= 114.6 Q -.165(ve)-.275 G 2.75(ra).165 G 2.75(ta)-2.75 G=0A= (ll points in time maintains a tar)-2.75 E=0A= (get reception rate, and the recei)-.198 E -.165(ve)-.275 G 2.75(ri).165=0A= G 2.75(sa)-2.75 G(llo)-2.75 E(wed)-.275 E(to join the ne)72 127.6 Q=0A= (xt w)-.165 E -2.475 -.22(av e)-.11 H(channel if joining w)2.97 E=0A= (ould increase its reception rate to at most its tar)-.11 E(get)-.198 E=0A= (reception rate.)72 140.6 Q(The tar)5.5 E=0A= (get rate is continually updated based on a set of measured parameters.)=0A= -.198 E(The)5.5 E(primary parameters are the a)72 153.6 Q -.165(ve)-.22=0A= G(rage loss probability LOSSP and the a).165 E -.165(ve)-.22 G=0A= (rage round-trip time).165 E(AR)72 166.6 Q(TT)-.66 E(.)-.814 E F1 2.75=0A= (3.2.2.1. A)72 209.2 R -.11(ve)-1.1 G(rage loss pr).11 E(obability)-.198=0A= E F0(The a)72 225.8 Q -.165(ve)-.22 G=0A= (rage loss probability LOSSP of the recei).165 E -.165(ve)-.275 G 2.75=0A= (ri).165 G 2.75(sm)-2.75 G(aintained in a manner v)-2.75 E=0A= (ery similar to that)-.165 E=0A= (described in TFRC [5]. The calculation of LOSSP as the in)72 238.8 Q=0A= -.165(ve)-.44 G(rse of a weighted a).165 E -.165(ve)-.22 G(rage is e)=0A= .165 E(xactly)-.165 E=0A= (as described in TFRC, with the same de\214nition of loss e)72 251.8 Q=0A= -.165(ve)-.275 G 2.75(nts. One).165 F(dif)2.75 E=0A= (ference is the number of)-.275 E(inter)72 264.8 Q=0A= (-loss periods that are used to compute the a)-.22 E -.165(ve)-.22 G=0A= (rage loss e).165 E -.165(ve)-.275 G(nt rate:).165 E 11(oI)77.5 281.4 S=0A= 2.75(nT)-11 G(FRC, it is recommended that the loss probability be a)=0A= -2.75 E -.165(ve)-.22 G(raged o).165 E -.165(ve)-.165 G 2.75(rt).165 G=0A= (he pre)-2.75 E(vious 8 loss)-.275 E -2.365 -.275(ev e)94 294.4 T(nts.)=0A= .275 E 11(oI)77.5 311 S 2.75(nW)-11 G=0A= (EBRC, it is recommended that the loss probability LOSSP be a)-2.75 E=0A= -.165(ve)-.22 G(raged o).165 E -.165(ve)-.165 G 2.75(rt).165 G(he)-2.75=0A= E(pre)94 324 Q=0A= (vious channel periods that include 8 channel periods where there w)=0A= -.275 E(as at least one loss)-.11 E -2.365 -.275(ev e)94 337 T 2.75=0A= (nt. A).275 F(channel period is de\214ned as the time between when a w)=0A= 2.75 E -2.475 -.22(av e)-.11 H(channel is joined till)2.97 E=0A= (the time the ne)94 350 Q(xt w)-.165 E -2.475 -.22(av e)-.11 H=0A= (channel is joined.)2.97 E=0A= (Thus, since a channel period is generally around)5.5 E=0A= (TSD seconds, the loss probability will generally be a)94 363 Q -.165=0A= (ve)-.22 G(raged o).165 E -.165(ve)-.165 G 2.75(ra).165 G 2.75(tl)-2.75=0A= G(east 8*TSD seconds.)-2.75 E(The other dif)72 379.6 Q=0A= (ference is in the weights:)-.275 E 11(oI)77.5 396.2 S 2.75(nT)-11 G(FR\=0A= C, it is recommended that the weights \(1, 1, 1, 1, 0.875, 0.625, 0.375\=0A= , 0.125\) be used)-2.75 E(for the 8 inter)94 409.2 Q=0A= (-loss periods \(from most recent to most distant past\).)-.22 E 11(oI)=0A= 77.5 425.8 S 2.75(nW)-11 G(EBRC, it is recommended that these same 8 v)=0A= -2.75 E(alues be used for the loss e)-.275 E -.165(ve)-.275 G=0A= (nts that are).165 E(1,2,...,8 lossy channel periods into the past.)94=0A= 438.8 Q(In addition, it is recommended that the weight)5.5 E=0A= (for each inter)94 451.8 Q(-loss period be multiplicati)-.22 E -.165(ve)=0A= -.275 G(ly adjusted by the f).165 E(actor P^max{0,LOSS_NWC-)-.11 E=0A= (NWC-1}, where NWC is the number of w)94 464.8 Q -2.475 -.22(av e)-.11 H=0A= (channels the recei)2.97 E -.165(ve)-.275 G 2.75(ri).165 G 2.75(sc)-2.75=0A= G(urrently subscribed to)-2.75 E(and LOSS_NWC is the number of w)94=0A= 477.8 Q -2.475 -.22(av e)-.11 H(channels the recei)2.97 E -.165(ve)-.275=0A= G 2.75(rw).165 G(as subscribed to at the end)-2.86 E(of the inter)94=0A= 490.8 Q(-loss period in question.)-.22 E F1 2.75(3.2.2.2. A)72 533.4 R=0A= -.11(ve)-1.1 G(rage r).11 E(ound-trip time)-.198 E F0(The recei)72 550 Q=0A= -.165(ve)-.275 G 2.75(rm).165 G(aintains an a)-2.75 E -.165(ve)-.22 G=0A= (rage round-trip time, AR).165 E(TT)-.66 E 2.75(,a)-.814 G 2.75(st)-2.75=0A= G(he e)-2.75 E(xponentially weighted mo)-.165 E(ving)-.165 E -2.475 -.22=0A= (av e)72 563 T(rage of R).22 E(TT measurements.)-.66 E=0A= (Each time the recei)5.5 E -.165(ve)-.275 G 2.75(rj).165 G=0A= (oins a channel \(either the base channel at)-2.75 E(the be)72 576 Q=0A= (ginning or w)-.165 E -2.475 -.22(av e)-.11 H=0A= (channels continually\), it mak)2.97 E=0A= (es a measurement of the round-trip time R)-.11 E(TT)-.66 E(as follo)72=0A= 589 Q 2.75(ws. When)-.275 F(the recei)2.75 E -.165(ve)-.275 G 2.75(rs)=0A= .165 G=0A= (ends the join for the channel it records the current time X and sets)=0A= -2.75 E 2.75(aB)72 602 S(oolean v)-2.75 E(ariable JOINING to true.)-.275=0A= E(When the \214rst pack)5.5 E(et is recei)-.11 E -.165(ve)-.275 G 2.75=0A= (df).165 G(rom the channel the)-2.75 E(recei)72 615 Q -.165(ve)-.275 G=0A= 2.75(rr).165 G(ecords the current time Y and resets the v)-2.75 E=0A= (alue of JOINING to f)-.275 E 2.75(alse. If)-.11 F(it is the base)2.75 E=0A= (channel that has been joined, R)72 628 Q(TT is set to Y - X.)-.66 E=0A= (Otherwise, when the recei)5.5 E -.165(ve)-.275 G 2.75(rr).165 G(ecei)=0A= -2.75 E -.165(ve)-.275 G 2.75(sas).165 G(econd)-2.75 E(pack)72 641 Q=0A= (et from the channel it records the current time Z.)-.11 E(Then, the v)=0A= 5.5 E(alue of R)-.275 E(TT is set to 3*Y/2 -)-.66 E(Z/2 - X.)72 654 Q=0A= (\(Note that this v)5.5 E(alue can be ne)-.275 E -.055(ga)-.165 G(ti)=0A= .055 E -.165(ve)-.275 G(.\)).165 E(When the kth R)72 680 Q=0A= (TT measurement is made, the v)-.66 E(alue of AR)-.275 E=0A= (TT is updated to max{ARR)-.66 E(T/2,)-.66 E(\(1-delta\)*AR)72 693 Q=0A= (TT + delta*R)-.66 E(TT}, where delta =3D alpha/\(1-\(1-alpha\)^k\).)-.66=0A= E(The recommended v)5.5 E(alue for)-.275 E(alpha is 0.1.)72 706 Q=0A= (Luby/Go)72 769 Q 265.152(yal Section)-.11 F 2.75(3.2.2.2. [P)2.75 F=0A= (age 9])-.165 E EP=0A= %%Page: 10 10=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E=0A= (It is RECOMMENDED that the R)72 85 Q=0A= (TT measurement be adjusted appropriately for pack)-.66 E(ets lost)-.11=0A= E(between the \214rst and second recei)72 98 Q -.165(ve)-.275 G 2.75(dp)=0A= .165 G(ack)-2.75 E(ets.)-.11 E/F1 11/Times-Bold@0 SF 2.75(3.2.2.3. Rate)=0A= 72 140.6 R(Equation)2.75 E F0(The recei)72 157.2 Q -.165(ve)-.275 G 2.75=0A= (rc).165 G=0A= (alculates the reception rate REQN based on the TCP-equation as follo)=0A= -2.75 E(ws: REQN =3D)-.275 E(CONNEC/\(AR)72 170.2 Q=0A= (TT*sqrt{LOSSP}\(0.816 + 7.35*LOSSP*\(1+32*LOSSP^2\)\)\).)-.66 E=0A= (This equation)5.5 E(comes from TFRC [5].)72 183.2 Q F1 2.75=0A= (3.2.2.4. Epochs)72 212.8 R F0(The recei)72 229.4 Q -.165(ve)-.275 G=0A= 2.75(rm).165 G(ak)-2.75 E=0A= (es decisions on whether or not to join another w)-.11 E -2.475 -.22=0A= (av e)-.11 H(channel at equally spaced)2.97 E=0A= (units of time called epochs.)72 242.4 Q=0A= (The duration of an epoch in seconds, EL, is set to be a small fraction)=0A= 5.5 E(of TSD, so that decisions to join a channel can be made at a much\=0A= \214ner granularity than TSD.)72 255.4 Q(A)5.5 E=0A= (standard setting is EL =3D TSD/20.)72 268.4 Q=0A= (Thus, if TSD =3D 10, then EL =3D 0.5.)5.5 E F1 2.75(3.2.2.5. A)72 311 R=0A= -.11(ve)-1.1 G(rage r).11 E(eception rate)-.198 E F0(There are tw)72=0A= 327.6 Q 2.75(oa)-.11 G -.165(ve)-2.97 G(raged v).165 E=0A= (ersions of reception rate maintained by the recei)-.165 E -.165(ve)=0A= -.275 G(r: TRR_P).165 E 2.75(,t)-1.221 G(he true)-2.75 E=0A= (reception rate, and ARR_P)72 340.6 Q 2.75(,t)-1.221 G=0A= (he anticipated reception rate.)-2.75 E(These are used for dif)5.5 E=0A= (ferent purposes)-.275 E(and thus are calculated quite dif)72 353.6 Q=0A= (ferently)-.275 E(.)-.715 E=0A= (The true reception rate TRR_P is used to ensure that the recei)72 379.6=0A= Q -.165(ve)-.275 G 2.75(rd).165 G(oes not increase its reception)-2.75 E=0A= (rate too quickly abo)72 392.6 Q .33 -.165(ve i)-.165 H=0A= (ts current reception rate.).165 E=0A= (TRR_P is calculated based on the measurement)5.5 E(of RR_P)72 405.6 Q=0A= 2.75(,w)-1.221 G(here RR_P is the recei)-2.75 E -.165(ve)-.275 G 2.75=0A= (rr).165 G(eception rate in pack)-2.75 E(ets per second measured at the)=0A= -.11 E(be)72 418.6 Q(ginning of an epoch a)-.165 E -.165(ve)-.22 G=0A= (raged o).165 E -.165(ve)-.165 G 2.75(rt).165 G=0A= (he epoch that just ended.)-2.75 E(TRR_P is initialized to BCR_P)5.5 E=0A= (.)-1.221 E(TRR_P is updated to \(1-beta\)*TRR_P + beta*RR_P at the be)=0A= 72 431.6 Q(ginning of each epoch after RR_P is)-.165 E=0A= (measured for the pre)72 444.6 Q(vious epoch.)-.275 E=0A= (The anticipated reception rate ARR_P is the recei)72 470.6 Q -.165(ve)=0A= -.275 G(r').165 E 2.75(se)-.605 G=0A= (stimate of the total instantaneous rate of)-2.75 E=0A= (the currently subscribed channels.)72 483.6 Q(It is used to compare ag)=0A= 5.5 E(ainst the tar)-.055 E(get rate to decide whether)-.198 E=0A= (or not the recei)72 496.6 Q -.165(ve)-.275 G 2.75(rc).165 G=0A= (an increase its reception rate by joining the ne)-2.75 E(xt channel.)=0A= -.165 E(ARR_P is)5.5 E(calculated based on a measurement IRR_P and on t\=0A= he number of subscribed w)72 509.6 Q -2.475 -.22(av e)-.11 H(channels)=0A= 2.97 E 2.75(NWC. IRR_P)72 522.6 R(is what the ideal recei)2.75 E -.165=0A= (ve)-.275 G 2.75(rr).165 G(eception rate should ha)-2.75 E .33 -.165=0A= (ve b)-.22 H(een in pack).165 E(ets per second)-.11 E=0A= (measured at the be)72 535.6 Q(ginning of the epoch a)-.165 E -.165(ve)=0A= -.22 G(raged o).165 E -.165(ve)-.165 G 2.75(rt).165 G(he pre)-2.75 E=0A= (vious epoch, i.e., IRR_P includes)-.275 E(both recei)72 548.6 Q -.165=0A= (ve)-.275 G 2.75(da).165 G(nd lost pack)-2.75 E 2.75(ets. ARR_P)-.11 F=0A= (is initialized to BCR_P and NWC is initialized to 0.)2.75 E(ARR_P)72=0A= 561.6 Q 2.75(,I)-1.221 G(RR_P and NWC are updated as follo)-2.75 E(ws:)=0A= -.275 E 11(oA)77.5 578.2 S 2.75(tt)-11 G(he be)-2.75 E=0A= (ginning of each epoch, IRR_P is measured o)-.165 E -.165(ve)-.165 G=0A= 2.75(rt).165 G(he pre)-2.75 E(vious epoch and then)-.275 E=0A= (ARR_P is updated to P^\(EL/TSD\)*\(1-g)94 591.2 Q(amma\)*ARR_P + g)=0A= -.055 E(amma*IRR_P)-.055 E(.)-1.221 E 11(oW)77.5 607.8 S=0A= (hen a join is made to a w)-11 E -2.475 -.22(av e)-.11 H=0A= (channel, NWC is updated to NWC+1 and then ARR_P is)2.97 E(multiplicati)=0A= 94 620.8 Q -.165(ve)-.275 G(ly increased by the f).165 E=0A= (actor \(\(1/P\)^\(NWC+1\)-1\)/\(\(1/P\)^NWC-1\).)-.11 E(\(Joins happen)=0A= 5.5 E(at epoch boundaries; this adjustment is in addition to the adjust\=0A= ment abo)94 633.8 Q -.165(ve)-.165 G(.\)).165 E 11(oW)77.5 650.4 S=0A= (hen a ne)-11 E 2.75(wt)-.275 G(ime slot inde)-2.75 E 2.75(xi)-.165 G=0A= 2.75(sd)-2.75 G(etected, ARR_P is additi)-2.75 E -.165(ve)-.275 G=0A= (ly increased by \(1-P\)*BCR_P to).165 E=0A= (account for the change in rate on the base channel.)94 663.4 Q=0A= (In addition, for each w)5.5 E -2.475 -.22(av e)-.11 H(channel that)2.97=0A= E(has just been detected to be quiescent and thus had a lea)94 676.4 Q=0A= .33 -.165(ve m)-.22 H(essage sent, ARR_P is).165 E(additi)94 689.4 Q=0A= -.165(ve)-.275 G(ly decreased by BCR_P and NWC is decremented by 1.).165=0A= E(\(Normally)5.5 E 2.75(,ac)-.715 G(hange of time)-2.75 E(slot inde)94=0A= 702.4 Q 2.75(xi)-.165 G(ndicates that one w)-2.75 E -2.475 -.22(av e)=0A= -.11 H(channel has become quiescent.\))2.97 E(Luby/Go)72 769 Q 259.652=0A= (yal Section)-.11 F 2.75(3.2.2.5. [P)2.75 F(age 10])-.165 E EP=0A= %%Page: 11 11=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E=0A= (Consider for the moment what happens if g)72 85 Q=0A= (amma =3D 0 and ARR_P is an accurate estimate of the)-.055 E=0A= (total rate of the subscribed channels.)72 98 Q=0A= (The adjustments to ARR_P upon joining and lea)5.5 E(ving w)-.22 E=0A= -2.475 -.22(av e)-.11 H(channels, with the passage of epochs, and with \=0A= the detection of time slot changes will then cause)72 111 Q=0A= (ARR_P to remain an accurate estimate.)72 124 Q(In practice, g)5.5 E=0A= (amma MUST be positi)-.055 E -.165(ve)-.275 G 2.75(;a).165 G(llo)-2.75 E=0A= (wing an)-.275 E(in\215uence of IRR_P pre)72 137 Q -.165(ve)-.275 G=0A= (nts ARR_P from drifting a).165 E -.11(wa)-.165 G 2.75(yf).11 G=0A= (rom being an accurate estimate of the)-2.75 E(total subscribed rate.)72=0A= 150 Q(The moti)72 176 Q -.275(va)-.275 G=0A= (tion for separate estimates TRR_P and ARR_P is as follo).275 E 2.75=0A= (ws. TRR_P)-.275 F(alone is not)2.75 E=0A= (appropriate for comparison with the TFRC-inspired tar)72 189 Q=0A= (get rate because there is a considerable lag)-.198 E=0A= (before it re\215ects the rate increase resulting from joining the ne)72=0A= 202 Q(xt w)-.165 E -2.475 -.22(av e)-.11 H 2.75(channel. Ho)2.97 F(we)=0A= -.275 E -.165(ve)-.275 G .88 -.44(r, T).165 H(RR_P).44 E(is needed beca\=0A= use ARR_P alone does not re\215ect the amount of data getting to the re\=0A= cei)72 215 Q -.165(ve)-.275 G .88 -.44(r, a).165 H(nd a).44 E(lar)72 228=0A= Q(ge g)-.198 E(ap between the subscribed rate and the recei)-.055 E=0A= -.165(ve)-.275 G 2.75(dr).165 G(ate should not be allo)-2.75 E(wed.)=0A= -.275 E(The recommended v)72 254 Q(alues of beta and g)-.275 E=0A= (amma depend on whether the recei)-.055 E -.165(ve)-.275 G 2.75(ri).165=0A= G 2.75(si)-2.75 G 2.75(ns)-2.75 G(lo)-2.75 E 2.75(ws)-.275 G(tart)-2.75=0A= E 2.75(mode. Slo)72 267 R 2.75(ws)-.275 G(tart is indicated by TRR_P < \=0A= SSR_P/zeta^2, where SSR_P is de\214ned in the)-2.75 E(follo)72 280 Q(wi\=0A= ng subsection and zeta =3D \(\(1/P\)^\(NWC+2\)-1\)/\(\(1/P\)^\(NWC+1\)-1\=0A= \).)-.275 E(It is recommended that)5.5 E(beta and g)72 293 Q=0A= (amma equal P/\(1+P\) in slo)-.055 E 2.75(ws)-.275 G=0A= (tart and equal \(2/3\)*\(-c/2 + sqrt\(c^2/4 + 3*c/2\), where c =3D)-2.75=0A= E 2.75(1-c)72 306 S(os\(6.283*EL/TSD\), otherwise.)-2.75 E/F1 11=0A= /Times-Bold@0 SF 2.75(3.2.2.6. Slo)72 348.6 R 2.75(ws)-.11 G(tart)-2.75=0A= E F0(WEBRC uses a slo)72 365.2 Q 2.75(ws)-.275 G=0A= (tart mechanism to quickly ramp up its rate at both the be)-2.75 E=0A= (ginning of the)-.165 E(session and in the middle of a session when the\=0A= rate drops precipitously)72 378.2 Q 5.5(.T)-.715 G 2.75(oe)-6.38 G=0A= (nact this, the)-2.75 E(recei)72 391.2 Q -.165(ve)-.275 G 2.75(rm).165 G=0A= (aintains the follo)-2.75 E(wing parameters:)-.275 E 11(oS)77.5 407.8 S=0A= (SMINR_P is the minimum allo)-11 E(wed slo)-.275 E 2.75(ws)-.275 G=0A= (tart threshold rate in pack)-2.75 E(ets per second.)-.11 E(The)5.5 E=0A= (recommended v)94 420.8 Q(alue for SSMINR_P is BCR_P*\(1+1/P+1/P^2\).)=0A= -.275 E 11(oS)77.5 437.4 S(SR_P is the slo)-11 E 2.75(ws)-.275 G=0A= (tart threshold rate in pack)-2.75 E(ets per second.)-.11 E=0A= (SSR_P is initialized to)5.5 E(MRR_P*P^2.)94 450.4 Q F1 2.75(3.2.2.7. T)=0A= 72 493 R(ar)-1.012 E(get rate)-.11 E F0(The tar)72 509.6 Q(get rate TRA)=0A= -.198 E(TE is computed as TRA)-1.221 E(TE =3D min{max{SSR_P)-1.221 E 2.75=0A= (,R)-1.221 G(EQN},)-2.75 E=0A= (TRR_P*\(\(1/P\)^\(NWC+3\)-1\)/\(\(1/P\)^\(NWC+1\)-1\), MRR_P}.)72 522.6=0A= Q F1 2.75(3.2.3. Recei)72 565.2 R -.11(ve)-.11 G 2.75(re).11 G -.11(ve)=0A= -2.915 G(nts).11 E F0(There are v)72 581.8 Q(arious recei)-.275 E -.165=0A= (ve)-.275 G 2.75(re).165 G -.165(ve)-3.025 G=0A= (nts, some of which are triggered by the passing of time on the).165 E=0A= (recei)72 594.8 Q -.165(ve)-.275 G .88 -.44(r, a).165 H(nd others by e)=0A= .44 E -.165(ve)-.275 G(nts such as pack).165 E=0A= (et reception, detection of pack)-.11 E(et loss, reception of a)-.11 E=0A= (\214rst pack)72 607.8 Q(et from a channel, and e)-.11 E=0A= (xceptional time-outs.)-.165 E F1 2.75(3.2.3.1. Epoch)72 650.4 R(change)=0A= 2.75 E F0(This is an e)72 667 Q -.165(ve)-.275 G(nt that is dri).165 E=0A= -.165(ve)-.275 G 2.75(nb).165 G 2.75(yt)-2.75 G=0A= (he passage of time on the recei)-2.75 E -.165(ve)-.275 G .88 -.44(r, w)=0A= .165 H(hich occurs each EL).44 E 2.75(seconds. When)72 680 R(this happe\=0A= ns, TRR_P and ARR_P are computed as described in Section 3.2.2.5.)2.75 E=0A= (Immediately after these updates, a decision is made about whether to j\=0A= oin a w)72 693 Q -2.475 -.22(av e)-.11 H(channel as)2.97 E=0A= (described in Section 3.2.3.3.)72 706 Q(Luby/Go)72 769 Q 259.652=0A= (yal Section)-.11 F 2.75(3.2.3.1. [P)2.75 F(age 11])-.165 E EP=0A= %%Page: 12 12=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E/F1 11/Times-Bold@0 SF 2.75=0A= (3.2.3.2. T)72 85 R(ime slot change)-.198 E F0(This is an e)72 101.6 Q=0A= -.165(ve)-.275 G(nt that is dri).165 E -.165(ve)-.275 G 2.75(nb).165 G=0A= 2.75(yt)-2.75 G(he reception of a pack)-2.75 E(et with a ne)-.11 E 2.75=0A= (wC)-.275 G(TSI v)-2.75 E 2.75(alue. When)-.275 F 2.75(ap)2.75 G(ack)=0A= -2.75 E(et)-.11 E(with a ne)72 114.6 Q 2.75(wC)-.275 G(TSI =3D i is = recei)=0A= -2.75 E -.165(ve)-.275 G(d, a lea).165 E .33 -.165(ve i)-.22 H 2.75(ss)=0A= .165 G(ent for w)-2.75 E -2.475 -.22(av e)-.11 H(channel i-1 modulo T)=0A= 2.97 E 2.75(,N)-.814 G(WC is updated)-2.75 E(to NWC-1, and ARR_P is upd\=0A= ated to ARR_P - P*BCR_P as described in Section 3.2.2.5.)72 127.6 Q F1=0A= 2.75(3.2.3.3. J)72 157.2 R(oin a wa)-.165 E .22 -.11(ve c)-.275 H=0A= (hannel).11 E F0(At the be)72 173.8 Q=0A= (ginning of each epoch, after updating the v)-.165 E=0A= (alues of ARR_P and TRR_P as described in)-.275 E=0A= (Section 3.2.2.5, the recei)72 186.8 Q -.165(ve)-.275 G 2.75(rd).165 G=0A= (ecides whether or not to join a w)-2.75 E -2.475 -.22(av e)-.11 H=0A= (channel as follo)2.97 E(ws:)-.275 E 11(oI)77.5 203.4 S 2.75(ft)-11 G=0A= (here is a loss e)-2.75 E -.165(ve)-.275 G(nt in progress \(LOSS_EVENT \=0A= =3D true\) or there is a join of a channel in).165 E=0A= (progress \(JOINING =3D true\), then no join of a channel is = attempted.)94=0A= 216.4 Q 11(oT)77.5 233 S(he recei)-11 E -.165(ve)-.275 G 2.75(rc).165 G=0A= (alculates REQN as described in Section 3.2.2.3.)-2.75 E 11(oT)77.5=0A= 249.6 S(he recei)-11 E -.165(ve)-.275 G 2.75(rc).165 G(alculates TRA)=0A= -2.75 E(TE as described in Section 3.2.2.7.)-1.221 E 11(oT)77.5 266.2 S=0A= (he recei)-11 E -.165(ve)-.275 G 2.75(rc).165 G(hecks if TRA)-2.75 E=0A= (TE > ARR_P*\(\(1/P\)^{NWC+2}-1\)/\(\(1/P\)^{NWC+1}-1\).)-1.221 E=0A= (In this)5.5 E(case, subscribing to an additional w)94 279.2 Q -2.475=0A= -.22(av e)-.11 H(channel is not e)2.97 E=0A= (xpected to increase the subscribed)-.165 E(rate o)94 292.2 Q -.165(ve)=0A= -.165 G 2.75(rt).165 G(he tar)-2.75 E(get rate.)-.198 E=0A= (Suppose CTSI =3D i and NWC =3D J.)5.5 E(If the inequality is true then = the)=0A= 5.5 E(recei)94 305.2 Q -.165(ve)-.275 G 2.75(rj).165 G(oins the w)-2.75=0A= E -2.475 -.22(av e)-.11 H(channel with CN =3D i+J modulo T)2.97 E = 2.75(,u)=0A= -.814 G(pdates NWC to NWC+1, and)-2.75 E(then updates AR)94 318.2 Q(TT \=0A= to ARR_P*\(\(1/P\)^{NWC+1}-1\)/\(\(1/P\)^NWC-1\) as described in Sectio\=0A= n)-.66 E(3.2.2.5.)94 331.2 Q F1 2.75(3.2.3.4. Loss)72 360.8 R -2.53=0A= -.165(ev e)2.75 H(nt).165 E F0(Each time the recei)72 377.4 Q -.165(ve)=0A= -.275 G 2.75(rd).165 G(etects a lost pack)-2.75 E=0A= (et \(based on the sequence numbers in the pack)-.11 E(ets scoped)-.11 E=0A= (by the channel number\), the recei)72 390.4 Q -.165(ve)-.275 G 2.75(rr)=0A= .165 G(ecords the start of a ne)-2.75 E 2.75(wl)-.275 G(oss e)-2.75 E=0A= -.165(ve)-.275 G(nt, and sets a Boolean).165 E -.275(va)72 403.4 S=0A= (riable LOSS_EVENT to true that will automatically reset to f).275 E=0A= (alse after AR)-.11 E(TT seconds.)-.66 E(All)5.5 E(subsequent pack)72=0A= 416.4 Q(et loss for a period of AR)-.11 E=0A= (TT seconds is considered as part of the same loss e)-.66 E -.165(ve)=0A= -.275 G(nt.).165 E(When a start of a loss e)72 429.4 Q -.165(ve)-.275 G=0A= (nt is detected, the v).165 E(alue of SSR_P is updated to max{SSMINR_P)=0A= -.275 E(,)-1.221 E(TRR_P*P^2}.)72 442.4 Q=0A= (It is RECOMMENDED that the recei)72 468.4 Q -.165(ve)-.275 G 2.75(ra)=0A= .165 G(ccount for simple misordering of pack)-2.75 E(ets without)-.11 E=0A= (inferring a loss.)72 481.4 Q F1 2.75(3.2.3.5. Exceptional)72 524 R=0A= (timeouts)2.75 E F0(These are timeouts when the pack)72 540.6 Q=0A= (et reception beha)-.11 E(vior is f)-.22 E=0A= (ar from what it should be and should)-.11 E(trigger a drastic e)72=0A= 553.6 Q -.165(ve)-.275 G(nt by the recei).165 E -.165(ve)-.275 G 3.96=0A= -.605(r. E).165 H(xception timeouts include e).605 E -.165(ve)-.275 G=0A= (nts such as when no).165 E(pack)72 566.6 Q(ets are recei)-.11 E -.165=0A= (ve)-.275 G 2.75(df).165 G=0A= (or a long time, there is no change in the current time slot inde)-2.75=0A= E 2.75(xf)-.165 G(or a long)-2.75 E(time, or no \214rst or second pack)=0A= 72 579.6 Q(et is recei)-.11 E -.165(ve)-.275 G 2.75(da).165 G=0A= (fter join to channel for a long time.)-2.75 E F1(4.)72 622.2 Q/F2 14=0A= /Times-Bold@0 SF -.35(Ap)5.5 G(plicability Statement).35 E F0(WEBRC is \=0A= intended to be a congestion control scheme that can be used in a comple\=0A= te protocol)72 638.8 Q(instantiation that deli)72 651.8 Q -.165(ve)-.275=0A= G(rs objects and streams \(both reliable content deli).165 E -.165(ve)=0A= -.275 G(ry and streaming of).165 E(multimedia information\).)72 664.8 Q=0A= 2.75(AW)5.5 G=0A= (EBRC session comprises a logically related set of one or more)-2.75 E(\=0A= channels originating at a single sender that are used for some period o\=0A= f time to carry pack)72 677.8 Q(ets)-.11 E(pertaining to the transmissi\=0A= on of one or more objects that can be of interest to recei)72 690.8 Q=0A= -.165(ve)-.275 G 2.75(rs. WEBRC).165 F=0A= (congestion control is to be performed o)72 703.8 Q -.165(ve)-.165 G=0A= 2.75(rt).165 G(he aggre)-2.75 E -.055(ga)-.165 G(te of all pack).055 E=0A= (ets sent to the session.)-.11 E(Luby/Go)72 769 Q 284.402(yal Section)=0A= -.11 F 2.75(4. [P)2.75 F(age 12])-.165 E EP=0A= %%Page: 13 13=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E=0A= (WEBRC is most applicable for deli)72 85 Q -.165(ve)-.275 G=0A= (ry of objects or streams of substantial length, i.e., objects or).165 E=0A= (streams that range in length from hundreds of kilobytes to man)72 98 Q=0A= 2.75(yg)-.165 G(ig)-2.75 E(abytes, and whose transfer)-.055 E=0A= (time is in the order of tens of seconds or more.)72 111 Q/F1 11=0A= /Times-Bold@0 SF(4.1.)72 150 Q/F2 13/Times-Bold@0 SF(En)5.5 E(vir)-.52 E=0A= (onmental Requir)-.234 E(ements and Considerations)-.234 E F0=0A= (WEBRC can be used with both multicast and unicast netw)72 179.6 Q 2.75=0A= (orks. Ho)-.11 F(we)-.275 E -.165(ve)-.275 G .88 -.44(r, t).165 H=0A= (he scope of this).44 E(document is limited to multicast.)72 192.6 Q=0A= (WEBRC requires connecti)5.5 E(vity between a sender and recei)-.275 E=0A= -.165(ve)-.275 G(rs,).165 E -.22(bu)72 205.6 S 2.75(td).22 G=0A= (oes not require connecti)-2.75 E(vity from recei)-.275 E -.165(ve)-.275=0A= G(rs to the sender).165 E(.)-.605 E(WEBRC inherently w)72 231.6 Q=0A= (orks with all types of netw)-.11 E(orks, including LANs, W)-.11 E=0A= (ANs, Intranets, the)-1.32 E(Internet, asymmetric netw)72 244.6 Q=0A= (orks, wireless netw)-.11 E(orks, and satellite netw)-.11 E 2.75=0A= (orks. Thus,)-.11 F(the inherent ra)2.75 E(w)-.165 E=0A= (scalability of WEBRC is unlimited.)72 257.6 Q(Ho)5.5 E(we)-.275 E -.165=0A= (ve)-.275 G .88 -.44(r, i).165 H 2.75(ns).44 G(ome netw)-2.75 E(ork en)=0A= -.11 E(vironments v)-.44 E(arying reception)-.275 E(rates to recei)72=0A= 270.6 Q -.165(ve)-.275 G(rs may not be adv).165 E=0A= (antageous, e.g., when the netw)-.275 E(ork is a satellite netw)-.11 E=0A= (ork with a)-.11 E<8c78>72 283.6 Q=0A= (ed amount of bandwidth dedicated to a session.)-.165 E(Recei)72 309.6 Q=0A= -.165(ve)-.275 G(rs join and lea).165 E .33 -.165(ve c)-.22 H=0A= (hannels using the appropriate multicast join and lea).165 E .33 -.165=0A= (ve m)-.22 H 2.75(essages. F).165 F(or)-.165 E=0A= (IPv4 multicast, IGMP messages are used by recei)72 322.6 Q -.165(ve)=0A= -.275 G(rs to join and lea).165 E .33 -.165(ve c)-.22 H 2.75(hannels. F)=0A= .165 F(or IPv6,)-.165 E(MLDv2 messages are used by recei)72 335.6 Q=0A= -.165(ve)-.275 G(rs to join and lea).165 E .33 -.165(ve c)-.22 H 2.75=0A= (hannels. This).165 F(is the only dependenc)2.75 E(y)-.165 E=0A= (of WEBRC on the IP v)72 348.6 Q(ersion.)-.165 E(WEBRC requires recei)72=0A= 374.6 Q -.165(ve)-.275 G=0A= (rs to be able to uniquely identify and demultiple).165 E 2.75(xp)-.165=0A= G(ack)-2.75 E(ets associated)-.11 E(with a session in order to ef)72=0A= 387.6 Q(fecti)-.275 E -.165(ve)-.275 G(ly perform congestion control o)=0A= .165 E -.165(ve)-.165 G 2.75(ra).165 G(ll pack)-2.75 E=0A= (ets associated with)-.11 E(the session.)72 400.6 Q(Ho)5.5 E 2.75(wr)=0A= -.275 G(ecei)-2.75 E -.165(ve)-.275 G(rs achie).165 E .33 -.165(ve t)=0A= -.275 H(his is outside the scope of this document.).165 E=0A= (WEBRC is presumed to be used with an underlying netw)72 426.6 Q=0A= (ork or transport service that is a `)-.11 E(`best)-.814 E(ef)72 439.6 Q=0A= (fort')-.275 E 2.75('s)-.814 G(ervice that does not guarantee pack)-2.75=0A= E(et reception, pack)-.11 E(et reception order)-.11 E 2.75(,a)-.44 G=0A= (nd which does)-2.75 E(not ha)72 452.6 Q .33 -.165(ve a)-.22 H .33 -.165=0A= (ny s).165 H(upport for \215o).165 E 2.75(wo)-.275 G 2.75(rc)-2.75 G=0A= (ongestion control.)-2.75 E -.165(Fo)5.5 G 2.75(re).165 G=0A= (xample, the An)-2.915 E(y-Source Multicast)-.165 E(\(ASM\) model of IP\=0A= multicast as de\214ned in RFC1112 [4] is such a best ef)72 465.6 Q=0A= (fort netw)-.275 E(ork service.)-.11 E(While the basic service pro)72=0A= 478.6 Q(vided by RFC1112 is lar)-.165 E(gely scalable, pro)-.198 E=0A= (viding congestion control or)-.165 E=0A= (reliability should be done carefully to a)72 491.6 Q -.22(vo)-.22 G=0A= (id se).22 E -.165(ve)-.275 G=0A= (re scalability limitations, especially in the).165 E=0A= (presence of heterogeneous sets of recei)72 504.6 Q -.165(ve)-.275 G=0A= (rs.).165 E(There are currently tw)72 530.6 Q 2.75(om)-.11 G=0A= (odels of multicast deli)-2.75 E -.165(ve)-.275 G(ry).165 E 2.75(,t)=0A= -.715 G(he An)-2.75 E(y-Source Multicast \(ASM\) model as)-.165 E(de\=0A= \214ned in RFC1112 [4] and the Source-Speci\214c Multicast \(SSM\) mode\=0A= l as de\214ned in [6].)72 543.6 Q(WEBRC w)72 556.6 Q=0A= (orks with both multicast models, b)-.11 E(ut in a slightly dif)-.22 E=0A= (ferent w)-.275 E(ay with some)-.11 E(what)-.275 E(dif)72 569.6 Q=0A= (ferent en)-.275 E(vironmental concerns.)-.44 E=0A= (When using ASM, a sender S sends pack)5.5 E(ets to a multicast)-.11 E(\=0A= group G, and the WEBRC channel address consists of the pair \(S,G\), wh\=0A= ere S is the IP address of)72 582.6 Q=0A= (the sender and G is a multicast group address.)72 595.6 Q=0A= (When using SSM, a sender S sends pack)5.5 E(ets to an)-.11 E(SSM chann\=0A= el \(S,G\), and the WEBRC channel address coincides with the SSM channe\=0A= l address.)72 608.6 Q 2.75(As)72 634.6 S=0A= (ender can locally allocate unique SSM channel addresses, and this mak)=0A= -2.75 E(es allocation of)-.11 E(channel addresses easy with SSM.)72=0A= 647.6 Q 1.76 -.88(To a)5.5 H=0A= (llocate channel addresses using ASM, the sender must).88 E(uniquely ch\=0A= ose the ASM multicast group address across the scope of the group, and \=0A= this mak)72 660.6 Q(es)-.11 E=0A= (allocation of WEBRC channel addresses more dif)72 673.6 Q=0A= (\214cult with ASM.)-.275 E=0A= (WEBRC channels and SSM channels coincide, and thus the recei)72 699.6 Q=0A= -.165(ve)-.275 G 2.75(rw).165 G(ill only recei)-2.75 E .33 -.165(ve p)=0A= -.275 H(ack).165 E(ets)-.11 E(sent to the requested WEBRC channel.)72=0A= 712.6 Q -.44(Wi)5.5 G(th ASM, the recei).44 E -.165(ve)-.275 G 2.75(rj)=0A= .165 G(oins a channel by joining a)-2.75 E=0A= (multicast group G, and all pack)72 725.6 Q(ets sent to G, re)-.11 E=0A= -.055(ga)-.165 G(rdless of the sender).055 E 2.75(,m)-.44 G(ay be recei)=0A= -2.75 E -.165(ve)-.275 G 2.75(db).165 G 2.75(yt)-2.75 G(he)-2.75 E=0A= (Luby/Go)72 769 Q 276.152(yal Section)-.11 F 2.75(4.1. [P)2.75 F=0A= (age 13])-.165 E EP=0A= %%Page: 14 14=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E(recei)72 85 Q -.165(ve)-.275=0A= G 3.96 -.605(r. T).165 H(hus, SSM has compelling security adv).605 E=0A= (antages o)-.275 E -.165(ve)-.165 G 2.75(rA).165 G(SM for pre)-2.75 E=0A= -.165(ve)-.275 G(ntion of denial of).165 E(service attacks.)72 98 Q=0A= (In either case, recei)5.5 E -.165(ve)-.275 G=0A= (rs SHOULD use mechanisms to \214lter out pack).165 E(ets from)-.11 E=0A= (unw)72 111 Q(anted sources.)-.11 E(WEBRC assumes that the pack)72 137 Q=0A= (et route between the sender and a particular recei)-.11 E -.165(ve)=0A= -.275 G 2.75(ri).165 G 2.75(st)-2.75 G(he same)-2.75 E=0A= (for all channels associated with a session.)72 150 Q -.165(Fo)5.5 G=0A= 2.75(rS).165 G(SM this assumption is true because the multicast)-2.75 E=0A= (tree is a shortest path tree from each recei)72 163 Q -.165(ve)-.275 G=0A= 2.75(rt).165 G 2.75(ot)-2.75 G=0A= (he sender and generally this path changes)-2.75 E(infrequently)72 176 Q=0A= 5.5(.F)-.715 G=0A= (or ASM there are some issues that if not properly considered may in)=0A= -5.665 E -.275(va)-.44 G(lidate this).275 E 2.75(assumption. W)72 189 R=0A= (ith ASM, the pack)-.44 E(et route between the sender and recei)-.11 E=0A= -.165(ve)-.275 G(rs may initially be).165 E(through the Rendezv)72 202 Q=0A= (ous Point \(RP\) and then switch o)-.22 E -.165(ve)-.165 G 2.75(rt).165=0A= G 2.75(ot)-2.75 G(he shortest path to the sender as)-2.75 E(pack)72 215=0A= Q(ets start \215o)-.11 E(wing in a channel.)-.275 E=0A= (The \214rst issue is that the RP may not be the same for all)5.5 E=0A= (channels associated with a session, and thus the \214rst pack)72 228 Q=0A= (ets sent to the channels may follo)-.11 E 2.75(wa)-.275 G=0A= (route that depends on the RP of the channel.)72 241 Q=0A= (This depends on the RP con\214guration for the sender)5.5 E(.)-.605 E=0A= (If the sender re)72 254 Q(gisters all channels associated with the ses\=0A= sion with the same RP then the)-.165 E(assumption is true, b)72 267 Q=0A= (ut if the sender re)-.22 E(gisters dif)-.165 E=0A= (ferent channels with dif)-.275 E(ferent RPs then the)-.275 E=0A= (assumption may not be true.)72 280 Q=0A= (Thus, it is RECOMMENDED that the sender re)5.5 E(gister all channels)=0A= -.165 E(associated with a session with the same RP)72 293 Q 5.5(.A)=0A= -1.221 G(nother issue is that when the channel switches o)-5.5 E -.165=0A= (ve)-.165 G(r).165 E(from the RP to the sender)72 306 Q=0A= (-based tree then the route to the recei)-.22 E -.165(ve)-.275 G=0A= (rs may v).165 E(ary within a channel.)-.275 E=0A= (Furthermore, this may cause either the receipt of duplicate pack)72 319=0A= Q(ets at recei)-.11 E -.165(ve)-.275 G(rs or loss of pack).165 E(ets)=0A= -.11 E(depending on the smoothness of the switcho)72 332 Q -.165(ve)=0A= -.165 G 3.96 -.605(r. T).165 H(hus, it is RECOMMENDED that the RP be)=0A= .605 E(placed as close as possible to the sender)72 345 Q 5.5(.T)-.605 G=0A= (he best location for the RP is that it be the \214rst-hop)-5.5 E=0A= (router closest to the sender)72 358 Q 2.75(,i)-.44 G 2.75(nw)-2.75 G=0A= (hich case the path to the sender and the path to the RP is the same)=0A= -2.75 E(for each recei)72 371 Q -.165(ve)-.275 G 2.75(ra).165 G=0A= (nd the problems mentioned abo)-2.75 E .33 -.165(ve a)-.165 H=0A= (re eliminated.).165 E(The consequences of this)5.5 E=0A= (assumption not being true are that the recei)72 384 Q -.165(ve)-.275 G=0A= 2.75(rr).165 G(eaction to congestion may not be appropriate.)-2.75 E=0A= (Generally)72 397 Q 2.75(,t)-.715 G(he WEBRC recei)-2.75 E -.165(ve)=0A= -.275 G 2.75(rw).165 G(ill act conserv)-2.75 E(ati)-.275 E -.165(ve)=0A= -.275 G(ly and reduce its reception rate too much if).165 E=0A= (this assumption is not true, b)72 410 Q=0A= (ut there can be cases where the recei)-.22 E -.165(ve)-.275 G=0A= (rs will act inappropriately).165 E(.)-.715 E(Some netw)72 436 Q=0A= (orks are not amenable to the WEBRC congestion control protocol.)-.11 E=0A= (In particular)5.5 E 2.75(,f)-.44 G(or a)-2.75 E=0A= (satellite or wireless netw)72 449 Q=0A= (ork, there may be no mechanism for recei)-.11 E -.165(ve)-.275 G=0A= (rs to ef).165 E(fecti)-.275 E -.165(ve)-.275 G(ly reduce their).165 E=0A= (reception rate since there may be a \214x)72 462 Q=0A= (ed transmission rate allocated to the session.)-.165 E/F1 11=0A= /Times-Bold@0 SF(5.)72 504.6 Q/F2 14/Times-Bold@0 SF -.14(Pa)5.5 G(ck)=0A= .14 E(et Header Fields)-.14 E F0 -.165(Pa)72 521.2 S(ck).165 E(ets sent\=0A= to a session using WEBRC MUST include Congestion Control Information \=0A= \214elds as)-.11 E(speci\214ed in this section. This document speci\214\=0A= es short and long formats for the Congestion)72 534.2 Q(Control Informa\=0A= tion, and it is RECOMMENDED that protocol instantiations use one of the\=0A= se tw)72 547.2 Q(o)-.11 E 2.75(formats. Other)72 560.2 R=0A= (formats for the Congestion Control Information \214elds MA)2.75 E 2.75=0A= (Yb)-1.155 G 2.75(eu)-2.75 G(sed by protocol)-2.75 E(instantiations, b)=0A= 72 573.2 Q(ut all protocol instantiations are REQ)-.22 E=0A= (UIRED to use these \214elds in a format that is)-.11 E=0A= (compatible with the interpretations of these \214elds.)72 586.2 Q=0A= (Thus, if a protocol does use a dif)5.5 E(ferent format)-.275 E(for the\=0A= \214elds in the Congestion Control Information then it MUST specify th\=0A= e lengths and)72 599.2 Q(positions of these \214elds within the pack)72=0A= 612.2 Q(et header)-.11 E(.)-.605 E(All inte)72 638.2 Q=0A= (ger \214elds are carried in "big-endian" or "netw)-.165 E=0A= (ork order" format, that is, most signi\214cant)-.11 E=0A= (byte \(octet\) \214rst.)72 651.2 Q=0A= (All constants, unless otherwise speci\214ed, are e)5.5 E=0A= (xpressed in base ten.)-.165 E F1(5.1.)72 693.8 Q/F3 13/Times-Bold@0 SF=0A= (Short F)5.5 E(ormat Congestion Contr)-.325 E(ol Inf)-.234 E(ormation)=0A= -.325 E F0=0A= (The short format for the Congestion Control Information is sho)72 710.4=0A= Q(wn in Fig. 1.)-.275 E(The total length of)5.5 E=0A= (the short format is 32-bits.)72 723.4 Q(Luby/Go)72 769 Q 276.152=0A= (yal Section)-.11 F 2.75(5.1. [P)2.75 F(age 14])-.165 E EP=0A= %%Page: 15 15=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E/F1 8/Courier@0 SF 91.2(0123)=0A= 81.6 85 S 4.8(01234567890123456789012345678901)81.6 98 S=0A= (+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+)76.8=0A= 111 Q 28.8(|C)76.8 124 S 19.2(TSI |)-28.8 F(Channel Number|)4.8 E=0A= (Packet Sequence Number)19.2 E(|)24 E=0A= (+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+)76.8=0A= 137 Q/F2 10/Times-Roman@0 SF=0A= (Fig. 1 - Short format for Congestion Control Information)72 176 Q F0(T\=0A= he function of each \214eld in the Congestion Control Information is th\=0A= e follo)72 205.6 Q(wing.)-.275 E(Current T)83 235.2 Q(ime Slot Inde)=0A= -.385 E 2.75(x\()-.165 G(CTSI\): 8 bits)-2.75 E(CTSI indicates the inde)=0A= 105 251.8 Q 2.75(xo)-.165 G 2.75(ft)-2.75 G(he current time slot.)-2.75=0A= E(This must be sent in each pack)5.5 E(et within)-.11 E(the session.)105=0A= 264.8 Q(The Current T)5.5 E(ime Slot Inde)-.385 E 2.75(xi)-.165 G=0A= (ncreases by one modulo T each TSD seconds)-2.75 E(at the sender)105=0A= 277.8 Q 2.75(,w)-.44 G(here T is the number of time slots associated wi\=0A= th the session and TSD is)-2.75 E(the time slot duration.)105 290.8 Q=0A= (Note that T is also the number of w)5.5 E -2.475 -.22(av e)-.11 H=0A= (channels associated with)2.97 E=0A= (the session, and thus T MUST be at most 255.)105 303.8 Q=0A= (Channel Number \(CN\): 8 bits)83 333.4 Q=0A= (CN is the channel number that this pack)105 350 Q(et belongs to.)-.11 E=0A= (CN for the base channel is T)5.5 E 2.75(,a)-.814 G(nd)-2.75 E=0A= (the CNs for the w)105 363 Q -2.475 -.22(av e)-.11 H=0A= (channels are 0 through T)2.97 E 2.75(-1. Thus,)-1.012 F=0A= (T+1 channels in total are used,)2.75 E(and thus T MUST be at most 255.)=0A= 105 376 Q -.165(Pa)83 405.6 S(ck).165 E=0A= (et Sequence Number \(PSN\): 16 bits)-.11 E(The PSN of each pack)105=0A= 422.2 Q(et is scoped by its CN v)-.11 E 2.75(alue. The)-.275 F=0A= (sequence numbers of consecuti)2.75 E -.165(ve)-.275 G(pack)105 435.2 Q=0A= (ets sent to the base channel are numbered consecuti)-.11 E -.165(ve)=0A= -.275 G(ly modulo 2^16.).165 E(The same)5.5 E=0A= (sequence of PSNs are used for each w)105 448.2 Q -2.475 -.22(av e)-.11=0A= H(channel in each c)2.97 E 2.75(ycle. The)-.165 F(sequence numbers)2.75=0A= E(of consecuti)105 461.2 Q .33 -.165(ve p)-.275 H(ack).165 E=0A= (ets sent to a w)-.11 E -2.475 -.22(av e)-.11 H=0A= (channel are numbered consecuti)2.97 E -.165(ve)-.275 G(ly modulo 2^16)=0A= .165 E(within each c)105 474.2 Q(ycle, ending with the last pack)-.165 E=0A= (et sent to the channel before the channel goes)-.11 E=0A= (quiescent with PSN =3D 2^16-1.)105 487.2 Q/F3 11/Times-Bold@0 SF(5.2.)72=0A= 529.8 Q/F4 13/Times-Bold@0 SF(Long F)5.5 E(ormat Congestion Contr)-.325=0A= E(ol Inf)-.234 E(ormation)-.325 E F0=0A= (The long format for the Congestion Control Information is sho)72 546.4=0A= Q(wn in Fig. 2.)-.275 E(The total length of)5.5 E=0A= (the long format is 64-bits.)72 559.4 Q(Luby/Go)72 769 Q 276.152=0A= (yal Section)-.11 F 2.75(5.2. [P)2.75 F(age 15])-.165 E EP=0A= %%Page: 16 16=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E/F1 8/Courier@0 SF 91.2(0123)=0A= 81.6 85 S 4.8(01234567890123456789012345678901)81.6 98 S=0A= (+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+)76.8=0A= 111 Q 62.4(|C)76.8 124 S 62.4(TSI |)-62.4 F(Channel Number)38.4 E(|)43.2=0A= E(+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+)76.8=0A= 137 Q 100.8(|P)76.8 150 S(acket Sequence Number)-100.8 E(|)96 E=0A= (+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+)76.8=0A= 163 Q/F2 10/Times-Roman@0 SF=0A= (Fig. 1 - Long format for Congestion Control Information)72 202 Q F0(Th\=0A= e meaning of each \214eld for the long format is the same as for the sh\=0A= ort format, the only)72 231.6 Q(dif)72 244.6 Q=0A= (ference is that each \214eld is twice as long.)-.275 E(Current T)83=0A= 274.2 Q(ime Slot Inde)-.385 E 2.75(x\()-.165 G(CTSI\): 16 bits)-2.75 E=0A= (CTSI indicates the inde)105 290.8 Q 2.75(xo)-.165 G 2.75(ft)-2.75 G=0A= (he current time slot.)-2.75 E(This must be sent in each pack)5.5 E=0A= (et within)-.11 E(the session.)105 303.8 Q(The Current T)5.5 E=0A= (ime Slot Inde)-.385 E 2.75(xi)-.165 G=0A= (ncreases by one modulo T each TSD seconds)-2.75 E(at the sender)105=0A= 316.8 Q 2.75(,w)-.44 G(here T is the number of time slots associated wi\=0A= th the session and TSD is)-2.75 E(the time slot duration.)105 329.8 Q=0A= (Note that T is also the number of w)5.5 E -2.475 -.22(av e)-.11 H=0A= (channels associated with)2.97 E=0A= (the session, and thus T MUST be at most 65 535.)105 342.8 Q=0A= (Channel Number \(CN\): 16 bits)83 372.4 Q=0A= (CN is the channel number that this pack)105 389 Q(et belongs to.)-.11 E=0A= (CN for the base channel is T)5.5 E 2.75(,a)-.814 G(nd)-2.75 E=0A= (the CNs for the w)105 402 Q -2.475 -.22(av e)-.11 H=0A= (channels are 0 through T)2.97 E 2.75(-1. Thus,)-1.012 F=0A= (T+1 channels in total are used,)2.75 E=0A= (and thus T MUST be at most 65 535.)105 415 Q -.165(Pa)83 444.6 S(ck)=0A= .165 E(et Sequence Number \(PSN\): 32 bits)-.11 E(The PSN of each pack)=0A= 105 461.2 Q(et is scoped by its CN v)-.11 E 2.75(alue. The)-.275 F=0A= (sequence numbers of consecuti)2.75 E -.165(ve)-.275 G(pack)105 474.2 Q=0A= (ets sent to the base channel are numbered consecuti)-.11 E -.165(ve)=0A= -.275 G(ly modulo 2^32.).165 E(The same)5.5 E=0A= (sequence of PSNs are used for each w)105 487.2 Q -2.475 -.22(av e)-.11=0A= H(channel in each c)2.97 E 2.75(ycle. The)-.165 F(sequence numbers)2.75=0A= E(of consecuti)105 500.2 Q .33 -.165(ve p)-.275 H(ack).165 E=0A= (ets sent to a w)-.11 E -2.475 -.22(av e)-.11 H=0A= (channel are numbered consecuti)2.97 E -.165(ve)-.275 G(ly modulo 2^32)=0A= .165 E(within each c)105 513.2 Q(ycle, ending with the last pack)-.165 E=0A= (et sent to the channel before the channel goes)-.11 E=0A= (quiescent with PSN =3D 2^32-1.)105 526.2 Q/F3 11/Times-Bold@0 SF(6.)72=0A= 568.8 Q/F4 14/Times-Bold@0 SF(Requir)5.5 E(ements Fr)-.252 E=0A= (om Other Building Blocks)-.252 E F0=0A= (As described in RFC3048, WEBRC is a b)72 585.4 Q=0A= (uilding block that is intended to be used, in conjunction)-.22 E=0A= (with other b)72 598.4 Q=0A= (uilding blocks, to help specify a protocol instantiation.)-.22 E=0A= (WEBRC does not pro)72 615 Q(vide higher le)-.165 E -.165(ve)-.275 G=0A= 2.75(ls).165 G(ession support, e.g., ho)-2.75 E 2.75(wr)-.275 G(ecei)=0A= -2.75 E -.165(ve)-.275 G(rs obtain the necessary).165 E=0A= (session description and ho)72 628 Q 2.75(wt)-.275 G(he recei)-2.75 E=0A= -.165(ve)-.275 G(rs demultiple).165 E 2.75(xr)-.165 G(ecei)-2.75 E -.165=0A= (ve)-.275 G 2.75(dp).165 G(ack)-2.75 E(ets based on their session.)-.11=0A= E(There is support pro)72 641 Q(vided by other b)-.165 E=0A= (uilding blocks that can be used in conjunction with WEBRC)-.22 E=0A= (to pro)72 654 Q(vide some of this support.)-.165 E -.165(Fo)5.5 G 2.75=0A= (re).165 G(xample, LCT [11] can pro)-2.915 E(vide some of the higher le)=0A= -.165 E -.165(ve)-.275 G 2.75(li).165 G(n-)-2.75 E=0A= (band session support that may be needed by recei)72 667 Q -.165(ve)=0A= -.275 G(rs, and the WEBRC Congestion Control).165 E=0A= (Information \(CCI\) required in each pack)72 680 Q=0A= (et can be carried in the CCI \214eld of the LCT header [11].)-.11 E=0A= (WEBRC does not pro)72 696.6 Q(vide an)-.165 E 2.75(yt)-.165 G=0A= (ype of reliability)-2.75 E 2.75(,a)-.715 G=0A= (nd in particular does not pro)-2.75 E(vide support for)-.165 E=0A= (retransmission of loss pack)72 709.6 Q 2.75(ets. Reliability)-.11 F=0A= (can be added by independent means, such as by the)2.75 E=0A= (use of FEC codes as described in [12] and speci\214ed in the FEC b)72=0A= 722.6 Q(uilding block [13].)-.22 E(Luby/Go)72 769 Q 284.402(yal Section)=0A= -.11 F 2.75(6. [P)2.75 F(age 16])-.165 E EP=0A= %%Page: 17 17=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E/F1 11/Times-Bold@0 SF(7.)72=0A= 85 Q/F2 14/Times-Bold@0 SF(Security Considerations)5.5 E F0=0A= (WEBRC can be subject to denial-of-service attacks by attack)72 101.6 Q=0A= (ers that try to confuse the congestion)-.11 E=0A= (control mechanism for recei)72 114.6 Q -.165(ve)-.275 G=0A= (rs by injecting for).165 E(ged pack)-.198 E=0A= (ets into the multicast stream.)-.11 E(This attack)5.5 E(most adv)72=0A= 127.6 Q(ersely af)-.165 E(fects netw)-.275 E(ork elements and recei)-.11=0A= E -.165(ve)-.275 G(rs do).165 E(wnstream of the attack, and much less)=0A= -.275 E(signi\214cantly the rest of the netw)72 140.6 Q=0A= (ork and other recei)-.11 E -.165(ve)-.275 G 2.75(rs. Because).165 F=0A= (of this and because of the)2.75 E=0A= (potential attacks due to the use of FEC described abo)72 153.6 Q -.165=0A= (ve)-.165 G 2.75(,i).165 G 2.75(ti)-2.75 G 2.75(sR)-2.75 G=0A= (ECOMMENDED that some form)-2.75 E(of pack)72 166.6 Q=0A= (et authentication such as TESLA [15] be used to protect ag)-.11 E=0A= (ainst such attacks and that)-.055 E(Re)72 179.6 Q -.165(ve)-.275 G=0A= (rse P).165 E(ath F)-.165 E(orw)-.165 E=0A= (arding checks be enabled in all netw)-.11 E=0A= (ork routers and switches along the path)-.11 E=0A= (from the sender to recei)72 192.6 Q -.165(ve)-.275 G=0A= (rs to limit the possibility of a bad agent injecting for).165 E=0A= (ged pack)-.198 E(ets into the)-.11 E(multicast tree data path.)72 205.6=0A= Q 2.75(Ar)72 231.6 S(ecei)-2.75 E -.165(ve)-.275 G 2.75(rw).165 G=0A= (ith an incorrect or corrupted implementation of WEBRC may af)-2.75 E=0A= (fect health of the)-.275 E(netw)72 244.6 Q=0A= (ork in the path between the sender and the recei)-.11 E -.165(ve)-.275=0A= G .88 -.44(r, a).165 H(nd may also af).44 E(fect the reception rates of)=0A= -.275 E(other recei)72 257.6 Q -.165(ve)-.275 G=0A= (rs joined to the session.).165 E=0A= (It is therefore RECOMMENDED that recei)5.5 E -.165(ve)-.275 G=0A= (rs be required).165 E(to identify themselv)72 270.6 Q(es as le)-.165 E=0A= (gitimate before the)-.165 E 2.75(yr)-.165 G(ecei)-2.75 E .33 -.165=0A= (ve t)-.275 H(he session description needed to join the).165 E(session.)=0A= 72 283.6 Q(Another vulnerability of ALC is the potential of recei)72=0A= 309.6 Q -.165(ve)-.275 G(rs obtaining an incorrect session).165 E=0A= (description for the session.)72 322.6 Q=0A= (The consequences of this could be that le)5.5 E(gitimate recei)-.165 E=0A= -.165(ve)-.275 G(rs with the).165 E=0A= (wrong session description are unable to correctly recei)72 335.6 Q .33=0A= -.165(ve t)-.275 H(he session content, or that recei).165 E -.165(ve)=0A= -.275 G(rs).165 E(inadv)72 348.6 Q(ertently try to recei)-.165 E .33=0A= -.165(ve a)-.275 H 2.75(tam).165 G(uch higher rate than the)-2.75 E 2.75=0A= (ya)-.165 G(re capable of, thereby disrupting)-2.75 E(traf)72 361.6 Q=0A= (\214c in portions of the netw)-.275 E 2.75(ork. T)-.11 F 2.75(oa)-.88 G=0A= -.22(vo)-2.97 G(id these problems, it is RECOMMENDED that the).22 E=0A= (recei)72 374.6 Q -.165(ve)-.275 G 2.75(ra).165 G=0A= (uthenticate the session description, for e)-2.75 E=0A= (xample by using either the ESP \(with enabled)-.165 E(authentication s\=0A= ervice\) [9] or AH [8] protocols of IPSEC [7] to ensure the authenticit\=0A= y of the)72 387.6 Q(session description.)72 400.6 Q F1(8.)72 443.2 Q F2=0A= (IAN)5.5 E 3.5(AC)-.28 G(onsiderations)-3.5 E F0=0A= (No information in this speci\214cation is subject to IAN)72 459.8 Q=0A= 2.75(Ar)-.385 G -.165(eg)-2.75 G(istration.).165 E F1(9.)72 502.4 Q F2=0A= (Intellectual Pr)5.5 E(operty Issues)-.252 E F0(Digital F)72 532 Q=0A= (ountain maintains all rights to the WEBRC technology)-.165 E 2.75(,b)=0A= -.715 G(ut will grant a ro)-2.97 E(yalty-free)-.11 E -.11(wo)72 545 S=0A= (rldwide license to the WEBRC technology once it has been standardized.)=0A= .11 E(WEBRC may be used with other protocols which are proprietary)72=0A= 571 Q 2.75(,o)-.715 G 2.75(rh)-2.75 G -2.475 -.22(av e)-2.75 H=0A= (pending or granted)2.97 E(patents.)72 584 Q F1(10.)72 626.6 Q F2(Refer)=0A= 5.5 E(ences)-.252 E F0([1] Bradner)72 656.2 Q 2.75(,S)-.44 G(., `)-2.75=0A= E(`The Internet Standards Process -- Re)-.814 E(vision 3,)-.275 E 1.628=0A= -.814('' R)-.77 H(FC2026, October 1996.).814 E([2] Bradner)72 682.2 Q=0A= 2.75(,S)-.44 G(., `)-2.75 E(`K)-.814 E .33 -.165(ey w)-.275 H=0A= (ords for use in RFCs to Indicate Requirement Le).055 E -.165(ve)-.275 G=0A= (ls,).165 E 1.628 -.814('' R)-.77 H(FC2119, March).814 E(1997.)72 695.2=0A= Q([3] Byers, J.W)72 721.2 Q(., Frumin, M., Horn, G., Luby)-1.012 E 2.75=0A= (,M)-.715 G(., Mitzenmacher)-2.75 E 2.75(,M)-.44 G(., Roetter)-2.75 E=0A= 2.75(,A)-.44 G(., and Sha)-2.75 E -.165(ve)-.22 G .88 -.44(r, W).165 H=0A= (.,)-.572 E(Luby/Go)72 769 Q 278.902(yal Section)-.11 F 2.75(10. [P)2.75=0A= F(age 17])-.165 E EP=0A= %%Page: 18 18=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E -.814(``)72 85 S=0A= (FLID-DL: Congestion Control for Layered Multicast,).814 E 1.628 -.814=0A= ('' P)-.77 H(roc. Second International W).814 E(orkshop on)-.88 E(Netw)=0A= 72 98 Q(ork)-.11 E(ed Group Communications \(NGC 2000\), P)-.11 E=0A= (alo Alto, CA, No)-.165 E -.165(ve)-.165 G(mber 2000, pp. 71-81.).165 E=0A= ([4] Deering, S., `)72 124 Q(`Host Extensions for IP Multicasting,)-.814=0A= E 1.628 -.814('' R)-.77 H(FC1112, August 1989.).814 E([5] Handle)72 150=0A= Q 1.43 -.715(y, M)-.165 H(., P).715 E(adh)-.165 E(ye, J., Flo)-.055 E=0A= (yd, S., W)-.11 E(idmer)-.44 E 2.75(,J)-.44 G(., `)-2.75 E=0A= (`TCP Friendly Rate Control \(TFRC\): Protocol)-.814 E(Speci\214cation,)=0A= 72 163 Q 1.628 -.814('' I)-.77 H=0A= (nternet Draft draft-ietf-tsvwg-tfrc-03, July 2001, a w).814 E=0A= (ork in progress.)-.11 E([6] Holbrook, H. W)72 189 Q(., `)-1.012 E 1.76=0A= -.88(`A C)-.814 H(hannel Model for Multicast,).88 E 1.628 -.814('' P)=0A= -.77 H(h.D. Dissertation, Stanford Uni).814 E -.165(ve)-.275 G(rsity)=0A= .165 E(,)-.715 E=0A= (Department of Computer Science, Stanford, California, August 2001.)72=0A= 202 Q([7] K)72 228 Q(ent, S., Atkinson, R., `)-.275 E=0A= (`Security Architecture for the Internet Protocol,)-.814 E 1.628 -.814=0A= ('' R)-.77 H(FC2401,).814 E(No)72 241 Q -.165(ve)-.165 G(mber 1998.).165=0A= E([8] K)72 267 Q(ent, S., Atkinson, R., `)-.275 E=0A= (`IP Authentication Header)-.814 E -.77(,')-.44 G 2.75('R)-.044 G=0A= (FC2406, No)-2.75 E -.165(ve)-.165 G(mber 1998.).165 E([9] K)72 293 Q=0A= (ent, S., Atkinson, R., `)-.275 E(`IP Encapsulating Security P)-.814 E=0A= (ayload \(ESP\),)-.165 E 1.628 -.814('' R)-.77 H(FC2406, No).814 E -.165=0A= (ve)-.165 G(mber).165 E(1998.)72 306 Q([10] Luby)72 332 Q 2.75(,M)-.715=0A= G(., Gemmell, J., V)-2.75 E(icisano, L., Rizzo, L., Cro)-.66 E=0A= (wcroft, J., `)-.275 E -.88(`A)-.814 G(synchronous Layered).88 E=0A= (Coding protocol instantiation,)72 345 Q 1.628 -.814('' I)-.77 H=0A= (nternet Draft draft-ietf-rmt-pi-alc-06, February 2002, a w).814 E=0A= (ork in)-.11 E(progress.)72 358 Q([11] Luby)72 384 Q 2.75(,M)-.715 G=0A= (., Gemmell, J., V)-2.75 E(icisano, L., Rizzo, L., Handle)-.66 E 1.43=0A= -.715(y, M)-.165 H(., Cro).715 E(wcroft, J., `)-.275 E(`Layered Coding)=0A= -.814 E -.385(Tr)72 397 S(ansport b).385 E(uilding block,)-.22 E 1.628=0A= -.814('' I)-.77 H=0A= (nternet Draft draft-ietf-rmt-bb-lct-04.txt, February 2002, a w).814 E=0A= (ork in)-.11 E(progress.)72 410 Q([12] Luby)72 436 Q 2.75(,M)-.715 G=0A= (., Gemmell, V)-2.75 E(icisano, L., J., Rizzo, L., Handle)-.66 E 1.43=0A= -.715(y, M)-.165 H(., Cro).715 E(wcroft, J., `)-.275 E(`The Use of)-.814=0A= E -.165(Fo)72 449 S(rw).165 E=0A= (ard Error Correction in Reliable Multicast,)-.11 E 1.628 -.814('' I)=0A= -.77 H(nternet Draft draft-ietf-rmt-info-fec-02.txt,).814 E=0A= (February 2002, a w)72 462 Q(ork in progress.)-.11 E([13] Luby)72 488 Q=0A= 2.75(,M)-.715 G(., Gemmell, J., V)-2.75 E=0A= (icisano, L., Rizzo, L., Handle)-.66 E 1.43 -.715(y, M)-.165 H(., Cro)=0A= .715 E(wcroft, J., `)-.275 E(`F)-.814 E(orw)-.165 E(ard Error)-.11 E=0A= (Correction b)72 501 Q(uilding block,)-.22 E 1.628 -.814('' I)-.77 H=0A= (nternet Draft draft-ietf-rmt-bb-fec-06.txt, February 2002.).814 E=0A= ([14] Mankin, A., Romano)72 527 Q 1.43 -.715(w, A)-.275 H(., Bradner)=0A= .715 E 2.75(,S)-.44 G(., P)-2.75 E(axson V)-.165 E(., `)-1.419 E=0A= (`IETF Criteria for Ev)-.814 E(aluating Reliable)-.275 E(Multicast T)72=0A= 540 Q(ransport and Application Protocols,)-.385 E 1.628 -.814('' R)-.77=0A= H(FC2357, June 1998.).814 E([15] Perrig, A., Canetti, R., Song, D., T)72=0A= 566 Q(yg)-.88 E(ar)-.055 E 2.75(,J)-.44 G(.D., `)-2.75 E(`Ef)-.814 E=0A= (\214cient and Secure Source Authentication)-.275 E(for Multicast,)72=0A= 579 Q 1.628 -.814('' N)-.77 H(etw).814 E(ork and Distrib)-.11 E=0A= (uted System Security Symposium, NDSS 2001, pp. 35-46,)-.22 E=0A= (February 2001.)72 592 Q([16] V)72 618 Q(icisano, L., Rizzo, L., Cro)=0A= -.66 E(wcroft, J., "TCP-lik)-.275 E 2.75(eC)-.11 G=0A= (ongestion Control for Layered Multicast)-2.75 E(Data T)72 631 Q(ransfe\=0A= r", Proc. IEEE Infocom '98, San Francisco, CA, March-April 1998, pp. 99\=0A= 6-1003.)-.385 E([17] Whetten, B., V)72 657 Q(icisano, L., K)-.66 E=0A= (ermode, R., Handle)-.275 E 1.43 -.715(y, M)-.165 H(., Flo).715 E=0A= (yd, S., Luby)-.11 E 2.75(,M)-.715 G(., `)-2.75 E(`Reliable)-.814 E=0A= (Multicast T)72 670 Q(ransport Building Blocks for One-to-Man)-.385 E=0A= 2.75(yB)-.165 G(ulk-Data T)-2.75 E(ransfer)-.385 E -.77(,')-.44 G 2.75=0A= ('R)-.044 G(FC3048, January)-2.75 E(2001.)72 683 Q(Luby/Go)72 769 Q=0A= 278.902(yal Section)-.11 F 2.75(10. [P)2.75 F(age 18])-.165 E EP=0A= %%Page: 19 19=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E/F1 11/Times-Bold@0 SF(11.)72=0A= 85 Q/F2 14/Times-Bold@0 SF -.7(Au)5.5 G(thors' Addr).7 E(esses)-.252 E=0A= F0(Michael Luby)80.25 101.6 Q(luby@digitalfountain.com)80.25 114.6 Q=0A= (Digital F)80.25 127.6 Q(ountain)-.165 E(39141 Ci)80.25 140.6 Q=0A= (vic Center Dri)-.275 E -.165(ve)-.275 G(Suite 300)80.25 153.6 Q=0A= (Fremont, CA, USA, 94538)80.25 166.6 Q -1.705 -.66(Vi v)80.25 192.6 T=0A= (ek Go).495 E(yal)-.11 E(vi)80.25 205.6 Q -.165(ve)-.275 G=0A= (k@digitalfountain.com).165 E(Digital F)80.25 218.6 Q(ountain)-.165 E=0A= (39141 Ci)80.25 231.6 Q(vic Center Dri)-.275 E -.165(ve)-.275 G=0A= (Suite 300)80.25 244.6 Q(Fremont, CA, USA, 94538)80.25 257.6 Q(Luby/Go)=0A= 72 769 Q 278.902(yal Section)-.11 F 2.75(11. [P)2.75 F(age 19])-.165 E=0A= EP=0A= %%Page: 20 20=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E/F1 11/Times-Bold@0 SF(12.)72=0A= 85 Q/F2 14/Times-Bold@0 SF(Full Copyright Statement)5.5 E F0(Cop)72=0A= 101.6 Q(yright \(C\) The Internet Society \(2001\).)-.11 E=0A= (All Rights Reserv)5.5 E(ed.)-.165 E(This document and translations of \=0A= it may be copied and furnished to others, and deri)72 118.2 Q -.275(va)=0A= -.275 G(ti).275 E .33 -.165(ve w)-.275 H(orks).055 E=0A= (that comment on or otherwise e)72 131.2 Q=0A= (xplain it or assist in its implementation may be prepared, copied,)=0A= -.165 E(published and distrib)72 144.2 Q=0A= (uted, in whole or in part, without restriction of an)-.22 E 2.75(yk)=0A= -.165 G(ind, pro)-2.75 E(vided that the)-.165 E(abo)72 157.2 Q .33 -.165=0A= (ve c)-.165 H(op).165 E(yright notice and this paragraph are included o\=0A= n all such copies and deri)-.11 E -.275(va)-.275 G(ti).275 E .33 -.165=0A= (ve w)-.275 H(orks.).055 E(Ho)72 170.2 Q(we)-.275 E -.165(ve)-.275 G .88=0A= -.44(r, t).165 H(his document itself may not be modi\214ed in an).44 E=0A= 2.75(yw)-.165 G(ay)-2.86 E 2.75(,s)-.715 G(uch as by remo)-2.75 E=0A= (ving the)-.165 E(cop)72 183.2 Q(yright notice or references to the Int\=0A= ernet Society or other Internet or)-.11 E -.055(ga)-.198 G(nizations, e)=0A= .055 E(xcept as)-.165 E(needed for the purpose of de)72 196.2 Q -.165=0A= (ve)-.275 G(loping Internet standards in which case the procedures for)=0A= .165 E(cop)72 209.2 Q=0A= (yrights de\214ned in the Internet languages other than English.)-.11 E=0A= (The limited permissions granted abo)72 225.8 Q .33 -.165(ve a)-.165 H=0A= (re perpetual and will not be re).165 E -.22(vo)-.275 G -.11(ke).22 G=0A= 2.75(db).11 G 2.75(yt)-2.75 G(he Internet)-2.75 E=0A= (Society or its successors or assigns.)72 238.8 Q=0A= (This document and the information contained herein is pro)72 255.4 Q=0A= (vided on an "AS IS" basis and THE)-.165 E=0A= (INTERNET SOCIETY AND THE INTERNET ENGINEERING T)72 268.4 Q=0A= (ASK FORCE DISCLAIMS)-1.023 E(ALL W)72 281.4 Q=0A= (ARRANTIES, EXPRESS OR IMPLIED, INCLUDING B)-1.32 E(UT NO)-.11 E 2.75=0A= (TL)-.44 G(IMITED T)-2.75 E 2.75(OA)-.198 G(NY)-2.75 E -1.32(WA)72 294.4=0A= S(RRANTY THA)1.32 E 2.75(TT)-1.221 G(HE USE OF THE INFORMA)-2.75 E=0A= (TION HEREIN WILL NO)-1.221 E 2.75(TI)-.44 G(NFRINGE)-2.75 E=0A= (ANY RIGHTS OR ANY IMPLIED W)72 307.4 Q(ARRANTIES OF MERCHANT)-1.32 E=0A= (ABILITY OR FITNESS)-1.023 E(FOR A P)72 320.4 Q(AR)-1.012 E=0A= (TICULAR PURPOSE.")-.66 E(Luby/Go)72 769 Q 278.902(yal Section)-.11 F=0A= 2.75(12. [P)2.75 F(age 20])-.165 E EP=0A= %%Page: 21 21=0A= %%BeginPageSetup=0A= BP=0A= %%EndPageSetup=0A= /F0 11/Times-Roman@0 SF(INTERNET)72 49 Q 65.168(-DRAFT Expires:)-1.012 F=0A= (September 2002)2.75 E(March 2002)106.935 E(Luby/Go)72 769 Q 278.902=0A= (yal Section)-.11 F 2.75(12. [P)2.75 F(age 21])-.165 E EP=0A= %%Trailer=0A= end=0A= %%EOF=0A= ------=_NextPart_000_0000_01C1C306.34596B20 Content-Type: text/plain; name="draft-ietf-rmt-bb-webrc-01.txt" Content-Disposition: attachment; filename="draft-ietf-rmt-bb-webrc-01.txt" Content-Transfer-Encoding: quoted-printable Internet Engineering Task Force RMT WG=0A= INTERNET-DRAFT M. Luby/Digital Fountain=0A= draft-ietf-rmt-bb-webrc-01.txt V. Goyal/Digital Fountain=0A= 1 March 2002=0A= Expires: September 2002=0A= =0A= =0A= Wave and Equation Based Rate Control building block=0A= =0A= =0A= =0A= Status of this Document=0A= =0A= This document is an Internet-Draft and is in full conformance with all=0A= provisions of Section 10 of RFC2026.=0A= =0A= Internet-Drafts are working documents of the Internet Engineering Task=0A= Force (IETF), its areas, and its working groups. Note that other groups=0A= may also distribute working documents as Internet-Drafts.=0A= =0A= Internet-Drafts are valid for a maximum of six months and may be=0A= updated, replaced, or obsoleted by other documents at any time. It is=0A= inappropriate to use Internet-Drafts as reference material or to cite=0A= them other than as a "work in progress".=0A= =0A= The list of current Internet-Drafts can be accessed at=0A= http://www.ietf.org/ietf/1id-abstracts.txt=0A= =0A= To view the list Internet-Draft Shadow Directories, see=0A= http://www.ietf.org/shadow.html.=0A= =0A= This document is a product of the IETF RMT WG. Comments should be=0A= addressed to the authors, or the WG's mailing list at rmt@lbl.gov.=0A= =0A= =0A= Abstract=0A= =0A= =0A= Wave and Equation Based Rate Control (WEBRC) provides rate and=0A= congestion control for data delivery. WEBRC is specifically=0A= designed to support protocols using IP multicast, but could=0A= also be used to provide support to protocols that use unicast.=0A= WEBRC provides multiple rate congestion controlled delivery to=0A= receivers, i.e., different receivers joined to the same=0A= session may be receiving packets at different rates depending=0A= =0A= =0A= =0A= Luby/Goyal [Page 1]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= on their individual bandwidth connection to the sender and on=0A= competing traffic along that connection. WEBRC requires no=0A= feedback from receivers to the sender, i.e., it is a=0A= completely receiver-driven congestion control protocol. Thus,=0A= WEBRC is designed to scale to potentially massive numbers of=0A= receivers attached to a session from a single sender.=0A= Furthermore, because each individual receiver adjusts to the=0A= available bandwidth between the sender and that receiver,=0A= there is the potential to deliver data to each individual=0A= receiver at the fastest possible rate for that receiver, even=0A= in a highly heterogeneous network architecture, using a single=0A= sender.=0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= Luby/Goyal [Page 2]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= Table of Contents=0A= =0A= =0A= 1. Introduction. . . . . . . . . . . . . . . . . . . . . . 4=0A= 2. Rationale . . . . . . . . . . . . . . . . . . . . . . . 5=0A= 3. Functionality . . . . . . . . . . . . . . . . . . . . . 6=0A= 3.1. Sender Operation . . . . . . . . . . . . . . . . . . 8=0A= 3.1.1. Sender inputs and initialization. . . . . . . . . 8=0A= 3.1.2. Sending packets to the session. . . . . . . . . . 10=0A= 3.2. Receiver Operation . . . . . . . . . . . . . . . . . 11=0A= 3.2.1. Receiver inputs and initialization. . . . . . . . 11=0A= 3.2.2. Receiver measurements and calculations. . . . . . 12=0A= 3.2.2.1. Average loss probability . . . . . . . . . . . 12=0A= 3.2.2.2. Average round-trip time. . . . . . . . . . . . 13=0A= 3.2.2.3. Rate Equation. . . . . . . . . . . . . . . . . 14=0A= 3.2.2.4. Epochs . . . . . . . . . . . . . . . . . . . . 14=0A= 3.2.2.5. Average reception rate . . . . . . . . . . . . 14=0A= 3.2.2.6. Slow start . . . . . . . . . . . . . . . . . . 16=0A= 3.2.2.7. Target rate. . . . . . . . . . . . . . . . . . 16=0A= 3.2.3. Receiver events . . . . . . . . . . . . . . . . . 16=0A= 3.2.3.1. Epoch change . . . . . . . . . . . . . . . . . 16=0A= 3.2.3.2. Time slot change . . . . . . . . . . . . . . . 17=0A= 3.2.3.3. Join a wave channel. . . . . . . . . . . . . . 17=0A= 3.2.3.4. Loss event . . . . . . . . . . . . . . . . . . 17=0A= 3.2.3.5. Exceptional timeouts . . . . . . . . . . . . . 18=0A= 4. Applicability Statement . . . . . . . . . . . . . . . . 18=0A= 4.1. Environmental Requirements and=0A= Considerations. . . . . . . . . . . . . . . . . . . . . . 18=0A= 5. Packet Header Fields. . . . . . . . . . . . . . . . . . 20=0A= 5.1. Short Format Congestion Control Information. . . . . 21=0A= 5.2. Long Format Congestion Control Information . . . . . 22=0A= 6. Requirements From Other Building Blocks . . . . . . . . 23=0A= 7. Security Considerations . . . . . . . . . . . . . . . . 23=0A= 8. IANA Considerations . . . . . . . . . . . . . . . . . . 24=0A= 9. Intellectual Property Issues. . . . . . . . . . . . . . 24=0A= 10. References . . . . . . . . . . . . . . . . . . . . . . 24=0A= 11. Authors' Addresses . . . . . . . . . . . . . . . . . . 26=0A= 12. Full Copyright Statement . . . . . . . . . . . . . . . 27=0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= Luby/Goyal [Page 3]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= 1. Introduction=0A= =0A= This document specifies Wave and Equation Based Rate Control (WEBRC).=0A= WEBRC is a congestion control building block that is designed to be=0A= massively scalable when used with the IP multicast network service.=0A= WEBRC is also suitable as the basis for unicast congestion control, but=0A= this is outside the scope of this document. WEBRC is designed to=0A= compete fairly with TCP and similar congestion-controlled sessions.=0A= WEBRC can be used as a congestion control protocol for any type of data=0A= delivery, including reliable content delivery and streaming delivery.=0A= =0A= This document describes a building block as defined in RFC3048 [17] that=0A= conforms to RFC2357 [14].=0A= =0A= =0A= WEBRC is a receiver-driven congestion control protocol in the spirit of=0A= [16] and [3]. This means that all measurements and decisions to raise or=0A= lower the reception rate are made by each individual receiver, and these=0A= decisions are acted upon by sending join and leave messages for channels=0A= to the network. A receiver using WEBRC adjusts its reception rate=0A= without regard for other concerns such as reliability. This is=0A= different from TCP, where the congestion control protocol and the=0A= reliability protocol are intricately interwoven with one another.=0A= =0A= WEBRC takes the same basic equation-based approach as TFRC [5]. In=0A= particular, each WEBRC receiver measures parameters that are plugged=0A= into a TCP-like equation to compute the receiver target reception rate,=0A= and adjusts its reception rate up and down to closely approximate the=0A= target reception rate. The sender sends packets to multiple channels=0A= called wave channels. Each wave channel follows the same pattern of=0A= packet rate transmission spread out over equally spaced intervals of=0A= time. The pattern of each wave is that it starts at a high rate that=0A= decreases gradually and continually over a long interval of time.=0A= (Picture an infinite sequence of waves.) The receiver increases its=0A= reception rate by joining the next wave channel earlier in the descent=0A= of the wave than it joined the previous wave channel, and the receiver=0A= decreases its reception rate by joining the next wave channel later in=0A= the descent of the wave than it joined the previous wave channel.=0A= =0A= WEBRC is designed for applications that use a fixed packet size, and=0A= vary their packet reception rate in response to congestion. WEBRC is=0A= designed to be reasonably fair when competing for bandwidth with TCP=0A= flows, where a flow is ``reasonably fair'' if its reception rate is=0A= generally within a factor of two of the reception rate of a TCP flow=0A= under the same conditions. However WEBRC has a much lower variation of=0A= throughput over time compared to TCP, which makes it more suitable for=0A= applications such as telephony or streaming media where a relatively=0A= smooth rate is of importance. Furthermore, WEBRC is designed to be=0A= =0A= =0A= =0A= Luby/Goyal Section 1. [Page 4]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= massively scalable in the sense that the sender is insensitive to the=0A= number of receivers joined to a multicast session. The penalty of=0A= having smoother throughput than TCP while competing fairly for bandwidth=0A= is that WEBRC responds slower than TCP to changes in available=0A= bandwidth.=0A= =0A= The receiver measures and performs the calculation of congestion control=0A= parameters (e.g., the average loss probability) and makes decisions on=0A= how to increase or decrease its rate based on these parameters. The=0A= receiver-based approach is well suited to an application where the=0A= sender is handling many concurrent connections and therefore WEBRC is=0A= suitable as a building block for multicast congestion control.=0A= =0A= The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",=0A= "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this=0A= document are to be interpreted as described in RFC2119.=0A= =0A= =0A= =0A= 2. Rationale=0A= =0A= WEBRC provides congestion control for massively scalable protocols using=0A= the IP multicast network service. The congestion control that WEBRC=0A= provides is common to a variety of applications, including reliable=0A= content delivery and streaming applications.=0A= =0A= WEBRC is designed to provide congestion control for all packets that are=0A= sent to a session. A session comprises multiple channels originating at=0A= a single sender that are used for some period of time to carry packets=0A= pertaining to the transmission of one or more objects that can be of=0A= interest to receivers. The logic behind defining a session as=0A= originating from a single sender is that this is the right granularity=0A= to regulate packet traffic via congestion control. The rationale for=0A= providing congestion control that uses multiple channels within the same=0A= session is that this allows the data on the channels to be layered, and=0A= a receiver joins and leaves channels in a layered order during its=0A= participation in the session. There are advantages to layered data for=0A= streaming, where more data can be sent to the lower layers and=0A= incrementally important data to higher layers. For reliable content=0A= delivery, as described in [12] an application can send in packets=0A= encoded data generated from an object in such a way that the arrival of=0A= enough packets by a receiver is sufficient to reliably reconstruct=0A= original object. A primary advantage of WEBRC is that each receiver=0A= controls it reception rate independent of other receivers. Thus, for=0A= example, a receiver with a slow connection to the sender does not slow=0A= down the receivers with faster connections.=0A= =0A= There are coding techniques that provide massively scalable reliability=0A= =0A= =0A= =0A= Luby/Goyal Section 2. [Page 5]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= and asynchronous delivery which are compatible with WEBRC. When=0A= combined the result is a massively scalable, reliable, asynchronous=0A= content delivery protocol that is network friendly. WEBRC also provides=0A= congestion control that suitable for streaming applications.=0A= =0A= WEBRC avoids using techniques that are not massively scalable. For=0A= example, WEBRC does not provide any mechanisms for sending information=0A= from receivers to senders, although this does not rule out protocols=0A= that both use WEBRC and do require sending information from receivers to=0A= senders.=0A= =0A= WEBRC provides congestion control that can be tuned for different=0A= applications that may have differing application requirements. For=0A= example, a content delivery protocol may aggressively strive to use all=0A= available bandwidth between receivers and the sender, and thus it must=0A= drastically back off its rate when there is competing traffic. On the=0A= other hand, a streaming delivery protocol may strive to maintain a=0A= constant rate instead of trying to use all available bandwidth, and thus=0A= it may not back off its rate as fast when there is competing traffic.=0A= =0A= WEBRC does not provide any support beyond congestion control, and thus=0A= it must be combined with other building blocks to provide a complete=0A= protocol instantiation. For example, WEBRC does not provide any means=0A= that can be used to identify which session each received packet belongs=0A= to. As another example, WEBRC does not provide support for identifying=0A= which object each packet is carrying information about.=0A= =0A= =0A= =0A= 3. Functionality=0A= =0A= A WEBRC session comprises a logically related set of channels=0A= originating from a single sender that are used for some period of time=0A= to carry data packets with a header carrying WEBRC Congestion Control=0A= Information. At the network layer, a channel can be uniquely identified=0A= by a (sender IP address, multicast group address) pair.=0A= =0A= For each WEBRC session, the channels within the session are mapped=0A= uniquely to consecutive channel numbers. In each packet sent to a=0A= channel, the channel number that corresponds to the channel is carried=0A= in the WEBRC Congestion Control Information. A WEBRC receiver uses the=0A= channel number to determine which channel within a session a packet is=0A= received from. When packets are received, they are first checked to see=0A= that they belong to the appropriate session before WEBRC is applied.=0A= For example, if LCT [11] is being used with the session, then the sender=0A= IP address together with the Transport Session Identifier supported by=0A= LCT would be used to determine which session a received packet belongs=0A= to. The particular details of how this filtering is performed it=0A= =0A= =0A= =0A= Luby/Goyal Section 3. [Page 6]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= outside the scope of this document. In the remainder of this document,=0A= channels will be referred to by their channel number in the scope of the=0A= session.=0A= =0A= At the sender, time is partitioned into time slots, each of duration TSD=0A= seconds. There are a fixed number T of time slot indices associated=0A= with a session. As time progresses, the current time slot index=0A= increases by one modulo T each TSD seconds. The current time slot index=0A= CTSI is carried in the WEBRC Congestion Control Information. This=0A= allows receivers to perform very coarse grained synchronization within a=0A= session.=0A= =0A= WEBRC congestion control is achieved by having the sender send packets=0A= associated with a given session to several different channels.=0A= Individual receivers dynamically join and leave these channels according=0A= to the network congestion as experienced by the receiver. These=0A= congestion control adjustments are performed at each receiver=0A= independently of all other receivers, and in the case of multicast=0A= without any impact on the sender. A packet sequence number is carried=0A= in the WEBRC Congestion Control Information. The packet sequence=0A= numbers are consecutively numbered per channel and are used by receivers=0A= to measure packet loss.=0A= =0A= The channels associated with a session consist of one base channel and T=0A= wave channels. The packet rate for each channel varies over time. For=0A= the base channel, packets are sent to the channel at a low rate BCR_P at=0A= the beginning of a time slot and this rate gradually decreases to=0A= P*BCR_P at the end of the time slot, where P < 1 is a constant defined=0A= later. This pattern for the base channel repeats over each time slot.=0A= For each wave channel i, packets are sent to channel i starting at a=0A= high rate MWCR_P and decreasing over time by a fixed fraction P per time=0A= slot until a rate of BCR_P is reached at the end of time slot i. Then,=0A= for a period of time called the quiescent period, no packets are sent to=0A= wave channel i, and thereafter the whole cycle repeats itself, where the=0A= duration of the cycle is T*TSD seconds. Thus, the wave channels are=0A= going through the same cyclic pattern of packet rate transmission spaced=0A= out evenly by TSD seconds.=0A= =0A= Before joining a session, the receivers MUST obtain enough of the=0A= session description to start the session. This MUST include the=0A= relevant session parameters needed by a receiver to participate in the=0A= session and perform WEBRC congestion control. The session description=0A= is determined by the sender and is typically communicated to the=0A= receivers out of band. How receivers obtain the session description is=0A= outside the scope of this document.=0A= =0A= When a receiver initiates a session, it first joins the base channel.=0A= The packets in the base channel help the receiver orient itself in terms=0A= =0A= =0A= =0A= Luby/Goyal Section 3. [Page 7]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= of what the current time slot index is, which in turn allows the=0A= receiver to know the relative rates on the wave channels. The receiver=0A= remains joined to the base channel for the duration of its participation=0A= in the session.=0A= =0A= After joining a session the receiver adjusts its rate upwards and=0A= downwards by joining wave channels in sequence, from the lowest rate=0A= wave channel and moving towards the higher rate wave channels. Since=0A= the relative ordering among the channels with respect to their rate=0A= depends on the current time slot index, it is important that the=0A= receiver continually monitor the current time slot index contained in=0A= received packets. The reception rate at the receiver is determined by=0A= how early each wave channel is joined by the receiver: the earlier the=0A= receiver joins a channel with respect to when its wave started, the=0A= higher the reception rate. When the receiver wants to decrease its=0A= rate, it joins the next wave channel at a later time relative to when it=0A= joined the previous wave. When the receiver wants to increase its rate,=0A= it joins the next wave channel at an earlier time relative to when it=0A= joined the previous wave.=0A= =0A= Once the receiver is joined to a wave channel, the receiver remains=0A= joined to the wave channel until the channel goes quiescent, at which=0A= point the receiver MUST leave the channel.=0A= =0A= The way the receiver adjusts its reception rate is inspired by TFRC [5].=0A= The receiver at all points in time maintains a target reception rate,=0A= and the receiver is allowed to join the next wave channel if after=0A= joining its anticipated reception rate would be at most its target=0A= reception rate. The target rate is continually updated based on a set=0A= of measured parameters. The primary parameters are the average loss=0A= probability (measured in much the same way as described in TFRC) and the=0A= average round-trip time (measured as described below).=0A= =0A= =0A= =0A= 3.1. Sender Operation=0A= =0A= The sender operation is by design much simpler than the receiver=0A= operation.=0A= =0A= =0A= =0A= 3.1.1. Sender inputs and initialization=0A= =0A= The primary input to the sender for the session is MSR_b. MSR_b is the=0A= maximum sender transmission rate in bits per second at any point in time=0A= in aggregate to all channels. This is also the maximum rate in bits per=0A= second that any receiver could receive data from the session at any=0A= =0A= =0A= =0A= Luby/Goyal Section 3.1.1. [Page 8]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= point in time.=0A= =0A= The secondary inputs to the sender are listed below. These are=0A= secondary inputs because in general the values for these inputs will be=0A= fixed to default values that will not change, or because they are set=0A= based on non-WEBRC considerations.=0A= =0A= o LENP_B is the length of packets in bytes sent to the session. The=0A= value of LENP_B depends on the complete protocol, but in general=0A= this should be set to as high a value as possible without exceeding=0A= the MTU size for the network that would cause fragmentation.=0A= =0A= o BCR_P is the transmission rate on the base channel at the beginning=0A= of a time slot in packets per second. The default value for BCR_P=0A= is 1.=0A= =0A= o TSD is the time slot duration measured in seconds. The default=0A= value for TSD is 10.=0A= =0A= o QD is the minimum quiescent period duration measured in seconds.=0A= The default value for QD is 300.=0A= =0A= o P is the multiplicative drop in every channel rate over each time=0A= slot. The default value for P is 0.75.=0A= =0A= From these inputs the following fixed sender parameters can be derived=0A= as follows.=0A= =0A= o MSR_P =3D MSR_b/(8*LENP_B) is the maximum sender transmission rate in=0A= packets per second.=0A= =0A= o BCR_b =3D 8*LENP_B*BCR_P is the rate of the base channel at the=0A= beginning of a time slot in bits per second.=0A= =0A= o N =3D floor{log_{1/P}((1+(1-P)/P^2)*(MSR_P/BCR_P))}-1 is the number = of=0A= active time slots for a wave channel. A wave channel is active in=0A= any time slot if it is not quiescent. N is also the number of wave=0A= channels active in every time slot.=0A= =0A= o Q =3D ceil(QD/TSD) is the number of quiescent time slots for a wave=0A= channel.=0A= =0A= o T =3D N + Q is the total number of time slots in a cycle. T is also=0A= the total number of wave channels.=0A= =0A= o For the base channel CN =3D T and for the wave channels CN =3D=0A= 0,1,...,T-1. The sender has the description of the channels=0A= assigned to the session and the mapping between the channels and the=0A= =0A= =0A= =0A= Luby/Goyal Section 3.1.1. [Page 9]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= CNs.=0A= =0A= o C =3D TSD*T is the total duration in seconds of a cycle.=0A= =0A= =0A= =0A= 3.1.2. Sending packets to the session=0A= =0A= The sender keeps track of the current time slot index CTSI. The value=0A= of CTSI is incremented by 1 modulo T each TSD seconds. The value of=0A= CTSI is placed into each packet in the format described in Section 5.=0A= For each packet sent to the session, the sender also places the channel=0A= number CN of the channel into the packets in the format described in=0A= Section 5. Recall that CN =3D T for the base channel and CN =3D = 0,1,...,T-1=0A= for the wave channels.=0A= =0A= For each packet sent to the session, the sender calculates and places a=0A= packet sequence number PSN into the packet. The value of PSN is scoped=0A= by CN, and the value of PSN is consecutively increasing within each=0A= channel. Furthermore, for each wave channel, the last packet sent=0A= before the channel becomes quiescent must have the maximum possible PSN=0A= value. When the short format for Congestion Control Information is used=0A= (see Section 5.1), this implies that for any wave channel the PSN values=0A= are 65 535, 65 534, 65 533, ..., going backward in time starting at the=0A= last packet sent to the channel just before the channel becomes=0A= quiescent. Similarly, when the long format for Congestion Control=0A= Information is used (see Section 5.2), the PSN for the final packet of=0A= any wave is 4 294 967 295. The PSN of the initial packet of a wave thus=0A= depends on TSD, P, BCR_P and MSR_P. The format for the PSN within=0A= packets is described in Section 5.=0A= =0A= Packets are sent to the base channel at a rate of BCR_P packets per=0A= second at the beginning of a time slot decreasing at a constant relative=0A= rate till the rate reaches P*BCR_P at the end of a time slot. To=0A= explain the packet rate for wave channels, suppose for simplicity that=0A= MSR_P/BCR_P =3D 1 + (1/P) + (1/P)^2 + ... + (1/P)^N =3D=0A= ((1/P)^{N+1}-1)*P/(1-P). For all i =3D 0,1,...,T-1, the rate for wave=0A= channel i behaves as follows. At the beginning of time slot i+Q+1=0A= modulo T, packets are sent to wave channel i at the rate of=0A= BCR_P*(1/P)^N. The rate of packets sent to wave channel i steadily=0A= decreases geometrically by a factor of P during each of the next N time=0A= slots until the end of time slot i when the rate reaches BCR_P. During=0A= the Q time slots i+1 modulo T, i+2 modulo T, ..., i+Q modulo T, wave=0A= channel i is quiescent, i.e., no packets are sent to the channel.=0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= Luby/Goyal Section 3.1.2. [Page 10]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= 3.2. Receiver Operation=0A= =0A= The bulk of the complexity in WEBRC is in the receiver operation. For=0A= ease of explanation, suppose for the moment that during the reception=0A= there is no packet loss and packets are arriving at exactly the rate at=0A= which they were sent. The sender transmission rate to the channels is=0A= designed so that the receiver reception rate behaves as follows.=0A= =0A= Upon entering a session, the receiver immediately joins the base=0A= channel. When the receiver wants to increase its rate, it consecutively=0A= joins wave channels from the lowest rate channel to the highest rate=0A= channel. (Recall that the designations of lowest to highest change at=0A= time slot boundaries.) When the receiver wants to maintain its current=0A= reception rate and it is already joined to NWC wave channels, if the=0A= receiver joins channel i-1+NWC modulo T sometime during time slot i then=0A= the receiver joins channel i+NWC modulo T TSD seconds later in time slot=0A= i+1. As each wave channel becomes quiescent the receiver leaves the=0A= channel.=0A= =0A= Suppose the receiver wants to decrease its rate till it is joined to=0A= just the base channel. Assume that a receiver is joined to the NWC wave=0A= channels i, i+1 modulo T, ..., i+NWC-1 modulo T at the beginning of time=0A= slot i. Then, the aggregate packet reception rate of the receiver over=0A= the next NWC time slots will behave as follows. At the beginning of=0A= time slot i the receiver reception rate is BCR_P*(1 + (1/P) + (1/P)^2 +=0A= ... + (1/P)^NWC). Then the receiver reception rate decreases by a=0A= factor of P over the duration of each time slot, and at the end of each=0A= time slot the reception rate decreases by an additive amount of P*BCR_P.=0A= At the end of time slot i+NWC-1 mod T, the receiver reception rate is=0A= BCR_P*(1+P), and at the beginning of time slot i+NWC mod T the receiver=0A= is joined only to the base channel and its reception rate is BCR_P.=0A= =0A= =0A= =0A= 3.2.1. Receiver inputs and initialization=0A= =0A= Before joining a session the receiver MUST know the mapping between the=0A= CNs and the channels. Upon joining the session, it should have the=0A= values of LENP_B, BCR_P, TSD, P, N, Q and T. Some of these values may=0A= be obtained or measured once the receiver has joined the session. For=0A= example, the receiver MAY obtain LENP_B and T from the first packet=0A= received from the base channel, and the receiver MAY measure BCR_P once=0A= it is joined to the base channel. The values of P, Q and TSD MAY be=0A= fixed to default values built into the receiver that do not change from=0A= session to session, and the value of N MAY be computed as T-Q.=0A= =0A= When a receiver first joins a session, it MUST first join just the base=0A= channel and start receiving packets to determine the current time slot=0A= =0A= =0A= =0A= Luby/Goyal Section 3.2.1. [Page 11]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= index. If during the course of the session the receiver continually=0A= loses a high fraction of the packets from the base channel even when the=0A= receiver is only joined to the base channel, the receiver SHOULD leave=0A= the session.=0A= =0A= The receiver MAY also have other individually set parameters that may be=0A= used to determine its behavior. Two such parameters are:=0A= =0A= o MRR_b is the maximum receiver reception rate in bits per second.=0A= This may be used to determine the maximum reception rate this=0A= receiver is willing to reach. Thus, the maximum reception rate that=0A= the receiver can possibly achieve in the session is the minimum of=0A= MSR_b and MRR_b. A recommended value of MRR_b for a receiver is the=0A= bandwidth capacity of the last link to the receiver. MRR_P is the=0A= maximum receiver reception rate in packets per second, i.e., MRR_P =3D=0A= MRR_b/(8*LENP_B).=0A= =0A= o CONNEC is the number of virtual connections that the receiver is=0A= making to the sender to receive data from the session. This affects=0A= how competitive the receiver will be compared to other receivers=0A= (both TCP and WEBRC receivers) in competing for bandwidth over=0A= bottleneck links. The standard default value for CONNEC is 1.=0A= =0A= =0A= =0A= 3.2.2. Receiver measurements and calculations=0A= =0A= As outlined in the introduction, the way a receiver adjusts its=0A= reception rate is inspired by TFRC [5]. The receiver at all points in=0A= time maintains a target reception rate, and the receiver is allowed to=0A= join the next wave channel if joining would increase its reception rate=0A= to at most its target reception rate. The target rate is continually=0A= updated based on a set of measured parameters. The primary parameters=0A= are the average loss probability LOSSP and the average round-trip time=0A= ARTT.=0A= =0A= =0A= =0A= 3.2.2.1. Average loss probability=0A= =0A= The average loss probability LOSSP of the receiver is maintained in a=0A= manner very similar to that described in TFRC [5]. The calculation of=0A= LOSSP as the inverse of a weighted average is exactly as described in=0A= TFRC, with the same definition of loss events. One difference is the=0A= number of inter-loss periods that are used to compute the average loss=0A= event rate:=0A= =0A= =0A= =0A= =0A= =0A= Luby/Goyal Section 3.2.2.1. [Page 12]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= o In TFRC, it is recommended that the loss probability be averaged=0A= over the previous 8 loss events.=0A= =0A= o In WEBRC, it is recommended that the loss probability LOSSP be=0A= averaged over the previous channel periods that include 8 channel=0A= periods where there was at least one loss event. A channel period=0A= is defined as the time between when a wave channel is joined till=0A= the time the next wave channel is joined. Thus, since a channel=0A= period is generally around TSD seconds, the loss probability will=0A= generally be averaged over at least 8*TSD seconds.=0A= =0A= The other difference is in the weights:=0A= =0A= o In TFRC, it is recommended that the weights (1, 1, 1, 1, 0.875,=0A= 0.625, 0.375, 0.125) be used for the 8 inter-loss periods (from most=0A= recent to most distant past).=0A= =0A= o In WEBRC, it is recommended that these same 8 values be used for the=0A= loss events that are 1,2,...,8 lossy channel periods into the past.=0A= In addition, it is recommended that the weight for each inter-loss=0A= period be multiplicatively adjusted by the factor P^max{0,LOSS_NWC-=0A= NWC-1}, where NWC is the number of wave channels the receiver is=0A= currently subscribed to and LOSS_NWC is the number of wave channels=0A= the receiver was subscribed to at the end of the inter-loss period=0A= in question.=0A= =0A= =0A= =0A= 3.2.2.2. Average round-trip time=0A= =0A= The receiver maintains an average round-trip time, ARTT, as the=0A= exponentially weighted moving average of RTT measurements. Each time=0A= the receiver joins a channel (either the base channel at the beginning=0A= or wave channels continually), it makes a measurement of the round-trip=0A= time RTT as follows. When the receiver sends the join for the channel=0A= it records the current time X and sets a Boolean variable JOINING to=0A= true. When the first packet is received from the channel the receiver=0A= records the current time Y and resets the value of JOINING to false. If=0A= it is the base channel that has been joined, RTT is set to Y - X.=0A= Otherwise, when the receiver receives a second packet from the channel=0A= it records the current time Z. Then, the value of RTT is set to 3*Y/2 -=0A= Z/2 - X. (Note that this value can be negative.)=0A= =0A= When the kth RTT measurement is made, the value of ARTT is updated to=0A= max{ARRT/2, (1-delta)*ARTT + delta*RTT}, where delta =3D=0A= alpha/(1-(1-alpha)^k). The recommended value for alpha is 0.1.=0A= =0A= It is RECOMMENDED that the RTT measurement be adjusted appropriately for=0A= =0A= =0A= =0A= Luby/Goyal Section 3.2.2.2. [Page 13]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= packets lost between the first and second received packets.=0A= =0A= =0A= =0A= 3.2.2.3. Rate Equation=0A= =0A= The receiver calculates the reception rate REQN based on the TCP-=0A= equation as follows: REQN =3D CONNEC/(ARTT*sqrt{LOSSP}(0.816 +=0A= 7.35*LOSSP*(1+32*LOSSP^2))). This equation comes from TFRC [5].=0A= =0A= =0A= 3.2.2.4. Epochs=0A= =0A= The receiver makes decisions on whether or not to join another wave=0A= channel at equally spaced units of time called epochs. The duration of=0A= an epoch in seconds, EL, is set to be a small fraction of TSD, so that=0A= decisions to join a channel can be made at a much finer granularity than=0A= TSD. A standard setting is EL =3D TSD/20. Thus, if TSD =3D 10, then EL = =3D=0A= 0.5.=0A= =0A= =0A= =0A= 3.2.2.5. Average reception rate=0A= =0A= There are two averaged versions of reception rate maintained by the=0A= receiver: TRR_P, the true reception rate, and ARR_P, the anticipated=0A= reception rate. These are used for different purposes and thus are=0A= calculated quite differently.=0A= =0A= The true reception rate TRR_P is used to ensure that the receiver does=0A= not increase its reception rate too quickly above its current reception=0A= rate. TRR_P is calculated based on the measurement of RR_P, where RR_P=0A= is the receiver reception rate in packets per second measured at the=0A= beginning of an epoch averaged over the epoch that just ended. TRR_P is=0A= initialized to BCR_P. TRR_P is updated to (1-beta)*TRR_P + beta*RR_P at=0A= the beginning of each epoch after RR_P is measured for the previous=0A= epoch.=0A= =0A= The anticipated reception rate ARR_P is the receiver's estimate of the=0A= total instantaneous rate of the currently subscribed channels. It is=0A= used to compare against the target rate to decide whether or not the=0A= receiver can increase its reception rate by joining the next channel.=0A= ARR_P is calculated based on a measurement IRR_P and on the number of=0A= subscribed wave channels NWC. IRR_P is what the ideal receiver=0A= reception rate should have been in packets per second measured at the=0A= beginning of the epoch averaged over the previous epoch, i.e., IRR_P=0A= includes both received and lost packets. ARR_P is initialized to BCR_P=0A= and NWC is initialized to 0. ARR_P, IRR_P and NWC are updated as=0A= =0A= =0A= =0A= Luby/Goyal Section 3.2.2.5. [Page 14]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= follows:=0A= =0A= o At the beginning of each epoch, IRR_P is measured over the previous=0A= epoch and then ARR_P is updated to P^(EL/TSD)*(1-gamma)*ARR_P +=0A= gamma*IRR_P.=0A= =0A= o When a join is made to a wave channel, NWC is updated to NWC+1 and=0A= then ARR_P is multiplicatively increased by the factor=0A= ((1/P)^(NWC+1)-1)/((1/P)^NWC-1). (Joins happen at epoch boundaries;=0A= this adjustment is in addition to the adjustment above.)=0A= =0A= o When a new time slot index is detected, ARR_P is additively=0A= increased by (1-P)*BCR_P to account for the change in rate on the=0A= base channel. In addition, for each wave channel that has just been=0A= detected to be quiescent and thus had a leave message sent, ARR_P is=0A= additively decreased by BCR_P and NWC is decremented by 1.=0A= (Normally, a change of time slot index indicates that one wave=0A= channel has become quiescent.)=0A= =0A= Consider for the moment what happens if gamma =3D 0 and ARR_P is an=0A= accurate estimate of the total rate of the subscribed channels. The=0A= adjustments to ARR_P upon joining and leaving wave channels, with the=0A= passage of epochs, and with the detection of time slot changes will then=0A= cause ARR_P to remain an accurate estimate. In practice, gamma MUST be=0A= positive; allowing an influence of IRR_P prevents ARR_P from drifting=0A= away from being an accurate estimate of the total subscribed rate.=0A= =0A= The motivation for separate estimates TRR_P and ARR_P is as follows.=0A= TRR_P alone is not appropriate for comparison with the TFRC-inspired=0A= target rate because there is a considerable lag before it reflects the=0A= rate increase resulting from joining the next wave channel. However,=0A= TRR_P is needed because ARR_P alone does not reflect the amount of data=0A= getting to the receiver, and a large gap between the subscribed rate and=0A= the received rate should not be allowed.=0A= =0A= The recommended values of beta and gamma depend on whether the receiver=0A= is in slow start mode. Slow start is indicated by TRR_P < SSR_P/zeta^2,=0A= where SSR_P is defined in the following subsection and zeta =3D=0A= ((1/P)^(NWC+2)-1)/((1/P)^(NWC+1)-1). It is recommended that beta and=0A= gamma equal P/(1+P) in slow start and equal (2/3)*(-c/2 + sqrt(c^2/4 +=0A= 3*c/2), where c =3D 1 - cos(6.283*EL/TSD), otherwise.=0A= =0A= =0A= =0A= 3.2.2.6. Slow start=0A= =0A= WEBRC uses a slow start mechanism to quickly ramp up its rate at both=0A= the beginning of the session and in the middle of a session when the=0A= =0A= =0A= =0A= Luby/Goyal Section 3.2.2.6. [Page 15]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= rate drops precipitously. To enact this, the receiver maintains the=0A= following parameters:=0A= =0A= o SSMINR_P is the minimum allowed slow start threshold rate in packets=0A= per second. The recommended value for SSMINR_P is=0A= BCR_P*(1+1/P+1/P^2).=0A= =0A= o SSR_P is the slow start threshold rate in packets per second. SSR_P=0A= is initialized to MRR_P*P^2.=0A= =0A= =0A= =0A= 3.2.2.7. Target rate=0A= =0A= The target rate TRATE is computed as TRATE =3D min{max{SSR_P, REQN},=0A= TRR_P*((1/P)^(NWC+3)-1)/((1/P)^(NWC+1)-1), MRR_P}.=0A= =0A= =0A= =0A= 3.2.3. Receiver events=0A= =0A= There are various receiver events, some of which are triggered by the=0A= passing of time on the receiver, and others by events such as packet=0A= reception, detection of packet loss, reception of a first packet from a=0A= channel, and exceptional time-outs.=0A= =0A= =0A= =0A= 3.2.3.1. Epoch change=0A= =0A= This is an event that is driven by the passage of time on the receiver,=0A= which occurs each EL seconds. When this happens, TRR_P and ARR_P are=0A= computed as described in Section 3.2.2.5. Immediately after these=0A= updates, a decision is made about whether to join a wave channel as=0A= described in Section 3.2.3.3.=0A= =0A= =0A= =0A= =0A= 3.2.3.2. Time slot change=0A= =0A= This is an event that is driven by the reception of a packet with a new=0A= CTSI value. When a packet with a new CTSI =3D i is received, a leave is=0A= sent for wave channel i-1 modulo T, NWC is updated to NWC-1, and ARR_P=0A= is updated to ARR_P - P*BCR_P as described in Section 3.2.2.5.=0A= =0A= =0A= =0A= =0A= =0A= =0A= Luby/Goyal Section 3.2.3.2. [Page 16]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= 3.2.3.3. Join a wave channel=0A= =0A= At the beginning of each epoch, after updating the values of ARR_P and=0A= TRR_P as described in Section 3.2.2.5, the receiver decides whether or=0A= not to join a wave channel as follows:=0A= =0A= o If there is a loss event in progress (LOSS_EVENT =3D true) or there = is=0A= a join of a channel in progress (JOINING =3D true), then no join of a=0A= channel is attempted.=0A= =0A= o The receiver calculates REQN as described in Section 3.2.2.3.=0A= =0A= o The receiver calculates TRATE as described in Section 3.2.2.7.=0A= =0A= o The receiver checks if TRATE >=0A= ARR_P*((1/P)^{NWC+2}-1)/((1/P)^{NWC+1}-1). In this case,=0A= subscribing to an additional wave channel is not expected to=0A= increase the subscribed rate over the target rate. Suppose CTSI =3D = i=0A= and NWC =3D J. If the inequality is true then the receiver joins the=0A= wave channel with CN =3D i+J modulo T, updates NWC to NWC+1, and then=0A= updates ARTT to ARR_P*((1/P)^{NWC+1}-1)/((1/P)^NWC-1) as described=0A= in Section 3.2.2.5.=0A= =0A= =0A= 3.2.3.4. Loss event=0A= =0A= Each time the receiver detects a lost packet (based on the sequence=0A= numbers in the packets scoped by the channel number), the receiver=0A= records the start of a new loss event, and sets a Boolean variable=0A= LOSS_EVENT to true that will automatically reset to false after ARTT=0A= seconds. All subsequent packet loss for a period of ARTT seconds is=0A= considered as part of the same loss event. When a start of a loss event=0A= is detected, the value of SSR_P is updated to max{SSMINR_P, TRR_P*P^2}.=0A= =0A= It is RECOMMENDED that the receiver account for simple misordering of=0A= packets without inferring a loss.=0A= =0A= =0A= =0A= 3.2.3.5. Exceptional timeouts=0A= =0A= These are timeouts when the packet reception behavior is far from what=0A= it should be and should trigger a drastic event by the receiver.=0A= Exception timeouts include events such as when no packets are received=0A= for a long time, there is no change in the current time slot index for a=0A= long time, or no first or second packet is received after join to=0A= channel for a long time.=0A= =0A= =0A= =0A= =0A= Luby/Goyal Section 3.2.3.5. [Page 17]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= 4. Applicability Statement=0A= =0A= WEBRC is intended to be a congestion control scheme that can be used in=0A= a complete protocol instantiation that delivers objects and streams=0A= (both reliable content delivery and streaming of multimedia=0A= information). A WEBRC session comprises a logically related set of one=0A= or more channels originating at a single sender that are used for some=0A= period of time to carry packets pertaining to the transmission of one or=0A= more objects that can be of interest to receivers. WEBRC congestion=0A= control is to be performed over the aggregate of all packets sent to the=0A= session.=0A= =0A= WEBRC is most applicable for delivery of objects or streams of=0A= substantial length, i.e., objects or streams that range in length from=0A= hundreds of kilobytes to many gigabytes, and whose transfer time is in=0A= the order of tens of seconds or more.=0A= =0A= =0A= 4.1. Environmental Requirements and Considerations=0A= =0A= =0A= WEBRC can be used with both multicast and unicast networks. However,=0A= the scope of this document is limited to multicast. WEBRC requires=0A= connectivity between a sender and receivers, but does not require=0A= connectivity from receivers to the sender.=0A= =0A= WEBRC inherently works with all types of networks, including LANs, WANs,=0A= Intranets, the Internet, asymmetric networks, wireless networks, and=0A= satellite networks. Thus, the inherent raw scalability of WEBRC is=0A= unlimited. However, in some network environments varying reception=0A= rates to receivers may not be advantageous, e.g., when the network is a=0A= satellite network with a fixed amount of bandwidth dedicated to a=0A= session.=0A= =0A= Receivers join and leave channels using the appropriate multicast join=0A= and leave messages. For IPv4 multicast, IGMP messages are used by=0A= receivers to join and leave channels. For IPv6, MLDv2 messages are used=0A= by receivers to join and leave channels. This is the only dependency of=0A= WEBRC on the IP version.=0A= =0A= WEBRC requires receivers to be able to uniquely identify and demultiplex=0A= packets associated with a session in order to effectively perform=0A= congestion control over all packets associated with the session. How=0A= receivers achieve this is outside the scope of this document.=0A= =0A= WEBRC is presumed to be used with an underlying network or transport=0A= service that is a ``best effort'' service that does not guarantee packet=0A= reception, packet reception order, and which does not have any support=0A= =0A= =0A= =0A= Luby/Goyal Section 4.1. [Page 18]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= for flow or congestion control. For example, the Any-Source Multicast=0A= (ASM) model of IP multicast as defined in RFC1112 [4] is such a best=0A= effort network service. While the basic service provided by RFC1112 is=0A= largely scalable, providing congestion control or reliability should be=0A= done carefully to avoid severe scalability limitations, especially in=0A= the presence of heterogeneous sets of receivers.=0A= =0A= There are currently two models of multicast delivery, the Any-Source=0A= Multicast (ASM) model as defined in RFC1112 [4] and the Source-Specific=0A= Multicast (SSM) model as defined in [6]. WEBRC works with both multicast=0A= models, but in a slightly different way with somewhat different=0A= environmental concerns. When using ASM, a sender S sends packets to a=0A= multicast group G, and the WEBRC channel address consists of the pair=0A= (S,G), where S is the IP address of the sender and G is a multicast=0A= group address. When using SSM, a sender S sends packets to an SSM=0A= channel (S,G), and the WEBRC channel address coincides with the SSM=0A= channel address.=0A= =0A= A sender can locally allocate unique SSM channel addresses, and this=0A= makes allocation of channel addresses easy with SSM. To allocate=0A= channel addresses using ASM, the sender must uniquely chose the ASM=0A= multicast group address across the scope of the group, and this makes=0A= allocation of WEBRC channel addresses more difficult with ASM.=0A= =0A= WEBRC channels and SSM channels coincide, and thus the receiver will=0A= only receive packets sent to the requested WEBRC channel. With ASM, the=0A= receiver joins a channel by joining a multicast group G, and all packets=0A= sent to G, regardless of the sender, may be received by the receiver.=0A= Thus, SSM has compelling security advantages over ASM for prevention of=0A= denial of service attacks. In either case, receivers SHOULD use=0A= mechanisms to filter out packets from unwanted sources.=0A= =0A= WEBRC assumes that the packet route between the sender and a particular=0A= receiver is the same for all channels associated with a session. For=0A= SSM this assumption is true because the multicast tree is a shortest=0A= path tree from each receiver to the sender and generally this path=0A= changes infrequently. For ASM there are some issues that if not=0A= properly considered may invalidate this assumption. With ASM, the=0A= packet route between the sender and receivers may initially be through=0A= the Rendezvous Point (RP) and then switch over to the shortest path to=0A= the sender as packets start flowing in a channel. The first issue is=0A= that the RP may not be the same for all channels associated with a=0A= session, and thus the first packets sent to the channels may follow a=0A= route that depends on the RP of the channel. This depends on the RP=0A= configuration for the sender. If the sender registers all channels=0A= associated with the session with the same RP then the assumption is=0A= true, but if the sender registers different channels with different RPs=0A= then the assumption may not be true. Thus, it is RECOMMENDED that the=0A= =0A= =0A= =0A= Luby/Goyal Section 4.1. [Page 19]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= sender register all channels associated with a session with the same RP.=0A= Another issue is that when the channel switches over from the RP to the=0A= sender-based tree then the route to the receivers may vary within a=0A= channel. Furthermore, this may cause either the receipt of duplicate=0A= packets at receivers or loss of packets depending on the smoothness of=0A= the switchover. Thus, it is RECOMMENDED that the RP be placed as close=0A= as possible to the sender. The best location for the RP is that it be=0A= the first-hop router closest to the sender, in which case the path to=0A= the sender and the path to the RP is the same for each receiver and the=0A= problems mentioned above are eliminated. The consequences of this=0A= assumption not being true are that the receiver reaction to congestion=0A= may not be appropriate. Generally, the WEBRC receiver will act=0A= conservatively and reduce its reception rate too much if this assumption=0A= is not true, but there can be cases where the receivers will act=0A= inappropriately.=0A= =0A= Some networks are not amenable to the WEBRC congestion control protocol.=0A= In particular, for a satellite or wireless network, there may be no=0A= mechanism for receivers to effectively reduce their reception rate since=0A= there may be a fixed transmission rate allocated to the session.=0A= =0A= =0A= =0A= 5. Packet Header Fields=0A= =0A= Packets sent to a session using WEBRC MUST include Congestion Control=0A= Information fields as specified in this section. This document specifies=0A= short and long formats for the Congestion Control Information, and it is=0A= RECOMMENDED that protocol instantiations use one of these two formats.=0A= Other formats for the Congestion Control Information fields MAY be used=0A= by protocol instantiations, but all protocol instantiations are REQUIRED=0A= to use these fields in a format that is compatible with the=0A= interpretations of these fields. Thus, if a protocol does use a=0A= different format for the fields in the Congestion Control Information=0A= then it MUST specify the lengths and positions of these fields within=0A= the packet header.=0A= =0A= All integer fields are carried in "big-endian" or "network order"=0A= format, that is, most significant byte (octet) first. All constants,=0A= unless otherwise specified, are expressed in base ten.=0A= =0A= =0A= =0A= 5.1. Short Format Congestion Control Information=0A= =0A= The short format for the Congestion Control Information is shown in Fig.=0A= 1. The total length of the short format is 32-bits.=0A= =0A= =0A= =0A= =0A= Luby/Goyal Section 5.1. [Page 20]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= 0 1 2 3=0A= 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1=0A= +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+=0A= | CTSI | Channel Number| Packet Sequence Number |=0A= +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+=0A= =0A= =0A= Fig. 1 - Short format for Congestion Control Information=0A= =0A= =0A= The function of each field in the Congestion Control Information is the=0A= following.=0A= =0A= =0A= Current Time Slot Index (CTSI): 8 bits=0A= =0A= CTSI indicates the index of the current time slot. This must be=0A= sent in each packet within the session. The Current Time Slot=0A= Index increases by one modulo T each TSD seconds at the sender,=0A= where T is the number of time slots associated with the session=0A= and TSD is the time slot duration. Note that T is also the number=0A= of wave channels associated with the session, and thus T MUST be=0A= at most 255.=0A= =0A= =0A= Channel Number (CN): 8 bits=0A= =0A= CN is the channel number that this packet belongs to. CN for the=0A= base channel is T, and the CNs for the wave channels are 0 through=0A= T-1. Thus, T+1 channels in total are used, and thus T MUST be at=0A= most 255.=0A= =0A= =0A= Packet Sequence Number (PSN): 16 bits=0A= =0A= The PSN of each packet is scoped by its CN value. The sequence=0A= numbers of consecutive packets sent to the base channel are=0A= numbered consecutively modulo 2^16. The same sequence of PSNs are=0A= used for each wave channel in each cycle. The sequence numbers of=0A= consecutive packets sent to a wave channel are numbered=0A= consecutively modulo 2^16 within each cycle, ending with the last=0A= packet sent to the channel before the channel goes quiescent with=0A= PSN =3D 2^16-1.=0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= Luby/Goyal Section 5.1. [Page 21]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= 5.2. Long Format Congestion Control Information=0A= =0A= The long format for the Congestion Control Information is shown in Fig.=0A= 2. The total length of the long format is 64-bits.=0A= =0A= =0A= 0 1 2 3=0A= 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1=0A= +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+=0A= | CTSI | Channel Number |=0A= +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+=0A= | Packet Sequence Number |=0A= +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+=0A= =0A= =0A= Fig. 1 - Long format for Congestion Control Information=0A= =0A= =0A= The meaning of each field for the long format is the same as for the=0A= short format, the only difference is that each field is twice as long.=0A= =0A= =0A= Current Time Slot Index (CTSI): 16 bits=0A= =0A= CTSI indicates the index of the current time slot. This must be=0A= sent in each packet within the session. The Current Time Slot=0A= Index increases by one modulo T each TSD seconds at the sender,=0A= where T is the number of time slots associated with the session=0A= and TSD is the time slot duration. Note that T is also the number=0A= of wave channels associated with the session, and thus T MUST be=0A= at most 65 535.=0A= =0A= =0A= Channel Number (CN): 16 bits=0A= =0A= CN is the channel number that this packet belongs to. CN for the=0A= base channel is T, and the CNs for the wave channels are 0 through=0A= T-1. Thus, T+1 channels in total are used, and thus T MUST be at=0A= most 65 535.=0A= =0A= =0A= Packet Sequence Number (PSN): 32 bits=0A= =0A= The PSN of each packet is scoped by its CN value. The sequence=0A= numbers of consecutive packets sent to the base channel are=0A= numbered consecutively modulo 2^32. The same sequence of PSNs are=0A= used for each wave channel in each cycle. The sequence numbers of=0A= consecutive packets sent to a wave channel are numbered=0A= =0A= =0A= =0A= Luby/Goyal Section 5.2. [Page 22]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= consecutively modulo 2^32 within each cycle, ending with the last=0A= packet sent to the channel before the channel goes quiescent with=0A= PSN =3D 2^32-1.=0A= =0A= =0A= =0A= 6. Requirements From Other Building Blocks=0A= =0A= As described in RFC3048, WEBRC is a building block that is intended to=0A= be used, in conjunction with other building blocks, to help specify a=0A= protocol instantiation.=0A= =0A= WEBRC does not provide higher level session support, e.g., how receivers=0A= obtain the necessary session description and how the receivers=0A= demultiplex received packets based on their session. There is support=0A= provided by other building blocks that can be used in conjunction with=0A= WEBRC to provide some of this support. For example, LCT [11] can=0A= provide some of the higher level in-band session support that may be=0A= needed by receivers, and the WEBRC Congestion Control Information (CCI)=0A= required in each packet can be carried in the CCI field of the LCT=0A= header [11].=0A= =0A= WEBRC does not provide any type of reliability, and in particular does=0A= not provide support for retransmission of loss packets. Reliability can=0A= be added by independent means, such as by the use of FEC codes as=0A= described in [12] and specified in the FEC building block [13].=0A= =0A= =0A= 7. Security Considerations=0A= =0A= WEBRC can be subject to denial-of-service attacks by attackers that try=0A= to confuse the congestion control mechanism for receivers by injecting=0A= forged packets into the multicast stream. This attack most adversely=0A= affects network elements and receivers downstream of the attack, and=0A= much less significantly the rest of the network and other receivers.=0A= Because of this and because of the potential attacks due to the use of=0A= FEC described above, it is RECOMMENDED that some form of packet=0A= authentication such as TESLA [15] be used to protect against such=0A= attacks and that Reverse Path Forwarding checks be enabled in all=0A= network routers and switches along the path from the sender to receivers=0A= to limit the possibility of a bad agent injecting forged packets into=0A= the multicast tree data path.=0A= =0A= A receiver with an incorrect or corrupted implementation of WEBRC may=0A= affect health of the network in the path between the sender and the=0A= receiver, and may also affect the reception rates of other receivers=0A= joined to the session. It is therefore RECOMMENDED that receivers be=0A= required to identify themselves as legitimate before they receive the=0A= =0A= =0A= =0A= Luby/Goyal Section 7. [Page 23]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= session description needed to join the session.=0A= =0A= Another vulnerability of ALC is the potential of receivers obtaining an=0A= incorrect session description for the session. The consequences of this=0A= could be that legitimate receivers with the wrong session description=0A= are unable to correctly receive the session content, or that receivers=0A= inadvertently try to receive at a much higher rate than they are capable=0A= of, thereby disrupting traffic in portions of the network. To avoid=0A= these problems, it is RECOMMENDED that the receiver authenticate the=0A= session description, for example by using either the ESP (with enabled=0A= authentication service) [9] or AH [8] protocols of IPSEC [7] to ensure=0A= the authenticity of the session description.=0A= =0A= =0A= =0A= 8. IANA Considerations=0A= =0A= No information in this specification is subject to IANA registration.=0A= =0A= =0A= =0A= 9. Intellectual Property Issues=0A= =0A= =0A= Digital Fountain maintains all rights to the WEBRC technology, but will=0A= grant a royalty-free worldwide license to the WEBRC technology once it=0A= has been standardized.=0A= =0A= WEBRC may be used with other protocols which are proprietary, or have=0A= pending or granted patents.=0A= =0A= =0A= =0A= 10. References=0A= =0A= =0A= [1] Bradner, S., ``The Internet Standards Process -- Revision 3,''=0A= RFC2026, October 1996.=0A= =0A= [2] Bradner, S., ``Key words for use in RFCs to Indicate Requirement=0A= Levels,'' RFC2119, March 1997.=0A= =0A= [3] Byers, J.W., Frumin, M., Horn, G., Luby, M., Mitzenmacher, M.,=0A= Roetter, A., and Shaver, W., ``FLID-DL: Congestion Control for Layered=0A= Multicast,'' Proc. Second International Workshop on Networked Group=0A= Communications (NGC 2000), Palo Alto, CA, November 2000, pp. 71-81.=0A= =0A= [4] Deering, S., ``Host Extensions for IP Multicasting,'' RFC1112,=0A= =0A= =0A= =0A= Luby/Goyal Section 10. [Page 24]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= August 1989.=0A= =0A= [5] Handley, M., Padhye, J., Floyd, S., Widmer, J., ``TCP Friendly Rate=0A= Control (TFRC): Protocol Specification,'' Internet Draft draft-ietf-=0A= tsvwg-tfrc-03, July 2001, a work in progress.=0A= =0A= [6] Holbrook, H. W., ``A Channel Model for Multicast,'' Ph.D.=0A= Dissertation, Stanford University, Department of Computer Science,=0A= Stanford, California, August 2001.=0A= =0A= [7] Kent, S., Atkinson, R., ``Security Architecture for the Internet=0A= Protocol,'' RFC2401, November 1998.=0A= =0A= [8] Kent, S., Atkinson, R., ``IP Authentication Header,'' RFC2406,=0A= November 1998.=0A= =0A= [9] Kent, S., Atkinson, R., ``IP Encapsulating Security Payload (ESP),''=0A= RFC2406, November 1998.=0A= =0A= [10] Luby, M., Gemmell, J., Vicisano, L., Rizzo, L., Crowcroft, J.,=0A= ``Asynchronous Layered Coding protocol instantiation,'' Internet Draft=0A= draft-ietf-rmt-pi-alc-06, February 2002, a work in progress.=0A= =0A= [11] Luby, M., Gemmell, J., Vicisano, L., Rizzo, L., Handley, M.,=0A= Crowcroft, J., ``Layered Coding Transport building block,'' Internet=0A= Draft draft-ietf-rmt-bb-lct-04.txt, February 2002, a work in progress.=0A= =0A= [12] Luby, M., Gemmell, Vicisano, L., J., Rizzo, L., Handley, M.,=0A= Crowcroft, J., ``The Use of Forward Error Correction in Reliable=0A= Multicast,'' Internet Draft draft-ietf-rmt-info-fec-02.txt, February=0A= 2002, a work in progress.=0A= =0A= [13] Luby, M., Gemmell, J., Vicisano, L., Rizzo, L., Handley, M.,=0A= Crowcroft, J., ``Forward Error Correction building block,'' Internet=0A= Draft draft-ietf-rmt-bb-fec-06.txt, February 2002.=0A= =0A= [14] Mankin, A., Romanow, A., Bradner, S., Paxson V., ``IETF Criteria=0A= for Evaluating Reliable Multicast Transport and Application Protocols,''=0A= RFC2357, June 1998.=0A= =0A= [15] Perrig, A., Canetti, R., Song, D., Tygar, J.D., ``Efficient and=0A= Secure Source Authentication for Multicast,'' Network and Distributed=0A= System Security Symposium, NDSS 2001, pp. 35-46, February 2001.=0A= =0A= [16] Vicisano, L., Rizzo, L., Crowcroft, J., "TCP-like Congestion=0A= Control for Layered Multicast Data Transfer", Proc. IEEE Infocom '98,=0A= San Francisco, CA, March-April 1998, pp. 996-1003.=0A= =0A= =0A= =0A= =0A= Luby/Goyal Section 10. [Page 25]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= [17] Whetten, B., Vicisano, L., Kermode, R., Handley, M., Floyd, S.,=0A= Luby, M., ``Reliable Multicast Transport Building Blocks for One-to-Many=0A= Bulk-Data Transfer,'' RFC3048, January 2001.=0A= =0A= =0A= =0A= 11. Authors' Addresses=0A= =0A= Michael Luby=0A= luby@digitalfountain.com=0A= Digital Fountain=0A= 39141 Civic Center Drive=0A= Suite 300=0A= Fremont, CA, USA, 94538=0A= =0A= Vivek Goyal=0A= vivek@digitalfountain.com=0A= Digital Fountain=0A= 39141 Civic Center Drive=0A= Suite 300=0A= Fremont, CA, USA, 94538=0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= Luby/Goyal Section 11. [Page 26]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= 12. Full Copyright Statement=0A= =0A= Copyright (C) The Internet Society (2001). All Rights Reserved.=0A= =0A= This document and translations of it may be copied and furnished to=0A= others, and derivative works that comment on or otherwise explain it or=0A= assist in its implementation may be prepared, copied, published and=0A= distributed, in whole or in part, without restriction of any kind,=0A= provided that the above copyright notice and this paragraph are included=0A= on all such copies and derivative works. However, this document itself=0A= may not be modified in any way, such as by removing the copyright notice=0A= or references to the Internet Society or other Internet organizations,=0A= except as needed for the purpose of developing Internet standards in=0A= which case the procedures for copyrights defined in the Internet=0A= languages other than English.=0A= =0A= The limited permissions granted above are perpetual and will not be=0A= revoked by the Internet Society or its successors or assigns.=0A= =0A= This document and the information contained herein is provided on an "AS=0A= IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING TASK=0A= FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT=0A= LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT=0A= INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR=0A= FITNESS FOR A PARTICULAR PURPOSE."=0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= Luby/Goyal Section 12. [Page 27]=0A= =0C=0A= INTERNET-DRAFT Expires: September 2002 March 2002=0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= Luby/Goyal Section 12. [Page 28]=0A= ------=_NextPart_000_0000_01C1C306.34596B20 Content-Type: text/plain; name="WEBRC 00 -- 01 changes.txt" Content-Disposition: attachment; filename="WEBRC 00 -- 01 changes.txt" Content-Transfer-Encoding: quoted-printable WEBRC draft 00 --> 01 changes: (1) Changed to conform with the "Building Block" format specified in the = Author's Guidelines. - Environmental Requirements and Considerations beefed up. - Security Considerations beefed up. - Requirements from other Building Blocks section added - Applicability added - Other sections broken down into Functionality and Rationale (2) Tagged info that must appear in packet as Congestion Control = Information (CCI) (3) Specified two formats for CCI, a short and long format (4) Specified that session description must be obtained by receivers, = specifed what info but said how is out of band and out of scope (5) References cleaned up (6) Cleaned up and added detail to the description of the protocol - rearranged section 3(.0) - packet sequence numbers are increasing - 3.1.1: - changed base channel rate from 8kbps to 1 packet per second - rephrased "P" - corrected expression for N - rearranged section 3.2(.0) - 3.2.2.1: put in details on the weight - 3.2.2.2: put in details on ARTT - 3.2.2.5: put in details on ARR_P, TRR_P - 3.2.2.6: made SSMINR_P a recommended value - 3.2.3.2: Changed an off-by-one inconsistency between draft and {figures,ns code}. - 3.2.3.4: Recommend the handling of misordered packets. - 3.2.1 first paragraph: capitalized some MUST, SHOULD, etc. ------=_NextPart_000_0000_01C1C306.34596B20-- _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Tue Mar 5 21:44:05 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA11498 for ; Tue, 5 Mar 2002 21:44:05 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id VAA25229 for rmt-archive@odin.ietf.org; Tue, 5 Mar 2002 21:44:07 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id VAA23953; Tue, 5 Mar 2002 21:20:51 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id VAA23926 for ; Tue, 5 Mar 2002 21:20:49 -0500 (EST) Received: from cisco.com (cypher.cisco.com [171.69.11.18]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA10044 for ; Tue, 5 Mar 2002 21:20:46 -0500 (EST) Received: (from lorenzo@localhost) by cisco.com (8.8.8/2.6/Cisco List Logging/8.8.8) id SAA09260 for rmt@ietf.org; Tue, 5 Mar 2002 18:20:18 -0800 (PST) Date: Tue, 5 Mar 2002 18:20:18 -0800 From: Lorenzo Vicisano To: rmt@ietf.org Message-ID: <20020305182018.H14888@cisco.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.3.23i Subject: [Rmt] call for agenda items Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org As usual we have two slots for the RMT meeting: MONDAY, March 18, 2002 1415-1515 Afternoon Sessions II Salon E/F TSV rmt Reliable Multicast Transport WG THURSDAY, March 21, 2002 0900-1130 Morning Sessions Salon B TSV rmt Reliable Multicast Transport WG Please submit requests for agenda items ASAP. Usual recommendations about presentation content and length apply. thank you, the chairs. _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Tue Mar 5 22:52:34 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA14665 for ; Tue, 5 Mar 2002 22:52:33 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id WAA28926 for rmt-archive@odin.ietf.org; Tue, 5 Mar 2002 22:52:36 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id WAA28507; Tue, 5 Mar 2002 22:44:00 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id WAA28473 for ; Tue, 5 Mar 2002 22:43:58 -0500 (EST) Received: from mx.webfountain.com (mx.webfountain.com [63.161.54.39]) by ietf.org (8.9.1a/8.9.1a) with SMTP id WAA14427 for ; Tue, 5 Mar 2002 22:43:53 -0500 (EST) Received: (qmail 20732 invoked from network); 6 Mar 2002 03:43:25 -0000 Received: from mail.intranet (10.1.1.37) by mx.webfountain.com with SMTP; 6 Mar 2002 03:43:25 -0000 Received: by mail.intranet (8.12.1/8.12.1/Debian -5) with SMTP id g263hglk029917; Tue, 5 Mar 2002 19:43:42 -0800 From: "Michael Luby" To: "Lorenzo Vicisano" , Cc: "Michael Luby" Subject: RE: [Rmt] call for agenda items Date: Tue, 5 Mar 2002 19:44:21 -0800 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4807.1700 Importance: Normal In-Reply-To: <20020305182018.H14888@cisco.com> Content-Transfer-Encoding: 7bit Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org Content-Transfer-Encoding: 7bit LCT BB, FEC BB, FEC INFO and ALC PI (5-20 minutes) - 5 minutes if this is just a summary of what stage they are in the process of becoming Experimental/Informational RFCs. - This could be expanded to a 20 minute presentation if it is desired to go through the deltas between these versions and the previous versions. - Setting up the required IANA registration for FEC BB and how this works WEBRC BB issues (15-20 minutes) A discussion of the potential weaknesses and whether or not there should be some testing to get around them might be appropriate. -RP and MSDP issues and workarounds -Authentication issues and how for example TESLA could fit in to solve this -IGMP and PIM control message processing scalability issue --Perhaps a working group proposal of what the test environment should be to ensure that everyone is comfortable with the scalability of processing these control messages, i.e., different vendors could test their routers/switches for this and publish the results as one of the metrics that is important with respect to multicast deployment. -----Original Message----- From: rmt-admin@ietf.org [mailto:rmt-admin@ietf.org]On Behalf Of Lorenzo Vicisano Sent: Tuesday, March 05, 2002 6:20 PM To: rmt@ietf.org Subject: [Rmt] call for agenda items As usual we have two slots for the RMT meeting: MONDAY, March 18, 2002 1415-1515 Afternoon Sessions II Salon E/F TSV rmt Reliable Multicast Transport WG THURSDAY, March 21, 2002 0900-1130 Morning Sessions Salon B TSV rmt Reliable Multicast Transport WG Please submit requests for agenda items ASAP. Usual recommendations about presentation content and length apply. thank you, the chairs. _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Wed Mar 6 11:27:56 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA16176 for ; Wed, 6 Mar 2002 11:27:56 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id LAA21652 for rmt-archive@odin.ietf.org; Wed, 6 Mar 2002 11:28:00 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id KAA18769; Wed, 6 Mar 2002 10:58:27 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id KAA18739 for ; Wed, 6 Mar 2002 10:58:24 -0500 (EST) Received: from ebene.inrialpes.fr (ebene.inrialpes.fr [194.199.18.70]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA14402 for ; Wed, 6 Mar 2002 10:58:19 -0500 (EST) Received: from pop-serv.inrialpes.fr (pop-serv.inrialpes.fr [194.199.18.66]) by ebene.inrialpes.fr (8.11.6/8.11.6) with ESMTP id g26FvKF22836; Wed, 6 Mar 2002 16:57:21 +0100 (MET) Received: from inrialpes.fr (IDENT:roca@iseran.inrialpes.fr [194.199.24.100]) by pop-serv.inrialpes.fr (8.11.3/8.11.3/ImagV2) with ESMTP id g26FvKM10542; Wed, 6 Mar 2002 16:57:23 +0100 (MET) Message-ID: <3C863C61.12B60299@inrialpes.fr> Date: Wed, 06 Mar 2002 16:57:21 +0100 From: Vincent Roca Organization: INRIA X-Mailer: Mozilla 4.79 [en] (X11; U; Linux 2.2.14-5.0 i686) X-Accept-Language: en MIME-Version: 1.0 To: Michael Luby CC: vincent.roca@inrialpes.fr, rmt@ietf.org, Lorenzo Vicisano , Roger Kermode Subject: Re: [Rmt] I-D ACTION:draft-ietf-rmt-pi-alc-06.txt,.ps References: Content-Type: multipart/mixed; boundary="------------E0C82134CF79534F938A01DE" Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org This is a multi-part message in MIME format. --------------E0C82134CF79534F938A01DE Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Mike and al. Here are some comments on the ALC draft-06. Beyond the few comments made, I really like it! cheers, vincent ---------------------- http://www.inrialpes.fr/planete/people/roca/ INRIA Rhone-Alpes - projet planete vincent.roca@inrialpes.fr Zirst; 655 av. de l'Europe; Montbonnot phone (+33) 4.76.61.52.16 38334 ST ISMIER cedex - France fax (+33) 4.76.61.52.52 --------------E0C82134CF79534F938A01DE Content-Type: text/plain; charset=us-ascii; name="comments_alc_draft_02mar06.txt" Content-Disposition: inline; filename="comments_alc_draft_02mar06.txt" Content-Transfer-Encoding: 7bit Comments on draft-ietf-rmt-pi-alc-06.txt ---------------------------------------- Main comments ------------- * section 2.1, p 10: A session is now identified by the (sender IP addr, TSI). It's fine as long as the sender's IP addr of an incoming packet (e.g. as obtained with a recvfrom() syscall) is actually that of the sender. But it's a problem if the sender's site uses NAT (the sender does not easily know in advance its IP address). It will also be broken if a SIMPLISTIC reflector is used (e.g. to easily provide multicast connectivity to hosts that do not have access to ASM/SSM routing). The ALC draft may relax a little bit this constraint by saying that the TSI can be the result of a hash between the sender's IP address and a unique local session identifier. In that case, if we accept the small probability that two sessions from two different senders use the same TSI, the session is only identified by the TSI. So there could be two different modes: - (sender IP addr, TSI) - TSI only The sender chooses one and announces it in the session description. * section 4.4, p28: It is said that when authentication is used, then it should be checked immediately, before step 1 (header parsing). I think on the contrary that authentication should be checked just before step 3 (CCI processing) since: - it requires going through the LCT header, which is not possible until this header has been parsed/checked - step 2 (TSI check) will automatically (and efficiently!) discard all packets that are not part of the session * section 2.4, p 12: I think it's a bit restrictive to require that all the objects be known in advance with multi-object sessions (this is assumed when saying that the set of TOIs and the object length must be communicated to receivers before joining the session). In our implementation this info is communicated in-band dynamically. It gives more flexibility to application designers and increases the ALC field of application. The constraint here is to have a means to communicate this information in-band. We do it with a "private" ALC header extension that indicates the set of TOIs in the session (once it is known). * section 2, p9 vs. section 4.1, p21 There is a confusion. It is said that ALC assumes there is a single sender (p9) and later (p21) "the object that is being transmitted from several senders concurrently" (maybe this latter sentence only applies to LCT?) Anyway why should ALC be restricted to a single source ? Because it can be used with SSM ? Because having several sources sending to the same channel creates problems and inefficiencies in the congestion control building block of each receiver ? Maybe we can say that ALCv1 is primarily designed for sessions with a single sender, and that the case of multiple senders contributing to the same session may be possible but leads to problems that are not addressed in this document. Suggestions / clarification needed ---------------------------------- * section 2.1, p10: Why is it said that each TOI "should be globally unique across all sessions" ? As long as a session is uniquely identified, then (sender addr, TSI, TOI) is unique, no matter if TOI is the same or not. In some implementations the TOI is a counter, not a hash! * Is TSI == 0 a valid TSI? I think yes (and the draft does not forbid it). In that case having a TSI field in each packet wastes 32 bit. So I'd replace the sentence: "For ALC the length of the TSI field is required to be non-zero" by: "For ALC the length of the TSI field is required to be non-zero if TSI > 0. If the 'S' flag is 0 (no TSI field), it is assumed that TSI == 0." * Same kind of remark for TOI. As TOI = 0 is valid for sessions with several objects, we can assume that when the 'O' flag is 0 (no TOI field), TOI == 0. * A suggestion I already made some time ago: Change the "source block number" (of the FPI) to "source block ID" as its goal is to identify the source block. "Source block number" may be confused with the number of source blocks" (even if it is clarified in the text). * section 4.4, p28: Why should a receiver experiencing several authentication failures leave the session ? Why not just receiving the base layer as suggested above ? In most cases if a receiver experiences an authentication failure, there is a very high probability that more failures will take place during some time... Typos / minor ------------- * p4, 2nd paragraph: "the goal is to deliver" instead of "is to delivery" * p4, bottom: "protocol" instead of "protcol" * p18, ERT paragraph: replace '/' in "session / object" by "or" which is more explicit * p12, 4th bullet: "the address and port number used..." instead of "addressES" and "numberS" * p27, middle: replace "A receiver may be concurrently joined to..." by: "A receiver may concurrently join..." * p28, middle: add a final 's' to "check" in sentence "...that the rx immediately checkS the authenticity..." * p28, bottom: it is said that a MD5 check be made before delivery to an application. For the point of view of ALC, whatever is above ALC is the application. So I'd say: "...It is therefore recommended that an application built on top of ALC make an integrity check on received content, e.g. ..." --------------E0C82134CF79534F938A01DE-- _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Wed Mar 6 12:36:32 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA20914 for ; Wed, 6 Mar 2002 12:36:32 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id MAA29141 for rmt-archive@odin.ietf.org; Wed, 6 Mar 2002 12:36:36 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA26958; Wed, 6 Mar 2002 12:17:31 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id MAA26930 for ; Wed, 6 Mar 2002 12:17:29 -0500 (EST) Received: from cisco.com (cypher.cisco.com [171.69.11.18]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA19557 for ; Wed, 6 Mar 2002 12:17:24 -0500 (EST) Received: (from lorenzo@localhost) by cisco.com (8.8.8/2.6/Cisco List Logging/8.8.8) id JAA02975 for rmt@ietf.org; Wed, 6 Mar 2002 09:16:58 -0800 (PST) Date: Wed, 6 Mar 2002 09:16:58 -0800 From: Lorenzo Vicisano To: rmt@ietf.org Subject: Re: [Rmt] call for agenda items Message-ID: <20020306091658.B1690@cisco.com> References: <20020305182018.H14888@cisco.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20020305182018.H14888@cisco.com> User-Agent: Mutt/1.3.23i Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org correction: the 1st slot is on *TUESDAY*, not Monday. Sorry for the confusion, thanks, Lorenzo On Tue, Mar 05, 2002 at 06:20:18PM -0800, Lorenzo Vicisano wrote: > As usual we have two slots for the RMT meeting: > > MONDAY, March 18, 2002 > 1415-1515 Afternoon Sessions II > Salon E/F TSV rmt Reliable Multicast Transport WG > > THURSDAY, March 21, 2002 > 0900-1130 Morning Sessions > Salon B TSV rmt Reliable Multicast Transport WG > > Please submit requests for agenda items ASAP. > Usual recommendations about presentation content and > length apply. > > thank you, > the chairs. > > _______________________________________________ > Rmt mailing list > Rmt@ietf.org > https://www1.ietf.org/mailman/listinfo/rmt _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Wed Mar 6 15:46:32 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA04706 for ; Wed, 6 Mar 2002 15:46:31 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id PAA17038 for rmt-archive@odin.ietf.org; Wed, 6 Mar 2002 15:46:34 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA16244; Wed, 6 Mar 2002 15:37:21 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA16208 for ; Wed, 6 Mar 2002 15:37:19 -0500 (EST) Received: from mx.webfountain.com (mx.webfountain.com [63.161.54.39]) by ietf.org (8.9.1a/8.9.1a) with SMTP id PAA04121 for ; Wed, 6 Mar 2002 15:37:14 -0500 (EST) Received: (qmail 25698 invoked from network); 6 Mar 2002 20:36:46 -0000 Received: from mail.intranet (10.1.1.37) by mx.webfountain.com with SMTP; 6 Mar 2002 20:36:46 -0000 Received: by mail.intranet (8.12.1/8.12.1/Debian -5) with SMTP id g26Kb5lk015297; Wed, 6 Mar 2002 12:37:05 -0800 From: "Michael Luby" To: Cc: , "Lorenzo Vicisano" , "Roger Kermode" , "Michael Luby" Subject: RE: [Rmt] I-D ACTION:draft-ietf-rmt-pi-alc-06.txt,.ps Date: Wed, 6 Mar 2002 12:38:38 -0800 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) Importance: Normal In-Reply-To: <3C863C61.12B60299@inrialpes.fr> X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4807.1700 Content-Transfer-Encoding: 7bit Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org Content-Transfer-Encoding: 7bit Vincent, I've copies your attached comments below and have interspersed my replies (starting with ***) into your comments. Mike -----Original Message----- From: Vincent.Roca@inrialpes.fr [mailto:Vincent.Roca@inrialpes.fr] Sent: Wednesday, March 06, 2002 7:57 AM To: Michael Luby Cc: vincent.roca@inrialpes.fr; rmt@ietf.org; Lorenzo Vicisano; Roger Kermode Subject: Re: [Rmt] I-D ACTION:draft-ietf-rmt-pi-alc-06.txt,.ps Mike and al. Here are some comments on the ALC draft-06. Beyond the few comments made, I really like it! cheers, vincent ---------------------- http://www.inrialpes.fr/planete/people/roca/ INRIA Rhone-Alpes - projet planete vincent.roca@inrialpes.fr Zirst; 655 av. de l'Europe; Montbonnot phone (+33) 4.76.61.52.16 38334 ST ISMIER cedex - France fax (+33) 4.76.61.52.52 Comments on draft-ietf-rmt-pi-alc-06.txt ---------------------------------------- Main comments ------------- * section 2.1, p 10: A session is now identified by the (sender IP addr, TSI). It's fine as long as the sender's IP addr of an incoming packet (e.g. as obtained with a recvfrom() syscall) is actually that of the sender. But it's a problem if the sender's site uses NAT (the sender does not easily know in advance its IP address). It will also be broken if a SIMPLISTIC reflector is used (e.g. to easily provide multicast connectivity to hosts that do not have access to ASM/SSM routing). The ALC draft may relax a little bit this constraint by saying that the TSI can be the result of a hash between the sender's IP address and a unique local session identifier. In that case, if we accept the small probability that two sessions from two different senders use the same TSI, the session is only identified by the TSI. So there could be two different modes: - (sender IP addr, TSI) - TSI only The sender chooses one and announces it in the session description. *** The idea of having the session scoped by the sender IP address was an idea that came out of an informal meeting with several people back in the fall at ICSI, and the consensus was that the session should be scoped by the sender IP address. This has been in the ALC draft since version 3, so I'm surprised that this is coming up now. If we want to revisit this decision, it will have to be taken up in the entire RMT working group, because this not only affects ALC, but also GRA and several other drafts within the group. I personally hate the idea of using an unscoped TSI as a global identifier, because this brings problems similar to the problems with ASM and group address allocation, and having the group address scoped by the sender IP address as part of the channel address is exactly what SSM does, and this is a model that there is consensus on as having a shot of working. * section 4.4, p28: It is said that when authentication is used, then it should be checked immediately, before step 1 (header parsing). I think on the contrary that authentication should be checked just before step 3 (CCI processing) since: - it requires going through the LCT header, which is not possible until this header has been parsed/checked - step 2 (TSI check) will automatically (and efficiently!) discard all packets that are not part of the session *** Ok * section 2.4, p 12: I think it's a bit restrictive to require that all the objects be known in advance with multi-object sessions (this is assumed when saying that the set of TOIs and the object length must be communicated to receivers before joining the session). In our implementation this info is communicated in-band dynamically. It gives more flexibility to application designers and increases the ALC field of application. The constraint here is to have a means to communicate this information in-band. We do it with a "private" ALC header extension that indicates the set of TOIs in the session (once it is known). *** Ok, we can relax this to say that the object lengths must be either communicated out of band or in band using a Header Extension, but that the mechanism for doing this is outside the scope of the ALC document. A solution that to me seems cleaner in the long run than using private Header Extensions would be to define a new FEC Encoding ID that extends FEC Encoding ID 128 to also include the object length. Are you up for writing a document that would do this as specified in the FEC building block draft? In either case, I will relax the requirement so that object lengths can be passed either out of band or in band through header extensions. * section 2, p9 vs. section 4.1, p21 There is a confusion. It is said that ALC assumes there is a single sender (p9) and later (p21) "the object that is being transmitted from several senders concurrently" (maybe this latter sentence only applies to LCT?) Anyway why should ALC be restricted to a single source ? Because it can be used with SSM ? Because having several sources sending to the same channel creates problems and inefficiencies in the congestion control building block of each receiver ? *** ALC is explicitly NOT restricted to a single source, BUT an ALC session IS restricted to a single source. Again, this has been in the draft for several versions now, so I'm surprised that it is coming up again now. The reason that a session is restricted to a single source is that because it makes no sense to perform congestion control on channels that are coming from geographically disparate sources. ALC explicitly allows a receiver to be joined to multiple sessions concurrently, and the same object (or objects) to be delivered in several session concurrently, i.e., the receiver can be joined to multiple sessions, BUT the congestion control MUST be performed on each session separately. That is the intent and I believe that is what it says. Maybe we can say that ALCv1 is primarily designed for sessions with a single sender, and that the case of multiple senders contributing to the same session may be possible but leads to problems that are not addressed in this document. *** I think all versions of ALC should be designed to work with sessions that are defined to be from a single sender, but the receiver is allowed to be joined to multiple sessions from different senders and this should be explicitly addressed at the application level. I think anything else is getting back to the problems found in ASM, and away from the advantages of a model that is like SSM. It is an application level issue to coordinate the different sessions, and if you try to do it at the transport level the same problems as come up with ASM arise again. Suggestions / clarification needed ---------------------------------- * section 2.1, p10: Why is it said that each TOI "should be globally unique across all sessions" ? As long as a session is uniquely identified, then (sender addr, TSI, TOI) is unique, no matter if TOI is the same or not. In some implementations the TOI is a counter, not a hash! *** Good point. It is strong enough to say that the TOI should be unique for each object within the session. I'll change it to this. The reason it was stated this way is for the case when a receiver joins multiple sessions to obtain packets pertaining to the same object, to not get confused about which packets pertain to which object, but this mapping between objects and TOIs can be handled at the application level and can be different for the same object between different sessions. * Is TSI == 0 a valid TSI? I think yes (and the draft does not forbid it). In that case having a TSI field in each packet wastes 32 bit. So I'd replace the sentence: "For ALC the length of the TSI field is required to be non-zero" by: "For ALC the length of the TSI field is required to be non-zero if TSI > 0. If the 'S' flag is 0 (no TSI field), it is assumed that TSI == 0." *** The TSI field is REQUIRED by LCT, and ALC uses this. LCT allows the flexibility of using for example the Source Port Number as the TSI (and this has been in the LCT draft for several versions), but LCT does REQUIRE that the TSI be carried in each data packet. What ALC does is to specify that the required TSI is explicitly carred in the LCT portion of the ALC header. The TSI does not have to be 32-bits, for example it can be 16-bits if you use a TOI that is also 16-bits. This is achieved by the setting S=0, O=0 and H=1. The requirement is that the length of the TSI and TOI is a multiple of 32-bits, but each can be 16-bits for example. What is disallowed is for S=0 and H=0, i.e., a zero length TSI. * Same kind of remark for TOI. As TOI = 0 is valid for sessions with several objects, we can assume that when the 'O' flag is 0 (no TOI field), TOI == 0. *** The TOI field can be omitted by setting O=0 and H=0. I don't see any difference between omitting the TOI and setting TOI = 0, so why change this? You can within an application of course set a variable TOI=0 in the case that the TOI is omitted, but this seems quite outside the scope of ALC. * A suggestion I already made some time ago: Change the "source block number" (of the FPI) to "source block ID" as its goal is to identify the source block. "Source block number" may be confused with the number of source blocks" (even if it is clarified in the text). *** This is a comment on the FEC building block, not on ALC. We already went over this and I don't see any compelling reason to change this. If we do change it, then the FEC building block (which is already through last call) would have to be modified, which I'm not in favor of doing. * section 4.4, p28: Why should a receiver experiencing several authentication failures leave the session ? Why not just receiving the base layer as suggested above ? In most cases if a receiver experiences an authentication failure, there is a very high probability that more failures will take place during some time... *** Ok Typos / minor ------------- * p4, 2nd paragraph: "the goal is to deliver" instead of "is to delivery" * p4, bottom: "protocol" instead of "protcol" * p18, ERT paragraph: replace '/' in "session / object" by "or" which is more explicit * p12, 4th bullet: "the address and port number used..." instead of "addressES" and "numberS" * p27, middle: replace "A receiver may be concurrently joined to..." by: "A receiver may concurrently join..." * p28, middle: add a final 's' to "check" in sentence "...that the rx immediately checkS the authenticity..." * p28, bottom: it is said that a MD5 check be made before delivery to an application. For the point of view of ALC, whatever is above ALC is the application. So I'd say: "...It is therefore recommended that an application built on top of ALC make an integrity check on received content, e.g. ..." *** All ok _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Wed Mar 6 16:53:10 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA08736 for ; Wed, 6 Mar 2002 16:53:10 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id QAA21810 for rmt-archive@odin.ietf.org; Wed, 6 Mar 2002 16:53:13 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id QAA19666; Wed, 6 Mar 2002 16:25:27 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id NAA04245 for ; Wed, 6 Mar 2002 13:41:58 -0500 (EST) Received: from CNRI.Reston.VA.US (localhost [127.0.0.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA26095; Wed, 6 Mar 2002 13:41:54 -0500 (EST) Message-Id: <200203061841.NAA26095@ietf.org> Mime-Version: 1.0 Content-Type: Multipart/Mixed; Boundary="NextPart" To: IETF-Announce: ; Cc: rmt@ietf.org From: Internet-Drafts@ietf.org Reply-to: Internet-Drafts@ietf.org Date: Wed, 06 Mar 2002 13:41:54 -0500 Subject: [Rmt] I-D ACTION:draft-ietf-rmt-pi-norm-04.txt Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org --NextPart A New Internet-Draft is available from the on-line Internet-Drafts directories. This draft is a work item of the Reliable Multicast Transport Working Group of the IETF. Title : NACK-Oriented Reliable Multicast Protocol (NORM) Author(s) : B. Adamson, C. Bormann, M. Handley, J. Macker Filename : draft-ietf-rmt-pi-norm-04.txt Pages : 43 Date : 05-Mar-02 This document describes the messages and procedures of the Nega- tive-acknowledgement (NACK) oriented reliable multicast (NORM) pro- tocol. This protocol is designed to provide end-to-end reliable transport of bulk data objects or streams over generic IP multicast routing and forwarding services. A URL for this Internet-Draft is: http://www.ietf.org/internet-drafts/draft-ietf-rmt-pi-norm-04.txt To remove yourself from the IETF Announcement list, send a message to ietf-announce-request with the word unsubscribe in the body of the message. Internet-Drafts are also available by anonymous FTP. Login with the username "anonymous" and a password of your e-mail address. After logging in, type "cd internet-drafts" and then "get draft-ietf-rmt-pi-norm-04.txt". A list of Internet-Drafts directories can be found in http://www.ietf.org/shadow.html or ftp://ftp.ietf.org/ietf/1shadow-sites.txt Internet-Drafts can also be obtained by e-mail. Send a message to: mailserv@ietf.org. In the body type: "FILE /internet-drafts/draft-ietf-rmt-pi-norm-04.txt". NOTE: The mail server at ietf.org can return the document in MIME-encoded form by using the "mpack" utility. To use this feature, insert the command "ENCODING mime" before the "FILE" command. To decode the response(s), you will need "munpack" or a MIME-compliant mail reader. Different MIME-compliant mail readers exhibit different behavior, especially when dealing with "multipart" MIME messages (i.e. documents which have been split up into multiple messages), so check your local documentation on how to manipulate these messages. 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: Multipart/Alternative; Boundary="OtherAccess" --OtherAccess Content-Type: Message/External-body; access-type="mail-server"; server="mailserv@ietf.org" Content-Type: text/plain Content-ID: <20020305135203.I-D@ietf.org> ENCODING mime FILE /internet-drafts/draft-ietf-rmt-pi-norm-04.txt --OtherAccess Content-Type: Message/External-body; name="draft-ietf-rmt-pi-norm-04.txt"; site="ftp.ietf.org"; access-type="anon-ftp"; directory="internet-drafts" Content-Type: text/plain Content-ID: <20020305135203.I-D@ietf.org> --OtherAccess-- --NextPart-- _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Thu Mar 7 11:15:04 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA03661 for ; Thu, 7 Mar 2002 11:15:03 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id LAA07214 for rmt-archive@odin.ietf.org; Thu, 7 Mar 2002 11:15:07 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA04752; Thu, 7 Mar 2002 11:00:47 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA04711 for ; Thu, 7 Mar 2002 11:00:44 -0500 (EST) Received: from ebene.inrialpes.fr (ebene.inrialpes.fr [194.199.18.70]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA02557 for ; Thu, 7 Mar 2002 11:00:39 -0500 (EST) Received: from pop-serv.inrialpes.fr (pop-serv.inrialpes.fr [194.199.18.66]) by ebene.inrialpes.fr (8.11.6/8.11.6) with ESMTP id g27FxrF22554; Thu, 7 Mar 2002 16:59:53 +0100 (MET) Received: from inrialpes.fr (IDENT:roca@iseran.inrialpes.fr [194.199.24.100]) by pop-serv.inrialpes.fr (8.11.3/8.11.3/ImagV2) with ESMTP id g27FxrM05260; Thu, 7 Mar 2002 16:59:54 +0100 (MET) Message-ID: <3C878E79.8B600078@inrialpes.fr> Date: Thu, 07 Mar 2002 16:59:53 +0100 From: Vincent Roca Organization: INRIA X-Mailer: Mozilla 4.79 [en] (X11; U; Linux 2.2.14-5.0 i686) X-Accept-Language: en MIME-Version: 1.0 To: Michael Luby CC: vincent.roca@inrialpes.fr, rmt@ietf.org, Lorenzo Vicisano , Roger Kermode Subject: Re: [Rmt] I-D ACTION:draft-ietf-rmt-pi-alc-06.txt,.ps References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Transfer-Encoding: 7bit Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org Content-Transfer-Encoding: 7bit Mike, here are some answers... > Main comments > ------------- > [...] > *** The idea of having the session scoped by the sender IP address was an > idea that came out of an informal meeting with several people back in the > fall at ICSI, and the consensus was that the session should be scoped by the > sender IP address. Ok if there is a consensus around it... Yet I assume that some implementations will offer the possibility to remove this check as it may create problems in some cases. You're right, it was already in LCT-v2 draft... At that time it didn't strike me. > * section 2.4, p 12: > I think it's a bit restrictive to require that all the objects be > known in advance with multi-object sessions (this is assumed when > saying that the set of TOIs and the object length must be > communicated to receivers before joining the session). > In our implementation this info is communicated in-band dynamically. > It gives more flexibility to application designers and increases > the ALC field of application. > The constraint here is to have a means to communicate this > information in-band. We do it with a "private" ALC header extension > that indicates the set of TOIs in the session (once it is known). > > *** Ok, we can relax this to say that the object lengths must be either > communicated out of band or in band using a Header Extension, but that the > mechanism for doing this is outside the scope of the ALC document. A > solution that to me seems cleaner in the long run than using private Header > Extensions would be to define a new FEC Encoding ID that extends FEC > Encoding ID 128 to also include the object length. Are you up for writing a > document that would do this as specified in the FEC building block draft? > In either case, I will relax the requirement so that object lengths can be > passed either out of band or in band through header extensions. Oups, from your answer I see that there was an ambiguity in my original comment. Sorry! The object length is not the only thing I was concerned about (anyway the FEC Object Transmission Info (FTI) associated with FEC encoding ID 128 already enables the sender to communicate this info in-band dynamically). But ok to relax this item too. My concern was more general: objects may not be known at all in advance, nor their number and consequently the associated set of TOIs. So I'd also relax the 7th bullet of section 2.4, p 12: "If the sessions carries packets for more than one object...". To summarize, I'd say the 3 following possibilities are acceptable: 1- the set of TOIs can be communicated out-of-band before a receiver joins the session 2- the set of TOIs can be communicated dynamically by an out-of-band mechanism that is out of the scope of the ALC draft 3- the set of TOIs can be communicated dynamically by an in-band mechanism. We are using solution 3- and we currently do that with a private ALC Header Extension (but we can change it). I'm not sure that specifying a new FEC Encoding ID for that is appropriate as this need is more related to the notion of "ALC session" than to that of FEC BB. Yes, I can write a small paragraph for that once everybody agrees. [...] > *** ALC is explicitly NOT restricted to a single source, BUT an ALC session > IS restricted to a single source. Ok, I fully agree with you! Let's use meta-ALC-sessions when there are multiple contributing senders. Maybe we can add a sentence clarifying that point in the document. > Suggestions / clarification needed > ---------------------------------- > > > * Is TSI == 0 a valid TSI? I think yes (and the draft does not > forbid it). In that case having a TSI field in each packet wastes > 32 bit. > So I'd replace the sentence: > "For ALC the length of the TSI field is required to be non-zero" > by: > "For ALC the length of the TSI field is required to be non-zero > if TSI > 0. If the 'S' flag is 0 (no TSI field), it is assumed that > TSI == 0." > > *** The TSI field is REQUIRED by LCT, and ALC uses this. LCT allows the > flexibility of using for example the Source Port Number as the TSI (and this > has been in the LCT draft for several versions), but LCT does REQUIRE that > the TSI be carried in each data packet. What ALC does is to specify that > the required TSI is explicitly carred in the LCT portion of the ALC header. > The TSI does not have to be 32-bits, for example it can be 16-bits if you > use a TOI that is also 16-bits. This is achieved by the setting S=0, O=0 > and H=1. The requirement is that the length of the TSI and TOI is a multiple > of 32-bits, but each can be 16-bits for example. What is disallowed is for > S=0 and H=0, i.e., a zero length TSI. S=0 and H=0 (zero length TSI) can be synonymous to TSI==0 in ALC which is a valid TSI. In that case there is no need to reserve any room in the ALC header to carry a TSI equal to 0. It does not contradict the fact that ALC requires a session identifier (there is one, 0), unless we say that a session identifier must be > 0. If TSI==0 is valid, we can assume that many ALC implementations will use this value (at least by default). > * Same kind of remark for TOI. > As TOI = 0 is valid for sessions with several objects, we can > assume that when the 'O' flag is 0 (no TOI field), TOI == 0. > > *** The TOI field can be omitted by setting O=0 and H=0. I don't see any > difference between omitting the TOI and setting TOI = 0, so why change this? > You can within an application of course set a variable TOI=0 in the case > that the TOI is omitted, but this seems quite outside the scope of ALC. The ALC draft says that "the TOI field must be used in all packets if more than one object is to be tx in a session" (p21). I don't agree. It can be said in the draft that O=0 and H=0 is synonymous to TOI == 0 and that this is valid even when more than one object is to be transmitted in a session. It can save some room in the ALC header (especially if TSI==0 too). vincent ---------------------- http://www.inrialpes.fr/planete/people/roca/ INRIA Rhone-Alpes - projet planete vincent.roca@inrialpes.fr Zirst; 655 av. de l'Europe; Montbonnot phone (+33) 4.76.61.52.16 38334 ST ISMIER cedex - France fax (+33) 4.76.61.52.52 _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Thu Mar 7 15:47:59 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA23185 for ; Thu, 7 Mar 2002 15:47:54 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id PAA27848 for rmt-archive@odin.ietf.org; Thu, 7 Mar 2002 15:47:57 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA26366; Thu, 7 Mar 2002 15:24:19 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id PAA26331 for ; Thu, 7 Mar 2002 15:24:16 -0500 (EST) Received: from mx.webfountain.com (mx.webfountain.com [63.161.54.39]) by ietf.org (8.9.1a/8.9.1a) with SMTP id PAA21307 for ; Thu, 7 Mar 2002 15:24:05 -0500 (EST) Received: (qmail 32636 invoked from network); 7 Mar 2002 20:23:38 -0000 Received: from mail.intranet (10.1.1.37) by mx.webfountain.com with SMTP; 7 Mar 2002 20:23:38 -0000 Received: by mail.intranet (8.12.1/8.12.1/Debian -5) with SMTP id g27KNwNV011760; Thu, 7 Mar 2002 12:23:58 -0800 From: "Michael Luby" To: "Vincent Roca" Cc: , "Lorenzo Vicisano" , "Roger Kermode" , "Michael Luby" Subject: Luby reply to Roca comments on [Rmt] I-D ACTION:draft-ietf-rmt-pi-alc-06.txt,.ps Date: Thu, 7 Mar 2002 12:25:33 -0800 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4807.1700 In-Reply-To: <3C878E79.8B600078@inrialpes.fr> Content-Transfer-Encoding: 7bit Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org Content-Transfer-Encoding: 7bit Vincent, More comments below. Mike -----Original Message----- From: rmt-admin@ietf.org [mailto:rmt-admin@ietf.org]On Behalf Of Vincent Roca Sent: Thursday, March 07, 2002 8:00 AM To: Michael Luby Cc: vincent.roca@inrialpes.fr; rmt@ietf.org; Lorenzo Vicisano; Roger Kermode Subject: Re: [Rmt] I-D ACTION:draft-ietf-rmt-pi-alc-06.txt,.ps Mike, here are some answers... > Main comments > ------------- > [...] > *** The idea of having the session scoped by the sender IP address was an > idea that came out of an informal meeting with several people back in the > fall at ICSI, and the consensus was that the session should be scoped by the > sender IP address. Ok if there is a consensus around it... Yet I assume that some implementations will offer the possibility to remove this check as it may create problems in some cases. You're right, it was already in LCT-v2 draft... At that time it didn't strike me. *** So I think we agree at this point that it can remain as it is. > * section 2.4, p 12: > I think it's a bit restrictive to require that all the objects be > known in advance with multi-object sessions (this is assumed when > saying that the set of TOIs and the object length must be > communicated to receivers before joining the session). > In our implementation this info is communicated in-band dynamically. > It gives more flexibility to application designers and increases > the ALC field of application. > The constraint here is to have a means to communicate this > information in-band. We do it with a "private" ALC header extension > that indicates the set of TOIs in the session (once it is known). > > *** Ok, we can relax this to say that the object lengths must be either > communicated out of band or in band using a Header Extension, but that the > mechanism for doing this is outside the scope of the ALC document. A > solution that to me seems cleaner in the long run than using private Header > Extensions would be to define a new FEC Encoding ID that extends FEC > Encoding ID 128 to also include the object length. Are you up for writing a > document that would do this as specified in the FEC building block draft? > In either case, I will relax the requirement so that object lengths can be > passed either out of band or in band through header extensions. Oups, from your answer I see that there was an ambiguity in my original comment. Sorry! The object length is not the only thing I was concerned about (anyway the FEC Object Transmission Info (FTI) associated with FEC encoding ID 128 already enables the sender to communicate this info in-band dynamically). But ok to relax this item too. My concern was more general: objects may not be known at all in advance, nor their number and consequently the associated set of TOIs. So I'd also relax the 7th bullet of section 2.4, p 12: "If the sessions carries packets for more than one object...". To summarize, I'd say the 3 following possibilities are acceptable: 1- the set of TOIs can be communicated out-of-band before a receiver joins the session 2- the set of TOIs can be communicated dynamically by an out-of-band mechanism that is out of the scope of the ALC draft 3- the set of TOIs can be communicated dynamically by an in-band mechanism. We are using solution 3- and we currently do that with a private ALC Header Extension (but we can change it). I'm not sure that specifying a new FEC Encoding ID for that is appropriate as this need is more related to the notion of "ALC session" than to that of FEC BB. Yes, I can write a small paragraph for that once everybody agrees. *** This seems ok to me. I suggest you write the paragraph. [...] > *** ALC is explicitly NOT restricted to a single source, BUT an ALC session > IS restricted to a single source. Ok, I fully agree with you! Let's use meta-ALC-sessions when there are multiple contributing senders. Maybe we can add a sentence clarifying that point in the document. *** I think there are several sentences that already make this point. If you want to suggest another sentence and where to place it, please do that. > Suggestions / clarification needed > ---------------------------------- > > > * Is TSI == 0 a valid TSI? I think yes (and the draft does not > forbid it). In that case having a TSI field in each packet wastes > 32 bit. > So I'd replace the sentence: > "For ALC the length of the TSI field is required to be non-zero" > by: > "For ALC the length of the TSI field is required to be non-zero > if TSI > 0. If the 'S' flag is 0 (no TSI field), it is assumed that > TSI == 0." > > *** The TSI field is REQUIRED by LCT, and ALC uses this. LCT allows the > flexibility of using for example the Source Port Number as the TSI (and this > has been in the LCT draft for several versions), but LCT does REQUIRE that > the TSI be carried in each data packet. What ALC does is to specify that > the required TSI is explicitly carred in the LCT portion of the ALC header. > The TSI does not have to be 32-bits, for example it can be 16-bits if you > use a TOI that is also 16-bits. This is achieved by the setting S=0, O=0 > and H=1. The requirement is that the length of the TSI and TOI is a multiple > of 32-bits, but each can be 16-bits for example. What is disallowed is for > S=0 and H=0, i.e., a zero length TSI. S=0 and H=0 (zero length TSI) can be synonymous to TSI==0 in ALC which is a valid TSI. In that case there is no need to reserve any room in the ALC header to carry a TSI equal to 0. It does not contradict the fact that ALC requires a session identifier (there is one, 0), unless we say that a session identifier must be > 0. If TSI==0 is valid, we can assume that many ALC implementations will use this value (at least by default). *** I think we need to discuss this point in the RMT working group IETF meeting (are you going to be there?). One philosophical point that I believe we may differ on that I want to call out is the following. A building block is in general going to leave open a lot of possibilities and be more open-ended then a protocol instantiation, because the building block is meant to be useful for a multitude of protocol instantiations. This means in particular that a protocol instantiation may restrict the use of any building block beyond what is stated in the building block spec. Thus, even though the LCT building block allows S=0 and H=0 as a valid setting and thus the TSI is not carried in the LCT header, but nevertheless LCT does explicitly require a TSI associated with the session that uniquely identifies it within the scope of the sender IP address. The current version of the ALC draft further restricts its use of LCT by explicitly requiring the TSI field to be in the LCT header. This prevents the problem of confusion when a receiver is subscribed to multiple ALC sessions from the same sender. Perhaps what we could do is to say that the TSI field is explicitly REQUIRED if the sender sources more than one ALC session, but that if the sender sources only one ALC session then the TSI field MAY be omitted and it is then implicitly defined to be zero. > * Same kind of remark for TOI. > As TOI = 0 is valid for sessions with several objects, we can > assume that when the 'O' flag is 0 (no TOI field), TOI == 0. > > *** The TOI field can be omitted by setting O=0 and H=0. I don't see any > difference between omitting the TOI and setting TOI = 0, so why change this? > You can within an application of course set a variable TOI=0 in the case > that the TOI is omitted, but this seems quite outside the scope of ALC. The ALC draft says that "the TOI field must be used in all packets if more than one object is to be tx in a session" (p21). I don't agree. It can be said in the draft that O=0 and H=0 is synonymous to TOI == 0 and that this is valid even when more than one object is to be transmitted in a session. It can save some room in the ALC header (especially if TSI==0 too). *** There are a couple of interpretations to what you say: (1) One of the objects is allowed to not have the TOI in the header of packets and then it is defined to be TOI == 0, but then all other objects in the session MUST have a TOI that is non-zero and all unique. This seems like a really kludgy hack that I really don't like. This means that you could never explicitly use TOI = 0 as the setting of the field as otherwise this would be confused with packets that carry no TOI field, you are only saving header space for packets associated with one of multiple objects, and it is overall confusing. (2) None of the objects in the session are required to have a TOI, and then some other means must be used by the receiver to distinguish one object from another within the session. This defeats the very purpose of the TOI, which is to allow the receiver to distinguish between different objects in the same session. Thus, with either of the two interpretations of your comments that I can come up with (and maybe you meant something else), I don't see the logic of your suggestion. Again, we can resolve this at the RMT working group meeting at the IETF. vincent ---------------------- http://www.inrialpes.fr/planete/people/roca/ INRIA Rhone-Alpes - projet planete vincent.roca@inrialpes.fr Zirst; 655 av. de l'Europe; Montbonnot phone (+33) 4.76.61.52.16 38334 ST ISMIER cedex - France fax (+33) 4.76.61.52.52 _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Fri Mar 8 00:38:14 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA14760 for ; Fri, 8 Mar 2002 00:38:14 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id AAA22411 for rmt-archive@odin.ietf.org; Fri, 8 Mar 2002 00:38:14 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id AAA22209; Fri, 8 Mar 2002 00:32:42 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id AAA22180 for ; Fri, 8 Mar 2002 00:32:40 -0500 (EST) Received: from prv-mail20.provo.novell.com (prv-mail20.provo.novell.com [137.65.81.122]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA14639 for ; Fri, 8 Mar 2002 00:32:39 -0500 (EST) Received: from INET-PRV-MTA by prv-mail20.provo.novell.com with Novell_GroupWise; Thu, 07 Mar 2002 22:31:39 -0700 Message-Id: X-Mailer: Novell GroupWise Internet Agent 6.0.1 Date: Thu, 07 Mar 2002 22:32:00 -0700 From: "Anumita Biswas" To: Subject: [RMT] PGM and network layer independence Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Disposition: inline Content-Transfer-Encoding: 7bit Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org Content-Transfer-Encoding: 7bit PGM does not appear to be completely network layer independent. The GSI is calculated based on the DNS name to have network layer independence. But the packet formats themselves are not IPv6 compliant. Even though there are fields like NLA AFI and NLA in the packets, the byte space allocated allows only for 4 bytes of NLA. What is the need then for an address family field ? And also it appears that "NLA" could as well be called as IP(v4) address! Probably, the issue here is even deeper... the fact that transport layer headers carry network layer information which is mandated by the very design of PGM protocol. Are my assumptions wrong here? Any comments? -AB _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Fri Mar 8 04:09:56 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA26451 for ; Fri, 8 Mar 2002 04:09:52 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id EAA13767 for rmt-archive@odin.ietf.org; Fri, 8 Mar 2002 04:09:53 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id DAA11895; Fri, 8 Mar 2002 03:49:19 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id DAA11858 for ; Fri, 8 Mar 2002 03:49:16 -0500 (EST) Received: from cisco.com (cypher.cisco.com [171.69.11.18]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA25939 for ; Fri, 8 Mar 2002 03:49:13 -0500 (EST) Received: (from lorenzo@localhost) by cisco.com (8.8.8/2.6/Cisco List Logging/8.8.8) id AAA25356; Fri, 8 Mar 2002 00:48:45 -0800 (PST) Date: Fri, 8 Mar 2002 00:48:44 -0800 From: Lorenzo Vicisano To: Anumita Biswas Cc: rmt@ietf.org Subject: Re: [RMT] PGM and network layer independence Message-ID: <20020308004844.A17973@cisco.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.3.23i Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org Anumita, On Thu, Mar 07, 2002 at 10:32:00PM -0700, Anumita Biswas wrote: > PGM does not appear to be completely network layer independent. The GSI > is calculated based on the DNS name to have network layer independence. > > > But the packet formats themselves are not IPv6 compliant. Even though > there are fields like NLA AFI and NLA in the packets, the byte space > allocated allows only for 4 bytes of NLA. What is the need then for an > address family field ? And also it appears that "NLA" could as well be > called as IP(v4) address! PGM does not specify NLAs to be 4 bytes. Their length is implicitly determined by the AFI. Hence all the headers that contain one or more NLAs have variable length. If any of the header specifications in rfc3208 assumes that NLA is for bytes, then that's a bug in the document (please let the authors know). > > Probably, the issue here is even deeper... the fact that transport > layer headers carry network layer information which is mandated by the > very design of PGM protocol. All transport protocols need to identify network-layer endpoints somehow. In PGM the choice was to communicate this information in-band and to use NLAs directly, but note that NLAs are opaque to the transport protocol. Finally, further discussion concerning PGM specifically should be carried out outside this mailing list. Thank you, Lorenzo > > Are my assumptions wrong here? Any comments? > > -AB > > > > _______________________________________________ > Rmt mailing list > Rmt@ietf.org > https://www1.ietf.org/mailman/listinfo/rmt _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Sat Mar 9 15:19:37 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA15914 for ; Sat, 9 Mar 2002 15:19:36 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id PAA09619 for rmt-archive@odin.ietf.org; Sat, 9 Mar 2002 15:18:31 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA08236; Sat, 9 Mar 2002 14:50:58 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA08205 for ; Sat, 9 Mar 2002 14:50:56 -0500 (EST) Received: from postal1.lbl.gov (postal1.lbl.gov [128.3.7.82]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA15421 for ; Sat, 9 Mar 2002 14:50:51 -0500 (EST) Received: from SpamWall.lbl.gov (localhost [127.0.0.1]) by postal1.lbl.gov (8.11.2/8.11.2) with ESMTP id g29Jor304398 for ; Sat, 9 Mar 2002 11:50:53 -0800 (PST) Received: from letters.cs.ucsb.edu (letters.cs.ucsb.edu [128.111.41.13]) by SpamWall.lbl.gov (8.11.2/8.11.2) with ESMTP id g29Jori04395 for ; Sat, 9 Mar 2002 11:50:53 -0800 (PST) Received: from cab.cs.ucsb.edu (cab [128.111.49.120]) by letters.cs.ucsb.edu (8.11.6+Sun/8.9.3) with ESMTP id g29Jopq11275 for ; Sat, 9 Mar 2002 11:50:52 -0800 (PST) From: "Kevin C. Almeroth" Received: (from almeroth@localhost) by cab.cs.ucsb.edu (8.11.6+Sun/8.11.6) id g29JopC06369 for rmt@lbl.gov; Sat, 9 Mar 2002 11:50:51 -0800 (PST) Date: Sat, 9 Mar 2002 11:50:51 -0800 (PST) Message-Id: <200203091950.g29JopC06369@cab.cs.ucsb.edu> To: rmt@lbl.gov Subject: [Rmt] MMNS 2002 CFP (one month and counting) Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org IFIP/IEEE International Conference on Management of Multimedia Networks and Services MMNS 2002 Managing IP Multimedia End-To-End October 6-9, 2002 Santa Barbara, CA USA http://www.cs.ucsb.edu/conferences/MMNS02 CALL FOR PAPERS With the popularity of the IP-based Internet, there has been a major shift towards building a worldwide, all-service and all-media network based on the next-generation of IP technology. Rich multimedia is increasingly being incorporated into Internet applications (like Voice-over-IP, streaming video, etc.). However, more work is needed to evolve the Internet and the underlying networking technologies into an infrastructure capable of supporting all-service and all-media traffic. Significant management functionality is required to ensure that these applications have a network that is both capable and available. Management requirements may ensue from the fact that multimedia traffic may traverse networks using a wide variety of technology and each network may be owned, operated, and controlled by a different organization. In order to provide end-to-end, high quality multimedia services, the network must provide robust management functionality. The IFIP/IEEE International Conference on Management of Multimedia Networks and Services will hold its fifth annual meeting Oct. 6 through Oct. 9, 2002 in Santa Barbara, CA. The IFIP/IEEE MMNS is a single-track conference and provides an intimate setting for discussion and debate. The program committee is soliciting original papers describing research in the area of management of multimedia networks and services. Topics of interest include, but are not limited to, the following: o End-to-end IP multimedia network and service management o VoIP service management o IP Video, streaming, interactive video service management o Provisioning of multimedia networks and services o Wireless and mobile multimedia network management o Optical multimedia network management o Cable multimedia network management o Multimedia network traffic engineering and optimization o Distributed multimedia service management o Network management models and architectures o Billing and security for multi-media services o Resource, performance and fault management o Multi-point, multicast services management o Network programmability for multimedia services o Policy-based management for multimedia services o QoS management o Multimedia traffic management o Multimedia content protection o Deployment of multimedia services o Active multimedia network management o Multimedia session management o Middleware support for management PAPER SUBMISSION Papers must be submitted electronically in postscript or PDF format. Detailed instructions are provided on the conference web site, http://www.cs.ucsb.edu/conferences/MMNS02. Paper length is limited to 20 double-spaced pages. The paper's cover page must include: title of paper, authors' names and affiliations, contact author's name and address (both postal and electronic), a short abstract, keywords, and submission area (from the list of relevant topics of interest). For any further information, please refer to the conference home page or direct your questions to almeroth@cs.ucsb.edu. IMPORTANT DATES Submission deadline: April 7, 2002 Notification of acceptance: June 30, 2002 Final version: July 28, 2002 Conference: October 6-9, 2002 _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Tue Mar 12 09:44:19 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA23565 for ; Tue, 12 Mar 2002 09:44:14 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id JAA15882 for rmt-archive@odin.ietf.org; Tue, 12 Mar 2002 09:44:16 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id JAA15428; Tue, 12 Mar 2002 09:36:25 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id JAA15401 for ; Tue, 12 Mar 2002 09:36:23 -0500 (EST) Received: from ebene.inrialpes.fr (ebene.inrialpes.fr [194.199.18.70]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA23302 for ; Tue, 12 Mar 2002 09:36:20 -0500 (EST) Received: from pop-serv.inrialpes.fr (pop-serv.inrialpes.fr [194.199.18.66]) by ebene.inrialpes.fr (8.11.6/8.11.6) with ESMTP id g2CEZVG28003; Tue, 12 Mar 2002 15:35:32 +0100 (MET) Received: from inrialpes.fr (IDENT:roca@iseran.inrialpes.fr [194.199.24.100]) by pop-serv.inrialpes.fr (8.11.3/8.11.3/ImagV2) with ESMTP id g2CEZWM19465; Tue, 12 Mar 2002 15:35:33 +0100 (MET) Message-ID: <3C8E1236.410B0636@inrialpes.fr> Date: Tue, 12 Mar 2002 15:35:34 +0100 From: Vincent Roca Organization: INRIA X-Mailer: Mozilla 4.79 [en] (X11; U; Linux 2.2.14-5.0 i686) X-Accept-Language: en MIME-Version: 1.0 To: Michael Luby CC: vincent.roca@inrialpes.fr, rmt@ietf.org, Lorenzo Vicisano , Roger Kermode Subject: Re: Luby reply to Roca comments on [Rmt] I-D ACTION:draft-ietf-rmt-pi-alc-06.txt,.ps References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Transfer-Encoding: 7bit Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org Content-Transfer-Encoding: 7bit Mike As you agree on the idea of adding in-band support to carry some object information, here are the two paragraphs that I suggest to add to the alc-06 document. cheers, vincent nb: please correct the english. I did my best but... ---------------------- http://www.inrialpes.fr/planete/people/roca/ INRIA Rhone-Alpes - projet planete vincent.roca@inrialpes.fr Zirst; 655 av. de l'Europe; Montbonnot phone (+33) 4.76.61.52.16 38334 ST ISMIER cedex - France fax (+33) 4.76.61.52.52 <<< Vincent: add this paragraph in section 2.4, p12, after removing the 3 items mentioned below from the initial list >>> Sometimes the objects that will be sent in a session are completely known beforehand. They can be communicated to the receivers, before they join the ALC session, in the session description which, in that case, includes three additional items: o An indication of whether or not the session carries packets for more than one object; o If the session carries packets for more than one object, the set of Transport Object IDs (TOIs) for the objects in the session. o The total length of the objects in the session in bytes. Sometimes the objects are not be known at all when the ALC session starts. Their number, their length and the associated set of TOIs will only be determined during the ALC session lifetime, for instance when the application built on top of ALC indicates that no more objects will be submitted. In such cases several possibilities are offered: o the set of TOIs can be communicated dynamically by an out-of-band mechanism that is out of the scope of this document. o the set of TOIs can be communicated dynamically, in-band, by an ALC header extension that is defined in section 4.2 of this document. In this case the format of this header extension MUST be communicated to receivers in the initial session description before they join the ALC session. o the total length of each object can be communicated dynamically by an out-of-band mechanism that is out of the scope of this document. o the total length of each object can be communicated dynamically, in-band, by an ALC header extension that is defined in section 4.2 of this document. In this case the format of this header extension MUST be communicated to receivers in the initial session description before they join the ALC session. <<< Vincent: add this paragraph in section 4.2, p24, after removing "For this version of ALC, there are no PI-specific extensions" >>> This version of ALC defines two PI-specific header extensions: EXT_FTI=64 FEC Object Transmission Information (FTI) extension The purpose of this extension is to carry in-band the FTI information of an object. Of course the FEC Encoding ID and FEC Encoding Name directly impact this extension format. Therefore a receiver is REQUIRED to obtain the FTI header extension format in the session description before joining the ALC session. EXT_STOI=65 (variable-length version) EXT_STOI=192 (32-bit version) Set of Transport Object Identifiers (TOIs) extension The purpose of this extension is to carry in-band the set of TOIs of the objects sent in the session. Since the nature of a TOI is implementation specific, this document does not specify the format of the STOI extension. A receiver is REQUIRED to obtain the STOI header extension format in the session description before joining the ALC session. For instance, when the TOI is a counter ranging from 0 for the first object to max_toi for the last object, and when max_toi can be encoded in a 24 bit field, a possible STOI extension format is the following: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | HET=192 (STOI)| max_toi in the session | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fig. 4 - Example of STOI header extension This document does not specify when a STOI extension should be included in a packet. In particular, transmissions being unreliable, several packets (for instance all the packets sent on the base layer) may include it. Sessions using the On-demand model may also require that a STOI extension be included in many packets. These considerations are out of the scope of this document. _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Tue Mar 12 11:51:03 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA27025 for ; Tue, 12 Mar 2002 11:51:02 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id LAA23894 for rmt-archive@odin.ietf.org; Tue, 12 Mar 2002 11:51:06 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA22868; Tue, 12 Mar 2002 11:33:12 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA22836 for ; Tue, 12 Mar 2002 11:33:10 -0500 (EST) Received: from ebene.inrialpes.fr (ebene.inrialpes.fr [194.199.18.70]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA26570 for ; Tue, 12 Mar 2002 11:33:06 -0500 (EST) Received: from pop-serv.inrialpes.fr (pop-serv.inrialpes.fr [194.199.18.66]) by ebene.inrialpes.fr (8.11.6/8.11.6) with ESMTP id g2CGWIG02318; Tue, 12 Mar 2002 17:32:18 +0100 (MET) Received: from inrialpes.fr (IDENT:roca@iseran.inrialpes.fr [194.199.24.100]) by pop-serv.inrialpes.fr (8.11.3/8.11.3/ImagV2) with ESMTP id g2CGWHM07232; Tue, 12 Mar 2002 17:32:20 +0100 (MET) Message-ID: <3C8E2D93.66DFC066@inrialpes.fr> Date: Tue, 12 Mar 2002 17:32:19 +0100 From: Vincent Roca Organization: INRIA X-Mailer: Mozilla 4.79 [en] (X11; U; Linux 2.2.14-5.0 i686) X-Accept-Language: en MIME-Version: 1.0 To: Michael Luby CC: vincent.roca@inrialpes.fr, rmt@ietf.org, Lorenzo Vicisano , Roger Kermode Subject: Re: Luby reply to Roca comments on [Rmt] I-D ACTION:draft-ietf-rmt-pi-alc-06.txt,.ps References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Transfer-Encoding: 7bit Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org Content-Transfer-Encoding: 7bit Mike Here is the 2nd part. vincent ---------------------- http://www.inrialpes.fr/planete/people/roca/ INRIA Rhone-Alpes - projet planete vincent.roca@inrialpes.fr Zirst; 655 av. de l'Europe; Montbonnot phone (+33) 4.76.61.52.16 38334 ST ISMIER cedex - France fax (+33) 4.76.61.52.52 > > *** ALC is explicitly NOT restricted to a single source, BUT an ALC > session > > IS restricted to a single source. > > Ok, I fully agree with you! Let's use meta-ALC-sessions when there > are multiple contributing senders. > Maybe we can add a sentence clarifying that point in the document. > > *** I think there are several sentences that already make this point. If > you want to suggest another sentence and where to place it, please do that. I went through the document once again but didn't find any reference to it, maybe I missed it! Anyway, let's add at the end of the 2nd paragraph of section 2, p9: The fact that an ALC session is restricted to a single sender does not preclude the possibility of having multiple contributing senders for the same set of objects, probably by defining multiple ALC sessions in an application built on top of ALC. How these meta-sessions are managed is an application specific issue that is out of the scope of this document." > S=0 and H=0 (zero length TSI) can be synonymous to TSI==0 in ALC which > is a valid TSI. In that case there is no need to reserve any room in > the ALC header to carry a TSI equal to 0. > It does not contradict the fact that ALC requires a session identifier > (there is one, 0), unless we say that a session identifier must > be > 0. > > If TSI==0 is valid, we can assume that many ALC implementations will > use this value (at least by default). > > *** I think we need to discuss this point in the RMT working group IETF > meeting (are you going to be there?). One philosophical point that I > believe we may differ on that I want to call out is the following. A > building block is in general going to leave open a lot of possibilities and > be more open-ended then a protocol instantiation, because the building block > is meant to be useful for a multitude of protocol instantiations. This > means in particular that a protocol instantiation may restrict the use of > any building block beyond what is stated in the building block spec. Thus, > even though the LCT building block allows S=0 and H=0 as a valid setting and > thus the TSI is not carried in the LCT header, but nevertheless LCT does > explicitly require a TSI associated with the session that uniquely > identifies it within the scope of the sender IP address. The current > version of the ALC draft further restricts its use of LCT by explicitly > requiring the TSI field to be in the LCT header. This prevents the problem > of confusion when a receiver is subscribed to multiple ALC sessions from the > same sender. Perhaps what we could do is to say that the TSI field is > explicitly REQUIRED if the sender sources more than one ALC session, but > that if the sender sources only one ALC session then the TSI field MAY be > omitted and it is then implicitly defined to be zero. I fully agree with everything you said. In particular I fully agree with the difference in openess between a BB and a PI. My point was just to know if TSI==0 was or not valid and what to do when TSI==0... I think your last sentence is ok. > The ALC draft says that "the TOI field must be used in all packets if > more than one object is to be tx in a session" (p21). I don't agree. > It can be said in the draft that O=0 and H=0 is synonymous to TOI == 0 > and that this is valid even when more than one object is to be > transmitted in a session. > It can save some room in the ALC header (especially if TSI==0 too). > > *** There are a couple of interpretations to what you say: > (1) One of the objects is allowed to not have the TOI in the header of > packets and then it is defined to be TOI == 0, but then all other objects in > the session MUST have a TOI that is non-zero and all unique. This seems > like a really kludgy hack that I really don't like. This means that you > could never explicitly use TOI = 0 as the setting of the field as otherwise > this would be confused with packets that carry no TOI field, you are only > saving header space for packets associated with one of multiple objects, and > it is overall confusing. > (2) None of the objects in the session are required to have a TOI, and then > some other means must be used by the receiver to distinguish one object from > another within the session. This defeats the very purpose of the TOI, which > is to allow the receiver to distinguish between different objects in the > same session. > Thus, with either of the two interpretations of your comments that I can > come up with (and maybe you meant something else), I don't see the logic of > your suggestion. Again, we can resolve this at the RMT working group > meeting at the IETF. Interpretation (1) is what I meant. The idea behind it is "do not send something that is not usefull". If you agree with this idea, then IMHO everything becomes natural. The parsing function at the receiver is straightforward: if (header->O_field == 0 and header->H_field == 0 in the incoming packet) TOI = 0; else TOI = header->toi; and we don't care anymore in the rest of the packet processing code wether the TOI field was or not present in the incoming packet. If the sender doesn't like it, he may add a 0'ed TOI field, and the final result will be the same for the receiver. But why should he do it ? A simple (artificial?) example: you have 2 objects, a big file (TOI==0) and an second small one (TOI==1) containing some meta-data on the file. Removing the TOI field of the first object may save some bytes (depending on the H flag in fact). And we can do exactly the same with the TSI field, when the sender wants to use TSI==0, which, as you explained (and I fully agree), may not be a good choice, unless the sender knows it will never source more than one ALC session. _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Tue Mar 12 15:01:53 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA04529 for ; Tue, 12 Mar 2002 15:01:52 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id PAA06728 for rmt-archive@odin.ietf.org; Tue, 12 Mar 2002 15:01:56 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA05644; Tue, 12 Mar 2002 14:55:14 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id OAA05617 for ; Tue, 12 Mar 2002 14:55:13 -0500 (EST) Received: from cisco.com (cypher.cisco.com [171.69.11.18]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA04173 for ; Tue, 12 Mar 2002 14:55:09 -0500 (EST) Received: (from lorenzo@localhost) by cisco.com (8.8.8/2.6/Cisco List Logging/8.8.8) id LAA28119; Tue, 12 Mar 2002 11:54:41 -0800 (PST) Date: Tue, 12 Mar 2002 11:54:41 -0800 From: Lorenzo Vicisano To: rmt@ietf.org Cc: roger.kermode@motorola.com Message-ID: <20020312115440.B20150@cisco.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.3.23i Subject: [Rmt] RMT draft agenda Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org This is the agenda so far. We will submit this in the next few hours, please let us know if you have any last-minute change. thanks, the chairs. Agenda Bashing Roger Kermode/Lorenzo Vicisano 5 mins NORM status updates Brian Adamson 15 mins draft-ietf-rmt-bb-norm-03.txt draft-ietf-rmt-pi-norm-04.txt LCT BB, FEC BB, FEC INFO and ALC PI updates Michael Luby 20 mins draft-ietf-rmt-bb-lct-04.txt draft-ietf-rmt-bb-fec-06.txt draft-ietf-rmt-info-fec-02.txt draft-ietf-rmt-pi-alc-06.txt WEBRC BB issues Michael Luby 15 mins _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Fri Mar 22 11:26:40 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA25229 for ; Fri, 22 Mar 2002 11:26:40 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id LAA13734 for rmt-archive@odin.ietf.org; Fri, 22 Mar 2002 11:26:42 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA12690; Fri, 22 Mar 2002 11:17:33 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id LAA12491 for ; Fri, 22 Mar 2002 11:17:24 -0500 (EST) Received: from postal1.lbl.gov (postal1.lbl.gov [128.3.7.82]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA23597 for ; Fri, 22 Mar 2002 10:23:01 -0500 (EST) Received: from SpamWall.lbl.gov (localhost [127.0.0.1]) by postal1.lbl.gov (8.11.2/8.11.2) with ESMTP id g2MFN3308832 for ; Fri, 22 Mar 2002 07:23:03 -0800 (PST) Received: from letters.cs.ucsb.edu (letters.cs.ucsb.edu [128.111.41.13]) by SpamWall.lbl.gov (8.11.2/8.11.2) with ESMTP id g2MFN3i08829 for ; Fri, 22 Mar 2002 07:23:03 -0800 (PST) Received: from cab.cs.ucsb.edu (cab [128.111.49.120]) by letters.cs.ucsb.edu (8.11.6+Sun/8.9.3) with ESMTP id g2MFN1O27701 for ; Fri, 22 Mar 2002 07:23:01 -0800 (PST) From: "Kevin C. Almeroth" Received: (from almeroth@localhost) by cab.cs.ucsb.edu (8.11.6+Sun/8.11.6) id g2MFN1913202 for rmt@lbl.gov; Fri, 22 Mar 2002 07:23:01 -0800 (PST) Date: Fri, 22 Mar 2002 07:23:01 -0800 (PST) Message-Id: <200203221523.g2MFN1913202@cab.cs.ucsb.edu> To: rmt@lbl.gov Subject: [Rmt] NOSSDAV 2002 Call for PARTICIPATION Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org The 12th International Workshop on Network and Operating Systems Support for Digital Audio and Video (NOSSDAV 2002) Sponsored by ACM SIGMM May 12-14, 2002 Miami Beach, Florida http://www.nossdav.org/2002 CALL FOR PARTICIPATION Information ----------- Registration and information on local arrangements is now available via the NOSSDAV WWW page. Please note that early registration and hotel rooms are only available until APRIL 11, 2002. Program Schedule ---------------- Sunday, May 12, 2002 5:00 PM - 8:00 PM Reception at Wyndham Miami Beach Resort Monday, May 13, 2002 9:00 AM - 10:30 AM Keynote by Raj Yavatkar, Intel 10:30 AM - 11:00 AM 30 min break 11:00 AM - 12:30 PM Session 1: Network Issues for Video and Games 12:30 AM - 1:30 PM LUNCH 1:30 PM - 2:30 PM Session 2: Potporri 2:30 PM - 3:00 PM 30 min break 3:00 PM - 4:30 PM Session 3: VoIP 5:00 PM - 8:00 PM Social Event Tuesday, May 14, 2002 9:00 AM - 10:30 AM Session 4: Mobile and Wireless System 10:30 AM - 11:00 AM 30 min break 11:00 AM - 12:00 PM Session 5: Distribution Overlays 12:00 PM - 1:00 PM LUNCH 1:00 PM - 2:30 PM Session 6: Video Streaming I 2:30 PM - 3:00 PM 30 min break 3:00 PM - 4:00 PM Session 7: Video Streaming II 5:00 PM - 8:00 PM Joint Reception with IWQoS List of Accepted Papers ----------------------- 1a. Adaptive Rate Control for Streaming Stored Fine-Grained Scalable Video Philippe de Cuetos, Keith W. Ross 1b. Probabilistic Congestion Control for Non-Adaptable Flows Joerg Widmer, Martin Mauve, Jan Peter Damm 1c. On the Impact of Delay on Real-Time Multiplayer Games Lothar Pantel 2a. Characterizing Locality, Evolution, and Life Span of Accesses in Enterprise Media Server Workloads Lucy Cherkasova, Minaxi Gupta 2b. The Design of a Transport Protocol for On-Demand Graphical Rendering Albert Harris, Robin Kravets 3a. SIP-based conference control framework Henning Schulzrinne, Petri Koskelainen 3b. Impact of link failures on VoIP performance Catherine Boutremans, Gianluca Iannaccone, Christophe Diot 3c. Comparison and Optimization of Packet Loss Repair Methods on VoIP Perceived Quality under Bursty Loss Wenyu Jiang, Henning Schulzrinne 4a. Understanding Bandwidth Allocation Techniques for the First-Hop in Wireless Networks Lili Qiu, Victor Bahl, Atul Adya 4b. Application Level Hand-off Support for Mobile Media Transcoding Sessions Sumit Roy, Bo Shen, Vijay Sundaram, Raj Kumar 4c. Integration of Dynamic Voltage Scaling and Soft Real-Time Scheduling for Open Mobile Systems Wanghong Yuan, Klara Nahrstedt 5a. Broadcast Federation: An Application-layer Broadcast Internetwork Yatin Chawathe, Mukund Seshadri 5b. Topology-Aware Overlay Networks for Group Communication Minseok Kwon, Sonia Fahmy 6a. A Demand Adaptive and Locality Aware (DALA) Streaming Media Server Cluster Architecture Zihui Ge, Ping Ji, Prashant Shenoy 6b. Optimal Proxy Management for Multimedia Streaming in Content Distribution Networks Chitra Venkatramani, Olivier Verscheure, Pascal Frossard, Kang-Won Lee 6c. Selecting among Replicated Batching Video-on-Demand Servers Meng Guo, Mostafa Ammar, Ellen Zegura 7a. A Framework for Scalable and Inexpensive Video Delivery Min-You Wu, Min-You Wu, Su-Jun Ma 7b. Distributing Streaming Media Content Using Cooperative Networking Philip Chou, Venkata Padmanabhan, Kunwadee Sripanidkulchai, Helen J. Wang _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Fri Mar 22 17:34:46 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA03284 for ; Fri, 22 Mar 2002 17:34:46 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id RAA25107 for rmt-archive@odin.ietf.org; Fri, 22 Mar 2002 17:34:50 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id RAA23927; Fri, 22 Mar 2002 17:24:07 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id SAA24049 for ; Thu, 21 Mar 2002 18:59:13 -0500 (EST) Received: from mx.webfountain.com (mx.webfountain.com [63.161.54.39]) by ietf.org (8.9.1a/8.9.1a) with SMTP id SAA00611 for ; Thu, 21 Mar 2002 18:59:08 -0500 (EST) Received: (qmail 17350 invoked from network); 21 Mar 2002 23:58:36 -0000 Received: from mail.intranet (10.1.1.37) by mx.webfountain.com with SMTP; 21 Mar 2002 23:58:36 -0000 Received: by mail.intranet (8.12.1/8.12.1/Debian -5) with SMTP id g2LNxJNV018376; Thu, 21 Mar 2002 15:59:20 -0800 From: "Michael Luby" To: "Rmt@ietf. org" Cc: "Michael Luby" Date: Thu, 21 Mar 2002 16:01:12 -0800 Message-ID: MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_NextPart_000_0000_01C1D0F1.9F8788C0" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) X-MIMEOLE: Produced By Microsoft MimeOLE V5.50.4807.1700 Importance: Normal Subject: [Rmt] Summary of ALC PI presentation Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org This is a multi-part message in MIME format. ------=_NextPart_000_0000_01C1D0F1.9F8788C0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Dear RMTers, Attached below are the slides on the presentation I gave in the RMT working group (in both .ppt and .ps formats). I would like to move forward with revising ALC PI as quickly as possible (WEBRC 01 has already been resubmitted), so I'm summarizing below what I think the sense of the group was on each of the ALC PI issues: Slide 3: Point 1: No quibbles with the clarification that a session is from one sender. Point 2: The sense of the group was that the ALC PI should be left as it is in the 06 draft and not accept the proposal, i.e., the ALC PI document should specify that a TSI is explicitly required in each packet. This sentiment was expressed by a couple of people publicly in the meeting, and in private by a few more afterwards. Slide 4: There was consensus that the proposal for relaxing the requirements on communication of the FEC Object Information to receivers was ok. No dissension on this. Slide 5: Point 1: The sense of the group was that the ALC PI should be left as it is in the 06 draft and not accept the proposal, i.e., the ALC PI document should specify that a TOI must be carried for each object in any session that carries more than one object. A couple of people came to the microphone to express discomfort with not doing this, and one person came up to the microphone and pointed out that there is a whole working group that is devoted to saving a few bits here and there in packet headers, but it should not be the approach of the RMT. Again, some private comments supporting leaving the draft as it is after the meeting. Point 2: There were no objections to the proposal to explicitly limit the scope of the TOI to the session, i.e., the TOI is scoped by the (IP sender address, TSI) pair. I'm going to revise the ALC PI over the next couple of days and resubmit it, so please send any additional comments soon to the list if you can. Thanks, Mike ------=_NextPart_000_0000_01C1D0F1.9F8788C0 Content-Type: application/vnd.ms-powerpoint; name="IETF MAR 2002 overall.ppt" Content-Disposition: attachment; filename="IETF MAR 2002 overall.ppt" Content-Transfer-Encoding: base64 0M8R4KGxGuEAAAAAAAAAAAAAAAAAAAAAPgADAP7/CQAGAAAAAAAAAAAAAAABAAAAOgAAAAAAAAAA EAAAPAAAAAEAAAD+////AAAAADsAAAD///////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////+A eh/wYQAAAMl5db2A/xtJmNOYcZQQzAHwKAAAAAgAAAAIAAAAAQABAAAAAAAgAAAAiQsAAIkLAAAC AAAAAgAAAP///wAAAAAAcgAAAPMAAAA/AAAAJwAAAOQAAAD8AAAAzwAAAE4AAAAAAAAAAAAAAAAA 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 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADwDoAxI2AAAB AOkDKAAAAIAWAADgEAAA4BAAAIAWAAAFAAAACgAAAAIAAAADAAAAAQAAAAAAAAEPAPIDtAIAAC8A yA/qAAAAMADSDwQAAAACAAAAAAC6DygAAAAYIBwgCP8UMDv/W/8IMAowDDAOMBAw5f8E/yQAKABb AFwAewBi/+H/EAC6D6YAAAABMAIwDP8O//swGv8b/x//Af+bMJww/TD+MJ0wnjAFMPwwGSAdIAn/ FTA9/13/CTALMA0wDzARMLAAMCAyIDMgAyHg/wX/QTBDMEUwRzBJMGMwgzCFMIcwjjChMKMwpTCn MKkwwzDjMOUw5zDuMPUw9jAhACUAKQAsAC4AOgA7AD8AXQB9AGH/Y/9k/2X/Z/9o/2n/av9r/2z/ bf9u/2//cP+e/5//DwDVB+QAAAAAALcPRAAAAFQAaQBtAGUAcwAgAE4AZQB3ACAAUgBvAG0AYQBu AAAAjK66ACjYEgAQ2BIAZKcMMAgAAAAAAAAAKNgSACAmDDAAAAQAEAC3D0QAAABBAHIAaQBhAGwA AABOAGUAdwAgAFIAbwBtAGEAbgAAAIyuugAo2BIAENgSAGSnDDAIAAAAAAAAACjYEgAgJgwwAMwE IiAAtw9EAAAAVwBpAG4AZwBkAGkAbgBnAHMAAABSAG8AbQBhAG4AAACMrroAKNgSABDYEgBkpwww CAAAAAAAAAAo2BIAICYMMALMBAIAAKQPDAAAAIAAYQAAAAEA/////wAApQ8MAAAAAAAACC4AAAAC AAAAAACrDx4AAAD/HwAABQBAAgAAAAAAACABIAFAAkACYANgA4AEgAQAAKkPCgAAAAcAAAACAAkE AABAAKMPbgAAAAUA//0/AAAAIiAAAGQAAAAAAAAAZAAAAAAAAAAAAEACAAAAAAIAAAD//+8AAAAA AP///////xgAAAAAAQAAAAUAACABIAEAAAAAAAUAAEACQAIAAAAAAAUAAGADYAMAAAAAAAUAAIAE gAQAAAAADwALBN4RAAAPAADw1hEAAAAABvDIEAAAAmAIABgCAABoAAAAEAAAAAEAAAAHAAAABQAA AAQAAAAOAAAAAgAAAAAAAAADAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAA BAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAE AAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQA AAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAA AAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAA AAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAA AAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAA AAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAA AAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAA BAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAE AAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAJAAAAAAAAAAQA AAAAAAAABAAAAAAAAAAFAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAA AAAAAAAEAAAAAAAAAAUAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAA AAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAA AAAABQAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAA AAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAA AAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAA BAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAE AAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQA AAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAA AAAAAAAEAAAAAAAAABUAAAAAAAAABQAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAA AAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAcAAAAAAAAAIAAAAAAAAAAEAAAAAAAAAAUAAAAA AAAABAAAAAAAAAAFAAAAAAAAAAQAAAAAAAAABQAAAAAAAAAEAAAAAAAAAAQAAAAAAAAAIwAAAAAA AAAEAAAAAAAAAAcAAAAAAAAABwAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAA AAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAA BAAAAAAAAAAEAAAAXwAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAUAAAAAAAAAJQAAAAAAAAAE AAAAAAAAAAQAAAAAAAAAGwAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAOAAAAAAAAABEA AAAAAAAAGAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAA AAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAFAAAAAAAAAAQAAAAAAAAABQAAAAAAAAAEAAAA AAAAAAUAAAAAAAAABAAAAAAAAAAFAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAIAAAAA AAAACAAAAAAAAAACAAAAAAAAABQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAA AAAEAAAAAAAAAAYAAAAAAAAABAAAAAAAAAAEAAAAAAAAACYAAAAAAAAAKAAAAAAAAAAEAAAAAAAA AAQAAAAAAAAAEgAAAAAAAAAjAAAAAAAAACYAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAA HgAAAAAAAAAEAAAAAAAAACwAAAAAAAAAGgAAAAAAAAAtAAAAAAAAADQAAAAAAAAABAAAAAAAAAAE AAAAAAAAAAQAAAAAAAAACAAAAAAAAAAcAAAAAAAAAA4AAAAAAAAAGgAAAAAAAAATAAAAAAAAAAUA AAAAAAAABAAAAAAAAAAIAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAABQAAAAgAAAAAAAAABAAA AAAAAAAEAAAAAAAAAAQAAAAAAAAACAAAAAAAAAARAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAA AAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABQAAAAAAAAAEAAAAAAAAAAQAAAAA AAAAEQAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABQAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAA AAAEAAAAAAAAAAQAAAAAAAAABQAAAAAAAAAFAAAAAAAAAAUAAAAAAAAABQAAAAAAAAAFAAAAAAAA AAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAA BAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABQAAAAAAAAAEAAAAAAAAAAUAAAAAAAAABAAAAAAAAAAI AAAAAAAAAAUAAAAAAAAAAgAAAAAAAAAEAAAAAAAAABoAAAAAAAAABAAAAAAAAAAcAAAAAAAAAAQA AAAAAAAAJAAAAAAAAAAEAAAAAAAAABMAAAAAAAAABAAAAAAAAAAEAAAAAAAAABoAAAAAAAAABAAA AAAAAAARAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAFAAAAAAAAAAQAAAAAAAAAIQAAAAAAAAAEAAAA AAAAABAAAAAAAAAACwAAAAAAAAAhAAAAAAAAAAQAAAAAAAAAEAAAAAAAAAALAAAAAAAAAAsAAAAA AAAABQAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAA AAAEAAAAAAAAABAAAAAAAAAABQAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAA AAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAAHAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAA BAAAAAAAAABNAAAAAAAAACcAAAAAAAAAAgAAAAAAAAAEAAAAAAAAAAUAAAAAAAAAAwAAAAAAAAAf AAAAAAAAAAQAAAAAAAAAJAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAwAAAAQA AAAEAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABgAAAAAAAAAFAAAAAAAAAAUAAAAAAAAABAAA AAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAA AAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAA AAAABAAAAAAAAAAEAAAAAAAAACwAAAAAAAAABAAAAAAAAAAFAAAAAAAAAAQAAAAAAAAABAAAAAAA AAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAA AAQAAAAAAAAABAAAAAAAAAAPAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAA BAAAAAAAAAAGAAAABgAAAAUAAAAHAAAABAAAAAAAAAAHAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAE AAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQA AAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAA AAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAA AAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAA AAAABAAAAAAAAAAlAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAA AAAEAAAAAAAAAAQAAAAIAAAABAAAAAAAAAAEAAAACgAAAAQAAAAOAAAABAAAAAAAAAAEAAAABgAA AAQAAAAAAAAABAAAAAAAAAAEAAAACwAAAAYAAAAMAAAABAAAAAAAAAAwAAAAAAAAAAQAAAAAAAAA AwAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAAAwAAAAAAAAAE AAAAAAAAAFMAAAAAAAAABAAAAAAAAAADAAAAAAAAADEAAAAAAAAABAAAAAoAAAAKAAAAAAAAAAoA AAAAAAAABQAAAAAAAAAFAAAAAAAAADwAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAMAAAAAAAAABAAA AAAAAAAEAAAAAAAAABMAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAMAAAAAAAAADwAAAAAAAAAUAAAA AAAAAAQAAAAAAAAAFQAAAAAAAAAUAAAAAAAAAB0AAAAAAAAABAAAAAAAAABHAAAAAAAAAA8AAAAA AAAATAAAAAAAAABMAAAAAAAAAEwAAAAAAAAATAAAAAAAAABQAAAABwAAAEcAAAAAAAAAKgAAAAAA AAAMAAAAAAAAAC4AAAAAAAAALgAAAAAAAAAuAAAAAAAAACkAAAAAAAAAKgAAAAAAAAAqAAAAAAAA ACoAAAAAAAAAKgAAAAAAAAAqAAAAAAAAACAAAAAAAAAABAAAAAAAAABTAAAAAAAAAAQAAAAAAAAA AwAAAAAAAAAEAAAAAAAAAAMAAAAAAAAABAAAAAAAAAADAAAAAAAAAAQAAAACAAAABAAAAAAAAAAE AAAAAAAAAAQAAAAAAAAABAAAAAgAAAAEAAAAAgAAAAQAAAAPAAAABAAAABAAAAAEAAAAHwAB8CwA AAByAAfwJAAAAAcHyXl1vYD/G0mY05hxlBDMAf8AaQAAACoAAAAAAAAAAADJACMBC/CKAAAAgAEB AAAAgQEBAAAIgwEAAAAIhkEBAAAAh8EeAAAAvwEUABQAwAEBAAAIwQEAAAEAxAEAAAAAywE4YwAA zQEAAAAAzgEAAAAA0AEAAAAA0QEAAAAA1wECAAAA/wEIAAgAAQICAAAIPwIAAAIATABhAHIAZwBl ACAAYwBvAG4AZgBlAHQAdABpAAAAgAAa8SAAAAAA/wAAAMwAAABrYQDcAIEAcUQAADPMMwD/AAAA AIAAAEAAHvEQAAAA/////wMAAAj//////////x8A8A8cAAAAAADzAxQAAAAEAAAABAAAAAAAAAAA AACAAAAAAA8A0Ac4AgAAHwD/AxQAAAACAAAEDAAAAAAAAAAAAAAAAgAAAA8A+gNnAAAAAAD+AwMA AAAAAQAAAP0DNAAAAEwAAABkAAAATAAAAGQAAABkpwwwCAAAAAAAAAAc2BIAAAAAAAAAAABU/f// pv///wEAAABwAPsDCAAAAAAAAABwCAAAcAD7AwgAAAABAAAAQAsAAB8ABwQ8AAAAAAD9AzQAAAAy AAAAZAAAADIAAABkAAAAZz8MMNjbEgAAAAAAeK+6AAAAAAAAAAAAAAAAAAAAAAAAARIAHwAIBDwA AAAAAP0DNAAAAEIAAABkAAAAQgAAAGQAAABnPwww2NsSAAAAAAB4r7oAAAAAAAAAAAAAAAAAAAAA AAAAEgAfAPoDZwAAAAAA/gMDAAAAAAEAAAD9AzQAAABLAAAAZAAAAEsAAABkAAAAZKcMMAgAAAAA AAAAHNgSAAAAAAAAAAAAQv3//5YAAAAAAAAAcAD7AwgAAAAAAAAAQAsAAHAA+wMIAAAAAQAAAHAI AAAfABQEHAAAAAAAFQQUAAAAZv1LHwDKmjte+qEeAMqaOwACAAEfABMEPAAAAAAA/QM0AAAAZAAA AGQAAABkAAAAZAAAAEjYEgD+PQwwNNgSABQAAAAAAAAAAAAAAAAAAAAAAAAAAAESAA8AiBNGAAAA DwCKEz4AAAAAALoPDgAAAF8AXwBfAFAAUABUADkAAACLEyAAAAAvAMgPDAAAADAA0g8EAAAAgAAA AAAAehcEAAAAAgQAAD8A2Q8MAAAAAADaDwQAAAAAACUATwDZDwwAAAAAANoPBAAAAAAAPQAPAPAP DB4AAAAA8wMUAAAAVQEAAAAAAAACAAAABQEAAAAAAAAAAJ8PBAAAAAYAAAAAAKgPFgAAAEZFQyBC QgtGRUMgSU5GTwtMQ1QgQkIAAKEPIAAAABcAAAAAAAAAAAAWAAAAAAACACgAAQAAAAAAggACACgA EACfDwQAAAAFAAAAAACoD6QAAAANQWxsIHNlbnQgdG8gSUVTRyBmb3IgY29uc2lkZXJhdGlvbiBh cyBSRkNzDUZFQyBCQiBhbmQgTENUIEJCIGFzIEV4cGVyaW1lbnRhbA1GRUMgSU5GTyBhcyBJbmZv cm1hdGlvbmFsDShhbHRob3VnaCBmb3Igc29tZSByZWFzb24gc3RpbGwgbm90IGluIHRoZSBSRkMg RWRpdG9yIFF1ZXVlKQAAqg8aAAAAJwAAAAAAAAAFAAAAAQAAAAMAeQAAAAAAAAAAAPMDFAAAAIUB AAACAAAAAgAAAC0BAAAAAAAAAACfDwQAAAAGAAAAAACoDxQAAABBTEMgUEkgc3RhdHVzIHJlcG9y dAAAoQ8eAAAAFQAAAAAAAAAAABQAAAAAAAIAKAABAAAAAAACACAAEACfDwQAAAAFAAAAAACoD3EA AAANTS4gTHVieSwgRGlnaXRhbCBGb3VudGFpbg1KLiBHZW1tZWxsLCBNaWNyb3NvZnQNTC4gVmlj aXNhbm8sIENpc2NvDUwuIFJpenpvLCBBQ0lSSSBhbmQgVS4gUGlzYQ1KLiBDcm93Y3JvZnQsIFVD TAAAqg8+AAAAHgAAAAAAAAAHAAAAAQAAAAMADwAAAAAAAAAIAAAAAQAAAAMAJwAAAAAAAAAJAAAA AQAAAAMABgAAAAAAAAAAAPMDFAAAAPsBAAAAAAAAAgAAAJIBAAAAAAAAAACfDwQAAAAAAAAAAACo DxYAAABQcm9wb3NlZCBjaGFuZ2VzIHRvIDA2EACfDwQAAAABAAAAAACgD3IEAABDAGwAYQByAGkA ZgB5ACAAdABoAGEAdAAgAHMAZQBzAHMAaQBvAG4AIABpAHMAIABmAHIAbwBtACAAYQAgAHMAaQBu AGcAbABlACAAcwBlAG4AZABlAHIALAAgAGEAbgBkACAAdABoAGEAdAAgAHIAZQBjAGUAaQB2AGkA bgBnACAAZgByAG8AbQAgAG0AdQBsAHQAaQBwAGwAZQAgAHMAZQBuAGQAZQByAHMAIABpAHMAIABh AHQAIAB0AGgAZQAgAGEAcABwAGwAaQBjAGEAdABpAG8AbgAgAGwAZQB2AGUAbAAgAGEAYgBvAHYA ZQAgAEEATABDAA0AVAByAGEAbgBzAG0AaQBzAHMAaQBvAG4AIABTAGUAcwBzAGkAbwBuACAASQBE AA0AMAA2ADoADQBUAFMASQAgAGUAeABwAGwAaQBjAGkAdABsAHkAIAByAGUAcQB1AGkAcgBlAGQA IAB0AG8AIABiAGUAIABpAG4AIABlAGEAYwBoACAAcABhAGMAawBlAHQALgANAFAAcgBvAHAAbwBz AGUAZAA6AA0AVABTAEkAIABlAHgAcABsAGkAYwBpAHQAbAB5ACAAcgBlAHEAdQBpAHIAZQBkACAA dABvACAAYgBlACAAaQBuACAAZQBhAGMAaAAgAHAAYQBjAGsAZQB0ACAAVQBOAEwARQBTAFMAIAB0 AGgAZQAgAHMAZQBuAGQAZQByACAAaQBzACAAcwB1AHAAcABvAHIAdABpAG4AZwAgAG8AbgBsAHkA IABvAG4AZQAgAHMAZQBzAHMAaQBvAG4ALAAgAGkAbgAgAHcAaABpAGMAaAAgAGMAYQBzAGUAIABU AFMASQAgAE0AQQBZACAAYgBlACAAbwBtAGkAdAB0AGUAZAAgAGEAbgBkACAAaQBzACAAdABoAGUA bgAgAHAAcgBlAHMAdQBtAGUAZAAgAHQAbwAgAGIAZQAgADAALgANAFAAcgBvADoAIABTAGEAdgBl AHMAIABzAHAAYQBjAGUAIABpAG4AIABwAGEAYwBrAGUAdAAgAGgAZQBhAGQAZQByACAAaQBmACAA bwBuAGwAeQAgAG8AbgBlACAAcwBlAHMAcwBpAG8AbgAgAHAAZQByACAAcwBlAG4AZABlAHIADQBD AG8AbgBzADoAIABMAEMAVAAgAHIAZQBxAHUAaQByAGUAcwAgAFQAUwBJACAAaQBuACAAaABlAGEA ZABlAHIAIAATICAAaQBzACAAdQBzAGUAZgB1AGwAIABmAG8AcgAgAG4AZQB0AHcAbwByAGsAIABl AGwAZQBtAGUAbgB0AHMAIAB0AG8AIABjAGwAYQBzAHMAaQBmAHkAIABjAGgAYQBuAG4AZQBsAHMA IABhAGMAYwBvAHIAZABpAG4AZwAgAHQAbwAgAHMAZQBzAHMAaQBvAG4AIAATICAAaQBuAGMAbwBu AHMAaQBzAHQAZQBuAHQAIABhAGMAcgBvAHMAcwAgAHMAZQBuAGQAZQByAHMAAAChD7IAAACVAAAA AAAAEAAATgAEAAAAAQAAEAAATgAuAAAAAgAAEAAATgAKAAAAAQAAEAAATgBpAQAAAgAAEAAATgCV AAAAAAAAAAQAAAAABAAAAQQuAAAAAAQAAAEECgAAAAAIAAABCJ4AAAAADAAAAQwFAAAAAAwEAAEM AIAA/jwAAAAADAQAAQwAAAABBgAAAAAMBAABDP8AAP6DAAAAAAwEAAEMAAAAAQEAAAAADAQAAQz/ AAD+AADzAxQAAACtAQAAAAAAAAIAAABMAQAAAAAAAAAAnw8EAAAAAAAAAAAAqA8WAAAAUHJvcG9z ZWQgY2hhbmdlcyB0byAwNhAAnw8EAAAAAQAAAAAAoA8gAwAARgBFAEMAIABPAGIAagBlAGMAdAAg AFQAcgBhAG4AcwBtAGkAcwBzAGkAbwBuACAASQBuAGYAbwAuACAAYwBvAG0AbQB1AG4AaQBjAGEA dABpAG8AbgANADAANgA6AA0ATQB1AHMAdAAgAGIAZQAgAGMAbwBtAG0AdQBuAGkAYwBhAHQAZQBk ACAAbwB1AHQALQBvAGYALQBiAGEAbgBkACAAYgBlAGYAbwByAGUAIAByAGUAYwBlAGkAdgBlAHIA IABqAG8AaQBuAHMAIAB0AGgAZQAgAHMAZQBzAHMAaQBvAG4ALgANAFAAcgBvAHAAbwBzAGUAZAA6 AA0AQwBhAG4AIABiAGUAIABjAG8AbQBtAHUAbgBpAGMAYQB0AGUAZAAgAG8AdQB0AC0AbwBmAC0A YgBhAG4AZAAgAGIAZQBmAG8AcgBlACAAagBvAGkAbgBpAG4AZwAgAHQAaABlACAAcwBlAHMAcwBp AG8AbgANAEMAYQBuACAAYgBlACAAZAB5AG4AYQBtAGkAYwBhAGwAbAB5ACAAYwBvAG0AbQB1AG4A aQBjAGEAdABlAGQAIABvAHUAdAAtAG8AZgAtAGIAYQBuAGQAIABhAHMAIAB0AGgAZQAgAHMAZQBz AHMAaQBvAG4AIABpAHMAIABwAHIAbwBnAHIAZQBzAHMAaQBuAGcADQBDAGEAbgAgAGIAZQAgAGMA bwBtAG0AdQBuAGkAYwBhAHQAZQBkACAAaQBuAC0AYgBhAG4AZAAgAGEAcwAgAHQAaABlACAAcwBl AHMAcwBpAG8AbgAgAGkAcwAgAHAAcgBvAGcAcgBlAHMAcwBpAG4AZwAgABMgIABhAGQAZAAgAGEA IABmAGkAeABlAGQAIABsAGUAbgBnAHQAaAAgAGEAbgBkACAAdgBhAHIAaQBhAGIAbABlACAAbABl AG4AZwB0AGgAIABIAGUAYQBkAGUAcgAgAGUAeAB0AGUAbgBzAGkAbwBuACAAdABvACAAYwBhAHIA cgB5ACAAdABoAGkAcwAgAGkAbgBmAG8AcgBtAGEAdABpAG8AbgAAAKEPbAAAACwAAAAAAAAAAAAE AAAAAQAAAAAARAAAAAIAAAAAAAoAAAABAAAAAAATAQAAAgAAAAAALAAAAAAAAgAYAAQAAAAABAIA AQQUAEQAAAAABAIAAQQUAAoAAAAACAIAAQgUABMBAAAADAIAAQwUAAAA8wMUAAAA+AEAAAAAAAAC AAAAjwEAAAAAAAAAAJ8PBAAAAAAAAAAAAKgPFgAAAFByb3Bvc2VkIGNoYW5nZXMgdG8gMDYQAJ8P BAAAAAEAAAAAAKAPZgQAAFQAcgBhAG4AcwBtAGkAcwBzAGkAbwBuACAATwBiAGoAZQBjAHQAIABJ AEQAIAATICAAdwBoAGUAbgAgAHIAZQBxAHUAaQByAGUAZAANADAANgA6AA0ASQBmACAAbQBvAHIA ZQAgAHQAaABhAG4AIABvAG4AZQAgAG8AYgBqAGUAYwB0ACAAYwBhAHIAcgBpAGUAZAAgAGkAbgAg AHQAaABlACAAcwBlAHMAcwBpAG8AbgAgAHQAaABlAG4AIABUAE8ASQAgAHIAZQBxAHUAaQByAGUA ZAAgAHQAbwAgAGIAZQAgAGMAYQByAHIAaQBlAGQAIABpAG4AIABlAGEAYwBoACAAcABhAGMAawBl AHQADQBQAHIAbwBwAG8AcwBhAGwAOgANAEkAZgAgAG0AbwByAGUAIAB0AGgAYQBuACAAbwBuAGUA IABvAGIAagBlAGMAdAAgAGMAYQByAHIAaQBlAGQAIABpAG4AIAB0AGgAZQAgAHMAZQBzAHMAaQBv AG4AIAB0AGgAZQBuACAAVABPAEkAIAByAGUAcQB1AGkAcgBlAGQAIABmAG8AcgAgAGEAbABsACAA YgB1AHQAIABvAG4AZQAgAG8AYgBqAGUAYwB0ACwAIABhAG4AZAAgAHQAaABlACAAbwBiAGoAZQBj AHQAIABuAG8AdAAgAGMAYQByAHIAeQBpAG4AZwAgAHQAaABlACAAVABPAEkAIABpAHMAIABwAHIA ZQBzAHUAbQBlAGQAIAB0AG8AIABoAGEAdgBlACAAVABPAEkAIAA9ACAAMAANAFAAcgBvADoAIABT AGEAdgBlAHMAIABzAG8AbQBlACAAcwBwAGEAYwBlACAAaQBuACAAcABhAGMAawBlAHQAIABoAGUA YQBkAGUAcgAgAGYAbwByACAAbwBuAGUAIABvAGIAagBlAGMAdAANAEMAbwBuADoAIABJAHMAIABp AG4AZQBsAGUAZwBhAG4AdAAgAGEAbgBkACAAaQBuAGMAbwBuAHMAaQBzAHQAZQBuAHQAIAATICAA YwBhAG4AIABjAGEAdQBzAGUAIABjAG8AbgBmAHUAcwBpAG8AbgAgAGkAZgAgAG8AbgBlACAAbwBm ACAAdABoAGUAIABvAGIAagBlAGMAdABzACAAZQB4AHAAbABpAGMAaQB0AGwAeQAgAGMAYQByAHIA aQBlAHMAIABUAE8ASQAgAD0AIAAwAA0AVAByAGEAbgBzAG0AaQBzAHMAaQBvAG4AIABPAGIAagBl AGMAdAAgAEkARAAgABMgIABzAGMAbwBwAGUADQAwADYAOgANAEkAbQBwAGwAaQBjAGkAdABsAHkA IABzAGMAbwBwAGUAZAAgAGcAbABvAGIAYQBsAGwAeQANAFAAcgBvAHAAbwBzAGUAZAA6AA0ARQB4 AHAAbABpAGMAaQB0AGwAeQAgAHMAYwBvAHAAZQAgAHcAaQB0AGgAaQBuACAAcwBlAHMAcwBpAG8A bgAAAKEPMgEAACcAAAAAAAAQAABOAAQAAAABAAAQAABOAF4AAAACAAAQAABOAAoAAAABAAAQAABO ADkBAAACAAAQAABOAB8AAAAAAAAQAABOAAQAAAABAAAQAABOABsAAAACAAAQAABOAAoAAAABAAAQ AABOACAAAAACAAAQAABOACcAAAAAAAIAFAAEAAAAAAQCAAEEEgBeAAAAAAQCAAEEEgAKAAAAAAgC AAEIEgCZAAAAAAwCAAEMEgADAAAAAAwGAAEMEgAAgAD+MwAAAAAMAgABDBIAAwAAAAAMBgABDBIA /wAA/mcAAAAADAIAAQwSAB8AAAAAEAIAARAUAAQAAAAAFAIAARQUABsAAAAAFAIAARQUAAoAAAAA GAIAARgUAB8AAAAAHAIAARwUAAEAAAAAHAIAARwSAAAA8wMUAAAArwEAAAAAAAACAAAATgEAAAAA AAAAAJ8PBAAAAAAAAAAAAKgPDwAAAE5leHQgZm9yIEFMQyBQSQAAoQ8cAAAAEAAAAAAAAAAAAA8A AAAAAAIAKAABAAAAAAAAABAAnw8EAAAAAQAAAAAAqA8uAAAARGVjaWRlIHRoZXNlIGlzc3VlcyBo ZXJlDVJldmlzaW9uIDA3DUxhc3QgY2FsbAAAoQ8kAAAALwAAAAAAAAAAACQAAAAAAAAACgAAAAAA gAACAAEAAAAAAAAAAADzAxQAAAC1AQAABgAAAAIAAABTAQAAAAAAAAAAnw8EAAAABgAAAAAAqA8r AAAAV2F2ZSAmIEVxdWF0aW9uIEJhc2VkIFJhdGUgQ29udHJvbAtXRUJSQyBCQhAAnw8EAAAABQAA AAAAqg8KAAAAAQAAAAEAAAAAAAAA8wMUAAAAtgEAAAAAAAACAAAAVAEAAAAAAAAAAJ8PBAAAAAAA AAAAAKgPGAAAAFdFQlJDIG91dHN0YW5kaW5nIGlzc3VlcxAAnw8EAAAAAQAAAAAAqA+GAQAATnVt YmVyIG9mIGNoYW5uZWxzIHVzZWQgcGVyIHNlc3Npb24gY2FuIGJlIGxhcmdlDURlcGVuZHMgb24g dGltZW91dCBmb3IgZ2VuZXJhbCBxdWVyeQ0xIGNoYW5uZWwgdXNlZCBmb3IgZWFjaCAxMCBzZWNv bmRzDTE4MCBzZWNvbmRzIGRlZmF1bHQgdmFsdWUNRGVwZW5kcyBvbiByYXRpbyBvZiBtYXhyYXRl IHRvIG1pbnJhdGUNOCBLYnBzIGlzIGRlZmF1bHQgbWlucmF0ZQ1udW1iZXIgb2YgZ3JvdXBzID0g bG9nXyg0LzMpIChtYXhyYXRlL21pbnJhdGUpDVRoaXMgaXMgYW4gaXNzdWUgZm9yIEFTTSwgbm90 IHNvIGZvciBTU00NU3RhdGUgaW5mb3JtYXRpb24gaW4gcm91dGVycy9zd2l0Y2hlcyBzaG91bGQg YmUgb2sgZm9yIHRoaXMgaWYgb2sgZm9yIG90aGVyIG11bHRpY2FzdCBhcHBzAAChD3YAAAAxAAAA AAAAAAAAJQAAAAEAAAAAAD0AAAACAAAAAAAnAAAAAQAAAAAASQAAAAIAAAAAAIQAAAABAAAAAAAx AAAAAAAAACUAAAAABAAAAQQ9AAAAAAQAAAEEJwAAAAAIAAABCEkAAAAACAAAAQiEAAAAAAgAAAEI AACqD2IAAACnAAAAAAAAAAgAAAABAAAAAwADAAAAAAAAAAgAAAABAAAAAwASAAAAAAAAAAgAAAAB AAAAAwAeAAAAAAAAAAcAAAABAAAAAwABAAAAAAAAAAcAAAABAAAAAwCGAAAAAAAAAAAA8wMUAAAA +QEAAAAAAAACAAAAkAEAAAAAAAAAAJ8PBAAAAAAAAAAAAKgPGAAAAFdFQlJDIG91dHN0YW5kaW5n IGlzc3VlcxAAnw8EAAAAAQAAAAAAqA8yAgAAQWJpbGl0eSBvZiBuZXR3b3JraW5nIGluZnJhc3Ry dWN0dXJlIHRvIHNjYWxlIHRvIGhhbmRsZSByZWNlaXZlcnMgdGhhdCBzZW5kIG9uZSBJR01QIGpv aW4gYW5kIGxlYXZlIG1lc3NhZ2UgZWFjaCAxMCBzZWNvbmRzDVBlcmZvcm1hbmNlIGZpZ3VyZXMg b24gdGhlIGFtb3VudCBvZiBJR01QIG1lc3NhZ2VzIHJvdXRlcnMvc3dpdGNoZXMgY2FuIGhhbmRs ZQ1XaGVuIGhlYXZ5IGRhdGEgcGFja2V0IGxvYWQNV2hhdCBpbXBhY3Qgb24gZGF0YSBwYWNrZXRz Pw1XaGF0IGltcGFjdCBvbiBJR01QIGxhdGVuY3kNU2FtZSBxdWVzdGlvbnMgZm9yIHJlc3VsdGlu ZyBQSU0gU00gbWVzc2FnZXMNTm8gcGVyZm9ybWFuY2UgZmlndXJlcyB5ZXQgZnJvbSB2ZW5kb3Jz DURpZ2l0YWwgRm91bnRhaW4gcmVwbGljYXRvciBwZXJmb3JtYW5jZSBmaWd1cmVzDVNpbWlsYXIg dG8gSUdNUCBqb2luIG1lc3NhZ2UgcHJvY2Vzc2luZw0xOSBtcyBsYXRlbmN5IGZvciBqb2luIG1l c3NhZ2VzIHdoZW4gMSByZWNlaXZlcg0yMCBtcyBsYXRlbmN5IGZvciBqb2luIG1lc3NhZ2VzIHdo ZW4gMTAsMDAwIHJlY2VpdmVycwAAoQ+QAAAAfAAAAAAAABAAAE4ATwAAAAEAABAAAE4AVQAAAAIA ABAAAE4ALQAAAAEAABAAAE4AWAAAAAAAABAAAE4AjgAAAAEAABAAAE4AfAAAAAAAAABPAAAAAAQA AAEEVQAAAAAEAAABBC0AAAAACAAAAQgoAAAAAAgEAAEI/wAA/jAAAAAACAAAAQiOAAAAAAgAAAEI AACqDxoAAACGAQAAAAAAAAsAAAABAAAAAwCiAAAAAAAAAAAA8wMUAAAA+gEAAAAAAAACAAAAkQEA AAAAAAAAAJ8PBAAAAAAAAAAAAKgPGAAAAFdFQlJDIG91dHN0YW5kaW5nIGlzc3VlcxAAnw8EAAAA AQAAAAAAqA8eAQAATG9uZyBJR01QIGFuZCBQSU0gU00gam9pbiBsYXRlbmN5IGNhbiBhZHZlcnNl bHkgYWZmZWN0IGFnZ3Jlc3NpdmVuZXNzDUxvc3NlcyBkdWUgdG8gYmVoYXZpb3Igb2YgbXVsdGlj YXN0IHByb3RvY29sIGNhbiBhZHZlcnNlbHkgYWZmZWN0IGFnZ3Jlc3NpdmVuZXNzDVJQcw1Td2l0 Y2hvdmVyIGZyb20gKCosRykgdG8gKFMsRykNTVNEUA1SZWNvbW1lbmRhdGlvbnM/DVVzZSBTU00N SWYgdXNpbmcgQVNNDVBsYWNlIFJQIG5lYXIgc2VuZGVyDURvIG5vdCB1c2UgTVNEUCBpZiBpdCBj YW4gYmUgYXZvaWRlZAAAoQ9gAAAAlwAAAAAAAAAAACgAAAABAAAAAAARAAAAAAAAAAAAFQAAAAEA AAAAADoAAAACAAAAAACXAAAAAAAAACgAAAAAAAAAEQAAAAAEAAABBBUAAAAACAAAAQg6AAAAAAwA AAEMAACqDxoAAACXAAAAAAAAAAQAAAABAAAAAwCEAAAAAAAAAAAA8wMUAAAA9AEAAAAAAAACAAAA jgEAAAAAAAAAAJ8PBAAAAAAAAAAAAKgPDQAAAFdFQlJDIHJvYWRtYXAQAJ8PBAAAAAEAAAAAAKAP rgEAAEYAdQBsAGwAIABUAGUAYwBoAG4AaQBjAGEAbAAgAHAAYQBwAGUAcgANAFQAYQByAGcAZQB0 ACAAZABhAHQAZQAgAEEAcAByAGkAbAAgADEADQBuAHMAIABjAG8AZABlACAAYQBuAGQAIAB3AGUA YgBzAGkAdABlACAAbwBmACAAcgBlAHMAdQBsAHQAcwAgAGEAdgBhAGkAbABhAGIAbABlAA0AVABh AHIAZwBlAHQAIABkAGEAdABlACAAQQBwAHIAaQBsACAAMQANAEQAdQBtAG0AeQAgAG4AZQB0ACAA cwBpAG0AdQBsAGEAdABpAG8AbgBzAA0AVABhAHIAZwBlAHQAIABkAGEAdABlACAASgB1AG4AZQAg ADEADQBSAGUAYQBsACAAdwBvAHIAbABkACAAZQB4AHAAZQByAGkAbQBlAG4AdABzAA0AVABhAHIA ZwBlAHQAIABkAGEAdABlACAAQQB1AGcAdQBzAHQAIAAxAA0AUgBGAEMAIAB0AGEAcgBnAGUAdAAg ABMgIABTAGUAcAB0AGUAbQBiAGUAcgAgADIAMAAwADIADQAAAKEPugAAABUAAAAAAAAAAAAUAAAA AQAAAAAAKQAAAAAAAAAAABQAAAABAAAAAAAWAAAAAAAAAAAAEwAAAAEAAAAAABcAAAAAAAAAAAAV AAAAAQAAAAAAHQAAAAAAAAAAABUAAAAAAAAAFAAAAAAAAAADAAAAAgQAAAMEJgAAAAAEAAABBBQA AAAABAAAAQQWAAAAAAgAAAEIEwAAAAAIAAABCBcAAAAADAAAAQwVAAAAAAwAAAEMHQAAAAAMAAAB DC8A8A84AAAAAADzAxQAAABWAQAAAAAAAAAAAAABAQAAAAAAAAAA8wMUAAAAhwEAAAAAAAAAAAAA AgEAAAAAAAABAAEEUAAAAAAAAAH///9/AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABIAAADqAwAAAAAPAPgDMggAAAIA 7wMYAAAAAQAAAAECBwkIAAAAAAAAAAAAAAAAAAAAYADwByAAAAD///8AAAAAAICAgAAAAAAAAMyZ ADMzzADMzP8AsrKyAGAA8AcgAAAAAAD/AP///wAAAAAA//8AAP+ZAAAA//8A/wAAAJaWlgBgAPAH IAAAAP//zAAAAAAAZmYzAICAAAAzmTMAgAAAAAAzzAD/zGYAYADwByAAAAD///8AAAAAADMzMwAA AAAA3d3dAICAgABNTU0A6urqAGAA8AcgAAAA////AAAAAACAgIAAAAAAAP/MZgAAAP8AzADMAMDA wABgAPAHIAAAAP///wAAAAAAgICAAAAAAADAwMAAAGb/AP8AAAAAmQAAYADwByAAAAD///8AAAAA AICAgAAAAAAAM5n/AJn/zADMAMwAsrKyAAAAow8+AAAAAQD//T8AAAAiIAAAZAAAAAAAAQBYAAAA AAAAAAAAQAIAAAAAAgAAAP//7wABAAEA////////JAAGPej+AAAQAKMPhgAAAAUA//0/AA8ApwAC AGQABj3o/gAAWAAeAAAAtAAAAEACAAAAAAIAAAD//+8AAQABAP///////xgAAAAA/gAAkgUAAA0A IiAAALABIAEAAAIAFACQBQAAEyACANACQAIAAAAAsAUAACIgAAAAAAAFzANgAwAAAgAOAIAFAAAT IOwEgAQAAAAAIACjD24AAAAFAP/9PwAAACIgAABkAAAAAAAAAFoAKAAAAAAAAABAAgAAAAACAAAA ///vAAAAAQD///////8MAAAAAAEAAAAFAAAgASABAAAAAAAFAABAAkACAAAAAAAFAABgA2ADAAAA AAAFAACABIAEAAAAAFAAow9SAAAABQAAAAEJAAAOAAEAAAAAAAAAAQABCQAADAABACABAAAAAAIA AQkAAAwAAQBAAgAAAAADAAEJAAAMAAEAYAMAAAAABAABCQAADAABAIAEAAAAAGAAow8MAAAAAQAA AAAAAAAAAAAAcACjDz4AAAAFAAAAAAAAAAAAAgAUAAEAAAAAAAAAAgASAAIAAAAAAAAAAgASAAMA AAAAAAAAAgAMAAQAAAAAAAAAAgAMAIAAow8+AAAABQAAAAAAAAAAAAIAEgABAAAAAAAAAAIAEAAC AAAAAAAAAAIAEAADAAAAAAAAAAIACgAEAAAAAAAAAAIACgAPAAwETAQAAA8AAvBEBAAAEAAI8AgA AAAEAAAABgQAAA8AA/DiAwAADwAE8CgAAAABAAnwEAAAAAAAAAAAAAAAAAAAAAAAAAACAArwCAAA AAAEAAAFAAAADwAE8GcBAAASAArwCAAAAAIEAAAACgAAIwEL8IoAAAB/AAEAAQCAABTnugCBAHhh AQCCAKKtAACDAHhhAQCEAKKtAAC/ABAAHwCAAQEAAACBAQEAAAiDAQAAAAiGQQEAAACHwR4AAAC/ AQUAFQDAAQEAAAjLAZwxAAD/AQEACQABAgIAAAg/AgAAAgBMAGEAcgBnAGUAIABjAG8AbgBmAGUA dAB0AGkAAAATACLxBgAAAL8DAAAABAAAEPAIAAAA8APgAdAUwA8PABHwEAAAAAAAwwsIAAAAAQAA AAIAugAPAA3whwAAAAAAnw8EAAAAAQAAAAAAqA87AAAAQm9keSBUZXh0DVNlY29uZCBMZXZlbA1U aGlyZCBMZXZlbA1Gb3VydGggTGV2ZWwNRmlmdGggTGV2ZWwAAKIPHgAAAAoAAAAAAA0AAAABAAwA AAACAA0AAAADAAwAAAAEAAAAqg8KAAAAPAAAAAEAAAAAAA8ABPAlAQAAEgAK8AgAAAADBAAAAAoA ADMBC/CQAAAAfwABAAEAgABc67oAgQB4YQEAggCirQAAgwB4YQEAhACirQAAhwAEAAAAvwAQAB8A gAEBAAAAgQEBAAAIgwEAAAAIhkEBAAAAh8EeAAAAvwEFABUAwAEBAAAIywGcMQAA/wEBAAkAAQIC AAAIPwIAAAIATABhAHIAZwBlACAAYwBvAG4AZgBlAHQAdABpAAAAEwAi8QYAAAC/AwAAAAQAABDw CAAAAJAAoAJAFNACDwAR8BAAAAAAAMMLCAAAAAAAAAABALoADwAN8D8AAAAAAJ8PBAAAAAAAAAAA AKgPCwAAAFNsaWRlIFRpdGxlAACiDwYAAAAMAAAAAAAAAKoPCgAAAAwAAAABAAAAAAAPAATwDgEA ABIACvAIAAAABgQAAAAKAAATAQvwhAAAAIAAcO66AIEAeGEBAIIAoq0AAIMAeGEBAIQAoq0AAL8A EgAfAIABAQAAAIEBAQAACIMBAAAACIZBAQAAAIfBHgAAAL8BBAAUAMABAQAACMsBnDEAAP8BAAAI AAECAgAACD8CAAACAEwAYQByAGcAZQAgAGMAbwBuAGYAZQB0AHQAaQAAABMAIvEGAAAAvwMAAAAE AAAQ8AgAAADADxARZxVOEA8ADfBMAAAAAACfDwQAAAAEAAAAAACgDwIAAAAqAAAAoQ8iAAAAAgAA AAAAADgAAAIAWgAyAAIAAAABAAcAAQABAAoAAGth/gAA2A8EAAAAAAAAAA8ABPBCAAAAEgAK8AgA AAABBAAAAAwAAHMAC/AqAAAAgQEAAAAIkwGOn4sAlAHevWgAvwESABIA/wEAAAgABAMJAAAAPwMB AAEAEADwByAAAAD///8ACB1YAGGP/QBxIAAAM2X7AM8OMABnZ2cAwc7/ACAAug8yAAAAUABPAEQA QwAgAHQAYQBsAGsAIAAoAHIAZQB2AGkAcwBlAGQAIAAxADAALwA5ADYAKQAPAPADswIAAAEA8QMI AAAAAAAAgAAADDAPAAwEcwIAAA8AAvBrAgAAUAAI8AgAAAADAAAAAwgAAA8AA/AJAgAADwAE8CgA AAABAAnwEAAAAAAAAAAAAAAAAAAAAAAAAAACAArwCAAAAAAIAAAFAAAADwAE8FkBAAASAArwCAAA AAIIAAAACgAAIwEL8IoAAAB/AAEAAQCAADDtaAOBABNiAQCCAO6tAACDABNiAQCEAO6tAAC/ABAA HwCAAQEAAACBAQEAAAiDAQAAAAiGQQEAAACHwR4AAAC/AQUAFQDAAQEAAAjLAZwxAAD/AQEACQAB AgIAAAg/AgAAAgBMAGEAcgBnAGUAIABjAG8AbgBmAGUAdAB0AGkAAAAAABDwCAAAALAKQAKgDs8U DwAR8BAAAAAAAMMLCAAAAAMAAAAGAmgDDwAN8IcAAAAAAJ8PBAAAAAIAAAAAAKgPOwAAAEJvZHkg VGV4dA1TZWNvbmQgTGV2ZWwNVGhpcmQgTGV2ZWwNRm91cnRoIExldmVsDUZpZnRoIExldmVsAACi Dx4AAAAKAAAAAAANAAAAAQAMAAAAAgANAAAAAwAMAAAABAAAAKoPCgAAADwAAAABAAAAAAAPAATw cAAAABIACvAIAAAAAwgAAAAKAACDAAvwMAAAAH8ABAAEAIcAAQAAAH8BAAABAL8BAQARAMABAQAA CMsBnDEAAP8BCAAJAD8CAQADAAAAEPAIAAAAtQHXAgsOHAoPABHwEAAAAAAAwwsIAAAAAgAAAAUA aAMPAATwQgAAABIACvAIAAAAAQgAAAAMAABzAAvwKgAAAIEBAAAACJMB3r1oAJQBjp+LAL8BEgAS AP8BAAAIAAQDCQAAAD8DAQABABAA8AcgAAAA////AAAAAACRkZEAAAAAAGGP/QAArgAA/AEoAM7O zgAPAMkPygAAAA8ADASaAAAADwAC8JIAAADgAAjwCAAAAAEAAAABDAAADwAD8DAAAAAPAATwKAAA AAEACfAQAAAAAAAAAAAAAAAAAAAAAABYAAIACvAIAAAAAAwAAAUAAAAPAATwQgAAABIACvAIAAAA AQwAAAAMAABzAAvwKgAAAIEBAAAACJMB3r1oAJQBjp+LAL8BEgASAP8BAAAIAAQDCQAAAD8DAQAB ABAA8AcgAAAA////AAAAAACRkZEAAAAAAGGP/QAArgAA/AEoAM7OzgAPAO4DwAIAAAIA7wMYAAAA AAAAAA8QAAAAAAAAAAAAgAEBAAAHAAAADwDZDwwAAAAAANoPBAAAAAAAJQAPAAwEXAIAAA8AAvBU AgAAMAAI8AgAAAADAAAAA7QFAA8AA/DsAQAADwAE8CgAAAABAAnwEAAAAAAAAAAAAAAAAAAAAAAA AAACAArwCAAAAAC0BQAFAAAADwAE8MYAAAASAArwCAAAAAK0BQAgAgAAMwEL8HIAAAB/AAAABACA AOymaAOBAHhhAQCCANazAACDAHhhAQCEANazAACFAAAAAACGAAAAAACHAAAAAACIAAAAAACJAAAA AACKAAAAAACLAAAAAAC/ABAAHwC/AQAAEQD/AQAACQA/AgAAAgABAwMEAACIAwAAAAAAABDwCAAA AKACsAHQFHAFDwAR8BAAAAAAAMMLCAAAAAAAAAAPAGgDDwAN8AwAAAAAAJ4PBAAAAAAAAAAPAATw 5gAAABIACvAIAAAAA7QFACACAAAzAQvwcgAAAH8AAAAEAIAAdKxoA4EAeGEBAIIA1rMAAIMAeGEB AIQA1rMAAIUAAAAAAIYAAAAAAIcAAAAAAIgAAAAAAIkAAAAAAIoAAAAAAIsAAAAAAL8AEAAfAL8B AAARAP8BAAAJAD8CAAACAAEDAgQAAIgDAAAAAAAAEPAIAAAAkAawAdAU4A0PABHwEAAAAAAAwwsI AAAAAQAAABAAaAMPAA3wLAAAAAAAng8EAAAAAQAAAAAApg8YAAAA+R4AADoC1gDzAR4BEAM7AgoE WAMDBVIEDwAE8EgAAAASAArwCAAAAAG0BQAADAAAgwAL8DAAAACBAQAAAAiDAQUAAAiTAY6fiwCU Ad69aAC/AR4AHwD/AQAACAAEAwkAAAA/AwEAAQAQAPAHIAAAAABsqgD///8AkZGRAPr9AAAAk+YA /66mAAusoQD6NWMADwDuA8ACAAACAO8DGAAAAAAAAAAPEAAAAAAAAAAAAIACAQAABwAAAA8A2Q8M AAAAAADaDwQAAAAAACUADwAMBFwCAAAPAALwVAIAAGAACPAIAAAAAwAAAARoBgAPAAPw7AEAAA8A BPAoAAAAAQAJ8BAAAAAAAAAAAAAAAAAAAAAAAAAAAgAK8AgAAAAAaAYABQAAAA8ABPDGAAAAEgAK 8AgAAAACaAYAIAIAADMBC/ByAAAAfwAAAAQAgABUEWwDgQB4YQEAggDWswAAgwB4YQEAhADWswAA hQAAAAAAhgAAAAAAhwAAAAAAiAAAAAAAiQAAAAAAigAAAAAAiwAAAAAAvwAQAB8AvwEAABEA/wEA AAkAPwIAAAIAAQMDBAAAiAMAAAAAAAAQ8AgAAACgArAB0BRwBQ8AEfAQAAAAAADDCwgAAAAAAAAA DwBsAw8ADfAMAAAAAACeDwQAAAAAAAAADwAE8OYAAAASAArwCAAAAANoBgAgAgAAMwEL8HIAAAB/ AAAABACAAOwubAOBAHhhAQCCANazAACDAHhhAQCEANazAACFAAAAAACGAAAAAACHAAAAAACIAAAA AACJAAAAAACKAAAAAACLAAAAAAC/ABAAHwC/AQAAEQD/AQAACQA/AgAAAgABAwIEAACIAwAAAAAA ABDwCAAAAJAGsAHQFOANDwAR8BAAAAAAAMMLCAAAAAEAAAAQAGwDDwAN8CwAAAAAAJ4PBAAAAAEA AAAAAKYPGAAAAPkeAAA6AtYA8wEeARADOwIKBFgDAwVSBA8ABPBIAAAAEgAK8AgAAAABaAYAAAwA AIMAC/AwAAAAgQEAAAAIgwEFAAAIkwGOn4sAlAHevWgAvwEeAB8A/wEAAAgABAMJAAAAPwMBAAEA EADwByAAAAAAbKoA////AJGRkQD6/QAAAJPmAP+upgALrKEA+jVjAA8A7gPwAQAAAgDvAxgAAAAB AAAADQ4AAAAAAAAAAACAAAAAAAcAAAAPAAwEoAEAAA8AAvCYAQAAAAEI8AgAAAADAAAAA1wIAA8A A/AwAQAADwAE8CgAAAABAAnwEAAAAAAAAAAAAAAAAAAAAAAAAAACAArwCAAAAABcCAAFAAAADwAE 8HgAAAASAArwCAAAAAJcCAAgAgAAYwAL8CQAAAB/AAAABACAAKAAFAS/AQAAAQD/AQAAAQABAwME AACIAwAAAAAAABDwCAAAAJAAoAJAFNACDwAR8BAAAAAAAMMLCAAAAAAAAAANABQEDwAN8AwAAAAA AJ4PBAAAAAAAAAAPAATweAAAABIACvAIAAAAA1wIACACAABjAAvwJAAAAH8AAAAEAIAAtAMUBL8B AAABAP8BAAABAAEDAgQAAIgDAAAAAAAAEPAIAAAAkAPgAWAVYA8PABHwEAAAAAAAwwsIAAAAAQAA AA4AFAQPAA3wDAAAAAAAng8EAAAAAQAAAA8ABPBIAAAAEgAK8AgAAAABXAgAAAwAAIMAC/AwAAAA gQEAAAAIgwEFAAAIkwGOn4sAlAHevWgAvwESABIA/wEAAAgABAMJAAAAPwMBAAEAEADwByAAAAD/ //8ACB1YAGGP/QBxIAAAM2X7AM8OMABnZ2cAwc7/AA8A7gPwAQAAAgDvAxgAAAABAAAADQ4AAAAA AAAAAACAAAAAAAcAAAAPAAwEoAEAAA8AAvCYAQAAgAAI8AgAAAADAAAAAxQHAA8AA/AwAQAADwAE 8CgAAAABAAnwEAAAAAAAAAAAAAAAAAAAAAAAAAACAArwCAAAAAAUBwAFAAAADwAE8HgAAAASAArw CAAAAAIUBwAgAgAAYwAL8CQAAAB/AAAABACAAJAJbAO/AQAAAQD/AQAAAQABAwMEAACIAwAAAAAA ABDwCAAAAJAAoAJAFNACDwAR8BAAAAAAAMMLCAAAAAAAAAANAGwDDwAN8AwAAAAAAJ4PBAAAAAAA AAAPAATweAAAABIACvAIAAAAAxQHACACAABjAAvwJAAAAH8AAAAEAIAACDZsA78BAAABAP8BAAAB AAEDAgQAAIgDAAAAAAAAEPAIAAAAAAPgAWAV0A4PABHwEAAAAAAAwwsIAAAAAQAAAA4AbAMPAA3w DAAAAAAAng8EAAAAAQAAAA8ABPBIAAAAEgAK8AgAAAABFAcAAAwAAIMAC/AwAAAAgQEAAAAIgwEF AAAIkwGOn4sAlAHevWgAvwESABIA/wEAAAgABAMJAAAAPwMBAAEAEADwByAAAAD///8ACB1YAGGP /QBxIAAAM2X7AM8OMABnZ2cAwc7/AA8A7gPwAQAAAgDvAxgAAAABAAAADQ4AAAAAAAAAAACAAAAA AAcAAAAPAAwEoAEAAA8AAvCYAQAAkAAI8AgAAAADAAAAA1AIAA8AA/AwAQAADwAE8CgAAAABAAnw EAAAAAAAAAAAAAAAAAAAAAAAAAACAArwCAAAAABQCAAFAAAADwAE8HgAAAASAArwCAAAAAJQCAAg AgAAYwAL8CQAAAB/AAAABACAAFxTbAO/AQAAAQD/AQAAAQABAwMEAACIAwAAAAAAABDwCAAAAJAA oAJAFNACDwAR8BAAAAAAAMMLCAAAAAAAAAANAGwDDwAN8AwAAAAAAJ4PBAAAAAAAAAAPAATweAAA ABIACvAIAAAAA1AIACACAABjAAvwJAAAAH8AAAAEAIAAGFRsA78BAAABAP8BAAABAAEDAgQAAIgD AAAAAAAAEPAIAAAAcALgAWAVQA4PABHwEAAAAAAAwwsIAAAAAQAAAA4AbAMPAA3wDAAAAAAAng8E AAAAAQAAAA8ABPBIAAAAEgAK8AgAAAABUAgAAAwAAIMAC/AwAAAAgQEAAAAIgwEFAAAIkwGOn4sA lAHevWgAvwESABIA/wEAAAgABAMJAAAAPwMBAAEAEADwByAAAAD///8ACB1YAGGP/QBxIAAAM2X7 AM8OMABnZ2cAwc7/AA8A7gPwAQAAAgDvAxgAAAABAAAADQ4AAAAAAAAAAACAAAAAAAcAAAAPAAwE oAEAAA8AAvCYAQAAoAAI8AgAAAADAAAAAxwHAA8AA/AwAQAADwAE8CgAAAABAAnwEAAAAAAAAAAA AAAAAAAAAAAAAAACAArwCAAAAAAcBwAFAAAADwAE8HgAAAASAArwCAAAAAIcBwAgAgAAYwAL8CQA AAB/AAAABACAAFBBbAO/AQAAAQD/AQAAAQABAwMEAACIAwAAAAAAABDwCAAAAJAAoAJAFNACDwAR 8BAAAAAAAMMLCAAAAAAAAAANAGwDDwAN8AwAAAAAAJ4PBAAAAAAAAAAPAATweAAAABIACvAIAAAA AxwHACACAABjAAvwJAAAAH8AAAAEAIAAmDZsA78BAAABAP8BAAABAAEDAgQAAIgDAAAAAAAAEPAI AAAAYAPgAWAVMA8PABHwEAAAAAAAwwsIAAAAAQAAAA4AbAMPAA3wDAAAAAAAng8EAAAAAQAAAA8A BPBIAAAAEgAK8AgAAAABHAcAAAwAAIMAC/AwAAAAgQEAAAAIgwEFAAAIkwGOn4sAlAHevWgAvwES ABIA/wEAAAgABAMJAAAAPwMBAAEAEADwByAAAAD///8ACB1YAGGP/QBxIAAAM2X7AM8OMABnZ2cA wc7/AA8A7gO9AgAAAgDvAxgAAAAAAAAADxAAAAAAAAAAAACAAAAAAAcAAAAPAAwEbQIAAA8AAvBl AgAAsAAI8AgAAAADAAAABTQHAA8AA/D9AQAADwAE8CgAAAABAAnwEAAAAAAAAAAAAAAAAAAAAAAA AAACAArwCAAAAAA0BwAFAAAADwAE8HgAAAASAArwCAAAAAI0BwAgAgAAYwAL8CQAAAB/AAAABACA AJBsbAO/AQAAAQD/AQAAAQABAwMEAACIAwAAAAAAABDwCAAAACAEAACAFnAIDwAR8BAAAAAAAMML CAAAAAAAAAAPAGwDDwAN8AwAAAAAAJ4PBAAAAAAAAAAPAATwRQEAABIACvAIAAAABDQHAAAKAAAj AQvwigAAAH8AAAAFAIAA6DZsA4EAeGEBAIIA1rMAAIMAeGEBAIQA1rMAAL8AEAAfAIABAQAAAIEB AQAACIMBAAAACIZBAQAAAIfBHgAAAL8BBAAVAMABAQAACMsBnDEAAP8BAAAJAAECAgAACD8CAAAC AEwAYQByAGcAZQAgAGMAbwBuAGYAZQB0AHQAaQAAAAAAEPAIAAAAoAiwAdAU8A8PAA3wiwAAAAAA nw8EAAAABQAAAAAAqA81AAAADU0uIEx1YnksIERpZ2l0YWwgRm91bnRhaW4NVi4gR295YWwsIERp Z2l0YWwgRm91bnRhaW4AAKoPGgAAAB4AAAAAAAAABQAAAAEAAAADABMAAAAAAAAAAACmDxgAAAD5 HgAAOgLWAPMBHgEQAzsCCgRYAwMFUgQPAATwSAAAABIACvAIAAAAATQHAAAMAACDAAvwMAAAAIEB AAAACIMBBQAACJMBjp+LAJQB3r1oAL8BEgASAP8BAAAIAAQDCQAAAD8DAQABABAA8AcgAAAA//// AAAAAACAgIAAAAAAAADMmQAzM8wAzMz/ALKysgAPAO4DEAIAAAIA7wMYAAAAAQAAAA0OAAAAAAAA AAAAgAAAAAAHAAAADwAMBMABAAAPAALwuAEAAMAACPAIAAAAAwAAAAM4BwAPAAPwUAEAAA8ABPAo AAAAAQAJ8BAAAAAAAAAAAAAAAAAAAAAAAAAAAgAK8AgAAAAAOAcABQAAAA8ABPB+AAAAEgAK8AgA AAACOAcAIAIAAHMAC/AqAAAABAAAAAAAfwAAAAQAgADgYWwDvwEAAAEA/wEAAAEAAQMDBAAAiAMA AAAAAAAQ8AgAAACQAKACQBTQAg8AEfAQAAAAAADDCwgAAAAAAAAADQBsAw8ADfAMAAAAAACeDwQA AAAAAAAADwAE8JIAAAASAArwCAAAAAM4BwAgAgAAcwAL8CoAAAAEAAAAAAB/AAAABACAAGBaaAO/ AQAAAQD/AQAAAQABAwIEAACIAwAAAAAAABDwCAAAAPAD4AHQFMAPDwAR8BAAAAAAAMMLCAAAAAEA AAAOAGwDDwAN8CAAAAAAAJ4PBAAAAAEAAAAAAKYPDAAAANgAAADYANQB8AMQBQ8ABPBIAAAAEgAK 8AgAAAABOAcAAAwAAIMAC/AwAAAAgQEAAAAIgwEFAAAIkwGOn4sAlAHevWgAvwESABIA/wEAAAgA BAMJAAAAPwMBAAEAEADwByAAAAD///8AAAAAAICAgAAAAAAAAMyZADMzzADMzP8AsrKyAA8A7gMQ AgAAAgDvAxgAAAABAAAADQ4AAAAAAAAAAACAAAAAAAcAAAAPAAwEwAEAAA8AAvC4AQAA0AAI8AgA AAADAAAAA1QIAA8AA/BQAQAADwAE8CgAAAABAAnwEAAAAAAAAAAAAAAAAAAAAAAAAAACAArwCAAA AABUCAAFAAAADwAE8H4AAAASAArwCAAAAAJUCAAgAgAAcwAL8CoAAAAEAAAAAAB/AAAABACAAChg bAO/AQAAAQD/AQAAAQABAwMEAACIAwAAAAAAABDwCAAAAJAAoAJAFNACDwAR8BAAAAAAAMMLCAAA AAAAAAANAGwDDwAN8AwAAAAAAJ4PBAAAAAAAAAAPAATwkgAAABIACvAIAAAAA1QIACACAABzAAvw KgAAAAQAAAAAAH8AAAAEAIAAnHNoA78BAAABAP8BAAABAAEDAgQAAIgDAAAAAAAAEPAIAAAA8APg AdAUwA8PABHwEAAAAAAAwwsIAAAAAQAAAA4AbAMPAA3wIAAAAAAAng8EAAAAAQAAAAAApg8MAAAA 2AAAANgA1AHwAxAFDwAE8EgAAAASAArwCAAAAAFUCAAADAAAgwAL8DAAAACBAQAAAAiDAQUAAAiT AY6fiwCUAd69aAC/ARIAEgD/AQAACAAEAwkAAAA/AwEAAQAQAPAHIAAAAP///wAAAAAAgICAAAAA AAAAzJkAMzPMAMzM/wCysrIADwDuAxACAAACAO8DGAAAAAEAAAANDgAAAAAAAAAAAIAAAAAABwAA AA8ADATAAQAADwAC8LgBAADwAAjwCAAAAAMAAAADWAgADwAD8FABAAAPAATwKAAAAAEACfAQAAAA AAAAAAAAAAAAAAAAAAAAAAIACvAIAAAAAFgIAAUAAAAPAATwfgAAABIACvAIAAAAAlgIACACAABz AAvwKgAAAAQAAAAAAH8AAAAEAIAADEhsA78BAAABAP8BAAABAAEDAwQAAIgDAAAAAAAAEPAIAAAA kACgAkAU0AIPABHwEAAAAAAAwwsIAAAAAAAAAA0AbAMPAA3wDAAAAAAAng8EAAAAAAAAAA8ABPCS AAAAEgAK8AgAAAADWAgAIAIAAHMAC/AqAAAABAAAAAAAfwAAAAQAgAAEhmwDvwEAAAEA/wEAAAEA AQMCBAAAiAMAAAAAAAAQ8AgAAADwA+AB0BTADw8AEfAQAAAAAADDCwgAAAABAAAADgBsAw8ADfAg AAAAAACeDwQAAAABAAAAAACmDwwAAADYAAAA2ADUAfADEAUPAATwSAAAABIACvAIAAAAAVgIAAAM AACDAAvwMAAAAIEBAAAACIMBBQAACJMBjp+LAJQB3r1oAL8BEgASAP8BAAAIAAQDCQAAAD8DAQAB ABAA8AcgAAAA////AAAAAACAgIAAAAAAAADMmQAzM8wAzMz/ALKysgAPAO4DEAIAAAIA7wMYAAAA AQAAAA0OAAAAAAAAAAAAgAAAAAAHAAAADwAMBMABAAAPAALwuAEAACAACPAIAAAAAwAAAANACAAP AAPwUAEAAA8ABPAoAAAAAQAJ8BAAAAAAAAAAAAAAAAAAAAAAAAAAAgAK8AgAAAAAQAgABQAAAA8A BPB+AAAAEgAK8AgAAAACQAgAIAIAAHMAC/AqAAAABAAAAAAAfwAAAAQAgAA8jWgDvwEAAAEA/wEA AAEAAQMDBAAAiAMAAAAAAAAQ8AgAAACQAKACQBTQAg8AEfAQAAAAAADDCwgAAAAAAAAADQBoAw8A DfAMAAAAAACeDwQAAAAAAAAADwAE8JIAAAASAArwCAAAAANACAAgAgAAcwAL8CoAAAAEAAAAAAB/ AAAABACAAHyRaAO/AQAAAQD/AQAAAQABAwIEAACIAwAAAAAAABDwCAAAAPAD4AHQFMAPDwAR8BAA AAAAAMMLCAAAAAEAAAAOAGgDDwAN8CAAAAAAAJ4PBAAAAAEAAAAAAKYPDAAAANgAAADYANQB8AMQ BQ8ABPBIAAAAEgAK8AgAAAABQAgAAAwAAIMAC/AwAAAAgQEAAAAIgwEFAAAIkwGOn4sAlAHevWgA vwESABIA/wEAAAgABAMJAAAAPwMBAAEAEADwByAAAAD///8AAAAAAICAgAAAAAAAAMyZADMzzADM zP8AsrKyAA8A8AMgAgAAAQDxAwgAAAAFAQAABwAMMA8ADATgAQAADwAC8NgBAABAAAjwCAAAAAMA AAADuAUADwAD8HABAAAPAATwKAAAAAEACfAQAAAAAAAAAAAAAAAAAAAAAAAAAAIACvAIAAAAALgF AAUAAAAPAATwXgAAABIACvAIAAAAArgFAAAKAABTAAvwHgAAAH8ABAAEAH8BAAABAL8BEQARAP8B CQAJAD8CAQABAAAAEPAIAAAAtQHYAgwOHAoPABHwEAAAAAAAwwsIAAAAAAAAAAsAaAMPAATw0gAA ABIACvAIAAAAA7gFAAAKAAAjAQvwbAAAAH8AAQABAIAAGPVoA4EAwGUBAIIA4LIAAIMAwGUBAIQA 4LIAAL8AEAAfAIEBBAAACIMBAAAACL8BAQARAMABAQAACNIBAAAAANMBAAAAANQBAAAAANUBAAAA AP8BAQAJAAECAgAACD8CAAACAAAAEPAIAAAAsApAAqAOzxQPABHwEAAAAAAAwwsIAAAAAQAAAAwA aAMPAA3wHgAAAAAAnw8EAAAAAgAAAAAAqg8KAAAAAQAAAAEAAAAAAA8ABPBIAAAAEgAK8AgAAAAB uAUAAAwAAIMAC/AwAAAAgQEAAAAIgwEFAAAIkwHevWgAlAGOn4sAvwEeAB8A/wEAAAgABAMJAAAA PwMBAAEAEADwByAAAAD///8AAAAAAICAgAAAAAAAAMyZADMzzADMzP8AsrKyAA8A8AMgAgAAAQDx AwgAAAAtAQAABwAMMA8ADATgAQAADwAC8NgBAABwAAjwCAAAAAMAAAADbAYADwAD8HABAAAPAATw KAAAAAEACfAQAAAAAAAAAAAAAAAAAAAAAAAAAAIACvAIAAAAAGwGAAUAAAAPAATwXgAAABIACvAI AAAAAmwGAAAKAABTAAvwHgAAAH8ABAAEAH8BAAABAL8BEQARAP8BCQAJAD8CAQABAAAAEPAIAAAA tQHYAgwOHAoPABHwEAAAAAAAwwsIAAAAAAAAAAsAbAMPAATw0gAAABIACvAIAAAAA2wGAAAKAAAj AQvwbAAAAH8AAQABAIAAAAlsA4EAwGUBAIIA4LIAAIMAwGUBAIQA4LIAAL8AEAAfAIEBBAAACIMB AAAACL8BAQARAMABAQAACNIBAAAAANMBAAAAANQBAAAAANUBAAAAAP8BAQAJAAECAgAACD8CAAAC AAAAEPAIAAAAsApAAqAOzxQPABHwEAAAAAAAwwsIAAAAAQAAAAwAbAMPAA3wHgAAAAAAnw8EAAAA AgAAAAAAqg8KAAAAAQAAAAEAAAAAAA8ABPBIAAAAEgAK8AgAAAABbAYAAAwAAIMAC/AwAAAAgQEA AAAIgwEFAAAIkwHevWgAlAGOn4sAvwEeAB8A/wEAAAgABAMJAAAAPwMBAAEAEADwByAAAAD///8A AAAAAICAgAAAAAAAAMyZADMzzADMzP8AsrKyAAAAchdoAAAAAQBAAAAAAABUPgAAD0EAABo2AABV ASAA4UEAAHZaAACFARAAqUQAAIcBEACeXAAArQEQAGlJAACvARAAWU0AALUBIABRTwAAFlIAAPQB EABeWAAA+AFAAGFLAAAuVAAARlYAAHFHAAAAAPUPHAAAAI4BAACvEQADAAAAAMZeAAABAAAA+wEA AAEAAAAPAOgDADYAAAEA6QMoAAAAgBYAAOAQAADgEAAAgBYAAAUAAAAKAAAAAgAAAAMAAAABAAAA AAAAAQ8A8gO0AgAALwDID+oAAAAwANIPBAAAAAIAAAAAALoPKAAAABggHCAI/xQwO/9b/wgwCjAM MA4wEDDl/wT/JAAoAFsAXAB7AGL/4f8QALoPpgAAAAEwAjAM/w7/+zAa/xv/H/8B/5swnDD9MP4w HgAAAA8AAABPbi1zY3JlZW4gU2hvdwAAHgAAABEAAABEaWdpdGFsIEZvdW50YWluAAAAAAMAAAC2 lwAAAwAAAF4AAAADAAAACwAAAAMAAAACAAAAAwAAAAAAAAADAAAAAAAAAAMAAAAyEQkACwAAAAAA AAALAAAAAAAAAAsAAAAAAAAACwAAAAAAAAAeEAAADwAAABAAAABUaW1lcyBOZXcgUm9tYW4ABgAA AEFyaWFsAAoAAABXaW5nZGluZ3MAGgAAAFBPREMgdGFsayAocmV2aXNlZCAxMC85NikAFwAAAEZF QyBCQiBGRUMgSU5GTyBMQ1QgQkIAFQAAAEFMQyBQSSBzdGF0dXMgcmVwb3J0ABcAAABQcm9wb3Nl ZCBjaGFuZ2VzIHRvIDA2ABcAAABQcm9wb3NlZCBjaGFuZ2VzIHRvIDA2ABcAAABQcm9wb3NlZCBj aGFuZ2VzIHRvIDA2ABAAAABOZXh0IGZvciBBTEMgUEkALAAAAFdhdmUgJiBFcXVhdGlvbiBCYXNl ZCBSYXRlIENvbnRyb2wgV0VCUkMgQkIAGQAAAFdFQlJDIG91dHN0YW5kaW5nIGlzc3VlcwAZAAAA V0VCUkMgb3V0c3RhbmRpbmcgaXNzdWVzABkAAABXRUJSQyBvdXRzdGFuZGluZyBpc3N1ZXMADgAA AFdFQlJDIHJvYWRtYXAADBAAAAYAAAAeAAAACwAAAEZvbnRzIFVzZWQAAwAAAAMAAAAeAAAAEAAA AERlc2lnbiBUZW1wbGF0ZQADAAAAAQAAAB4AAAANAAAAU2xpZGUgVGl0bGVzAAMAAAALAAAA//8C AAAAFwAAACYGDwAjAAAA9g8kAAAAFAAAAF/AkeOSlwAADAD0AwMAugBNaWNoYWVsIEx1YnkIAAAA TQBpAGMAaABhAGUAbAAgAEwAdQBiAHkA//8DAAAADwAAACYGDwAUAFROUFAEAAwAAQAAAAEAAAAA AAAABQAAAAsCAAAAAAUAAAAMAtACwAMFAAAABAENAAAABwAAAPwCAAD///8AAAAEAAAALQEAAAgA AAD6AgUAAQAAAAAAAAAEAAAALQEBAAQAAAAtAQAACQAAAB0GIQDwANACwAMAAAAABAAAAC0BAAAH AAAA/AIAAP///wAAAAQAAAAtAQIABAAAAPABAAAIAAAA+gIAAAAAAAAAAAAABAAAAC0BAAAHAAAA /AIBAAAAAAAAAAQAAAAtAQMABAAAAC0BAQAHAAAAGwS5ApIDoALYAgQAAAAtAQIABAAAAC0BAAAc AAAA+wIAAAAAAAAAAAAAAAAAAAAAAAAA8hIAknH1d0AAAADoAwo2TFP1d1VT9XcBAAAAAAAwAFIA bwBvAHQAIABFAG4AdAByAHkAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAWAAUB//////////8DAAAAEI2BZJtPzxGG6gCqALkp6AAAAAAAAAAAAAAAAPAVI9Mt0cEB cAAAAMAOAAAAAAAAUABpAGMAdAB1AHIAZQBzAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAABIAAgH/////AgAAAP////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAABAAAAAAAABDAHUAcgByAGUAbgB0ACAAVQBzAGUAcgAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAGgACAP///////////////wAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADoAAAAyAAAAAAAAAAUAUwB1AG0AbQBhAHIAeQBJ AG4AZgBvAHIAbQBhAHQAaQBvAG4AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAoAAIBAQAAAAUA AAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACwLAAAAAAAAAQAA AAIAAAADAAAABAAAAAUAAAAGAAAABwAAAP7///8JAAAACgAAAAsAAAAMAAAADQAAAA4AAAAPAAAA EAAAABEAAAASAAAAEwAAABQAAAAVAAAAFgAAABcAAAAYAAAAGQAAABoAAAAbAAAAHAAAAB0AAAAe AAAAHwAAACAAAAAhAAAAIgAAACMAAAAkAAAAJQAAACYAAAAnAAAAKAAAACkAAAAqAAAAKwAAACwA AAAtAAAALgAAAC8AAAAwAAAAMQAAADIAAAAzAAAANAAAADUAAAA2AAAANwAAAFMAAAA9AAAA//// /1IAAAD9/////v////7///////////////////////////////////////////////////////// //////////////////////////////////////////////////////7///9UAAAAVQAAAFYAAABX AAAAWAAAAFkAAABaAAAAWwAAAFwAAABdAAAAXgAAAF8AAABgAAAAYQAAAGIAAABjAAAAZAAAAGUA AABmAAAAZwAAAGgAAABpAAAAagAAAGsAAABsAAAAbQAAAG4AAAD+/////////3EAAAByAAAAcwAA AHQAAAB1AAAAOAAAAP////////////////////////////////////////////////////8BAAAA AgAAAAMAAAAEAAAABQAAAAYAAAAHAAAACAAAAAkAAAAKAAAACwAAAAwAAAANAAAADgAAAA8AAAAQ AAAAEQAAABIAAAATAAAAFAAAABUAAAAWAAAAFwAAABgAAAAZAAAAGgAAABsAAAAcAAAAHQAAAB4A AAAfAAAAIAAAACEAAAAiAAAAIwAAACQAAAAlAAAAJgAAACcAAAAoAAAAKQAAACoAAAArAAAALAAA AP7///8uAAAALwAAADAAAAAxAAAAMgAAADMAAAA0AAAANQAAADYAAAA3AAAAOAAAADkAAAD+//// /v////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// /////////////////////////////////////////////////////////////////////wAAV0VC UkMgcm9hZG1hcAAMEAAABgAAAB4AAAALAAAARm9udHMgVXNlZAADAAAAAwAAAB4AAAAQAAAARGVz aWduIFRlbXBsYXRlAAMAAAABAAAAHgAAAA0AAABTbGlkZSBUaXRsZXMAAwAAAAsAAAD//wIAAAAX AAAAJgYPACMAAAD2Dx4AAAAUAAAAX8CR45KXAAAGAPQDAwC6AE9rdXRhbggAAABPAGsAdQB0AGEA bgBoAGEAZQBsACAATAB1AGIAeQD//wMAAAAPAAAAJgYPABQAVE5QUAQADAABAAAAAQAAAAAAAAAF AAAACwIAAAAABQAAAAwC0ALAAwUAAAAEAQ0AAAAHAAAA/AIAAP///wAAAAQAAAAtAQAACAAAAPoC BQABAAAAAAAAAAQAAAAtAQEABAAAAC0BAAAJAAAAHQYhAPAA0ALAAwAAAAAEAAAALQEAAAcAAAD8 AgAA////AAAABAAAAC0BAgAEAAAA8AEAAAgAAAD6AgAAAAAAAAAAAAAEAAAALQEAAAcAAAD8AgEA AAAAAAAABAAAAC0BAwAEAAAALQEBAAcAAAAbBLkCkgOgAtgCBAAAAC0BAgAEAAAALQEAABwAAAD7 AgAAAAAAAAAAAAAAAAAAAAAAAADyEgCScfV3QAAAAOgDCjZMU/V3VVP1dwEAAAAAADAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP7/AAAFAAIA AAAAAAAAAAAAAAAAAAAAAAEAAAAC1c3VnC4bEJOXCAArLPmuMAAAAAADAAAQAAAAAQAAAIgAAAAD AAAAkAAAAA8AAACoAAAABAAAAMQAAAAGAAAAzAAAAAcAAADUAAAACAAAANwAAAAJAAAA5AAAAAoA AADsAAAAFwAAAPQAAAALAAAA/AAAABAAAAAEAQAAEwAAAAwBAAAWAAAAFAEAAA0AAAAcAQAADAAA AKACAAACAAAA5AQAAB4AAAAPAAAAT24tc2NyZWVuIFNob3cAAB4AAAARAAAARGlnaXRhbCBGb3Vu dGFpbgAAAAADAAAAtpcAAAMAAABeAAAAAwAAAAsAAAADAAAAAgAAAAMAAAAAAAAAAwAAAAAAAAAD AAAAMhEJAAsAAAAAAAAACwAAAAAAAAALAAAAAAAAAAsAAAAAAAAAHhAAAA8AAAAQAAAAVGltZXMg TmV3IFJvbWFuAAYAAABBcmlhbAAKAAAAV2luZ2RpbmdzABoAAABQT0RDIHRhbGsgKHJldmlzZWQg MTAvOTYpABcAAABGRUMgQkIgRkVDIElORk8gTENUIEJCABUAAABBTEMgUEkgc3RhdHVzIHJlcG9y dAAXAAAAUHJvcG9zZWQgY2hhbmdlcyB0byAwNgAXAAAAUHJvcG9zZWQgY2hhbmdlcyB0byAwNgAX AAAAUHJvcG9zZWQgY2hhbmdlcyB0byAwNgAQAAAATmV4dCBmb3IgQUxDIFBJACwAAABXYXZlICYg RXF1YXRpb24gQmFzZWQgUmF0ZSBDb250cm9sIFdFQlJDIEJCABkAAABXRUJSQyBvdXRzdGFuZGlu ZyBpc3N1ZXMAGQAAAFdFQlJDIG91dHN0YW5kaW5nIGlzc3VlcwAZAAAAV0VCUkMgb3V0c3RhbmRp bmcgaXNzdWVzAA4AAABXRUJSQyByb2FkbWFwAAwQAAAGAAAAHgAAAAsAAABGb250cyBVc2VkAAMA AAADAAAAHgAAABAAAABEZXNpZ24gVGVtcGxhdGUAAwAAAAEAAAAeAAAADQAAAFNsaWRlIFRpdGxl cwADAAAACwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA 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 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAPYPJAAAABQAAABfwJHj Nl8AAAwA9AMDALoATWljaGFlbCBMdWJ5CAAAAE0AaQBjAGgAYQBlAGwAIABMAHUAYgB5AAAAAAAA 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 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQAAAAIAAAADAAAABAAAAAUAAAAGAAAA BwAAAP7///8JAAAACgAAAAsAAAAMAAAADQAAAA4AAAAPAAAAEAAAABEAAAASAAAAEwAAABQAAAAV AAAAFgAAABcAAAAYAAAAGQAAABoAAAAbAAAAHAAAAB0AAAAeAAAAHwAAACAAAAAhAAAAIgAAACMA AAAkAAAAJQAAACYAAAAnAAAAKAAAACkAAAAqAAAAKwAAACwAAAAtAAAALgAAAC8AAAAwAAAAMQAA ADIAAAAzAAAANAAAADUAAAA2AAAANwAAAFMAAAA5AAAA/v////////////////////////////// //////////////////////////////////////////////////////////////////////////// ///////////////9////UgAAAP7///9UAAAAVQAAAFYAAABXAAAAWAAAAFkAAABaAAAAWwAAAFwA AABdAAAAXgAAAF8AAABgAAAAYQAAAGIAAABjAAAAZAAAAGUAAABmAAAAZwAAAGgAAABpAAAAagAA AGsAAABsAAAAbQAAAG4AAAD+/////v///3EAAAByAAAAcwAAAHQAAAB1AAAAOAAAAP////////// //////////////////////////////////////////9SAG8AbwB0ACAARQBuAHQAcgB5AAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAFgAFAf//////////AwAAABCN gWSbT88RhuoAqgC5KegAAAAAAAAAAAAAAAAAhMXwZM/BAXAAAAAADwAAAAAAAFAAaQBjAHQAdQBy AGUAcwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAASAAIB /////wIAAAD/////AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQAAAA AAAAQwB1AHIAcgBlAG4AdAAgAFUAcwBlAHIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAABoAAgD///////////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAA6AAAARAAAAAAAAAAFAFMAdQBtAG0AYQByAHkASQBuAGYAbwByAG0AYQB0AGkAbwBuAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAKAACAQEAAAAFAAAA/////wAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAsCwAAAAAAAFAAbwB3AGUAcgBQAG8AaQBuAHQAIABEAG8A YwB1AG0AZQBuAHQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAoAAIB////////////////AAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACAAAALaXAAAAAAAABQBEAG8AYwB1AG0A ZQBuAHQAUwB1AG0AbQBhAHIAeQBJAG4AZgBvAHIAbQBhAHQAaQBvAG4AAAAAAAAAAAAAADgAAgEE AAAA//////////8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAtAAAAMAMAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAP///////////////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA////////////////AAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAnTCeMAUw/DAZIB0gCf8VMD3/Xf8JMAswDTAP MBEwsAAwIDIgMyADIeD/Bf9BMEMwRTBHMEkwYzCDMIUwhzCOMKEwozClMKcwqTDDMOMw5TDnMO4w 9TD2MCEAJQApACwALgA6ADsAPwBdAH0AYf9j/2T/Zf9n/2j/af9q/2v/bP9t/27/b/9w/57/n/8P ANUH5AAAAAAAtw9EAAAAVABpAG0AZQBzACAATgBlAHcAIABSAG8AbQBhAG4AAACMrroAdNoSAFza EgBkpwwwCAAAAAAAAAB02hIAICYMMAAABAAQALcPRAAAAEEAcgBpAGEAbAAAAE4AZQB3ACAAUgBv AG0AYQBuAAAAjK66AHTaEgBc2hIAZKcMMAgAAAAAAAAAdNoSACAmDDAAzAQiIAC3D0QAAABXAGkA bgBnAGQAaQBuAGcAcwAAAFIAbwBtAGEAbgAAAIyuugB02hIAXNoSAGSnDDAIAAAAAAAAAHTaEgAg JgwwAswEAgAApA8MAAAAgABhAAAAAQD/////AAClDwwAAAAAAAAILgAAAAIAAAAAAKsPHgAAAP8f AAAFAEACAAAAAAAAIAEgAUACQAJgA2ADgASABAAAqQ8KAAAABwAAAAIACQQAAEAAow9uAAAABQD/ /T8AAAAiIAAAZAAAAAAAAABkAAAAAAAAAAAAQAIAAAAAAgAAAP//7wAAAAAA////////GAAAAAAB AAAABQAAIAEgAQAAAAAABQAAQAJAAgAAAAAABQAAYANgAwAAAAAABQAAgASABAAAAAAPAAsE3hEA AA8AAPDWEQAAAAAG8MgQAAACZAgAGAIAAGgAAAAQAAAAAQAAAAcAAAAFAAAABAAAABAAAAACAAAA AAAAAAMAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAA AAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAA AAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAA AAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAA BAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAE AAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQA AAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAA AAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAA AAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAA AAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAA AAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAkAAAAAAAAABAAAAAAAAAAEAAAAAAAA AAUAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAA BQAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAE AAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAFAAAAAAAAAAQA AAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAA AAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAA AAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAA AAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAA AAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAA AAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAA FQAAAAAAAAAFAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAE AAAAAAAAAAQAAAAAAAAABwAAAAAAAAAgAAAAAAAAAAQAAAAAAAAABQAAAAAAAAAEAAAAAAAAAAUA AAAAAAAABAAAAAAAAAAFAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAjAAAAAAAAAAQAAAAAAAAABwAA AAAAAAAHAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAA AAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAABf AAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABQAAAAAAAAAlAAAAAAAAAAQAAAAAAAAABAAAAAAA AAAbAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAA4AAAAAAAAAEQAAAAAAAAAYAAAAAAAA AAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAA BAAAAAAAAAAEAAAAAAAAAAUAAAAAAAAABAAAAAAAAAAFAAAAAAAAAAQAAAAAAAAABQAAAAAAAAAE AAAAAAAAAAUAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAAAgAAAAAAAAAIAAAAAAAAAAIA AAAAAAAAFAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABgAA AAAAAAAEAAAAAAAAAAQAAAAAAAAAJgAAAAAAAAAoAAAAAAAAAAQAAAAAAAAABAAAAAAAAAASAAAA AAAAACMAAAAAAAAAJgAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAeAAAAAAAAAAQAAAAA AAAALAAAAAAAAAAaAAAAAAAAAC0AAAAAAAAANAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAA AAAIAAAAAAAAABwAAAAAAAAADgAAAAAAAAAaAAAAAAAAABMAAAAAAAAABQAAAAAAAAAEAAAAAAAA AAgAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAFAAAACAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAA BAAAAAAAAAAIAAAAAAAAABEAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAE AAAAAAAAAAQAAAAAAAAABAAAAAAAAAAFAAAAAAAAAAQAAAAAAAAABAAAAAAAAAARAAAAAAAAAAQA AAAAAAAABAAAAAAAAAAFAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAA AAAAAAAFAAAAAAAAAAUAAAAAAAAABQAAAAAAAAAFAAAAAAAAAAUAAAAAAAAABAAAAAAAAAAEAAAA AAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAA AAAABAAAAAAAAAAFAAAAAAAAAAQAAAAAAAAABQAAAAAAAAAEAAAAAAAAAAgAAAAAAAAABQAAAAAA AAACAAAAAAAAAAQAAAAAAAAAGgAAAAAAAAAEAAAAAAAAABwAAAAAAAAABAAAAAAAAAAkAAAAAAAA AAQAAAAAAAAAEwAAAAAAAAAEAAAAAAAAAAQAAAAAAAAAGgAAAAAAAAAEAAAAAAAAABEAAAAAAAAA BAAAAAAAAAAEAAAAAAAAAAUAAAAAAAAABAAAAAAAAAAhAAAAAAAAAAQAAAAAAAAAEAAAAAAAAAAL AAAAAAAAACEAAAAAAAAABAAAAAAAAAAQAAAAAAAAAAsAAAAAAAAACwAAAAAAAAAFAAAAAAAAAAQA AAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAAEAAA AAAAAAAFAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAA AAAAAAQAAAAAAAAABAAAAAAAAAAcAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAE0AAAAA AAAAJwAAAAAAAAACAAAAAAAAAAQAAAAAAAAABQAAAAAAAAADAAAAAAAAAB8AAAAAAAAABAAAAAAA AAAkAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAADAAAABAAAAAQAAAAEAAAAAAAA AAQAAAAAAAAABAAAAAAAAAAGAAAAAAAAAAUAAAAAAAAABQAAAAAAAAAEAAAAAAAAAAQAAAAAAAAA BAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAE AAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQA AAAAAAAALAAAAAAAAAAEAAAAAAAAAAUAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAA AAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAA AAAAAA8AAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAYAAAAG AAAABQAAAAcAAAAEAAAAAAAAAAcAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAA AAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAA AAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAA BAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAE AAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAACUA AAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAA AAkAAAAEAAAAAAAAAAQAAAALAAAABAAAABAAAAAEAAAAAAAAAAQAAAAGAAAABAAAAAAAAAAEAAAA AAAAAAQAAAAMAAAABgAAAA0AAAAEAAAAAAAAADAAAAAAAAAABAAAAAAAAAADAAAAAAAAAAQAAAAA AAAABAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAAAAADAAAAAAAAAAQAAAAAAAAAUwAAAAAA AAAEAAAAAAAAAAMAAAAAAAAAMQAAAAAAAAAEAAAACwAAAAoAAAAAAAAACgAAAAAAAAAFAAAAAAAA AAUAAAAAAAAAPAAAAAAAAAAEAAAAAAAAAAQAAAAAAAAAAwAAAAAAAAAEAAAAAAAAAAQAAAAAAAAA EwAAAAAAAAAEAAAAAAAAAAQAAAAAAAAAAwAAAAAAAAAPAAAAAAAAABQAAAAAAAAABAAAAAAAAAAV AAAAAAAAABQAAAAAAAAAHQAAAAAAAAAEAAAAAAAAAEcAAAAAAAAADwAAAAAAAABMAAAAAAAAAEwA AAAAAAAATAAAAAAAAABMAAAAAAAAAFAAAAAHAAAARwAAAAAAAAAqAAAAAAAAAAwAAAAAAAAALgAA AAAAAAAuAAAAAAAAAC4AAAAAAAAAKQAAAAAAAAAqAAAAAAAAACoAAAAAAAAAKgAAAAAAAAAqAAAA AAAAACoAAAAAAAAAIAAAAAAAAAAEAAAAAAAAAFMAAAAAAAAABAAAAAAAAAADAAAAAAAAAAQAAAAA AAAAAwAAAAAAAAAEAAAAAAAAAAMAAAAAAAAABAAAAAIAAAAEAAAAAAAAAAQAAAAAAAAABAAAAAAA AAAEAAAACQAAAAQAAAACAAAABAAAAA8AAAAEAAAACAAAAAQAAAAfAAHwLAAAAHIAB/AkAAAABwfJ eXW9gP8bSZjTmHGUEMwB/wBpAAAAKgAAAAAAAAAAAMkAIwEL8IoAAACAAQEAAACBAQEAAAiDAQAA AAiGQQEAAACHwR4AAAC/ARQAFADAAQEAAAjBAQAAAQDEAQAAAADLAThjAADNAQAAAADOAQAAAADQ AQAAAADRAQAAAADXAQIAAAD/AQgACAABAgIAAAg/AgAAAgBMAGEAcgBnAGUAIABjAG8AbgBmAGUA dAB0AGkAAACAABrxIAAAAAD/AAAAzAAAAGthANwAgQBxRAAAM8wzAP8AAAAAgAAAQAAe8RAAAAD/ ////AwAACP//////////HwDwDxwAAAAAAPMDFAAAAAQAAAAEAAAAAAAAAAAAAIAAAAAADwDQBzgC AAAfAP8DFAAAAAIAAAQMAAAAAAAAAAAAAAACAAAADwD6A2cAAAAAAP4DAwAAAAABAAAA/QM0AAAA TAAAAGQAAABMAAAAZAAAAGSnDDAIAAAAAAAAAGjaEgAAAAAAAAAAAFT9//+m////AQAAAHAA+wMI AAAAAAAAAHAIAABwAPsDCAAAAAEAAABACwAAHwAHBDwAAAAAAP0DNAAAADIAAABkAAAAMgAAAGQA AABnPwwwJN4SAAAAAAB4r7oAAAAAAAAAAAAAAAAAAAAAAAABEgAfAAgEPAAAAAAA/QM0AAAAQgAA AGQAAABCAAAAZAAAAGc/DDAk3hIAAAAAAHivugAAAAAAAAAAAAAAAAAAAAAAAAASAB8A+gNnAAAA AAD+AwMAAAAAAQAAAP0DNAAAAEsAAABkAAAASwAAAGQAAABkpwwwCAAAAAAAAABo2hIAAAAAAAAA AABC/f//lgAAAAAAAABwAPsDCAAAAAAAAABACwAAcAD7AwgAAAABAAAAcAgAAB8AFAQcAAAAAAAV BBQAAABm/UsfAMqaO176oR4Aypo7AAIAAR8AEwQ8AAAAAAD9AzQAAABkAAAAZAAAAGQAAABkAAAA lNoSAP49DDCA2hIAFAAAAAAAAAAAAAAAAAAAAAAAAAAAARIADwCIE0YAAAAPAIoTPgAAAAAAug8O AAAAXwBfAF8AUABQAFQAOQAAAIsTIAAAAC8AyA8MAAAAMADSDwQAAACAAAAAAAB6FwQAAAACBAAA PwDZDwwAAAAAANoPBAAAAAAAJQBPANkPDAAAAAAA2g8EAAAAAAA9AA8A8A/6HQAAAADzAxQAAABV AQAAAAAAAAIAAAAFAQAAAAAAAAAAnw8EAAAABgAAAAAAqA8WAAAARkVDIEJCC0ZFQyBJTkZPC0xD VCBCQgAAoQ8gAAAAFwAAAAAAAAAAABYAAAAAAAIAKAABAAAAAACCAAIAKAAQAJ8PBAAAAAUAAAAA AKgPpAAAAA1BbGwgc2VudCB0byBJRVNHIGZvciBjb25zaWRlcmF0aW9uIGFzIFJGQ3MNRkVDIEJC IGFuZCBMQ1QgQkIgYXMgRXhwZXJpbWVudGFsDUZFQyBJTkZPIGFzIEluZm9ybWF0aW9uYWwNKGFs dGhvdWdoIGZvciBzb21lIHJlYXNvbiBzdGlsbCBub3QgaW4gdGhlIFJGQyBFZGl0b3IgUXVldWUp AACqDxoAAAAnAAAAAAAAAAUAAAABAAAAAwB5AAAAAAAAAAAA8wMUAAAAhQEAAAIAAAACAAAALQEA AAAAAAAAAJ8PBAAAAAYAAAAAAKgPFAAAAEFMQyBQSSBzdGF0dXMgcmVwb3J0AAChDx4AAAAVAAAA AAAAAAAAFAAAAAAAAgAoAAEAAAAAAAIAIAAQAJ8PBAAAAAUAAAAAAKgPcQAAAA1NLiBMdWJ5LCBE aWdpdGFsIEZvdW50YWluDUouIEdlbW1lbGwsIE1pY3Jvc29mdA1MLiBWaWNpc2FubywgQ2lzY28N TC4gUml6em8sIEFDSVJJIGFuZCBVLiBQaXNhDUouIENyb3djcm9mdCwgVUNMAACqDz4AAAAeAAAA AAAAAAcAAAABAAAAAwAPAAAAAAAAAAgAAAABAAAAAwAnAAAAAAAAAAkAAAABAAAAAwAGAAAAAAAA AAAA8wMUAAAA+wEAAAAAAAACAAAAkgEAAAAAAAAAAJ8PBAAAAAAAAAAAAKgPFgAAAFByb3Bvc2Vk IGNoYW5nZXMgdG8gMDYQAJ8PBAAAAAEAAAAAAKAPcgQAAEMAbABhAHIAaQBmAHkAIAB0AGgAYQB0 ACAAcwBlAHMAcwBpAG8AbgAgAGkAcwAgAGYAcgBvAG0AIABhACAAcwBpAG4AZwBsAGUAIABzAGUA bgBkAGUAcgAsACAAYQBuAGQAIAB0AGgAYQB0ACAAcgBlAGMAZQBpAHYAaQBuAGcAIABmAHIAbwBt ACAAbQB1AGwAdABpAHAAbABlACAAcwBlAG4AZABlAHIAcwAgAGkAcwAgAGEAdAAgAHQAaABlACAA YQBwAHAAbABpAGMAYQB0AGkAbwBuACAAbABlAHYAZQBsACAAYQBiAG8AdgBlACAAQQBMAEMADQBU AHIAYQBuAHMAbQBpAHMAcwBpAG8AbgAgAFMAZQBzAHMAaQBvAG4AIABJAEQADQAwADYAOgANAFQA UwBJACAAZQB4AHAAbABpAGMAaQB0AGwAeQAgAHIAZQBxAHUAaQByAGUAZAAgAHQAbwAgAGIAZQAg AGkAbgAgAGUAYQBjAGgAIABwAGEAYwBrAGUAdAAuAA0AUAByAG8AcABvAHMAZQBkADoADQBUAFMA SQAgAGUAeABwAGwAaQBjAGkAdABsAHkAIAByAGUAcQB1AGkAcgBlAGQAIAB0AG8AIABiAGUAIABp AG4AIABlAGEAYwBoACAAcABhAGMAawBlAHQAIABVAE4ATABFAFMAUwAgAHQAaABlACAAcwBlAG4A ZABlAHIAIABpAHMAIABzAHUAcABwAG8AcgB0AGkAbgBnACAAbwBuAGwAeQAgAG8AbgBlACAAcwBl AHMAcwBpAG8AbgAsACAAaQBuACAAdwBoAGkAYwBoACAAYwBhAHMAZQAgAFQAUwBJACAATQBBAFkA IABiAGUAIABvAG0AaQB0AHQAZQBkACAAYQBuAGQAIABpAHMAIAB0AGgAZQBuACAAcAByAGUAcwB1 AG0AZQBkACAAdABvACAAYgBlACAAMAAuAA0AUAByAG8AOgAgAFMAYQB2AGUAcwAgAHMAcABhAGMA ZQAgAGkAbgAgAHAAYQBjAGsAZQB0ACAAaABlAGEAZABlAHIAIABpAGYAIABvAG4AbAB5ACAAbwBu AGUAIABzAGUAcwBzAGkAbwBuACAAcABlAHIAIABzAGUAbgBkAGUAcgANAEMAbwBuAHMAOgAgAEwA QwBUACAAcgBlAHEAdQBpAHIAZQBzACAAVABTAEkAIABpAG4AIABoAGUAYQBkAGUAcgAgABMgIABp AHMAIAB1AHMAZQBmAHUAbAAgAGYAbwByACAAbgBlAHQAdwBvAHIAawAgAGUAbABlAG0AZQBuAHQA cwAgAHQAbwAgAGMAbABhAHMAcwBpAGYAeQAgAGMAaABhAG4AbgBlAGwAcwAgAGEAYwBjAG8AcgBk AGkAbgBnACAAdABvACAAcwBlAHMAcwBpAG8AbgAgABMgIABpAG4AYwBvAG4AcwBpAHMAdABlAG4A dAAgAGEAYwByAG8AcwBzACAAcwBlAG4AZABlAHIAcwAAAKEPsgAAAJUAAAAAAAAQAABOAAQAAAAB AAAQAABOAC4AAAACAAAQAABOAAoAAAABAAAQAABOAGkBAAACAAAQAABOAJUAAAAAAAAABAAAAAAE AAABBC4AAAAABAAAAQQKAAAAAAgAAAEIngAAAAAMAAABDAUAAAAADAQAAQwAgAD+PAAAAAAMBAAB DAAAAAEGAAAAAAwEAAEM/wAA/oMAAAAADAQAAQwAAAABAQAAAAAMBAABDP8AAP4AAPMDFAAAAK0B AAAAAAAAAgAAAEwBAAAAAAAAAACfDwQAAAAAAAAAAACoDxYAAABQcm9wb3NlZCBjaGFuZ2VzIHRv IDA2EACfDwQAAAABAAAAAACgDyADAABGAEUAQwAgAE8AYgBqAGUAYwB0ACAAVAByAGEAbgBzAG0A aQBzAHMAaQBvAG4AIABJAG4AZgBvAC4AIABjAG8AbQBtAHUAbgBpAGMAYQB0AGkAbwBuAA0AMAA2 ADoADQBNAHUAcwB0ACAAYgBlACAAYwBvAG0AbQB1AG4AaQBjAGEAdABlAGQAIABvAHUAdAAtAG8A ZgAtAGIAYQBuAGQAIABiAGUAZgBvAHIAZQAgAHIAZQBjAGUAaQB2AGUAcgAgAGoAbwBpAG4AcwAg AHQAaABlACAAcwBlAHMAcwBpAG8AbgAuAA0AUAByAG8AcABvAHMAZQBkADoADQBDAGEAbgAgAGIA ZQAgAGMAbwBtAG0AdQBuAGkAYwBhAHQAZQBkACAAbwB1AHQALQBvAGYALQBiAGEAbgBkACAAYgBl AGYAbwByAGUAIABqAG8AaQBuAGkAbgBnACAAdABoAGUAIABzAGUAcwBzAGkAbwBuAA0AQwBhAG4A IABiAGUAIABkAHkAbgBhAG0AaQBjAGEAbABsAHkAIABjAG8AbQBtAHUAbgBpAGMAYQB0AGUAZAAg AG8AdQB0AC0AbwBmAC0AYgBhAG4AZAAgAGEAcwAgAHQAaABlACAAcwBlAHMAcwBpAG8AbgAgAGkA cwAgAHAAcgBvAGcAcgBlAHMAcwBpAG4AZwANAEMAYQBuACAAYgBlACAAYwBvAG0AbQB1AG4AaQBj AGEAdABlAGQAIABpAG4ALQBiAGEAbgBkACAAYQBzACAAdABoAGUAIABzAGUAcwBzAGkAbwBuACAA aQBzACAAcAByAG8AZwByAGUAcwBzAGkAbgBnACAAEyAgAGEAZABkACAAYQAgAGYAaQB4AGUAZAAg AGwAZQBuAGcAdABoACAAYQBuAGQAIAB2AGEAcgBpAGEAYgBsAGUAIABsAGUAbgBnAHQAaAAgAEgA ZQBhAGQAZQByACAAZQB4AHQAZQBuAHMAaQBvAG4AIAB0AG8AIABjAGEAcgByAHkAIAB0AGgAaQBz ACAAaQBuAGYAbwByAG0AYQB0AGkAbwBuAAAAoQ9iAAAALAAAAAAAAAAAAAQAAAABAAAAAABEAAAA AgAAAAAACgAAAAEAAAAAABMBAAACAAAAAAAsAAAAAAAAAAQAAAAABAAAAQREAAAAAAQAAAEECgAA AAAIAAABCBMBAAAADAAAAQwAAPMDFAAAAPgBAAAAAAAAAgAAAI8BAAAAAAAAAACfDwQAAAAAAAAA AACoDxYAAABQcm9wb3NlZCBjaGFuZ2VzIHRvIDA2EACfDwQAAAABAAAAAACgD2YEAABUAHIAYQBu AHMAbQBpAHMAcwBpAG8AbgAgAE8AYgBqAGUAYwB0ACAASQBEACAAEyAgAHcAaABlAG4AIAByAGUA cQB1AGkAcgBlAGQADQAwADYAOgANAEkAZgAgAG0AbwByAGUAIAB0AGgAYQBuACAAbwBuAGUAIABv AGIAagBlAGMAdAAgAGMAYQByAHIAaQBlAGQAIABpAG4AIAB0AGgAZQAgAHMAZQBzAHMAaQBvAG4A IAB0AGgAZQBuACAAVABPAEkAIAByAGUAcQB1AGkAcgBlAGQAIAB0AG8AIABiAGUAIABjAGEAcgBy AGkAZQBkACAAaQBuACAAZQBhAGMAaAAgAHAAYQBjAGsAZQB0AA0AUAByAG8AcABvAHMAYQBsADoA DQBJAGYAIABtAG8AcgBlACAAdABoAGEAbgAgAG8AbgBlACAAbwBiAGoAZQBjAHQAIABjAGEAcgBy AGkAZQBkACAAaQBuACAAdABoAGUAIABzAGUAcwBzAGkAbwBuACAAdABoAGUAbgAgAFQATwBJACAA cgBlAHEAdQBpAHIAZQBkACAAZgBvAHIAIABhAGwAbAAgAGIAdQB0ACAAbwBuAGUAIABvAGIAagBl AGMAdAAsACAAYQBuAGQAIAB0AGgAZQAgAG8AYgBqAGUAYwB0ACAAbgBvAHQAIABjAGEAcgByAHkA aQBuAGcAIAB0AGgAZQAgAFQATwBJACAAaQBzACAAcAByAGUAcwB1AG0AZQBkACAAdABvACAAaABh AHYAZQAgAFQATwBJACAAPQAgADAADQBQAHIAbwA6ACAAUwBhAHYAZQBzACAAcwBvAG0AZQAgAHMA cABhAGMAZQAgAGkAbgAgAHAAYQBjAGsAZQB0ACAAaABlAGEAZABlAHIAIABmAG8AcgAgAG8AbgBl ACAAbwBiAGoAZQBjAHQADQBDAG8AbgA6ACAASQBzACAAaQBuAGUAbABlAGcAYQBuAHQAIABhAG4A ZAAgAGkAbgBjAG8AbgBzAGkAcwB0AGUAbgB0ACAAEyAgAGMAYQBuACAAYwBhAHUAcwBlACAAYwBv AG4AZgB1AHMAaQBvAG4AIABpAGYAIABvAG4AZQAgAG8AZgAgAHQAaABlACAAbwBiAGoAZQBjAHQA cwAgAGUAeABwAGwAaQBjAGkAdABsAHkAIABjAGEAcgByAGkAZQBzACAAVABPAEkAIAA9ACAAMAAN AFQAcgBhAG4AcwBtAGkAcwBzAGkAbwBuACAATwBiAGoAZQBjAHQAIABJAEQAIAATICAAcwBjAG8A cABlAA0AMAA2ADoADQBJAG0AcABsAGkAYwBpAHQAbAB5ACAAcwBjAG8AcABlAGQAIABnAGwAbwBi AGEAbABsAHkADQBQAHIAbwBwAG8AcwBlAGQAOgANAEUAeABwAGwAaQBjAGkAdABsAHkAIABzAGMA bwBwAGUAIAB3AGkAdABoAGkAbgAgAHMAZQBzAHMAaQBvAG4AAAChDzIBAAAnAAAAAAAAEAAATgAE AAAAAQAAEAAATgBeAAAAAgAAEAAATgAKAAAAAQAAEAAATgA5AQAAAgAAEAAATgAfAAAAAAAAEAAA TgAEAAAAAQAAEAAATgAbAAAAAgAAEAAATgAKAAAAAQAAEAAATgAgAAAAAgAAEAAATgAnAAAAAAAC ABQABAAAAAAEAgABBBIAXgAAAAAEAgABBBIACgAAAAAIAgABCBIAmQAAAAAMAgABDBIAAwAAAAAM BgABDBIAAIAA/jMAAAAADAIAAQwSAAMAAAAADAYAAQwSAP8AAP5nAAAAAAwCAAEMEgAfAAAAABAC AAEQFAAEAAAAABQCAAEUFAAbAAAAABQCAAEUFAAKAAAAABgCAAEYFAAfAAAAABwCAAEcFAABAAAA ABwCAAEcEgAAAPMDFAAAAK8BAAAAAAAAAgAAAE4BAAAAAAAAAACfDwQAAAAAAAAAAACoDw8AAABO ZXh0IGZvciBBTEMgUEkAAKEPHAAAABAAAAAAAAAAAAAPAAAAAAACACgAAQAAAAAAAAAQAJ8PBAAA AAEAAAAAAKgPLgAAAERlY2lkZSB0aGVzZSBpc3N1ZXMgaGVyZQ1SZXZpc2lvbiAwNw1MYXN0IGNh bGwAAKEPJAAAAC8AAAAAAAAAAAAkAAAAAAAAAAoAAAAAAIAAAgABAAAAAAAAAAAA8wMUAAAAtQEA AAYAAAACAAAAUwEAAAAAAAAAAJ8PBAAAAAYAAAAAAKgPKwAAAFdhdmUgJiBFcXVhdGlvbiBCYXNl ZCBSYXRlIENvbnRyb2wLV0VCUkMgQkIQAJ8PBAAAAAUAAAAAAKoPCgAAAAEAAAABAAAAAAAAAPMD FAAAALYBAAAAAAAAAgAAAFQBAAAAAAAAAACfDwQAAAAAAAAAAACoDxgAAABXRUJSQyBvdXRzdGFu ZGluZyBpc3N1ZXMQAJ8PBAAAAAEAAAAAAKgPhgEAAE51bWJlciBvZiBjaGFubmVscyB1c2VkIHBl ciBzZXNzaW9uIGNhbiBiZSBsYXJnZQ1EZXBlbmRzIG9uIHRpbWVvdXQgZm9yIGdlbmVyYWwgcXVl cnkNMSBjaGFubmVsIHVzZWQgZm9yIGVhY2ggMTAgc2Vjb25kcw0xODAgc2Vjb25kcyBkZWZhdWx0 IHZhbHVlDURlcGVuZHMgb24gcmF0aW8gb2YgbWF4cmF0ZSB0byBtaW5yYXRlDTggS2JwcyBpcyBk ZWZhdWx0IG1pbnJhdGUNbnVtYmVyIG9mIGdyb3VwcyA9IGxvZ18oNC8zKSAobWF4cmF0ZS9taW5y YXRlKQ1UaGlzIGlzIGFuIGlzc3VlIGZvciBBU00sIG5vdCBzbyBmb3IgU1NNDVN0YXRlIGluZm9y bWF0aW9uIGluIHJvdXRlcnMvc3dpdGNoZXMgc2hvdWxkIGJlIG9rIGZvciB0aGlzIGlmIG9rIGZv ciBvdGhlciBtdWx0aWNhc3QgYXBwcwAAoQ92AAAAMQAAAAAAAAAAACUAAAABAAAAAAA9AAAAAgAA AAAAJwAAAAEAAAAAAEkAAAACAAAAAACEAAAAAQAAAAAAMQAAAAAAAAAlAAAAAAQAAAEEPQAAAAAE AAABBCcAAAAACAAAAQhJAAAAAAgAAAEIhAAAAAAIAAABCAAAqg9iAAAApwAAAAAAAAAIAAAAAQAA AAMAAwAAAAAAAAAIAAAAAQAAAAMAEgAAAAAAAAAIAAAAAQAAAAMAHgAAAAAAAAAHAAAAAQAAAAMA AQAAAAAAAAAHAAAAAQAAAAMAhgAAAAAAAAAAAPMDFAAAAPkBAAAAAAAAAgAAAJABAAAAAAAAAACf DwQAAAAAAAAAAACoDxgAAABXRUJSQyBvdXRzdGFuZGluZyBpc3N1ZXMQAJ8PBAAAAAEAAAAAAKgP MgIAAEFiaWxpdHkgb2YgbmV0d29ya2luZyBpbmZyYXN0cnVjdHVyZSB0byBzY2FsZSB0byBoYW5k bGUgcmVjZWl2ZXJzIHRoYXQgc2VuZCBvbmUgSUdNUCBqb2luIGFuZCBsZWF2ZSBtZXNzYWdlIGVh Y2ggMTAgc2Vjb25kcw1QZXJmb3JtYW5jZSBmaWd1cmVzIG9uIHRoZSBhbW91bnQgb2YgSUdNUCBt ZXNzYWdlcyByb3V0ZXJzL3N3aXRjaGVzIGNhbiBoYW5kbGUNV2hlbiBoZWF2eSBkYXRhIHBhY2tl dCBsb2FkDVdoYXQgaW1wYWN0IG9uIGRhdGEgcGFja2V0cz8NV2hhdCBpbXBhY3Qgb24gSUdNUCBs YXRlbmN5DVNhbWUgcXVlc3Rpb25zIGZvciByZXN1bHRpbmcgUElNIFNNIG1lc3NhZ2VzDU5vIHBl cmZvcm1hbmNlIGZpZ3VyZXMgeWV0IGZyb20gdmVuZG9ycw1EaWdpdGFsIEZvdW50YWluIHJlcGxp Y2F0b3IgcGVyZm9ybWFuY2UgZmlndXJlcw1TaW1pbGFyIHRvIElHTVAgam9pbiBtZXNzYWdlIHBy b2Nlc3NpbmcNMTkgbXMgbGF0ZW5jeSBmb3Igam9pbiBtZXNzYWdlcyB3aGVuIDEgcmVjZWl2ZXIN MjAgbXMgbGF0ZW5jeSBmb3Igam9pbiBtZXNzYWdlcyB3aGVuIDEwLDAwMCByZWNlaXZlcnMAAKEP kAAAAHwAAAAAAAAQAABOAE8AAAABAAAQAABOAFUAAAACAAAQAABOAC0AAAABAAAQAABOAFgAAAAA AAAQAABOAI4AAAABAAAQAABOAHwAAAAAAAAATwAAAAAEAAABBFUAAAAABAAAAQQtAAAAAAgAAAEI KAAAAAAIBAABCP8AAP4wAAAAAAgAAAEIjgAAAAAIAAABCAAAqg8aAAAAhgEAAAAAAAALAAAAAQAA AAMAogAAAAAAAAAAAPMDFAAAAPoBAAAAAAAAAgAAAJEBAAAAAAAAAACfDwQAAAAAAAAAAACoDxgA AABXRUJSQyBvdXRzdGFuZGluZyBpc3N1ZXMQAJ8PBAAAAAEAAAAAAKgPFgEAAExvbmcgSUdNUCBh bmQgUElNIFNNIGpvaW4gbGF0ZW5jeSBjYW4gYWR2ZXJzZWx5IGFmZmVjdCB0aHJvdWdocHV0DUxv c3NlcyBkdWUgdG8gYmVoYXZpb3Igb2YgbXVsdGljYXN0IHByb3RvY29sIGNhbiBhZHZlcnNlbHkg YWZmZWN0IHRocm91Z2hwdXQNUlBzDVN3aXRjaG92ZXIgZnJvbSAoKixHKSB0byAoUyxHKQ1NU0RQ DVJlY29tbWVuZGF0aW9ucz8NVXNlIFNTTQ1JZiB1c2luZyBBU00NUGxhY2UgUlAgbmVhciBzZW5k ZXINRG8gbm90IHVzZSBNU0RQIGlmIGl0IGNhbiBiZSBhdm9pZGVkAAChD2AAAACPAAAAAAAAAAAA KAAAAAEAAAAAABEAAAAAAAAAAAAVAAAAAQAAAAAAOgAAAAIAAAAAAI8AAAAAAAAAKAAAAAAAAAAR AAAAAAQAAAEEFQAAAAAIAAABCDoAAAAADAAAAQwAAKoPGgAAAI8AAAAAAAAABAAAAAEAAAADAIQA AAAAAAAAAADzAxQAAAD0AQAAAAAAAAIAAACOAQAAAAAAAAAAnw8EAAAAAAAAAAAAqA8NAAAAV0VC UkMgcm9hZG1hcBAAnw8EAAAAAQAAAAAAoA+uAQAARgB1AGwAbAAgAFQAZQBjAGgAbgBpAGMAYQBs ACAAcABhAHAAZQByAA0AVABhAHIAZwBlAHQAIABkAGEAdABlACAAQQBwAHIAaQBsACAAMQANAG4A cwAgAGMAbwBkAGUAIABhAG4AZAAgAHcAZQBiAHMAaQB0AGUAIABvAGYAIAByAGUAcwB1AGwAdABz ACAAYQB2AGEAaQBsAGEAYgBsAGUADQBUAGEAcgBnAGUAdAAgAGQAYQB0AGUAIABBAHAAcgBpAGwA IAAxAA0ARAB1AG0AbQB5ACAAbgBlAHQAIABzAGkAbQB1AGwAYQB0AGkAbwBuAHMADQBUAGEAcgBn AGUAdAAgAGQAYQB0AGUAIABKAHUAbgBlACAAMQANAFIAZQBhAGwAIAB3AG8AcgBsAGQAIABlAHgA cABlAHIAaQBtAGUAbgB0AHMADQBUAGEAcgBnAGUAdAAgAGQAYQB0AGUAIABBAHUAZwB1AHMAdAAg ADEADQBSAEYAQwAgAHQAYQByAGcAZQB0ACAAEyAgAFMAZQBwAHQAZQBtAGIAZQByACAAMgAwADAA MgANAAAAoQ+6AAAAFQAAAAAAAAAAABQAAAABAAAAAAApAAAAAAAAAAAAFAAAAAEAAAAAABYAAAAA AAAAAAATAAAAAQAAAAAAFwAAAAAAAAAAABUAAAABAAAAAAAdAAAAAAAAAAAAFQAAAAAAAAAUAAAA AAAAAAMAAAACBAAAAwQmAAAAAAQAAAEEFAAAAAAEAAABBBYAAAAACAAAAQgTAAAAAAgAAAEIFwAA AAAMAAABDBUAAAAADAAAAQwdAAAAAAwAAAEMLwDwDzgAAAAAAPMDFAAAAFYBAAAAAAAAAAAAAAEB AAAAAAAAAADzAxQAAACHAQAAAAAAAAAAAAACAQAAAAAAAAEAAQRQAAAAAAAAAf///38AAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAEgAAAOoDAAAAAA8A7gMQAgAAAgDvAxgAAAABAAAADQ4AAAAAAAAAAACAAAAAAAcAAAAP AAwEwAEAAA8AAvC4AQAA8AAI8AgAAAADAAAAA1gIAA8AA/BQAQAADwAE8CgAAAABAAnwEAAAAAAA AAAAAAAAAAAAAAAAAAACAArwCAAAAABYCAAFAAAADwAE8H4AAAASAArwCAAAAAJYCAAgAgAAcwAL 8CoAAAAEAAAAAAB/AAAABACAAISubAO/AQAAAQD/AQAAAQABAwMEAACIAwAAAAAAABDwCAAAAJAA oAJAFNACDwAR8BAAAAAAAMMLCAAAAAAAAAANAGwDDwAN8AwAAAAAAJ4PBAAAAAAAAAAPAATwkgAA ABIACvAIAAAAA1gIACACAABzAAvwKgAAAAQAAAAAAH8AAAAEAIAAZLRsA78BAAABAP8BAAABAAED AgQAAIgDAAAAAAAAEPAIAAAA8APgAdAUwA8PABHwEAAAAAAAwwsIAAAAAQAAAA4AbAMPAA3wIAAA AAAAng8EAAAAAQAAAAAApg8MAAAA2AAAANgA1AHwAxAFDwAE8EgAAAASAArwCAAAAAFYCAAADAAA gwAL8DAAAACBAQAAAAiDAQUAAAiTAY6fiwCUAd69aAC/ARIAEgD/AQAACAAEAwkAAAA/AwEAAQAQ APAHIAAAAP///wAAAAAAgICAAAAAAAAAzJkAMzPMAMzM/wCysrIAAAByFxAAAAABABAAWl8AAPoB EABilQAAAAD1DxwAAACRAQAArxEAAzZfAAB6lwAAAQAAAPsBAAABAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAABAAAAAgAAAAMAAAAEAAAABQAAAAYAAAAHAAAACAAAAAkAAAAKAAAACwAAAAwAAAANAAAADgAA AA8AAAAQAAAAEQAAABIAAAATAAAAFAAAABUAAAAWAAAAFwAAABgAAAAZAAAAGgAAABsAAAAcAAAA HQAAAB4AAAAfAAAAIAAAACEAAAAiAAAAIwAAACQAAAAlAAAAJgAAACcAAAAoAAAAKQAAACoAAAAr AAAALAAAAP7///8uAAAALwAAADAAAAAxAAAAMgAAADMAAAA0AAAANQAAADYAAAA3AAAAOAAAADkA AAD+////OwAAAP7///////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////// //7/AAAFAAIAAAAAAAAAAAAAAAAAAAAAAAEAAADghZ/y+U9oEKuRCAArJ7PZMAAAAPwKAAARAAAA AQAAAJAAAAACAAAAmAAAAAMAAADQAAAABAAAAOwAAAAFAAAABAEAAAYAAAAQAQAABwAAABwBAAAI AAAAYAEAAAkAAAB4AQAAEgAAAIQBAAAKAAAAqAEAAAsAAAC0AQAADAAAAMABAAANAAAAzAEAAA4A AADYAQAADwAAAOABAAARAAAA6AEAAAIAAADkBAAAHgAAAC0AAABJRVRGIExDVCwgRkVDIEJCLCBG RUMgSW5mbywgTENDIFByZXNlbnRhdGlvbgByb3MeAAAAEwAAAElFVEYgRGVjZW1iZXIgMjAwMABF HgAAAA0AAABNaWNoYWVsIEx1YnkAIDIwHgAAAAEAAAAAaWNoHgAAAAEAAAAAaWNoHgAAADkAAABB bnRpIDA6RG91ZyB3b3JrOlByZXNlbnRhdGlvbnM6UE9EQyB0YWxrIChyZXZpc2VkIDEwLzk2KQBQ b2keAAAADQAAAE1pY2hhZWwgTHVieQBvcmseAAAABAAAADE2MgAeAAAAGQAAAE1pY3Jvc29mdCBQ b3dlclBvaW50IDQuMAB0aW9AAAAAaMQOY6usEgBAAAAAwOgE5hGGvgFAAAAAAHQrX8/IvAFAAAAA oNW68GTPwQEDAAAAIwAAAAMAAABiAgAARwAAAAoJAAD/////AwAAAAgAiRBnDAAAAQAJAAADfQQA AAYAWAAAAAAAEQAAACYGDwAYAP////8AABAAAAAAAAAAAADAAwAA0AIAAAkAAAAmBg8ACAD///// AgAAABcAAAAmBg8AIwD/////BAAbAFROUFAUAEQBugAyAAAA//9PABQAAABNAGkAAAAKAAAAJgYP AAoAVE5QUAAAAgD0AwkAAAAmBg8ACAD/////AwAAAA8AAAAmBg8AFABUTlBQBAAMAAEAAAABAAAA AAAAAAUAAAALAgAAAAAFAAAADALQAsADBQAAAAQBDQAAAAcAAAD8AgAA////AAAABAAAAC0BAAAI AAAA+gIFAAEAAAAAAAAABAAAAC0BAQAEAAAALQEAAAkAAAAdBiEA8ADQAsADAAAAAAQAAAAtAQAA BwAAAPwCAAD///8AAAAEAAAALQECAAQAAADwAQAACAAAAPoCAAAAAAAAAAAAAAQAAAAtAQAABwAA APwCAQAAAAAAAAAEAAAALQEDAAQAAAAtAQEABwAAABsEuQKSA6AC2AIEAAAALQECAAQAAAAtAQAA HAAAAPsCAAAAAAAAAAAAAAAAAAAAAAAAAPISAJJx9XdAAAAA6AMKNkxT9XdVU/V3AQAAAAAAMAAE AAAALQEEAAUAAAAJAgAAAAIFAAAAFAIAAAAAHAAAAPsC8/8AAAAAAAC8AgAAAAAAQAAiQXJpYWwA 9XdAAAAA2gQKnExT9XdVU/V3AQAAAAAAMAAEAAAALQEFAAQAAADwAQQABQAAAAkCAGthAgUAAAAU AgAAAAAFAAAALgEYAAAABQAAAAIBAQAAAAkAAAAyCrACgQMBAAAAMQAHAAUAAAAuAQEAAAAFAAAA AgECAAAABQAAAAIBAgAAAAQAAAAtAQMABAAAAC0BAQAHAAAAGwTpAHkDcABIAAQAAAAtAQIABAAA AC0BAAAFAAAACQIAa2ECBQAAABQCAAAAABwAAAD7Asv/AAAAAAAAvAIAAAAAAEAAIkFyaWFsAPV3 QAAAAOgDCjlMU/V3VVP1dwEAAAAAADAABAAAAC0BBAAEAAAA8AEFAAUAAAAJAgY96AIFAAAAFAIA AAAABQAAAC4BGAAAAAUAAAACAQEAAAAQAAAAMgqiAH4BBgAAAEZFQyBCQiAAIwAmABAAJgAmAAUA AAAuAQEAAAAFAAAAAgECAAAABQAAAAkCBj3oAgUAAAAUAgAAAAAFAAAALgEYAAAABQAAAAIBAQAA ABMAAAAyCtsAZAEIAAAARkVDIElORk8gACMAJgAPABAAJgAhACkABQAAAC4BAQAAAAUAAAACAQIA AAAFAAAACQIGPegCBQAAABQCAAAAAAUAAAAuARgAAAAFAAAAAgEBAAAAEAAAADIKEwF/AQYAAABM Q1QgQkIgACYAIAAQACYAJgAFAAAALgEBAAAABQAAAAIBAgAAAAUAAAACAQIAAAAEAAAALQEDAAQA AAAtAQEABwAAABsEUQJ5AxgBSAAEAAAALQECAAQAAAAtAQAABQAAAAkCBj3oAgUAAAAUAgAAAAAc AAAA+wLg/wAAAAAAALwCAAAAAABAACJBcmlhbAD1d0AAAABxBQoKTFP1d1VT9XcBAAAAAAAwAAQA AAAtAQUABAAAAPABBAAFAAAACQIAAAACBQAAABQCAAAAAAUAAAAuARgAAAAFAAAAAgEBAAAAQAAA ADIKZgGdACYAAABBbGwgc2VudCB0byBJRVNHIGZvciBjb25zaWRlcmF0aW9uIGFzIBcACQAKAAgA EgASABMACwAJAAsAEwAJAAgAFQAVABkACQALABMADQAJABIAEwATABIACgATABIADQASAAoACgAT ABMACQASABIACQAFAAAALgEBAAAABQAAAAIBAgAAAAUAAAAJAgAAAAIFAAAAFAIAAAAABQAAAC4B GAAAAAUAAAACAQEAAAANAAAAMgpmAdECBAAAAFJGQ3MWABQAFwASAAUAAAAuAQEAAAAFAAAAAgEC AAAABQAAAAkCAAAAAgUAAAAUAgAAAAAFAAAALgEYAAAABQAAAAIBAQAAADkAAAAyCpMBxQAhAAAA RkVDIEJCIGFuZCBMQ1QgQkIgYXMgRXhwZXJpbWVudGFsABQAFQAXAAkAFwAWAAkAEgAUABMACQAU ABcAFAAJABcAFgAJABIAEgAJABUAEgATABIADQAKABwAEgATAAsAEQAKAAUAAAAuAQEAAAAFAAAA AgECAAAABQAAAAkCAAAAAgUAAAAUAgAAAAAFAAAALgEYAAAABQAAAAIBAQAAAC0AAAAyCsABFQEZ AAAARkVDIElORk8gYXMgSW5mb3JtYXRpb25hbAAUABUAFwAJAAkAFgAUABgACQASABIACQAJABMA CwATAA0AHAASAAsACgATABMAEgAKAAUAAAAuAQEAAAAFAAAAAgECAAAABQAAAAkCAAAAAgUAAAAU AgAAAAAFAAAALgEYAAAABQAAAAIBAQAAAFgAAAAyCu0BUgA2AAAAKGFsdGhvdWdoIGZvciBzb21l IHJlYXNvbiBzdGlsbCBub3QgaW4gdGhlIFJGQyBFZGl0b3IgCwARAAoACwATABMAEwATABMACQAL ABMADQAJABIAFAAcABIACQANABIAEgASABMAEwAJABIACgAJAAkACgAJABMAEwALAAgACgATAAkA CwATABIACQAWABQAFwAJABUAEwAKAAsAEwANAAgABQAAAC4BAQAAAAUAAAACAQIAAAAFAAAACQIA AAACBQAAABQCAAAAAAUAAAAuARgAAAAFAAAAAgEBAAAAEAAAADIKDwK7AQYAAABRdWV1ZSkYABMA EwATABIACwAFAAAALgEBAAAABQAAAAIBAgAAAAUAAAACAQIAAAAEAAAALQEBAAQAAAAtAQMAHAAA APsCEAAHAAAAAAC8AgAAAAABAgIiU3lzdGVtAAAAAAoAAAAEAAAAAAAEAAAAAQAAAAAAMAAEAAAA LQEEAAQAAADwAQUADwAAACYGDwAUAFROUFAEAAwAAAAAAAAAAAAAAAAACQAAACYGDwAIAP////8B AAAAAwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP7/AAAFAAIAAAAAAAAAAAAAAAAAAAAAAAEA AAAC1c3VnC4bEJOXCAArLPmuMAAAAAADAAAQAAAAAQAAAIgAAAADAAAAkAAAAA8AAACoAAAABAAA AMQAAAAGAAAAzAAAAAcAAADUAAAACAAAANwAAAAJAAAA5AAAAAoAAADsAAAAFwAAAPQAAAALAAAA /AAAABAAAAAEAQAAEwAAAAwBAAAWAAAAFAEAAA0AAAAcAQAADAAAAKACAAACAAAA5AQAAA== ------=_NextPart_000_0000_01C1D0F1.9F8788C0 Content-Type: application/postscript; name="IETF MAR 2002 overall.ps" Content-Disposition: attachment; filename="IETF MAR 2002 overall.ps" Content-Transfer-Encoding: quoted-printable =1B%-12345X@PJL JOB=0A= @PJL SET RESOLUTION =3D 600=0A= @PJL SET ECONOMODE =3D OFF=0A= @PJL ENTER LANGUAGE =3D POSTSCRIPT =0A= %!PS-Adobe-3.0=0A= %%Title: Microsoft PowerPoint - IETF MAR 2002 overall.ppt=0A= %%Creator: Windows NT 4.0=0A= %%CreationDate: 15:56 3/21/2002=0A= %%Pages: (atend)=0A= %%BoundingBox: 13 12 780 598=0A= %%LanguageLevel: 2=0A= %%DocumentNeededFonts: (atend)=0A= %%DocumentSuppliedFonts: (atend)=0A= %%EndComments=0A= %%BeginProlog=0A= %%BeginResource: procset NTPSOct95 /NTPSOct95 100 dict dup begin/bd{bind def}bind def/ld{load = def}bd/ed{exch def} bd/a{currentpoint}bd/c/curveto ld/d/dup ld/e/eofill ld/f/fill = ld/tr/translate ld/gr/grestore ld/gs/gsave ld/j/setlinejoin ld/L/lineto ld/M/moveto ld/n /newpath ld/cp/closepath ld/rm/rmoveto ld/sl/setlinewidth ld/sd/setdash = ld/g /setgray ld/r/setrgbcolor ld/s/stroke ld/t/show ld/aw/awidthshow ld/im /imagemask ld/MS{moveto show}bd/SF{findfont exch scalefont = setfont}bd/SM{cmtx setmatrix}bd/MF{findfont exch makefont setfont}bd/CM{/cmtx matrix = currentmatrix def}bd/B{M exch dup 0 rlt exch 0 exch rlt neg 0 rlt}bd/CB{B cp = eoclip}bd/EA{1 index 0/G0 put 4 string 1 1 4 -1 roll{3 copy neg exch cvs dup 0 71 put = cvn 3 -1 roll exch put}for pop}bd/rlt/rlineto ld/L2?/languagelevel where{pop languagelevel 2 ge}{false}ifelse def end def=20 %%EndResource %%EndProlog=0A= %%BeginSetup=0A= [{0=0A= /languagelevel where{pop languagelevel 2 ge}{false}ifelse=0A= {1 dict dup/JobTimeout 4 -1 roll put setuserparams}=0A= {statusdict/setjobtimeout get exec}ifelse=0A= }stopped cleartomark=0A= [{120=0A= /languagelevel where{pop languagelevel 2 ge}{false}ifelse=0A= {1 dict dup/WaitTimeout 4 -1 roll put setuserparams}=0A= {statusdict/waittimeout 3 -1 roll put}ifelse=0A= }stopped cleartomark=0A= /#copies 1 def=0A= [{=0A= %%BeginFeature: *HPPaperPolicy PromptUser=0A= <> setpagedevice=0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *HPHalftone PrinterDefault=0A= =0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *HPCollate True=0A= <> setpagedevice=0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *Smoothing True=0A= << /PostRenderingEnhance true /PostRenderingEnhanceDetails << /REValue = 2 /Type 8 >> >> setpagedevice=0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *PageSize Letter=0A= <> setpagedevice=0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *MediaType None=0A= =0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *Duplex None=0A= <> setpagedevice=0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *HPwmFont HelveticaB=0A= /Helvetica-Bold findfont dup length dict begin {1 index /FID ne {def} {pop pop} ifelse} forall /Encoding ISOLatin1Encoding def currentdict=20 end /HPwmFont exch definefont pop=0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *HPwmFontSize pt48=0A= userdict /HPwmSize 48 put=0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *HPwmTextStyle Medium=0A= userdict /HPwmStyle .48 put=0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *HPwmTextAngle Deg45=0A= userdict /HPwmAngle 45 put=0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *HPwmText None=0A= =0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *HPwmLocation True=0A= userdict /HPwmLocation true put=0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *HPNup OneUp=0A= % Copyright (c) Hewlett-Packard Co 1997 userdict begin true setglobal /HPwm 5 dict dup begin /HPwmOn true def /HPwmOdd true = def end def false setglobal userdict /HPwmAngle known not {/HPwmAngle 45 def} if userdict /HPwmSize known not {/HPwmSize 48 def} if userdict /HPwmLocation known not {/HPwmLocation true def} if userdict /HPwmStyle known not {/HPwmStyle .48 def} if userdict /HPwmDuplex known not {/HPwmDuplex 0 def} if /HPwmEOP {HPwmDuplex 0 eq {true}{HPwmDuplex 1 eq HPwmOdd eq dup not = {erasepage}if true setglobal /HPwmOdd HPwmOdd not def false setglobal}ifelse} bind = def end << /EndPage {userdict begin userdict /HPwmText known HPwm /HPwmOn get and {initmatrix 0 setgray 1 setlinewidth true setstrokeadjust 0 setlinejoin 0 = setlinecap [] 0 setdash currentpagedevice /PageSize get aload pop 2 div exch 2 div exch = translate HPwmAngle rotate /HPwmFont HPwmSize selectfont HPwmText stringwidth 2 div neg exch 2 div neg exch HPwmSize .25 mul = sub moveto HPwmText false charpath userdict /HPwmStyle1 known {gsave 1 setgray HPwmStyle1 HPwmStyle add setlinewidth stroke = grestore} if 0 setgray HPwmStyle setlinewidth stroke HPwmLocation not {true setglobal HPwm /HPwmOn false put false = setglobal} if } if 2 eq {pop false}{pop HPwm begin HPwmEOP end} ifelse end } bind >> setpagedevice=0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *HPScalePatterns Scale=0A= /dtransform {GDIBWPatternDict /Width known {currentpagedevice = /HWResolution get 0 get 150 div mul exch currentpagedevice /HWResolution get 0 get 150 div mul = exch dtransform}{dtransform}ifelse} bind def=0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *Option1 False=0A= =0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *Option2 False=0A= =0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *Option3 True=0A= =0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *Option4 False=0A= =0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *Option5 False=0A= =0A= %%EndFeature=0A= } stopped cleartomark=0A= [{=0A= %%BeginFeature: *VMOption 4MB=0A= =0A= %%EndFeature=0A= } stopped cleartomark=0A= %%EndSetup=0A= NTPSOct95 begin=0A= %%Page: 1 1=0A= NTPSOct95 /PageSV save put=0A= 90 rotate 0 -612 translate=0A= 13.059 598.762 translate 72 600 div dup neg scale=0A= 0 0 transform .25 add round .25 sub exch .25 add round .25 sub exch = itransform translate=0A= n=0A= 5999 4499 195 193 B=0A= 1 g=0A= f=0A= %%IncludeFont: Helvetica-Bold=0A= [83 0 0 -83 0 0]/Helvetica-Bold MF=0A= 0 g=0A= (1)5797 4495 MS=0A= [333 0 0 -333 0 0]/Helvetica-Bold MF=0A= (F)2575 1208 MS (E)2778 1208 MS (C)3000 1208 MS ( )3240 1208 MS (B)3333 = 1208 MS (B)3574 1208 MS=0A= (F)2417 1560 MS (E)2620 1560 MS (C)2842 1560 MS ( )3082 1560 MS (I)3175 = 1560 MS (N)3268 1560 MS (F)3508 1560 MS (O)3711 1560 MS=0A= (L)2584 1912 MS (C)2787 1912 MS (T)3027 1912 MS ( )3230 1912 MS (B)3323 = 1912 MS (B)3564 1912 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (A)1173 2426 MS (l)1317 2426 MS (l)1373 2426 MS ( )1428 2426 MS (s)1484 = 2426 MS (e)1595 2426 MS (n)1706 2426 MS (t)1828 2426 MS ( )1895 2426 MS = (t)1951 2426 MS (o)2018 2426 MS ( )2139 2426 MS (I)2194 2426 MS (E)2250 = 2426 MS (S)2383 2426 MS (G)2516 2426 MS=0A= ( )2671 2426 MS (f)2727 2426 MS (o)2794 2426 MS (r)2916 2426 MS ( )2993 = 2426 MS (c)3049 2426 MS (o)3160 2426 MS (n)3282 2426 MS (s)3404 2426 MS = (i)3515 2426 MS (d)3571 2426 MS (e)3693 2426 MS (r)3804 2426 MS (a)3882 = 2426 MS (t)3993 2426 MS (i)4060 2426 MS=0A= (o)4115 2426 MS (n)4237 2426 MS ( )4359 2426 MS (a)4415 2426 MS (s)4526 = 2426 MS ( )4637 2426 MS=0A= (R)4693 2426 MS (F)4837 2426 MS (C)4959 2426 MS (s)5104 2426 MS=0A= (F)1428 2708 MS (E)1550 2708 MS (C)1683 2708 MS ( )1827 2708 MS (B)1883 = 2708 MS (B)2028 2708 MS ( )2172 2708 MS (a)2228 2708 MS (n)2339 2708 MS = (d)2461 2708 MS ( )2583 2708 MS (L)2639 2708 MS (C)2761 2708 MS (T)2905 = 2708 MS ( )3027 2708 MS (B)3083 2708 MS=0A= (B)3226 2708 MS ( )3370 2708 MS (a)3426 2708 MS (s)3537 2708 MS ( )3648 = 2708 MS (E)3704 2708 MS (x)3837 2708 MS (p)3948 2708 MS (e)4070 2708 MS = (r)4182 2708 MS (i)4260 2708 MS (m)4316 2708 MS (e)4494 2708 MS (n)4605 = 2708 MS (t)4727 2708 MS (a)4794 2708 MS=0A= (l)4905 2708 MS=0A= (F)1929 2991 MS (E)2051 2991 MS (C)2184 2991 MS ( )2328 2991 MS (I)2384 = 2991 MS (N)2440 2991 MS (F)2584 2991 MS (O)2706 2991 MS ( )2861 2991 MS = (a)2916 2991 MS (s)3027 2991 MS ( )3138 2991 MS (I)3194 2991 MS (n)3250 = 2991 MS (f)3372 2991 MS (o)3439 2991 MS=0A= (r)3561 2991 MS (m)3639 2991 MS (a)3817 2991 MS (t)3927 2991 MS (i)3994 = 2991 MS (o)4050 2991 MS (n)4172 2991 MS (a)4294 2991 MS (l)4404 2991 MS=0A= (\()708 3273 MS (a)775 3273 MS (l)886 3273 MS (t)942 3273 MS (h)1009 = 3273 MS (o)1131 3273 MS (u)1253 3273 MS (g)1374 3273 MS (h)1496 3273 MS = ( )1618 3273 MS (f)1673 3273 MS (o)1740 3273 MS (r)1862 3273 MS ( )1940 = 3273 MS (s)1996 3273 MS (o)2107 3273 MS=0A= (m)2229 3273 MS (e)2407 3273 MS ( )2518 3273 MS (r)2574 3273 MS (e)2651 = 3273 MS (a)2762 3273 MS (s)2873 3273 MS (o)2984 3273 MS (n)3106 3273 MS = ( )3228 3273 MS (s)3284 3273 MS (t)3395 3273 MS (i)3462 3273 MS (l)3518 = 3273 MS (l)3573 3273 MS ( )3628 3273 MS=0A= (n)3684 3273 MS (o)3806 3273 MS (t)3928 3273 MS ( )3994 3273 MS (i)4049 = 3273 MS (n)4105 3273 MS ( )4226 3273 MS (t)4282 3273 MS (h)4349 3273 MS = (e)4471 3273 MS ( )4581 3273 MS (R)4636 3273 MS (F)4780 3273 MS (C)4902 = 3273 MS ( )5046 3273 MS (E)5102 3273 MS=0A= (d)5235 3273 MS (i)5357 3273 MS (t)5413 3273 MS (o)5480 3273 MS (r)5602 = 3273 MS ( )5680 3273 MS=0A= (Q)2961 3483 MS (u)3117 3483 MS (e)3239 3483 MS (u)3350 3483 MS (e)3472 = 3483 MS (\))3583 3483 MS=0A= showpage=0A= %%Page: 2 2=0A= 90 rotate 0 -612 translate=0A= 13.059 598.762 translate 72 600 div dup neg scale=0A= 0 0 transform .25 add round .25 sub exch .25 add round .25 sub exch = itransform translate=0A= n=0A= 5999 4499 195 193 B=0A= 1 g=0A= f=0A= [83 0 0 -83 0 0]/Helvetica-Bold MF=0A= 0 g=0A= (2)5797 4495 MS=0A= [333 0 0 -333 0 0]/Helvetica-Bold MF=0A= (A)1583 1208 MS (L)1823 1208 MS (C)2026 1208 MS ( )2266 1208 MS (P)2359 = 1208 MS (I)2582 1208 MS ( )2675 1208 MS (s)2768 1208 MS (t)2953 1208 MS = (a)3064 1208 MS (t)3249 1208 MS (u)3360 1208 MS (s)3564 1208 MS ( )3749 = 1208 MS (r)3842 1208 MS (e)3972 1208 MS=0A= (p)4157 1208 MS (o)4360 1208 MS (r)4564 1208 MS (t)4694 1208 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (M)2008 2426 MS (.)2175 2426 MS ( )2230 2426 MS (L)2286 2426 MS (u)2408 = 2426 MS (b)2530 2426 MS (y)2652 2426 MS (,)2761 2426 MS ( )2817 2426 MS = (D)2873 2426 MS (i)3017 2426 MS (g)3073 2426 MS (i)3194 2426 MS (t)3250 = 2426 MS (a)3317 2426 MS (l)3428 2426 MS=0A= ( )3483 2426 MS (F)3539 2426 MS (o)3661 2426 MS (u)3782 2426 MS (n)3904 = 2426 MS (t)4025 2426 MS (a)4092 2426 MS (i)4203 2426 MS (n)4259 2426 MS=0A= (J)2156 2708 MS (.)2267 2708 MS ( )2323 2708 MS=0A= (G)2378 2708 MS (e)2534 2708 MS (m)2645 2708 MS (m)2823 2708 MS (e)3001 = 2708 MS (l)3112 2708 MS (l)3167 2708 MS=0A= (,)3222 2708 MS ( )3278 2708 MS (M)3333 2708 MS (i)3499 2708 MS (c)3555 = 2708 MS (r)3666 2708 MS (o)3744 2708 MS (s)3866 2708 MS (o)3977 2708 MS = (f)4099 2708 MS (t)4166 2708 MS=0A= (L)2340 2991 MS (.)2462 2991 MS ( )2517 2991 MS=0A= (V)2573 2991 MS (i)2706 2991 MS (c)2762 2991 MS (i)2872 2991 MS (s)2928 = 2991 MS (a)3039 2991 MS (n)3150 2991 MS (o)3272 2991 MS=0A= (,)3394 2991 MS ( )3449 2991 MS (C)3505 2991 MS (i)3649 2991 MS (s)3705 = 2991 MS (c)3816 2991 MS (o)3927 2991 MS=0A= (L)1924 3273 MS (.)2046 3273 MS ( )2101 3273 MS (R)2157 3273 MS (i)2301 = 3273 MS (z)2357 3273 MS (z)2457 3273 MS (o)2557 3273 MS (,)2678 3273 MS = ( )2734 3273 MS (A)2790 3273 MS (C)2933 3273 MS (I)3077 3273 MS (R)3133 = 3273 MS (I)3277 3273 MS ( )3333 3273 MS=0A= (a)3389 3273 MS (n)3500 3273 MS (d)3622 3273 MS ( )3744 3273 MS (U)3800 = 3273 MS (.)3943 3273 MS ( )3998 3273 MS (P)4054 3273 MS (i)4187 3273 MS = (s)4243 3273 MS (a)4354 3273 MS=0A= (J)2350 3555 MS (.)2461 3555 MS ( )2517 3555 MS=0A= (C)2573 3555 MS (r)2717 3555 MS (o)2795 3555 MS (w)2916 3555 MS (c)3073 = 3555 MS (r)3184 3555 MS (o)3261 3555 MS (f)3383 3555 MS (t)3450 3555 MS=0A= (,)3516 3555 MS ( )3572 3555 MS (U)3628 3555 MS (C)3772 3555 MS (L)3916 = 3555 MS=0A= showpage=0A= %%Page: 3 3=0A= 90 rotate 0 -612 translate=0A= 13.059 598.762 translate 72 600 div dup neg scale=0A= 0 0 transform .25 add round .25 sub exch .25 add round .25 sub exch = itransform translate=0A= n=0A= 5999 4499 195 193 B=0A= 1 g=0A= f=0A= [83 0 0 -83 0 0]/Helvetica-Bold MF=0A= 0 g=0A= (3)5797 4495 MS=0A= [300 0 0 -300 0 0]/Helvetica-Bold MF=0A= (P)1511 741 MS (r)1711 741 MS (o)1828 741 MS (p)2011 741 MS (o)2194 741 = MS (s)2378 741 MS (e)2545 741 MS (d)2712 741 MS ( )2895 741 MS (c)2978 = 741 MS (h)3145 741 MS (a)3328 741 MS (n)3495 741 MS (g)3678 741 MS = (e)3861 741 MS (s)4028 741 MS=0A= ( )4195 741 MS (t)4278 741 MS (o)4378 741 MS ( )4561 741 MS (0)4644 741 = MS (6)4811 741 MS=0A= NTPSOct95 /FontSV save put=0A= %%BeginFont: Wingdings02009940=0A= %!FontType1-1.0: Wingdings02009940=0A= % Copyright (c) 1993 Microsoft Corporation.=0A= 11 dict begin=0A= /FontInfo 8 dict dup begin=0A= /FullName (Wingdings_Regular:_MS:_1995) def=0A= /FamilyName (Wingdings) def=0A= /Weight (Light) def=0A= /ItalicAngle 0 def=0A= /isFixedPitch false def=0A= /UnderlinePosition -200 def=0A= /UnderlineThickness 100 def=0A= end def=0A= /FontName /Wingdings02009940 def=0A= /PaintType 0 def=0A= /FontType 1 def=0A= /FontMatrix [1 2048 div 0 0 1 2048 div 0 0] def=0A= /FontBBox{0 -431 2786 1843}def=0A= /Encoding 256 array 225 EA=0A= 226 1 255 {1 index exch /.notdef put} for=0A= def currentdict end=0A= currentfile eexec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cleartomark=0A= %%EndFont=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 1324 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (C)942 1324 MS (l)1086 1324 MS (a)1142 1324 MS (r)1253 1324 MS (i)1331 = 1324 MS (f)1387 1324 MS (y)1455 1324 MS ( )1564 1324 MS (t)1620 1324 MS = (h)1687 1324 MS (a)1809 1324 MS (t)1920 1324 MS ( )1987 1324 MS (s)2043 = 1324 MS (e)2154 1324 MS (s)2265 1324 MS=0A= (s)2376 1324 MS (i)2487 1324 MS (o)2543 1324 MS (n)2665 1324 MS ( )2787 = 1324 MS (i)2842 1324 MS (s)2897 1324 MS ( )3008 1324 MS (f)3064 1324 MS = (r)3131 1324 MS (o)3209 1324 MS (m)3331 1324 MS ( )3509 1324 MS (a)3565 = 1324 MS ( )3675 1324 MS (s)3731 1324 MS=0A= (i)3842 1324 MS (n)3897 1324 MS (g)4019 1324 MS (l)4141 1324 MS (e)4197 = 1324 MS ( )4308 1324 MS (s)4364 1324 MS (e)4475 1324 MS (n)4586 1324 MS = (d)4708 1324 MS (e)4829 1324 MS (r)4940 1324 MS (,)5018 1324 MS ( )5074 = 1324 MS (a)5130 1324 MS (n)5241 1324 MS=0A= (d)5363 1324 MS ( )5484 1324 MS=0A= (t)942 1510 MS (h)1009 1510 MS (a)1131 1510 MS (t)1242 1510 MS ( )1309 = 1510 MS (r)1365 1510 MS (e)1443 1510 MS (c)1554 1510 MS (e)1665 1510 MS = (i)1776 1510 MS (v)1832 1510 MS (i)1943 1510 MS (n)1999 1510 MS (g)2121 = 1510 MS ( )2242 1510 MS (f)2298 1510 MS=0A= (r)2365 1510 MS (o)2443 1510 MS (m)2565 1510 MS ( )2742 1510 MS (m)2797 = 1510 MS (u)2975 1510 MS (l)3097 1510 MS (t)3153 1510 MS (i)3219 1510 MS = (p)3275 1510 MS (l)3396 1510 MS (e)3452 1510 MS ( )3563 1510 MS (s)3619 = 1510 MS (e)3729 1510 MS (n)3840 1510 MS=0A= (d)3962 1510 MS (e)4084 1510 MS (r)4195 1510 MS (s)4273 1510 MS ( )4384 = 1510 MS (i)4440 1510 MS (s)4496 1510 MS ( )4607 1510 MS (a)4662 1510 MS = (t)4773 1510 MS ( )4840 1510 MS (t)4896 1510 MS (h)4963 1510 MS (e)5085 = 1510 MS ( )5196 1510 MS=0A= (a)942 1697 MS (p)1053 1697 MS (p)1175 1697 MS (l)1297 1697 MS (i)1353 = 1697 MS (c)1409 1697 MS (a)1520 1697 MS (t)1631 1697 MS (i)1697 1697 MS = (o)1753 1697 MS (n)1874 1697 MS ( )1996 1697 MS (l)2051 1697 MS (e)2107 = 1697 MS (v)2218 1697 MS (e)2329 1697 MS=0A= (l)2440 1697 MS ( )2496 1697 MS (a)2552 1697 MS (b)2663 1697 MS (o)2784 = 1697 MS (v)2906 1697 MS (e)3017 1697 MS ( )3128 1697 MS (A)3184 1697 MS = (L)3328 1697 MS (C)3450 1697 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 1955 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (T)942 1955 MS (r)1064 1955 MS (a)1142 1955 MS (n)1253 1955 MS (s)1375 = 1955 MS (m)1486 1955 MS (i)1664 1955 MS (s)1720 1955 MS (s)1831 1955 MS = (i)1943 1955 MS (o)1999 1955 MS (n)2121 1955 MS ( )2242 1955 MS (S)2298 = 1955 MS (e)2431 1955 MS (s)2542 1955 MS=0A= (s)2653 1955 MS (i)2764 1955 MS (o)2820 1955 MS (n)2942 1955 MS ( )3064 = 1955 MS (I)3119 1955 MS (D)3175 1955 MS=0A= %%IncludeFont: Helvetica=0A= [167 0 0 -167 0 0]/Helvetica MF=0A= /IsChar{exch/CharStrings get exch known}bd/MapCh{3 -1 roll/Encoding get = 3 1 roll put}bd/MapDegree{dup 16#b0 exch/degree IsChar{/degree}{/ring}ifelse = MapCh} bd/MapBB{dup 16#a6 exch/brokenbar IsChar{/brokenbar}{/bar}ifelse = MapCh}bd /reencode{findfont begin currentdict dup length dict begin{1 index/FID = ne{def} {pop pop}ifelse}forall/FontName exch def dup length 0 ne{/Encoding = Encoding 256 array copy def 0 exch{dup type/nametype eq{Encoding 2 index 2 index put = pop 1 add}{exch pop}ifelse}forall}if pop currentdict dup end end/FontName get = exch definefont dup MapDegree = MapBB}bd/LATENC[0/grave/acute/circumflex/tilde/macron /breve/dotaccent/dieresis/ring/cedilla/hungarumlaut/ogonek/caron/dotlessi= /fi/fl /Lslash/lslash/Zcaron/zcaron/minus/.notdef/.notdef/.notdef/.notdef/.notde= f /.notdef/.notdef/.notdef/.notdef/.notdef/.notdef/space/exclam/quotedbl /numbersign/dollar/percent/ampersand/quotesingle/parenleft/parenright/ast= erisk /plus/comma/hyphen/period/slash/zero/one/two/three/four/five/six/seven/ei= ght /nine/colon/semicolon/less/equal/greater/question/at/A/B/C/D/E/F/G/H/I/J/= K/L/M /N/O/P/Q/R/S/T/U/V/W/X/Y/Z/bracketleft/backslash/bracketright/asciicircum= /underscore/grave/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/bra= celeft /bar/braceright/asciitilde/.notdef/.notdef/.notdef/quotesinglbase/florin /quotedblbase/ellipsis/dagger/daggerdbl/circumflex/perthousand/Scaron /guilsinglleft/OE/.notdef/.notdef/.notdef/.notdef/quoteleft/quoteright /quotedblleft/quotedblright/bullet/endash/emdash/tilde/trademark/scaron /guilsinglright/oe/.notdef/.notdef/Ydieresis/.notdef/exclamdown/cent/ster= ling /currency/yen/brokenbar/section/dieresis/copyright/ordfeminine/guillemotl= eft /logicalnot/hyphen/registered/macron/degree/plusminus/twosuperior/threesu= perior /acute/mu/paragraph/periodcentered/cedilla/onesuperior/ordmasculine /guillemotright/onequarter/onehalf/threequarters/questiondown/Agrave/Aacu= te /Acircumflex/Atilde/Adieresis/Aring/AE/Ccedilla/Egrave/Eacute/Ecircumflex= /Edieresis/Igrave/Iacute/Icircumflex/Idieresis/Eth/Ntilde/Ograve/Oacute /Ocircumflex/Otilde/Odieresis/multiply/Oslash/Ugrave/Uacute/Ucircumflex /Udieresis/Yacute/Thorn/germandbls/agrave/aacute/acircumflex/atilde/adier= esis /aring/ae/ccedilla/egrave/eacute/ecircumflex/edieresis/igrave/iacute /icircumflex/idieresis/eth/ntilde/ograve/oacute/ocircumflex/otilde/odiere= sis /divide/oslash/ugrave/uacute/ucircumflex/udieresis/yacute/thorn/ydieresis= ]def LATENC /_Helvetica /Helvetica reencode=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 2178 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (0)1204 2178 MS (6)1297 2178 MS (:)1390 2178 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 2394 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (T)1504 2394 MS (S)1606 2394 MS (I)1717 2394 MS ( )1763 2394 MS (e)1809 = 2394 MS (x)1902 2394 MS (p)1995 2394 MS (l)2097 2394 MS (i)2143 2394 MS = (c)2189 2394 MS (i)2282 2394 MS (t)2327 2394 MS (l)2383 2394 MS (y)2430 = 2394 MS ( )2521 2394 MS (r)2567 2394 MS=0A= (e)2632 2394 MS (q)2725 2394 MS (u)2827 2394 MS (i)2929 2394 MS (r)2975 = 2394 MS (e)3039 2394 MS (d)3132 2394 MS ( )3234 2394 MS (t)3280 2394 MS = (o)3336 2394 MS ( )3437 2394 MS (b)3482 2394 MS (e)3584 2394 MS ( )3677 = 2394 MS (i)3723 2394 MS (n)3769 2394 MS=0A= ( )3871 2394 MS (e)3917 2394 MS (a)4010 2394 MS (c)4102 2394 MS (h)4195 = 2394 MS ( )4297 2394 MS (p)4343 2394 MS (a)4444 2394 MS (c)4537 2394 MS = (k)4630 2394 MS (e)4723 2394 MS (t)4815 2394 MS (.)4871 2394 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 2611 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (P)1204 2611 MS (r)1315 2611 MS (o)1380 2611 MS (p)1482 2611 MS (o)1584 = 2611 MS (s)1686 2611 MS (e)1779 2611 MS (d)1872 2611 MS (:)1974 2611 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 2828 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (T)1504 2828 MS (S)1606 2828 MS (I)1717 2828 MS ( )1763 2828 MS (e)1809 = 2828 MS (x)1902 2828 MS (p)1995 2828 MS (l)2097 2828 MS (i)2143 2828 MS = (c)2189 2828 MS (i)2282 2828 MS (t)2327 2828 MS (l)2383 2828 MS (y)2430 = 2828 MS ( )2521 2828 MS (r)2567 2828 MS=0A= (e)2632 2828 MS (q)2725 2828 MS (u)2827 2828 MS (i)2929 2828 MS (r)2975 = 2828 MS (e)3039 2828 MS (d)3132 2828 MS ( )3234 2828 MS (t)3280 2828 MS = (o)3336 2828 MS ( )3437 2828 MS (b)3482 2828 MS (e)3584 2828 MS ( )3677 = 2828 MS (i)3723 2828 MS (n)3769 2828 MS=0A= ( )3871 2828 MS (e)3917 2828 MS (a)4010 2828 MS (c)4102 2828 MS (h)4195 = 2828 MS ( )4297 2828 MS (p)4343 2828 MS (a)4444 2828 MS (c)4537 2828 MS = (k)4630 2828 MS (e)4723 2828 MS (t)4815 2828 MS ( )4871 2828 MS (U)4917 = 2828 MS (N)5038 2828 MS (L)5159 2828 MS=0A= (E)5261 2828 MS (S)5371 2828 MS (S)5482 2828 MS ( )5593 2828 MS=0A= (t)1504 2984 MS (h)1560 2984 MS (e)1662 2984 MS ( )1755 2984 MS (s)1801 = 2984 MS (e)1893 2984 MS (n)1986 2984 MS (d)2088 2984 MS (e)2190 2984 MS = (r)2283 2984 MS ( )2348 2984 MS (i)2393 2984 MS (s)2438 2984 MS ( )2531 = 2984 MS (s)2577 2984 MS (u)2670 2984 MS=0A= (p)2772 2984 MS (p)2874 2984 MS (o)2976 2984 MS (r)3078 2984 MS (t)3142 = 2984 MS (i)3198 2984 MS (n)3244 2984 MS (g)3346 2984 MS ( )3447 2984 MS = (o)3493 2984 MS (n)3595 2984 MS (l)3697 2984 MS (y)3744 2984 MS ( )3834 = 2984 MS (o)3880 2984 MS (n)3983 2984 MS=0A= (e)4085 2984 MS ( )4178 2984 MS (s)4224 2984 MS (e)4317 2984 MS (s)4409 = 2984 MS (s)4502 2984 MS (i)4595 2984 MS (o)4641 2984 MS (n)4743 2984 MS = (,)4845 2984 MS ( )4891 2984 MS (i)4937 2984 MS (n)4983 2984 MS ( )5085 = 2984 MS (w)5128 2984 MS (h)5261 2984 MS=0A= (i)5362 2984 MS (c)5407 2984 MS (h)5500 2984 MS ( )5602 2984 MS=0A= (c)1504 3140 MS (a)1597 3140 MS (s)1690 3140 MS (e)1783 3140 MS ( )1876 = 3140 MS (T)1922 3140 MS (S)2024 3140 MS (I)2135 3140 MS ( )2181 3140 MS = (M)2226 3140 MS (A)2365 3140 MS (Y)2485 3140 MS ( )2596 3140 MS (b)2642 = 3140 MS (e)2744 3140 MS ( )2837 3140 MS=0A= (o)2883 3140 MS (m)2985 3140 MS (i)3133 3140 MS (t)3179 3140 MS (t)3234 = 3140 MS (e)3290 3140 MS (d)3383 3140 MS ( )3484 3140 MS (a)3530 3140 MS = (n)3623 3140 MS (d)3725 3140 MS ( )3827 3140 MS (i)3873 3140 MS (s)3919 = 3140 MS ( )4012 3140 MS (t)4057 3140 MS=0A= (h)4113 3140 MS (e)4215 3140 MS (n)4308 3140 MS ( )4410 3140 MS (p)4455 = 3140 MS (r)4557 3140 MS (e)4622 3140 MS (s)4715 3140 MS (u)4808 3140 MS = (m)4910 3140 MS (e)5058 3140 MS (d)5151 3140 MS ( )5253 3140 MS (t)5298 = 3140 MS (o)5354 3140 MS ( )5455 3140 MS=0A= (b)5501 3140 MS (e)5603 3140 MS ( )5696 3140 MS=0A= (0)1504 3296 MS (.)1597 3296 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 3513 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (P)1504 3513 MS (r)1615 3513 MS (o)1680 3513 MS (:)1782 3513 MS ( )1838 = 3513 MS=0A= (S)1884 3513 MS (a)1995 3513 MS (v)2088 3513 MS (e)2179 3513 MS (s)2272 = 3513 MS ( )2365 3513 MS (s)2411 3513 MS (p)2504 3513 MS (a)2606 3513 MS = (c)2699 3513 MS (e)2792 3513 MS ( )2885 3513 MS (i)2931 3513 MS (n)2977 = 3513 MS ( )3079 3513 MS (p)3125 3513 MS=0A= (a)3227 3513 MS (c)3319 3513 MS (k)3412 3513 MS (e)3504 3513 MS (t)3597 = 3513 MS ( )3653 3513 MS (h)3699 3513 MS (e)3801 3513 MS (a)3893 3513 MS = (d)3986 3513 MS (e)4088 3513 MS (r)4181 3513 MS ( )4245 3513 MS (i)4291 = 3513 MS (f)4336 3513 MS ( )4392 3513 MS=0A= (o)4437 3513 MS (n)4539 3513 MS (l)4641 3513 MS (y)4688 3513 MS ( )4779 = 3513 MS (o)4825 3513 MS (n)4927 3513 MS (e)5029 3513 MS ( )5122 3513 MS = (s)5168 3513 MS (e)5261 3513 MS (s)5354 3513 MS (s)5446 3513 MS (i)5539 = 3513 MS (o)5585 3513 MS (n)5687 3513 MS=0A= ( )5789 3513 MS=0A= (p)1504 3669 MS (e)1606 3669 MS (r)1699 3669 MS ( )1764 3669 MS (s)1810 = 3669 MS (e)1903 3669 MS (n)1996 3669 MS (d)2098 3669 MS (e)2199 3669 MS = (r)2292 3669 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 3886 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (C)1504 3886 MS (o)1625 3886 MS (n)1727 3886 MS (s)1829 3886 MS (:)1922 = 3886 MS ( )1978 3886 MS=0A= (L)2024 3886 MS (C)2126 3886 MS (T)2247 3886 MS ( )2349 3886 MS (r)2394 = 3886 MS (e)2458 3886 MS (q)2551 3886 MS (u)2653 3886 MS (i)2755 3886 MS = (r)2801 3886 MS (e)2866 3886 MS (s)2959 3886 MS ( )3052 3886 MS (T)3098 = 3886 MS (S)3200 3886 MS (I)3311 3886 MS=0A= ( )3357 3886 MS (i)3403 3886 MS (n)3448 3886 MS ( )3550 3886 MS (h)3596 = 3886 MS (e)3698 3886 MS (a)3791 3886 MS (d)3884 3886 MS (e)3986 3886 MS = (r)4079 3886 MS ( )4144 3886 MS=0A= LATENC /_Helvetica-Bold /Helvetica-Bold reencode=0A= [167 0 0 -167 0 0]/_Helvetica-Bold MF=0A= (\226)4189 3886 MS=0A= (i)4328 3886 MS (s)4374 3886 MS ( )4466 3886 MS (u)4512 3886 MS (s)4614 = 3886 MS (e)4707 3886 MS (f)4800 3886 MS (u)4856 3886 MS (l)4957 3886 MS = ( )5003 3886 MS (f)5048 3886 MS (o)5104 3886 MS (r)5206 3886 MS ( )5271 = 3886 MS=0A= (n)1504 4042 MS (e)1606 4042 MS (t)1699 4042 MS (w)1752 4042 MS (o)1885 = 4042 MS (r)1986 4042 MS (k)2051 4042 MS ( )2143 4042 MS (e)2189 4042 MS = (l)2282 4042 MS (e)2328 4042 MS (m)2420 4042 MS (e)2568 4042 MS (n)2661 = 4042 MS (t)2763 4042 MS (s)2819 4042 MS=0A= ( )2911 4042 MS (t)2957 4042 MS (o)3013 4042 MS ( )3115 4042 MS (c)3160 = 4042 MS (l)3253 4042 MS (a)3299 4042 MS (s)3391 4042 MS (s)3484 4042 MS = (i)3577 4042 MS (f)3622 4042 MS (y)3679 4042 MS ( )3769 4042 MS (c)3815 = 4042 MS (h)3908 4042 MS (a)4010 4042 MS=0A= (n)4103 4042 MS (n)4205 4042 MS (e)4307 4042 MS (l)4400 4042 MS (s)4446 = 4042 MS ( )4539 4042 MS (a)4585 4042 MS (c)4678 4042 MS (c)4770 4042 MS = (o)4863 4042 MS (r)4965 4042 MS (d)5030 4042 MS (i)5132 4042 MS (n)5178 = 4042 MS (g)5280 4042 MS ( )5381 4042 MS=0A= (t)5427 4042 MS (o)5483 4042 MS ( )5585 4042 MS=0A= (s)1504 4198 MS (e)1597 4198 MS (s)1690 4198 MS (s)1783 4198 MS (i)1876 = 4198 MS (o)1922 4198 MS (n)2024 4198 MS ( )2126 4198 MS=0A= (\226)2171 4198 MS=0A= (i)2310 4198 MS (n)2356 4198 MS (c)2457 4198 MS (o)2550 4198 MS (n)2652 = 4198 MS (s)2754 4198 MS (i)2847 4198 MS (s)2893 4198 MS (t)2985 4198 MS = (e)3041 4198 MS (n)3134 4198 MS (t)3235 4198 MS ( )3291 4198 MS (a)3337 = 4198 MS (c)3429 4198 MS (r)3522 4198 MS=0A= (o)3587 4198 MS (s)3689 4198 MS (s)3782 4198 MS ( )3875 4198 MS (s)3920 = 4198 MS (e)4013 4198 MS (n)4106 4198 MS (d)4208 4198 MS (e)4310 4198 MS = (r)4402 4198 MS (s)4467 4198 MS=0A= showpage=0A= %%Page: 4 4=0A= 90 rotate 0 -612 translate=0A= 13.059 598.762 translate 72 600 div dup neg scale=0A= 0 0 transform .25 add round .25 sub exch .25 add round .25 sub exch = itransform translate=0A= n=0A= 5999 4499 195 193 B=0A= 1 g=0A= f=0A= [83 0 0 -83 0 0]/Helvetica-Bold MF=0A= 0 g=0A= (4)5797 4495 MS=0A= [300 0 0 -300 0 0]/Helvetica-Bold MF=0A= (P)1511 741 MS (r)1711 741 MS (o)1828 741 MS (p)2011 741 MS (o)2194 741 = MS (s)2378 741 MS (e)2545 741 MS (d)2712 741 MS ( )2895 741 MS (c)2978 = 741 MS (h)3145 741 MS (a)3328 741 MS (n)3495 741 MS (g)3678 741 MS = (e)3861 741 MS (s)4028 741 MS=0A= ( )4195 741 MS (t)4278 741 MS (o)4378 741 MS ( )4561 741 MS (0)4644 741 = MS (6)4811 741 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 1193 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (F)942 1193 MS (E)1064 1193 MS (C)1197 1193 MS ( )1341 1193 MS (O)1397 = 1193 MS (b)1553 1193 MS (j)1674 1193 MS (e)1730 1193 MS (c)1841 1193 MS = (t)1952 1193 MS ( )2019 1193 MS (T)2075 1193 MS (r)2197 1193 MS (a)2275 = 1193 MS (n)2386 1193 MS (s)2508 1193 MS=0A= (m)2619 1193 MS (i)2797 1193 MS (s)2853 1193 MS (s)2963 1193 MS (i)3074 = 1193 MS (o)3130 1193 MS (n)3252 1193 MS ( )3374 1193 MS (I)3429 1193 MS = (n)3485 1193 MS (f)3606 1193 MS (o)3673 1193 MS (.)3794 1193 MS ( )3850 = 1193 MS (c)3905 1193 MS (o)4016 1193 MS=0A= (m)4138 1193 MS (m)4316 1193 MS (u)4494 1193 MS (n)4616 1193 MS (i)4738 = 1193 MS (c)4794 1193 MS (a)4905 1193 MS (t)5016 1193 MS (i)5083 1193 MS = (o)5139 1193 MS (n)5261 1193 MS=0A= [167 0 0 -167 0 0]/Helvetica MF=0A= /IsChar{exch/CharStrings get exch known}bd/MapCh{3 -1 roll/Encoding get = 3 1 roll put}bd/MapDegree{dup 16#b0 exch/degree IsChar{/degree}{/ring}ifelse = MapCh} bd/MapBB{dup 16#a6 exch/brokenbar IsChar{/brokenbar}{/bar}ifelse = MapCh}bd /reencode{findfont begin currentdict dup length dict begin{1 index/FID = ne{def} {pop pop}ifelse}forall/FontName exch def dup length 0 ne{/Encoding = Encoding 256 array copy def 0 exch{dup type/nametype eq{Encoding 2 index 2 index put = pop 1 add}{exch pop}ifelse}forall}if pop currentdict dup end end/FontName get = exch definefont dup MapDegree = MapBB}bd/LATENC[0/grave/acute/circumflex/tilde/macron /breve/dotaccent/dieresis/ring/cedilla/hungarumlaut/ogonek/caron/dotlessi= /fi/fl /Lslash/lslash/Zcaron/zcaron/minus/.notdef/.notdef/.notdef/.notdef/.notde= f /.notdef/.notdef/.notdef/.notdef/.notdef/.notdef/space/exclam/quotedbl /numbersign/dollar/percent/ampersand/quotesingle/parenleft/parenright/ast= erisk /plus/comma/hyphen/period/slash/zero/one/two/three/four/five/six/seven/ei= ght /nine/colon/semicolon/less/equal/greater/question/at/A/B/C/D/E/F/G/H/I/J/= K/L/M /N/O/P/Q/R/S/T/U/V/W/X/Y/Z/bracketleft/backslash/bracketright/asciicircum= /underscore/grave/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/bra= celeft /bar/braceright/asciitilde/.notdef/.notdef/.notdef/quotesinglbase/florin /quotedblbase/ellipsis/dagger/daggerdbl/circumflex/perthousand/Scaron /guilsinglleft/OE/.notdef/.notdef/.notdef/.notdef/quoteleft/quoteright /quotedblleft/quotedblright/bullet/endash/emdash/tilde/trademark/scaron /guilsinglright/oe/.notdef/.notdef/Ydieresis/.notdef/exclamdown/cent/ster= ling /currency/yen/brokenbar/section/dieresis/copyright/ordfeminine/guillemotl= eft /logicalnot/hyphen/registered/macron/degree/plusminus/twosuperior/threesu= perior /acute/mu/paragraph/periodcentered/cedilla/onesuperior/ordmasculine /guillemotright/onequarter/onehalf/threequarters/questiondown/Agrave/Aacu= te /Acircumflex/Atilde/Adieresis/Aring/AE/Ccedilla/Egrave/Eacute/Ecircumflex= /Edieresis/Igrave/Iacute/Icircumflex/Idieresis/Eth/Ntilde/Ograve/Oacute /Ocircumflex/Otilde/Odieresis/multiply/Oslash/Ugrave/Uacute/Ucircumflex /Udieresis/Yacute/Thorn/germandbls/agrave/aacute/acircumflex/atilde/adier= esis /aring/ae/ccedilla/egrave/eacute/ecircumflex/edieresis/igrave/iacute /icircumflex/idieresis/eth/ntilde/ograve/oacute/ocircumflex/otilde/odiere= sis /divide/oslash/ugrave/uacute/ucircumflex/udieresis/yacute/thorn/ydieresis= ]def LATENC /_Helvetica /Helvetica reencode=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 1436 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (0)1204 1436 MS (6)1297 1436 MS (:)1390 1436 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 1673 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (M)1504 1673 MS (u)1643 1673 MS (s)1745 1673 MS (t)1838 1673 MS ( )1893 = 1673 MS (b)1939 1673 MS (e)2041 1673 MS ( )2134 1673 MS (c)2180 1673 MS = (o)2273 1673 MS (m)2375 1673 MS (m)2522 1673 MS (u)2670 1673 MS (n)2772 = 1673 MS (i)2874 1673 MS (c)2920 1673 MS=0A= (a)3013 1673 MS (t)3106 1673 MS (e)3161 1673 MS (d)3254 1673 MS ( )3356 = 1673 MS (o)3402 1673 MS (u)3503 1673 MS (t)3605 1673 MS=0A= (-)3660 1673 MS=0A= (o)3716 1673 MS (f)3818 1673 MS=0A= (-)3873 1673 MS=0A= (b)3929 1673 MS (a)4031 1673 MS (n)4124 1673 MS (d)4226 1673 MS ( )4328 = 1673 MS (b)4374 1673 MS (e)4475 1673 MS (f)4568 1673 MS (o)4624 1673 MS = (r)4726 1673 MS (e)4790 1673 MS ( )4883 1673 MS (r)4929 1673 MS (e)4993 = 1673 MS (c)5086 1673 MS (e)5179 1673 MS=0A= (i)5272 1673 MS (v)5318 1673 MS (e)5409 1673 MS (r)5502 1673 MS ( )5567 = 1673 MS=0A= (j)1504 1849 MS (o)1550 1849 MS (i)1652 1849 MS (n)1698 1849 MS (s)1800 = 1849 MS ( )1893 1849 MS (t)1939 1849 MS (h)1995 1849 MS (e)2096 1849 MS = ( )2189 1849 MS (s)2235 1849 MS (e)2328 1849 MS (s)2420 1849 MS (s)2512 = 1849 MS (i)2605 1849 MS (o)2651 1849 MS=0A= (n)2753 1849 MS (.)2855 1849 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 2085 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (P)1204 2085 MS (r)1315 2085 MS (o)1380 2085 MS (p)1482 2085 MS (o)1584 = 2085 MS (s)1686 2085 MS (e)1779 2085 MS (d)1872 2085 MS (:)1974 2085 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 2321 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (C)1504 2321 MS (a)1625 2321 MS (n)1718 2321 MS ( )1820 2321 MS (b)1866 = 2321 MS (e)1968 2321 MS ( )2061 2321 MS (c)2107 2321 MS (o)2200 2321 MS = (m)2302 2321 MS (m)2449 2321 MS (u)2597 2321 MS (n)2699 2321 MS (i)2801 = 2321 MS (c)2847 2321 MS (a)2940 2321 MS=0A= (t)3033 2321 MS (e)3088 2321 MS (d)3181 2321 MS ( )3283 2321 MS (o)3329 = 2321 MS (u)3430 2321 MS (t)3532 2321 MS=0A= (-)3587 2321 MS=0A= (o)3643 2321 MS (f)3745 2321 MS=0A= (-)3800 2321 MS=0A= (b)3856 2321 MS (a)3958 2321 MS (n)4051 2321 MS (d)4153 2321 MS ( )4255 = 2321 MS (b)4301 2321 MS (e)4402 2321 MS (f)4495 2321 MS (o)4551 2321 MS = (r)4653 2321 MS (e)4717 2321 MS ( )4810 2321 MS (j)4856 2321 MS (o)4902 = 2321 MS (i)5004 2321 MS (n)5050 2321 MS=0A= (i)5152 2321 MS (n)5198 2321 MS (g)5300 2321 MS ( )5401 2321 MS (t)5447 = 2321 MS (h)5503 2321 MS (e)5605 2321 MS ( )5698 2321 MS=0A= (s)1504 2498 MS (e)1597 2498 MS (s)1690 2498 MS (s)1783 2498 MS (i)1876 = 2498 MS (o)1922 2498 MS (n)2024 2498 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 2734 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (C)1504 2734 MS (a)1625 2734 MS (n)1718 2734 MS ( )1820 2734 MS (b)1866 = 2734 MS (e)1968 2734 MS ( )2061 2734 MS (d)2107 2734 MS (y)2210 2734 MS = (n)2300 2734 MS (a)2402 2734 MS (m)2496 2734 MS (i)2644 2734 MS (c)2690 = 2734 MS (a)2783 2734 MS (l)2876 2734 MS=0A= (l)2922 2734 MS (y)2969 2734 MS ( )3059 2734 MS (c)3105 2734 MS (o)3198 = 2734 MS (m)3300 2734 MS (m)3448 2734 MS (u)3596 2734 MS (n)3698 2734 MS = (i)3800 2734 MS (c)3846 2734 MS (a)3939 2734 MS (t)4032 2734 MS (e)4087 = 2734 MS (d)4180 2734 MS ( )4282 2734 MS=0A= (o)4328 2734 MS (u)4429 2734 MS (t)4531 2734 MS=0A= (-)4586 2734 MS=0A= (o)4642 2734 MS (f)4744 2734 MS=0A= (-)4799 2734 MS=0A= (b)4855 2734 MS (a)4957 2734 MS (n)5050 2734 MS (d)5152 2734 MS ( )5254 = 2734 MS (a)5299 2734 MS (s)5391 2734 MS ( )5484 2734 MS (t)5530 2734 MS = (h)5586 2734 MS (e)5688 2734 MS ( )5780 2734 MS=0A= (s)1504 2910 MS (e)1597 2910 MS (s)1690 2910 MS (s)1783 2910 MS (i)1876 = 2910 MS (o)1922 2910 MS (n)2024 2910 MS ( )2126 2910 MS (i)2171 2910 MS = (s)2217 2910 MS ( )2310 2910 MS (p)2356 2910 MS (r)2458 2910 MS (o)2522 = 2910 MS (g)2624 2910 MS (r)2726 2910 MS=0A= (e)2791 2910 MS (s)2884 2910 MS (s)2977 2910 MS (i)3070 2910 MS (n)3116 = 2910 MS (g)3218 2910 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 3146 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (C)1504 3146 MS (a)1625 3146 MS (n)1718 3146 MS ( )1820 3146 MS (b)1866 = 3146 MS (e)1968 3146 MS ( )2061 3146 MS (c)2107 3146 MS (o)2200 3146 MS = (m)2302 3146 MS (m)2449 3146 MS (u)2597 3146 MS (n)2699 3146 MS (i)2801 = 3146 MS (c)2847 3146 MS (a)2940 3146 MS=0A= (t)3033 3146 MS (e)3088 3146 MS (d)3181 3146 MS ( )3283 3146 MS (i)3329 = 3146 MS (n)3374 3146 MS=0A= (-)3476 3146 MS=0A= (b)3532 3146 MS (a)3634 3146 MS (n)3727 3146 MS (d)3829 3146 MS ( )3931 = 3146 MS (a)3977 3146 MS (s)4069 3146 MS ( )4162 3146 MS (t)4208 3146 MS = (h)4264 3146 MS (e)4365 3146 MS ( )4458 3146 MS (s)4504 3146 MS (e)4597 = 3146 MS (s)4690 3146 MS (s)4782 3146 MS=0A= (i)4875 3146 MS (o)4921 3146 MS (n)5023 3146 MS ( )5125 3146 MS (i)5171 = 3146 MS (s)5217 3146 MS ( )5310 3146 MS=0A= (p)1504 3323 MS (r)1606 3323 MS (o)1671 3323 MS (g)1773 3323 MS (r)1875 = 3323 MS (e)1940 3323 MS (s)2033 3323 MS (s)2126 3323 MS (i)2218 3323 MS = (n)2264 3323 MS (g)2366 3323 MS ( )2468 3323 MS=0A= LATENC /_Helvetica-Bold /Helvetica-Bold reencode=0A= [167 0 0 -167 0 0]/_Helvetica-Bold MF=0A= (\226)2513 3323 MS=0A= (a)2652 3323 MS (d)2745 3323 MS (d)2847 3323 MS ( )2949 3323 MS (a)2995 = 3323 MS ( )3088 3323 MS (f)3133 3323 MS (i)3189 3323 MS (x)3235 3323 MS = (e)3327 3323 MS (d)3420 3323 MS ( )3521 3323 MS (l)3567 3323 MS (e)3613 = 3323 MS (n)3706 3323 MS (g)3808 3323 MS=0A= (t)3910 3323 MS (h)3966 3323 MS ( )4068 3323 MS (a)4113 3323 MS (n)4206 = 3323 MS (d)4308 3323 MS ( )4410 3323 MS (v)4455 3323 MS (a)4546 3323 MS = (r)4639 3323 MS (i)4704 3323 MS (a)4750 3323 MS (b)4843 3323 MS (l)4945 = 3323 MS (e)4991 3323 MS ( )5084 3323 MS=0A= (l)5130 3323 MS (e)5176 3323 MS (n)5269 3323 MS (g)5371 3323 MS (t)5473 = 3323 MS (h)5529 3323 MS ( )5631 3323 MS=0A= (H)1504 3499 MS (e)1625 3499 MS (a)1718 3499 MS (d)1811 3499 MS (e)1913 = 3499 MS (r)2006 3499 MS ( )2070 3499 MS (e)2116 3499 MS (x)2209 3499 MS = (t)2301 3499 MS (e)2357 3499 MS (n)2449 3499 MS (s)2551 3499 MS (i)2644 = 3499 MS (o)2690 3499 MS (n)2792 3499 MS=0A= ( )2894 3499 MS (t)2940 3499 MS (o)2996 3499 MS ( )3098 3499 MS (c)3143 = 3499 MS (a)3236 3499 MS (r)3329 3499 MS (r)3393 3499 MS (y)3457 3499 MS = ( )3548 3499 MS (t)3594 3499 MS (h)3650 3499 MS (i)3752 3499 MS (s)3798 = 3499 MS ( )3891 3499 MS (i)3937 3499 MS=0A= (n)3983 3499 MS (f)4085 3499 MS (o)4141 3499 MS (r)4243 3499 MS (m)4308 = 3499 MS (a)4455 3499 MS (t)4548 3499 MS (i)4604 3499 MS (o)4650 3499 MS = (n)4752 3499 MS=0A= showpage=0A= %%Page: 5 5=0A= 90 rotate 0 -612 translate=0A= 13.059 598.762 translate 72 600 div dup neg scale=0A= 0 0 transform .25 add round .25 sub exch .25 add round .25 sub exch = itransform translate=0A= n=0A= 5999 4499 195 193 B=0A= 1 g=0A= f=0A= [83 0 0 -83 0 0]/Helvetica-Bold MF=0A= 0 g=0A= (5)5797 4495 MS=0A= [300 0 0 -300 0 0]/Helvetica-Bold MF=0A= (P)1511 741 MS (r)1711 741 MS (o)1828 741 MS (p)2011 741 MS (o)2194 741 = MS (s)2378 741 MS (e)2545 741 MS (d)2712 741 MS ( )2895 741 MS (c)2978 = 741 MS (h)3145 741 MS (a)3328 741 MS (n)3495 741 MS (g)3678 741 MS = (e)3861 741 MS (s)4028 741 MS=0A= ( )4195 741 MS (t)4278 741 MS (o)4378 741 MS ( )4561 741 MS (0)4644 741 = MS (6)4811 741 MS=0A= [167 0 0 -167 0 0]/Wingdings02009940 MF=0A= (\203)754 999 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (T)942 999 MS (r)1044 999 MS (a)1109 999 MS (n)1202 999 MS (s)1304 999 = MS (m)1397 999 MS (i)1545 999 MS (s)1591 999 MS (s)1683 999 MS (i)1776 = 999 MS (o)1822 999 MS (n)1923 999 MS ( )2025 999 MS (O)2071 999 MS = (b)2201 999 MS (j)2303 999 MS=0A= (e)2349 999 MS (c)2441 999 MS (t)2534 999 MS ( )2589 999 MS (I)2635 999 = MS (D)2681 999 MS ( )2802 999 MS=0A= /IsChar{exch/CharStrings get exch known}bd/MapCh{3 -1 roll/Encoding get = 3 1 roll put}bd/MapDegree{dup 16#b0 exch/degree IsChar{/degree}{/ring}ifelse = MapCh} bd/MapBB{dup 16#a6 exch/brokenbar IsChar{/brokenbar}{/bar}ifelse = MapCh}bd /reencode{findfont begin currentdict dup length dict begin{1 index/FID = ne{def} {pop pop}ifelse}forall/FontName exch def dup length 0 ne{/Encoding = Encoding 256 array copy def 0 exch{dup type/nametype eq{Encoding 2 index 2 index put = pop 1 add}{exch pop}ifelse}forall}if pop currentdict dup end end/FontName get = exch definefont dup MapDegree = MapBB}bd/LATENC[0/grave/acute/circumflex/tilde/macron /breve/dotaccent/dieresis/ring/cedilla/hungarumlaut/ogonek/caron/dotlessi= /fi/fl /Lslash/lslash/Zcaron/zcaron/minus/.notdef/.notdef/.notdef/.notdef/.notde= f /.notdef/.notdef/.notdef/.notdef/.notdef/.notdef/space/exclam/quotedbl /numbersign/dollar/percent/ampersand/quotesingle/parenleft/parenright/ast= erisk /plus/comma/hyphen/period/slash/zero/one/two/three/four/five/six/seven/ei= ght /nine/colon/semicolon/less/equal/greater/question/at/A/B/C/D/E/F/G/H/I/J/= K/L/M /N/O/P/Q/R/S/T/U/V/W/X/Y/Z/bracketleft/backslash/bracketright/asciicircum= /underscore/grave/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/bra= celeft /bar/braceright/asciitilde/.notdef/.notdef/.notdef/quotesinglbase/florin /quotedblbase/ellipsis/dagger/daggerdbl/circumflex/perthousand/Scaron /guilsinglleft/OE/.notdef/.notdef/.notdef/.notdef/quoteleft/quoteright /quotedblleft/quotedblright/bullet/endash/emdash/tilde/trademark/scaron /guilsinglright/oe/.notdef/.notdef/Ydieresis/.notdef/exclamdown/cent/ster= ling /currency/yen/brokenbar/section/dieresis/copyright/ordfeminine/guillemotl= eft /logicalnot/hyphen/registered/macron/degree/plusminus/twosuperior/threesu= perior /acute/mu/paragraph/periodcentered/cedilla/onesuperior/ordmasculine /guillemotright/onequarter/onehalf/threequarters/questiondown/Agrave/Aacu= te /Acircumflex/Atilde/Adieresis/Aring/AE/Ccedilla/Egrave/Eacute/Ecircumflex= /Edieresis/Igrave/Iacute/Icircumflex/Idieresis/Eth/Ntilde/Ograve/Oacute /Ocircumflex/Otilde/Odieresis/multiply/Oslash/Ugrave/Uacute/Ucircumflex /Udieresis/Yacute/Thorn/germandbls/agrave/aacute/acircumflex/atilde/adier= esis /aring/ae/ccedilla/egrave/eacute/ecircumflex/edieresis/igrave/iacute /icircumflex/idieresis/eth/ntilde/ograve/oacute/ocircumflex/otilde/odiere= sis /divide/oslash/ugrave/uacute/ucircumflex/udieresis/yacute/thorn/ydieresis= ]def LATENC /_Helvetica-Bold /Helvetica-Bold reencode=0A= [167 0 0 -167 0 0]/_Helvetica-Bold MF=0A= (\226)2848 999 MS=0A= (w)2983 999 MS (h)3117 999 MS (e)3218 999 MS (n)3311 999 MS ( )3413 999 = MS (r)3458 999 MS (e)3523 999 MS (q)3616 999 MS (u)3718 999 MS (i)3820 = 999 MS (r)3866 999 MS (e)3930 999 MS (d)4023 999 MS=0A= [150 0 0 -150 0 0]/Helvetica MF=0A= LATENC /_Helvetica /Helvetica reencode=0A= [150 0 0 -150 0 0]/_Helvetica MF=0A= (\225)1054 1196 MS=0A= [150 0 0 -150 0 0]/Helvetica-Bold MF=0A= (0)1204 1196 MS (6)1287 1196 MS (:)1370 1196 MS=0A= [150 0 0 -150 0 0]/_Helvetica MF=0A= (\226)1354 1391 MS=0A= [150 0 0 -150 0 0]/Helvetica-Bold MF=0A= (I)1504 1391 MS (f)1546 1391 MS ( )1596 1391 MS (m)1638 1391 MS (o)1771 = 1391 MS (r)1863 1391 MS (e)1921 1391 MS ( )2004 1391 MS (t)2046 1391 MS = (h)2096 1391 MS (a)2188 1391 MS (n)2271 1391 MS ( )2363 1391 MS (o)2405 = 1391 MS (n)2496 1391 MS (e)2588 1391 MS=0A= ( )2671 1391 MS (o)2713 1391 MS (b)2805 1391 MS (j)2897 1391 MS (e)2939 = 1391 MS (c)3022 1391 MS (t)3105 1391 MS ( )3155 1391 MS (c)3197 1391 MS = (a)3280 1391 MS (r)3363 1391 MS (r)3421 1391 MS (i)3479 1391 MS (e)3521 = 1391 MS (d)3604 1391 MS ( )3696 1391 MS=0A= (i)3738 1391 MS (n)3780 1391 MS ( )3871 1391 MS (t)3913 1391 MS (h)3963 = 1391 MS (e)4055 1391 MS ( )4138 1391 MS (s)4180 1391 MS (e)4263 1391 MS = (s)4346 1391 MS (s)4429 1391 MS (i)4512 1391 MS (o)4554 1391 MS (n)4646 = 1391 MS ( )4738 1391 MS (t)4780 1391 MS=0A= (h)4830 1391 MS (e)4922 1391 MS (n)5005 1391 MS ( )5096 1391 MS (T)5138 = 1391 MS (O)5230 1391 MS (I)5346 1391 MS ( )5388 1391 MS=0A= (r)1504 1532 MS (e)1562 1532 MS (q)1645 1532 MS (u)1737 1532 MS (i)1829 = 1532 MS (r)1871 1532 MS (e)1929 1532 MS (d)2012 1532 MS ( )2104 1532 MS = (t)2146 1532 MS (o)2196 1532 MS ( )2288 1532 MS (b)2330 1532 MS (e)2422 = 1532 MS ( )2504 1532 MS (c)2546 1532 MS=0A= (a)2629 1532 MS (r)2712 1532 MS (r)2770 1532 MS (i)2828 1532 MS (e)2870 = 1532 MS (d)2953 1532 MS ( )3045 1532 MS (i)3087 1532 MS (n)3129 1532 MS = ( )3221 1532 MS (e)3263 1532 MS (a)3346 1532 MS (c)3429 1532 MS (h)3512 = 1532 MS ( )3604 1532 MS (p)3646 1532 MS=0A= (a)3738 1532 MS (c)3821 1532 MS (k)3904 1532 MS (e)3987 1532 MS (t)4070 = 1532 MS=0A= [150 0 0 -150 0 0]/_Helvetica MF=0A= (\225)1054 1726 MS=0A= [150 0 0 -150 0 0]/Helvetica-Bold MF=0A= (P)1204 1726 MS (r)1304 1726 MS (o)1362 1726 MS (p)1454 1726 MS (o)1546 = 1726 MS (s)1638 1726 MS (a)1721 1726 MS (l)1804 1726 MS (:)1846 1726 MS=0A= [150 0 0 -150 0 0]/_Helvetica MF=0A= (\226)1354 1921 MS=0A= [150 0 0 -150 0 0]/Helvetica-Bold MF=0A= (I)1504 1921 MS (f)1546 1921 MS ( )1596 1921 MS (m)1638 1921 MS (o)1771 = 1921 MS (r)1863 1921 MS (e)1921 1921 MS ( )2004 1921 MS (t)2046 1921 MS = (h)2096 1921 MS (a)2188 1921 MS (n)2271 1921 MS ( )2363 1921 MS (o)2405 = 1921 MS (n)2496 1921 MS (e)2588 1921 MS=0A= ( )2671 1921 MS (o)2713 1921 MS (b)2805 1921 MS (j)2897 1921 MS (e)2939 = 1921 MS (c)3022 1921 MS (t)3105 1921 MS ( )3155 1921 MS (c)3197 1921 MS = (a)3280 1921 MS (r)3363 1921 MS (r)3421 1921 MS (i)3479 1921 MS (e)3521 = 1921 MS (d)3604 1921 MS ( )3696 1921 MS=0A= (i)3738 1921 MS (n)3780 1921 MS ( )3871 1921 MS (t)3913 1921 MS (h)3963 = 1921 MS (e)4055 1921 MS ( )4138 1921 MS (s)4180 1921 MS (e)4263 1921 MS = (s)4346 1921 MS (s)4429 1921 MS (i)4512 1921 MS (o)4554 1921 MS (n)4646 = 1921 MS ( )4738 1921 MS (t)4780 1921 MS=0A= (h)4830 1921 MS (e)4922 1921 MS (n)5005 1921 MS ( )5096 1921 MS (T)5138 = 1921 MS (O)5230 1921 MS (I)5346 1921 MS ( )5388 1921 MS=0A= (r)1504 2062 MS (e)1562 2062 MS (q)1645 2062 MS (u)1737 2062 MS (i)1829 = 2062 MS (r)1871 2062 MS (e)1929 2062 MS (d)2012 2062 MS ( )2104 2062 MS = (f)2146 2062 MS (o)2196 2062 MS (r)2288 2062 MS ( )2346 2062 MS (a)2388 = 2062 MS (l)2471 2062 MS (l)2512 2062 MS=0A= ( )2554 2062 MS (b)2596 2062 MS (u)2688 2062 MS (t)2780 2062 MS ( )2829 = 2062 MS (o)2871 2062 MS (n)2963 2062 MS (e)3055 2062 MS ( )3138 2062 MS = (o)3179 2062 MS (b)3271 2062 MS (j)3363 2062 MS (e)3405 2062 MS (c)3487 = 2062 MS (t)3570 2062 MS (,)3620 2062 MS=0A= ( )3662 2062 MS (a)3704 2062 MS (n)3787 2062 MS (d)3879 2062 MS ( )3971 = 2062 MS (t)4013 2062 MS (h)4063 2062 MS (e)4155 2062 MS ( )4238 2062 MS = (o)4279 2062 MS (b)4371 2062 MS (j)4463 2062 MS (e)4504 2062 MS (c)4587 = 2062 MS (t)4670 2062 MS ( )4720 2062 MS=0A= (n)4762 2062 MS (o)4854 2062 MS (t)4946 2062 MS ( )4996 2062 MS (c)5038 = 2062 MS (a)5121 2062 MS (r)5204 2062 MS (r)5262 2062 MS (y)5321 2062 MS = (i)5403 2062 MS (n)5446 2062 MS (g)5538 2062 MS ( )5630 2062 MS=0A= (t)1504 2202 MS (h)1554 2202 MS (e)1646 2202 MS ( )1729 2202 MS (T)1771 = 2202 MS (O)1863 2202 MS (I)1980 2202 MS ( )2021 2202 MS (i)2063 2202 MS = (s)2105 2202 MS ( )2188 2202 MS (p)2230 2202 MS (r)2322 2202 MS (e)2380 = 2202 MS (s)2462 2202 MS (u)2545 2202 MS=0A= (m)2637 2202 MS (e)2770 2202 MS (d)2853 2202 MS ( )2945 2202 MS (t)2987 = 2202 MS (o)3037 2202 MS ( )3129 2202 MS (h)3171 2202 MS (a)3263 2202 MS = (v)3347 2202 MS (e)3429 2202 MS ( )3512 2202 MS (T)3554 2202 MS (O)3646 = 2202 MS (I)3763 2202 MS ( )3805 2202 MS=0A= (=3D)3846 2202 MS ( )3934 2202 MS (0)3976 2202 MS=0A= [150 0 0 -150 0 0]/_Helvetica MF=0A= (\226)1354 2397 MS=0A= [150 0 0 -150 0 0]/Helvetica-Bold MF=0A= (P)1504 2397 MS (r)1604 2397 MS (o)1662 2397 MS=0A= (:)1754 2397 MS ( )1804 2397 MS (S)1846 2397 MS (a)1946 2397 MS (v)2030 = 2397 MS (e)2111 2397 MS (s)2195 2397 MS ( )2278 2397 MS (s)2320 2397 MS = (o)2403 2397 MS (m)2496 2397 MS (e)2629 2397 MS ( )2712 2397 MS (s)2754 = 2397 MS (p)2837 2397 MS (a)2929 2397 MS=0A= (c)3012 2397 MS (e)3095 2397 MS ( )3178 2397 MS (i)3220 2397 MS (n)3262 = 2397 MS ( )3354 2397 MS (p)3396 2397 MS (a)3488 2397 MS (c)3571 2397 MS = (k)3654 2397 MS (e)3737 2397 MS (t)3820 2397 MS ( )3870 2397 MS (h)3912 = 2397 MS (e)4004 2397 MS (a)4087 2397 MS=0A= (d)4170 2397 MS (e)4262 2397 MS (r)4346 2397 MS ( )4404 2397 MS (f)4446 = 2397 MS (o)4496 2397 MS (r)4588 2397 MS ( )4646 2397 MS (o)4688 2397 MS = (n)4780 2397 MS (e)4872 2397 MS ( )4955 2397 MS (o)4997 2397 MS (b)5088 = 2397 MS (j)5180 2397 MS (e)5222 2397 MS=0A= (c)5305 2397 MS (t)5388 2397 MS=0A= [150 0 0 -150 0 0]/_Helvetica MF=0A= (\226)1354 2592 MS=0A= [150 0 0 -150 0 0]/Helvetica-Bold MF=0A= (C)1504 2592 MS (o)1612 2592 MS (n)1704 2592 MS=0A= (:)1796 2592 MS ( )1846 2592 MS (I)1888 2592 MS (s)1930 2592 MS ( )2013 = 2592 MS (i)2055 2592 MS (n)2096 2592 MS (e)2188 2592 MS (l)2271 2592 MS = (e)2313 2592 MS (g)2396 2592 MS (a)2487 2592 MS (n)2570 2592 MS (t)2662 = 2592 MS ( )2712 2592 MS (a)2754 2592 MS=0A= (n)2837 2592 MS (d)2929 2592 MS ( )3021 2592 MS (i)3063 2592 MS (n)3104 = 2592 MS (c)3196 2592 MS (o)3279 2592 MS (n)3371 2592 MS (s)3462 2592 MS = (i)3545 2592 MS (s)3587 2592 MS (t)3670 2592 MS (e)3720 2592 MS (n)3803 = 2592 MS (t)3895 2592 MS ( )3945 2592 MS=0A= LATENC /_Helvetica-Bold /Helvetica-Bold reencode=0A= [150 0 0 -150 0 0]/_Helvetica-Bold MF=0A= (\226)3987 2592 MS=0A= (c)4112 2592 MS (a)4195 2592 MS (n)4278 2592 MS ( )4370 2592 MS (c)4412 = 2592 MS (a)4495 2592 MS (u)4578 2592 MS (s)4670 2592 MS (e)4753 2592 MS = ( )4836 2592 MS (c)4878 2592 MS (o)4961 2592 MS (n)5053 2592 MS (f)5145 = 2592 MS (u)5195 2592 MS (s)5287 2592 MS=0A= (i)5370 2592 MS (o)5412 2592 MS (n)5504 2592 MS ( )5596 2592 MS (i)5638 = 2592 MS (f)5680 2592 MS ( )5729 2592 MS=0A= (o)1504 2733 MS (n)1596 2733 MS (e)1688 2733 MS ( )1771 2733 MS (o)1813 = 2733 MS (f)1905 2733 MS ( )1955 2733 MS (t)1997 2733 MS (h)2046 2733 MS = (e)2138 2733 MS ( )2221 2733 MS (o)2263 2733 MS (b)2355 2733 MS (j)2446 = 2733 MS (e)2487 2733 MS (c)2570 2733 MS=0A= (t)2653 2733 MS (s)2703 2733 MS ( )2786 2733 MS (e)2828 2733 MS (x)2911 = 2733 MS (p)2994 2733 MS (l)3086 2733 MS (i)3128 2733 MS (c)3170 2733 MS = (i)3253 2733 MS (t)3295 2733 MS (l)3345 2733 MS (y)3387 2733 MS ( )3470 = 2733 MS (c)3512 2733 MS (a)3595 2733 MS=0A= (r)3678 2733 MS (r)3736 2733 MS (i)3794 2733 MS (e)3836 2733 MS (s)3919 = 2733 MS ( )4002 2733 MS (T)4044 2733 MS (O)4136 2733 MS (I)4253 2733 MS = ( )4295 2733 MS (=3D)4337 2733 MS ( )4425 2733 MS (0)4466 2733 MS=0A= [167 0 0 -167 0 0]/Wingdings02009940 MF=0A= (\203)754 2947 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (T)942 2947 MS (r)1044 2947 MS (a)1109 2947 MS (n)1202 2947 MS (s)1304 = 2947 MS (m)1397 2947 MS (i)1545 2947 MS (s)1591 2947 MS (s)1683 2947 MS = (i)1776 2947 MS (o)1822 2947 MS (n)1923 2947 MS ( )2025 2947 MS (O)2071 = 2947 MS (b)2201 2947 MS (j)2303 2947 MS=0A= (e)2349 2947 MS (c)2441 2947 MS (t)2534 2947 MS ( )2589 2947 MS (I)2635 = 2947 MS (D)2681 2947 MS ( )2802 2947 MS=0A= LATENC /_Helvetica-Bold /Helvetica-Bold reencode=0A= [167 0 0 -167 0 0]/_Helvetica-Bold MF=0A= (\226)2848 2947 MS=0A= (s)2986 2947 MS (c)3079 2947 MS (o)3172 2947 MS (p)3274 2947 MS (e)3376 = 2947 MS=0A= [167 0 0 -167 0 0]/Helvetica MF=0A= LATENC /_Helvetica /Helvetica reencode=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 3164 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (0)1204 3164 MS (6)1297 3164 MS (:)1390 3164 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 3381 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (I)1504 3381 MS (m)1550 3381 MS (p)1698 3381 MS (l)1800 3381 MS (i)1846 = 3381 MS (c)1892 3381 MS (i)1985 3381 MS (t)2030 3381 MS (l)2086 3381 MS = (y)2133 3381 MS ( )2223 3381 MS (s)2269 3381 MS (c)2362 3381 MS (o)2455 = 3381 MS (p)2557 3381 MS (e)2659 3381 MS=0A= (d)2752 3381 MS ( )2854 3381 MS (g)2900 3381 MS (l)3002 3381 MS (o)3048 = 3381 MS (b)3150 3381 MS (a)3252 3381 MS (l)3345 3381 MS (l)3391 3381 MS = (y)3436 3381 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 3597 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (P)1204 3597 MS (r)1315 3597 MS (o)1380 3597 MS (p)1482 3597 MS (o)1584 = 3597 MS (s)1686 3597 MS (e)1779 3597 MS (d)1872 3597 MS (:)1974 3597 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 3814 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (E)1504 3814 MS (x)1615 3814 MS (p)1708 3814 MS (l)1810 3814 MS (i)1856 = 3814 MS (c)1902 3814 MS (i)1995 3814 MS (t)2041 3814 MS (l)2096 3814 MS = (y)2143 3814 MS ( )2233 3814 MS (s)2279 3814 MS (c)2372 3814 MS (o)2465 = 3814 MS (p)2567 3814 MS (e)2669 3814 MS=0A= ( )2762 3814 MS (w)2805 3814 MS (i)2939 3814 MS (t)2984 3814 MS (h)3039 = 3814 MS (i)3141 3814 MS (n)3187 3814 MS ( )3289 3814 MS (s)3334 3814 MS = (e)3426 3814 MS (s)3519 3814 MS (s)3612 3814 MS (i)3705 3814 MS (o)3751 = 3814 MS (n)3853 3814 MS=0A= showpage=0A= %%Page: 6 6=0A= 90 rotate 0 -612 translate=0A= 13.059 598.762 translate 72 600 div dup neg scale=0A= 0 0 transform .25 add round .25 sub exch .25 add round .25 sub exch = itransform translate=0A= n=0A= 5999 4499 195 193 B=0A= 1 g=0A= f=0A= [83 0 0 -83 0 0]/Helvetica-Bold MF=0A= 0 g=0A= (6)5797 4495 MS=0A= [333 0 0 -333 0 0]/Helvetica-Bold MF=0A= (N)2022 752 MS (e)2262 752 MS (x)2447 752 MS (t)2633 752 MS ( )2744 752 = MS (f)2837 752 MS (o)2949 752 MS (r)3152 752 MS ( )3282 752 MS (A)3375 = 752 MS (L)3615 752 MS (C)3819 752 MS ( )4059 752 MS (P)4152 752 MS = (I)4374 752 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 1293 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (D)942 1293 MS (e)1086 1293 MS (c)1198 1293 MS (i)1309 1293 MS (d)1365 = 1293 MS (e)1487 1293 MS ( )1598 1293 MS (t)1654 1293 MS (h)1721 1293 MS = (e)1843 1293 MS (s)1953 1293 MS (e)2064 1293 MS ( )2175 1293 MS (i)2231 = 1293 MS (s)2287 1293 MS (s)2398 1293 MS=0A= (u)2509 1293 MS (e)2631 1293 MS (s)2742 1293 MS ( )2853 1293 MS (h)2909 = 1293 MS (e)3031 1293 MS (r)3142 1293 MS (e)3220 1293 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 1575 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (R)942 1575 MS (e)1086 1575 MS (v)1198 1575 MS (i)1309 1575 MS (s)1365 = 1575 MS (i)1476 1575 MS (o)1532 1575 MS (n)1654 1575 MS ( )1776 1575 MS = (0)1832 1575 MS (7)1942 1575 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 1857 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (L)942 1857 MS (a)1064 1857 MS (s)1175 1857 MS (t)1286 1857 MS ( )1353 = 1857 MS (c)1409 1857 MS (a)1520 1857 MS (l)1631 1857 MS (l)1687 1857 MS=0A= showpage=0A= %%Page: 7 7=0A= 90 rotate 0 -612 translate=0A= 13.059 598.762 translate 72 600 div dup neg scale=0A= 0 0 transform .25 add round .25 sub exch .25 add round .25 sub exch = itransform translate=0A= n=0A= 5999 4499 195 193 B=0A= 1 g=0A= f=0A= [83 0 0 -83 0 0]/Helvetica-Bold MF=0A= 0 g=0A= (7)5797 4495 MS=0A= [300 0 0 -300 0 0]/Helvetica-Bold MF=0A= (W)536 1808 MS (a)819 1808 MS (v)986 1808 MS (e)1153 1808 MS ( )1320 = 1808 MS (&)1403 1808 MS ( )1620 1808 MS (E)1703 1808 MS (q)1903 1808 MS = (u)2086 1808 MS (a)2270 1808 MS (t)2437 1808 MS (i)2537 1808 MS (o)2620 = 1808 MS (n)2803 1808 MS ( )2986 1808 MS=0A= (B)3069 1808 MS (a)3287 1808 MS (s)3454 1808 MS (e)3621 1808 MS (d)3788 = 1808 MS ( )3971 1808 MS (R)4054 1808 MS (a)4270 1808 MS (t)4437 1808 MS = (e)4537 1808 MS ( )4704 1808 MS (C)4787 1808 MS (o)5004 1808 MS (n)5187 = 1808 MS (t)5370 1808 MS (r)5470 1808 MS=0A= (o)5587 1808 MS (l)5770 1808 MS=0A= (W)2369 2125 MS (E)2652 2125 MS (B)2852 2125 MS (R)3069 2125 MS (C)3286 = 2125 MS ( )3503 2125 MS (B)3586 2125 MS (B)3803 2125 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (M)2008 2976 MS (.)2175 2976 MS ( )2230 2976 MS (L)2286 2976 MS (u)2408 = 2976 MS (b)2530 2976 MS (y)2652 2976 MS (,)2761 2976 MS ( )2817 2976 MS = (D)2873 2976 MS (i)3017 2976 MS (g)3073 2976 MS (i)3194 2976 MS (t)3250 = 2976 MS (a)3317 2976 MS (l)3428 2976 MS=0A= ( )3483 2976 MS (F)3539 2976 MS (o)3661 2976 MS (u)3782 2976 MS (n)3904 = 2976 MS (t)4025 2976 MS (a)4092 2976 MS (i)4203 2976 MS (n)4259 2976 MS=0A= (V)1985 3258 MS (.)2118 3258 MS ( )2174 3258 MS=0A= (G)2229 3258 MS (o)2385 3258 MS (y)2507 3258 MS (a)2617 3258 MS (l)2728 = 3258 MS=0A= (,)2784 3258 MS ( )2839 3258 MS (D)2894 3258 MS (i)3038 3258 MS (g)3094 = 3258 MS (i)3216 3258 MS (t)3272 3258 MS (a)3339 3258 MS (l)3449 3258 MS = ( )3505 3258 MS (F)3560 3258 MS (o)3682 3258 MS (u)3803 3258 MS (n)3925 = 3258 MS (t)4047 3258 MS (a)4114 3258 MS=0A= (i)4225 3258 MS (n)4281 3258 MS=0A= showpage=0A= %%Page: 8 8=0A= 90 rotate 0 -612 translate=0A= 13.059 598.762 translate 72 600 div dup neg scale=0A= 0 0 transform .25 add round .25 sub exch .25 add round .25 sub exch = itransform translate=0A= n=0A= 5999 4499 195 193 B=0A= 1 g=0A= f=0A= [83 0 0 -83 0 0]/Helvetica-Bold MF=0A= 0 g=0A= (8)5797 4495 MS=0A= [300 0 0 -300 0 0]/Helvetica-Bold MF=0A= (W)1269 741 MS (E)1552 741 MS (B)1752 741 MS (R)1969 741 MS (C)2186 741 = MS ( )2403 741 MS (o)2486 741 MS (u)2669 741 MS (t)2852 741 MS (s)2952 = 741 MS (t)3119 741 MS (a)3219 741 MS (n)3386 741 MS (d)3569 741 MS = (i)3752 741 MS (n)3835 741 MS=0A= (g)4019 741 MS ( )4202 741 MS (i)4285 741 MS (s)4368 741 MS (s)4535 741 = MS (u)4702 741 MS (e)4886 741 MS (s)5053 741 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 1443 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (N)979 1443 MS (u)1123 1443 MS (m)1245 1443 MS (b)1423 1443 MS (e)1545 = 1443 MS (r)1656 1443 MS ( )1734 1443 MS (o)1790 1443 MS (f)1912 1443 MS = ( )1979 1443 MS (c)2035 1443 MS (h)2146 1443 MS (a)2268 1443 MS (n)2379 = 1443 MS (n)2501 1443 MS (e)2623 1443 MS=0A= (l)2734 1443 MS (s)2790 1443 MS ( )2901 1443 MS (u)2956 1443 MS (s)3078 = 1443 MS (e)3189 1443 MS (d)3300 1443 MS ( )3422 1443 MS (p)3478 1443 MS = (e)3600 1443 MS (r)3711 1443 MS ( )3789 1443 MS (s)3845 1443 MS (e)3956 = 1443 MS (s)4067 1443 MS (s)4178 1443 MS=0A= (i)4289 1443 MS (o)4345 1443 MS (n)4467 1443 MS ( )4589 1443 MS (c)4645 = 1443 MS (a)4756 1443 MS (n)4867 1443 MS ( )4989 1443 MS (b)5045 1443 MS = (e)5167 1443 MS ( )5278 1443 MS=0A= (l)979 1653 MS (a)1035 1653 MS (r)1146 1653 MS (g)1224 1653 MS (e)1346 = 1653 MS=0A= [167 0 0 -167 0 0]/Helvetica MF=0A= /IsChar{exch/CharStrings get exch known}bd/MapCh{3 -1 roll/Encoding get = 3 1 roll put}bd/MapDegree{dup 16#b0 exch/degree IsChar{/degree}{/ring}ifelse = MapCh} bd/MapBB{dup 16#a6 exch/brokenbar IsChar{/brokenbar}{/bar}ifelse = MapCh}bd /reencode{findfont begin currentdict dup length dict begin{1 index/FID = ne{def} {pop pop}ifelse}forall/FontName exch def dup length 0 ne{/Encoding = Encoding 256 array copy def 0 exch{dup type/nametype eq{Encoding 2 index 2 index put = pop 1 add}{exch pop}ifelse}forall}if pop currentdict dup end end/FontName get = exch definefont dup MapDegree = MapBB}bd/LATENC[0/grave/acute/circumflex/tilde/macron /breve/dotaccent/dieresis/ring/cedilla/hungarumlaut/ogonek/caron/dotlessi= /fi/fl /Lslash/lslash/Zcaron/zcaron/minus/.notdef/.notdef/.notdef/.notdef/.notde= f /.notdef/.notdef/.notdef/.notdef/.notdef/.notdef/space/exclam/quotedbl /numbersign/dollar/percent/ampersand/quotesingle/parenleft/parenright/ast= erisk /plus/comma/hyphen/period/slash/zero/one/two/three/four/five/six/seven/ei= ght /nine/colon/semicolon/less/equal/greater/question/at/A/B/C/D/E/F/G/H/I/J/= K/L/M /N/O/P/Q/R/S/T/U/V/W/X/Y/Z/bracketleft/backslash/bracketright/asciicircum= /underscore/grave/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/bra= celeft /bar/braceright/asciitilde/.notdef/.notdef/.notdef/quotesinglbase/florin /quotedblbase/ellipsis/dagger/daggerdbl/circumflex/perthousand/Scaron /guilsinglleft/OE/.notdef/.notdef/.notdef/.notdef/quoteleft/quoteright /quotedblleft/quotedblright/bullet/endash/emdash/tilde/trademark/scaron /guilsinglright/oe/.notdef/.notdef/Ydieresis/.notdef/exclamdown/cent/ster= ling /currency/yen/brokenbar/section/dieresis/copyright/ordfeminine/guillemotl= eft /logicalnot/hyphen/registered/macron/degree/plusminus/twosuperior/threesu= perior /acute/mu/paragraph/periodcentered/cedilla/onesuperior/ordmasculine /guillemotright/onequarter/onehalf/threequarters/questiondown/Agrave/Aacu= te /Acircumflex/Atilde/Adieresis/Aring/AE/Ccedilla/Egrave/Eacute/Ecircumflex= /Edieresis/Igrave/Iacute/Icircumflex/Idieresis/Eth/Ntilde/Ograve/Oacute /Ocircumflex/Otilde/Odieresis/multiply/Oslash/Ugrave/Uacute/Ucircumflex /Udieresis/Yacute/Thorn/germandbls/agrave/aacute/acircumflex/atilde/adier= esis /aring/ae/ccedilla/egrave/eacute/ecircumflex/edieresis/igrave/iacute /icircumflex/idieresis/eth/ntilde/ograve/oacute/ocircumflex/otilde/odiere= sis /divide/oslash/ugrave/uacute/ucircumflex/udieresis/yacute/thorn/ydieresis= ]def LATENC /_Helvetica /Helvetica reencode=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 1896 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (D)1242 1896 MS (e)1363 1896 MS (p)1456 1896 MS (e)1558 1896 MS (n)1651 = 1896 MS (d)1753 1896 MS (s)1855 1896 MS ( )1948 1896 MS (o)1994 1896 MS = (n)2096 1896 MS ( )2198 1896 MS (t)2243 1896 MS (i)2299 1896 MS (m)2345 = 1896 MS (e)2493 1896 MS (o)2586 1896 MS=0A= (u)2688 1896 MS (t)2789 1896 MS ( )2845 1896 MS (f)2890 1896 MS (o)2946 = 1896 MS (r)3048 1896 MS ( )3113 1896 MS (g)3159 1896 MS (e)3260 1896 MS = (n)3353 1896 MS (e)3455 1896 MS (r)3548 1896 MS (a)3613 1896 MS (l)3706 = 1896 MS ( )3751 1896 MS (q)3797 1896 MS=0A= (u)3899 1896 MS (e)4001 1896 MS (r)4094 1896 MS (y)4160 1896 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 2133 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (1)1504 2133 MS ( )1597 2133 MS (c)1643 2133 MS (h)1736 2133 MS (a)1838 = 2133 MS (n)1931 2133 MS (n)2033 2133 MS (e)2135 2133 MS (l)2228 2133 MS = ( )2274 2133 MS (u)2320 2133 MS (s)2421 2133 MS (e)2513 2133 MS (d)2606 = 2133 MS ( )2708 2133 MS (f)2754 2133 MS=0A= (o)2810 2133 MS (r)2911 2133 MS ( )2976 2133 MS (e)3022 2133 MS (a)3115 = 2133 MS (c)3207 2133 MS (h)3300 2133 MS ( )3402 2133 MS (1)3447 2133 MS = (0)3540 2133 MS ( )3633 2133 MS (s)3679 2133 MS (e)3772 2133 MS (c)3864 = 2133 MS (o)3957 2133 MS (n)4059 2133 MS=0A= (d)4161 2133 MS (s)4263 2133 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 2369 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (1)1504 2369 MS (8)1597 2369 MS (0)1690 2369 MS ( )1783 2369 MS (s)1829 = 2369 MS (e)1921 2369 MS (c)2014 2369 MS (o)2107 2369 MS (n)2209 2369 MS = (d)2311 2369 MS (s)2413 2369 MS ( )2505 2369 MS (d)2551 2369 MS (e)2653 = 2369 MS (f)2746 2369 MS (a)2802 2369 MS=0A= (u)2895 2369 MS (l)2996 2369 MS (t)3042 2369 MS ( )3097 2369 MS (v)3143 = 2369 MS (a)3234 2369 MS (l)3327 2369 MS (u)3373 2369 MS (e)3476 2369 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 2606 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (D)1242 2606 MS (e)1363 2606 MS (p)1456 2606 MS (e)1558 2606 MS (n)1651 = 2606 MS (d)1753 2606 MS (s)1855 2606 MS ( )1948 2606 MS (o)1994 2606 MS = (n)2096 2606 MS ( )2198 2606 MS (r)2243 2606 MS (a)2308 2606 MS (t)2400 = 2606 MS (i)2456 2606 MS (o)2502 2606 MS=0A= ( )2604 2606 MS (o)2650 2606 MS (f)2752 2606 MS ( )2807 2606 MS=0A= (m)2853 2606 MS (a)3001 2606 MS (x)3094 2606 MS (r)3186 2606 MS (a)3250 = 2606 MS (t)3343 2606 MS (e)3399 2606 MS ( )3491 2606 MS=0A= (t)3537 2606 MS (o)3593 2606 MS ( )3695 2606 MS=0A= (m)3740 2606 MS (i)3888 2606 MS (n)3934 2606 MS (r)4036 2606 MS (a)4101 = 2606 MS (t)4193 2606 MS (e)4248 2606 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 2842 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (8)1504 2842 MS ( )1597 2842 MS (K)1643 2842 MS (b)1764 2842 MS (p)1866 = 2842 MS (s)1968 2842 MS ( )2061 2842 MS (i)2107 2842 MS (s)2153 2842 MS = ( )2245 2842 MS (d)2291 2842 MS (e)2393 2842 MS (f)2485 2842 MS (a)2541 = 2842 MS (u)2634 2842 MS (l)2736 2842 MS=0A= (t)2781 2842 MS ( )2837 2842 MS=0A= (m)2883 2842 MS (i)3031 2842 MS (n)3077 2842 MS (r)3178 2842 MS (a)3243 = 2842 MS (t)3335 2842 MS (e)3391 2842 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 3079 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (n)1504 3079 MS (u)1606 3079 MS (m)1708 3079 MS (b)1856 3079 MS (e)1958 = 3079 MS (r)2051 3079 MS ( )2116 3079 MS (o)2162 3079 MS (f)2264 3079 MS = ( )2319 3079 MS (g)2365 3079 MS (r)2466 3079 MS (o)2531 3079 MS (u)2633 = 3079 MS (p)2735 3079 MS (s)2837 3079 MS=0A= ( )2930 3079 MS (=3D)2975 3079 MS ( )3073 3079 MS (l)3119 3079 MS = (o)3165 3079 MS (g)3267 3079 MS (_)3369 3079 MS (\()3461 3079 MS (4)3517 = 3079 MS (/)3610 3079 MS (3)3655 3079 MS (\))3748 3079 MS ( )3803 3079 MS = (\()3849 3079 MS=0A= (m)3905 3079 MS (a)4052 3079 MS (x)4145 3079 MS (r)4238 3079 MS (a)4302 = 3079 MS (t)4395 3079 MS (e)4450 3079 MS=0A= (/)4543 3079 MS=0A= (m)4589 3079 MS (i)4737 3079 MS (n)4783 3079 MS (r)4885 3079 MS (a)4950 = 3079 MS (t)5042 3079 MS (e)5098 3079 MS=0A= (\))5190 3079 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 3315 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (T)1242 3315 MS (h)1344 3315 MS (i)1446 3315 MS (s)1492 3315 MS ( )1585 = 3315 MS (i)1631 3315 MS (s)1677 3315 MS ( )1770 3315 MS (a)1816 3315 MS = (n)1909 3315 MS ( )2011 3315 MS (i)2056 3315 MS (s)2102 3315 MS (s)2194 = 3315 MS (u)2287 3315 MS (e)2389 3315 MS=0A= ( )2482 3315 MS (f)2528 3315 MS (o)2584 3315 MS (r)2685 3315 MS ( )2750 = 3315 MS (A)2796 3315 MS (S)2917 3315 MS (M)3028 3315 MS (,)3167 3315 MS = ( )3212 3315 MS (n)3258 3315 MS (o)3360 3315 MS (t)3462 3315 MS ( )3518 = 3315 MS (s)3564 3315 MS (o)3657 3315 MS=0A= ( )3759 3315 MS (f)3804 3315 MS (o)3860 3315 MS (r)3962 3315 MS ( )4026 = 3315 MS (S)4072 3315 MS (S)4183 3315 MS (M)4294 3315 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 3552 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (S)1242 3552 MS (t)1353 3552 MS (a)1409 3552 MS (t)1502 3552 MS (e)1557 = 3552 MS ( )1650 3552 MS (i)1696 3552 MS (n)1742 3552 MS (f)1843 3552 MS = (o)1899 3552 MS (r)2001 3552 MS (m)2066 3552 MS (a)2213 3552 MS (t)2306 = 3552 MS (i)2362 3552 MS (o)2408 3552 MS=0A= (n)2510 3552 MS ( )2612 3552 MS (i)2658 3552 MS (n)2704 3552 MS ( )2805 = 3552 MS (r)2851 3552 MS (o)2916 3552 MS (u)3018 3552 MS (t)3119 3552 MS = (e)3174 3552 MS (r)3267 3552 MS (s)3332 3552 MS (/)3425 3552 MS (s)3471 = 3552 MS (w)3561 3552 MS (i)3694 3552 MS=0A= (t)3739 3552 MS (c)3794 3552 MS (h)3887 3552 MS (e)3988 3552 MS (s)4081 = 3552 MS ( )4173 3552 MS (s)4219 3552 MS (h)4312 3552 MS (o)4414 3552 MS = (u)4516 3552 MS (l)4618 3552 MS (d)4664 3552 MS ( )4766 3552 MS (b)4812 = 3552 MS (e)4914 3552 MS ( )5007 3552 MS=0A= (o)5053 3552 MS (k)5154 3552 MS ( )5247 3552 MS (f)5293 3552 MS (o)5349 = 3552 MS (r)5450 3552 MS ( )5515 3552 MS=0A= (t)1242 3728 MS (h)1298 3728 MS (i)1400 3728 MS (s)1446 3728 MS ( )1539 = 3728 MS (i)1584 3728 MS (f)1630 3728 MS ( )1686 3728 MS (o)1732 3728 MS = (k)1833 3728 MS ( )1926 3728 MS (f)1971 3728 MS (o)2027 3728 MS (r)2129 = 3728 MS ( )2194 3728 MS (o)2239 3728 MS=0A= (t)2341 3728 MS (h)2397 3728 MS (e)2499 3728 MS (r)2592 3728 MS ( )2656 = 3728 MS (m)2702 3728 MS (u)2850 3728 MS (l)2952 3728 MS (t)2998 3728 MS = (i)3053 3728 MS (c)3099 3728 MS (a)3191 3728 MS (s)3284 3728 MS (t)3377 = 3728 MS ( )3432 3728 MS (a)3478 3728 MS=0A= (p)3571 3728 MS (p)3673 3728 MS (s)3775 3728 MS=0A= showpage=0A= %%Page: 9 9=0A= 90 rotate 0 -612 translate=0A= 13.059 598.762 translate 72 600 div dup neg scale=0A= 0 0 transform .25 add round .25 sub exch .25 add round .25 sub exch = itransform translate=0A= n=0A= 5999 4499 195 193 B=0A= 1 g=0A= f=0A= [83 0 0 -83 0 0]/Helvetica-Bold MF=0A= 0 g=0A= (9)5797 4495 MS=0A= [300 0 0 -300 0 0]/Helvetica-Bold MF=0A= (W)1269 741 MS (E)1552 741 MS (B)1752 741 MS (R)1969 741 MS (C)2186 741 = MS ( )2403 741 MS (o)2486 741 MS (u)2669 741 MS (t)2852 741 MS (s)2952 = 741 MS (t)3119 741 MS (a)3219 741 MS (n)3386 741 MS (d)3569 741 MS = (i)3752 741 MS (n)3835 741 MS=0A= (g)4019 741 MS ( )4202 741 MS (i)4285 741 MS (s)4368 741 MS (s)4535 741 = MS (u)4702 741 MS (e)4886 741 MS (s)5053 741 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 1424 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (A)979 1424 MS (b)1123 1424 MS (i)1245 1424 MS (l)1301 1424 MS (i)1356 = 1424 MS (t)1412 1424 MS (y)1479 1424 MS ( )1588 1424 MS (o)1644 1424 MS = (f)1766 1424 MS ( )1833 1424 MS (n)1888 1424 MS (e)2010 1424 MS (t)2121 = 1424 MS (w)2187 1424 MS (o)2344 1424 MS=0A= (r)2466 1424 MS (k)2544 1424 MS (i)2655 1424 MS (n)2711 1424 MS (g)2832 = 1424 MS ( )2954 1424 MS (i)3009 1424 MS (n)3065 1424 MS (f)3187 1424 MS = (r)3254 1424 MS (a)3332 1424 MS (s)3443 1424 MS (t)3554 1424 MS (r)3621 = 1424 MS (u)3699 1424 MS (c)3821 1424 MS=0A= (t)3932 1424 MS (u)3999 1424 MS (r)4121 1424 MS (e)4199 1424 MS ( )4310 = 1424 MS (t)4366 1424 MS (o)4433 1424 MS ( )4555 1424 MS (s)4611 1424 MS = (c)4722 1424 MS (a)4832 1424 MS (l)4943 1424 MS (e)4999 1424 MS ( )5110 = 1424 MS (t)5166 1424 MS (o)5233 1424 MS=0A= ( )5354 1424 MS=0A= (h)979 1610 MS (a)1101 1610 MS (n)1212 1610 MS (d)1334 1610 MS (l)1456 = 1610 MS (e)1512 1610 MS ( )1623 1610 MS (r)1679 1610 MS (e)1757 1610 MS = (c)1868 1610 MS (e)1979 1610 MS (i)2090 1610 MS (v)2146 1610 MS (e)2257 = 1610 MS (r)2368 1610 MS (s)2446 1610 MS=0A= ( )2557 1610 MS (t)2613 1610 MS (h)2680 1610 MS (a)2802 1610 MS (t)2913 = 1610 MS ( )2980 1610 MS (s)3036 1610 MS (e)3147 1610 MS (n)3258 1610 MS = (d)3380 1610 MS ( )3502 1610 MS (o)3558 1610 MS (n)3680 1610 MS (e)3802 = 1610 MS ( )3912 1610 MS (I)3967 1610 MS=0A= (G)4022 1610 MS (M)4178 1610 MS (P)4345 1610 MS ( )4478 1610 MS (j)4533 = 1610 MS (o)4589 1610 MS (i)4710 1610 MS (n)4766 1610 MS ( )4887 1610 MS = (a)4942 1610 MS (n)5053 1610 MS (d)5175 1610 MS ( )5297 1610 MS=0A= (l)979 1797 MS (e)1035 1797 MS (a)1146 1797 MS (v)1257 1797 MS (e)1368 = 1797 MS ( )1479 1797 MS (m)1535 1797 MS (e)1713 1797 MS (s)1824 1797 MS = (s)1936 1797 MS (a)2047 1797 MS (g)2158 1797 MS (e)2280 1797 MS ( )2391 = 1797 MS (e)2447 1797 MS (a)2558 1797 MS=0A= (c)2670 1797 MS (h)2781 1797 MS ( )2903 1797 MS (1)2959 1797 MS (0)3070 = 1797 MS ( )3181 1797 MS (s)3237 1797 MS (e)3348 1797 MS (c)3459 1797 MS = (o)3570 1797 MS (n)3692 1797 MS (d)3814 1797 MS (s)3937 1797 MS=0A= [167 0 0 -167 0 0]/Helvetica MF=0A= /IsChar{exch/CharStrings get exch known}bd/MapCh{3 -1 roll/Encoding get = 3 1 roll put}bd/MapDegree{dup 16#b0 exch/degree IsChar{/degree}{/ring}ifelse = MapCh} bd/MapBB{dup 16#a6 exch/brokenbar IsChar{/brokenbar}{/bar}ifelse = MapCh}bd /reencode{findfont begin currentdict dup length dict begin{1 index/FID = ne{def} {pop pop}ifelse}forall/FontName exch def dup length 0 ne{/Encoding = Encoding 256 array copy def 0 exch{dup type/nametype eq{Encoding 2 index 2 index put = pop 1 add}{exch pop}ifelse}forall}if pop currentdict dup end end/FontName get = exch definefont dup MapDegree = MapBB}bd/LATENC[0/grave/acute/circumflex/tilde/macron /breve/dotaccent/dieresis/ring/cedilla/hungarumlaut/ogonek/caron/dotlessi= /fi/fl /Lslash/lslash/Zcaron/zcaron/minus/.notdef/.notdef/.notdef/.notdef/.notde= f /.notdef/.notdef/.notdef/.notdef/.notdef/.notdef/space/exclam/quotedbl /numbersign/dollar/percent/ampersand/quotesingle/parenleft/parenright/ast= erisk /plus/comma/hyphen/period/slash/zero/one/two/three/four/five/six/seven/ei= ght /nine/colon/semicolon/less/equal/greater/question/at/A/B/C/D/E/F/G/H/I/J/= K/L/M /N/O/P/Q/R/S/T/U/V/W/X/Y/Z/bracketleft/backslash/bracketright/asciicircum= /underscore/grave/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/bra= celeft /bar/braceright/asciitilde/.notdef/.notdef/.notdef/quotesinglbase/florin /quotedblbase/ellipsis/dagger/daggerdbl/circumflex/perthousand/Scaron /guilsinglleft/OE/.notdef/.notdef/.notdef/.notdef/quoteleft/quoteright /quotedblleft/quotedblright/bullet/endash/emdash/tilde/trademark/scaron /guilsinglright/oe/.notdef/.notdef/Ydieresis/.notdef/exclamdown/cent/ster= ling /currency/yen/brokenbar/section/dieresis/copyright/ordfeminine/guillemotl= eft /logicalnot/hyphen/registered/macron/degree/plusminus/twosuperior/threesu= perior /acute/mu/paragraph/periodcentered/cedilla/onesuperior/ordmasculine /guillemotright/onequarter/onehalf/threequarters/questiondown/Agrave/Aacu= te /Acircumflex/Atilde/Adieresis/Aring/AE/Ccedilla/Egrave/Eacute/Ecircumflex= /Edieresis/Igrave/Iacute/Icircumflex/Idieresis/Eth/Ntilde/Ograve/Oacute /Ocircumflex/Otilde/Odieresis/multiply/Oslash/Ugrave/Uacute/Ucircumflex /Udieresis/Yacute/Thorn/germandbls/agrave/aacute/acircumflex/atilde/adier= esis /aring/ae/ccedilla/egrave/eacute/ecircumflex/edieresis/igrave/iacute /icircumflex/idieresis/eth/ntilde/ograve/oacute/ocircumflex/otilde/odiere= sis /divide/oslash/ugrave/uacute/ucircumflex/udieresis/yacute/thorn/ydieresis= ]def LATENC /_Helvetica /Helvetica reencode=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 2019 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (P)1242 2019 MS (e)1353 2019 MS (r)1446 2019 MS (f)1511 2019 MS (o)1567 = 2019 MS (r)1669 2019 MS (m)1734 2019 MS (a)1881 2019 MS (n)1974 2019 MS = (c)2076 2019 MS (e)2169 2019 MS ( )2261 2019 MS (f)2307 2019 MS (i)2363 = 2019 MS (g)2409 2019 MS (u)2511 2019 MS=0A= (r)2613 2019 MS (e)2677 2019 MS (s)2770 2019 MS ( )2863 2019 MS (o)2909 = 2019 MS (n)3011 2019 MS ( )3113 2019 MS (t)3158 2019 MS (h)3213 2019 MS = (e)3315 2019 MS ( )3408 2019 MS (a)3454 2019 MS (m)3547 2019 MS (o)3695 = 2019 MS (u)3797 2019 MS (n)3899 2019 MS=0A= (t)4001 2019 MS ( )4057 2019 MS (o)4103 2019 MS (f)4204 2019 MS ( )4260 = 2019 MS (I)4306 2019 MS (G)4351 2019 MS (M)4481 2019 MS (P)4620 2019 MS = ( )4731 2019 MS (m)4777 2019 MS (e)4924 2019 MS (s)5017 2019 MS (s)5110 = 2019 MS (a)5202 2019 MS (g)5295 2019 MS=0A= (e)5397 2019 MS (s)5490 2019 MS ( )5583 2019 MS=0A= (r)1242 2175 MS (o)1307 2175 MS (u)1409 2175 MS (t)1511 2175 MS (e)1567 = 2175 MS (r)1659 2175 MS (s)1724 2175 MS (/)1817 2175 MS (s)1862 2175 MS = (w)1952 2175 MS (i)2085 2175 MS (t)2130 2175 MS (c)2185 2175 MS (h)2278 = 2175 MS (e)2380 2175 MS (s)2473 2175 MS=0A= ( )2566 2175 MS (c)2612 2175 MS (a)2704 2175 MS (n)2797 2175 MS ( )2899 = 2175 MS (h)2945 2175 MS (a)3047 2175 MS (n)3139 2175 MS (d)3241 2175 MS = (l)3343 2175 MS (e)3389 2175 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 2392 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (W)1504 2392 MS (h)1662 2392 MS (e)1764 2392 MS (n)1857 2392 MS ( )1959 = 2392 MS (h)2005 2392 MS (e)2107 2392 MS (a)2199 2392 MS (v)2292 2392 MS = (y)2385 2392 MS ( )2476 2392 MS (d)2523 2392 MS (a)2625 2392 MS (t)2718 = 2392 MS (a)2774 2392 MS ( )2867 2392 MS=0A= (p)2913 2392 MS (a)3014 2392 MS (c)3107 2392 MS (k)3200 2392 MS (e)3292 = 2392 MS (t)3385 2392 MS ( )3440 2392 MS (l)3486 2392 MS (o)3531 2392 MS = (a)3633 2392 MS (d)3726 2392 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 2609 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (W)1504 2609 MS (h)1662 2609 MS (a)1764 2609 MS (t)1856 2609 MS ( )1912 = 2609 MS (i)1958 2609 MS (m)2004 2609 MS (p)2152 2609 MS (a)2254 2609 MS = (c)2346 2609 MS (t)2439 2609 MS ( )2494 2609 MS (o)2540 2609 MS (n)2642 = 2609 MS ( )2744 2609 MS (d)2790 2609 MS=0A= (a)2892 2609 MS (t)2985 2609 MS (a)3041 2609 MS ( )3133 2609 MS (p)3179 = 2609 MS (a)3281 2609 MS (c)3374 2609 MS (k)3466 2609 MS (e)3559 2609 MS = (t)3652 2609 MS (s)3707 2609 MS (?)3800 2609 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 2825 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (W)1504 2825 MS (h)1662 2825 MS (a)1764 2825 MS (t)1856 2825 MS ( )1912 = 2825 MS (i)1958 2825 MS (m)2004 2825 MS (p)2152 2825 MS (a)2254 2825 MS = (c)2346 2825 MS (t)2439 2825 MS ( )2494 2825 MS (o)2540 2825 MS (n)2642 = 2825 MS ( )2744 2825 MS (I)2790 2825 MS=0A= (G)2836 2825 MS (M)2965 2825 MS (P)3104 2825 MS ( )3215 2825 MS (l)3261 = 2825 MS (a)3307 2825 MS (t)3399 2825 MS (e)3454 2825 MS (n)3547 2825 MS = (c)3649 2825 MS (y)3743 2825 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 3042 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (S)1242 3042 MS (a)1353 3042 MS (m)1446 3042 MS (e)1594 3042 MS ( )1687 = 3042 MS (q)1733 3042 MS (u)1835 3042 MS (e)1937 3042 MS (s)2030 3042 MS = (t)2122 3042 MS (i)2178 3042 MS (o)2223 3042 MS (n)2325 3042 MS (s)2427 = 3042 MS ( )2520 3042 MS (f)2566 3042 MS=0A= (o)2622 3042 MS (r)2724 3042 MS ( )2788 3042 MS (r)2834 3042 MS (e)2899 = 3042 MS (s)2991 3042 MS (u)3084 3042 MS (l)3186 3042 MS (t)3231 3042 MS = (i)3287 3042 MS (n)3333 3042 MS (g)3435 3042 MS ( )3537 3042 MS (P)3583 = 3042 MS (I)3694 3042 MS (M)3740 3042 MS=0A= ( )3879 3042 MS (S)3925 3042 MS (M)4036 3042 MS ( )4175 3042 MS (m)4220 = 3042 MS (e)4368 3042 MS (s)4461 3042 MS (s)4554 3042 MS (a)4647 3042 MS = (g)4740 3042 MS (e)4842 3042 MS (s)4934 3042 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 3294 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (N)979 3294 MS (o)1123 3294 MS ( )1245 3294 MS (p)1301 3294 MS (e)1423 = 3294 MS (r)1534 3294 MS (f)1612 3294 MS (o)1679 3294 MS (r)1801 3294 MS = (m)1879 3294 MS (a)2057 3294 MS (n)2168 3294 MS (c)2290 3294 MS (e)2401 = 3294 MS ( )2512 3294 MS (f)2568 3294 MS=0A= (i)2635 3294 MS (g)2691 3294 MS (u)2812 3294 MS (r)2934 3294 MS (e)3012 = 3294 MS (s)3123 3294 MS ( )3234 3294 MS (y)3291 3294 MS (e)3401 3294 MS = (t)3512 3294 MS ( )3579 3294 MS (f)3635 3294 MS (r)3702 3294 MS (o)3780 = 3294 MS (m)3902 3294 MS ( )4080 3294 MS=0A= (v)4136 3294 MS (e)4247 3294 MS (n)4358 3294 MS (d)4480 3294 MS (o)4602 = 3294 MS (r)4724 3294 MS (s)4802 3294 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 3553 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (D)979 3553 MS (i)1123 3553 MS (g)1179 3553 MS (i)1301 3553 MS (t)1357 = 3553 MS (a)1424 3553 MS (l)1534 3553 MS ( )1590 3553 MS (F)1645 3553 MS = (o)1767 3553 MS (u)1888 3553 MS (n)2010 3553 MS (t)2132 3553 MS (a)2199 = 3553 MS (i)2310 3553 MS (n)2366 3553 MS=0A= ( )2487 3553 MS=0A= (r)2543 3553 MS (e)2621 3553 MS (p)2732 3553 MS (l)2854 3553 MS (i)2909 = 3553 MS (c)2965 3553 MS (a)3076 3553 MS (t)3187 3553 MS (o)3254 3553 MS = (r)3376 3553 MS ( )3454 3553 MS=0A= (p)3510 3553 MS (e)3632 3553 MS (r)3743 3553 MS (f)3821 3553 MS (o)3887 = 3553 MS (r)4009 3553 MS (m)4087 3553 MS (a)4265 3553 MS (n)4376 3553 MS = (c)4498 3553 MS (e)4609 3553 MS ( )4720 3553 MS (f)4776 3553 MS (i)4843 = 3553 MS (g)4898 3553 MS (u)5020 3553 MS=0A= (r)5142 3553 MS (e)5220 3553 MS (s)5331 3553 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 3775 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (S)1242 3775 MS (i)1353 3775 MS (m)1399 3775 MS (i)1547 3775 MS (l)1593 = 3775 MS (a)1639 3775 MS (r)1732 3775 MS ( )1797 3775 MS (t)1842 3775 MS = (o)1898 3775 MS ( )2000 3775 MS (I)2046 3775 MS (G)2091 3775 MS (M)2220 = 3775 MS (P)2359 3775 MS ( )2470 3775 MS=0A= (j)2516 3775 MS (o)2562 3775 MS (i)2664 3775 MS (n)2710 3775 MS ( )2812 = 3775 MS (m)2858 3775 MS (e)3006 3775 MS (s)3099 3775 MS (s)3191 3775 MS = (a)3284 3775 MS (g)3377 3775 MS (e)3479 3775 MS ( )3572 3775 MS (p)3618 = 3775 MS (r)3720 3775 MS (o)3785 3775 MS=0A= (c)3886 3775 MS (e)3979 3775 MS (s)4072 3775 MS (s)4164 3775 MS (i)4257 = 3775 MS (n)4303 3775 MS (g)4405 3775 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 3992 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (1)1242 3992 MS (9)1335 3992 MS ( )1428 3992 MS (m)1474 3992 MS (s)1622 = 3992 MS ( )1715 3992 MS (l)1761 3992 MS (a)1806 3992 MS (t)1898 3992 MS = (e)1954 3992 MS (n)2047 3992 MS (c)2149 3992 MS (y)2241 3992 MS ( )2332 = 3992 MS (f)2378 3992 MS (o)2434 3992 MS=0A= (r)2536 3992 MS ( )2601 3992 MS (j)2647 3992 MS (o)2693 3992 MS (i)2795 = 3992 MS (n)2841 3992 MS ( )2943 3992 MS (m)2989 3992 MS (e)3137 3992 MS = (s)3230 3992 MS (s)3323 3992 MS (a)3416 3992 MS (g)3509 3992 MS (e)3611 = 3992 MS (s)3704 3992 MS ( )3796 3992 MS=0A= (w)3839 3992 MS (h)3973 3992 MS (e)4074 3992 MS (n)4165 3992 MS ( )4267 = 3992 MS (1)4313 3992 MS ( )4406 3992 MS (r)4452 3992 MS (e)4517 3992 MS = (c)4609 3992 MS (e)4702 3992 MS (i)4795 3992 MS (v)4841 3992 MS (e)4932 = 3992 MS (r)5025 3992 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 4209 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (2)1242 4209 MS (0)1335 4209 MS ( )1428 4209 MS (m)1474 4209 MS (s)1622 = 4209 MS ( )1715 4209 MS (l)1761 4209 MS (a)1806 4209 MS (t)1898 4209 MS = (e)1954 4209 MS (n)2047 4209 MS (c)2149 4209 MS (y)2241 4209 MS ( )2332 = 4209 MS (f)2378 4209 MS (o)2434 4209 MS=0A= (r)2536 4209 MS ( )2601 4209 MS (j)2647 4209 MS (o)2693 4209 MS (i)2795 = 4209 MS (n)2841 4209 MS ( )2943 4209 MS (m)2989 4209 MS (e)3137 4209 MS = (s)3230 4209 MS (s)3323 4209 MS (a)3416 4209 MS (g)3509 4209 MS (e)3611 = 4209 MS (s)3704 4209 MS ( )3796 4209 MS=0A= (w)3839 4209 MS (h)3973 4209 MS (e)4074 4209 MS (n)4165 4209 MS ( )4267 = 4209 MS (1)4313 4209 MS (0)4406 4209 MS (,)4499 4209 MS (0)4545 4209 MS = (0)4638 4209 MS (0)4730 4209 MS ( )4823 4209 MS (r)4869 4209 MS (e)4933 = 4209 MS (c)5026 4209 MS (e)5118 4209 MS=0A= (i)5211 4209 MS (v)5257 4209 MS (e)5348 4209 MS (r)5441 4209 MS (s)5506 = 4209 MS=0A= showpage=0A= %%Page: 10 10=0A= 90 rotate 0 -612 translate=0A= 13.059 598.762 translate 72 600 div dup neg scale=0A= 0 0 transform .25 add round .25 sub exch .25 add round .25 sub exch = itransform translate=0A= n=0A= 5999 4499 195 193 B=0A= 1 g=0A= f=0A= [83 0 0 -83 0 0]/Helvetica-Bold MF=0A= 0 g=0A= (1)5751 4495 MS (0)5797 4495 MS=0A= [300 0 0 -300 0 0]/Helvetica-Bold MF=0A= (W)1269 741 MS (E)1552 741 MS (B)1752 741 MS (R)1969 741 MS (C)2186 741 = MS ( )2403 741 MS (o)2486 741 MS (u)2669 741 MS (t)2852 741 MS (s)2952 = 741 MS (t)3119 741 MS (a)3219 741 MS (n)3386 741 MS (d)3569 741 MS = (i)3752 741 MS (n)3835 741 MS=0A= (g)4019 741 MS ( )4202 741 MS (i)4285 741 MS (s)4368 741 MS (s)4535 741 = MS (u)4702 741 MS (e)4886 741 MS (s)5053 741 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 1443 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (L)979 1443 MS (o)1101 1443 MS (n)1223 1443 MS (g)1345 1443 MS ( )1467 = 1443 MS (I)1522 1443 MS (G)1577 1443 MS (M)1733 1443 MS (P)1899 1443 MS = ( )2032 1443 MS (a)2088 1443 MS (n)2199 1443 MS (d)2321 1443 MS ( )2443 = 1443 MS (P)2499 1443 MS (I)2631 1443 MS=0A= (M)2687 1443 MS ( )2853 1443 MS (S)2908 1443 MS (M)3041 1443 MS ( )3208 = 1443 MS (j)3263 1443 MS (o)3319 1443 MS (i)3440 1443 MS (n)3496 1443 MS = ( )3618 1443 MS (l)3673 1443 MS (a)3729 1443 MS (t)3840 1443 MS (e)3906 = 1443 MS (n)4017 1443 MS (c)4139 1443 MS=0A= (y)4251 1443 MS ( )4360 1443 MS (c)4417 1443 MS (a)4528 1443 MS (n)4639 = 1443 MS ( )4761 1443 MS=0A= (a)979 1653 MS (d)1090 1653 MS (v)1212 1653 MS (e)1323 1653 MS (r)1434 = 1653 MS (s)1513 1653 MS (e)1624 1653 MS (l)1735 1653 MS (y)1792 1653 MS = ( )1901 1653 MS (a)1958 1653 MS (f)2069 1653 MS (f)2136 1653 MS (e)2203 = 1653 MS (c)2314 1653 MS (t)2425 1653 MS=0A= ( )2492 1653 MS (t)2548 1653 MS (h)2615 1653 MS (r)2737 1653 MS (o)2815 = 1653 MS (u)2936 1653 MS (g)3058 1653 MS (h)3180 1653 MS (p)3302 1653 MS = (u)3424 1653 MS (t)3546 1653 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 1935 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (L)979 1935 MS (o)1101 1935 MS (s)1223 1935 MS (s)1334 1935 MS (e)1445 = 1935 MS (s)1557 1935 MS ( )1668 1935 MS (d)1724 1935 MS (u)1846 1935 MS = (e)1968 1935 MS ( )2079 1935 MS (t)2135 1935 MS (o)2202 1935 MS ( )2323 = 1935 MS (b)2379 1935 MS (e)2501 1935 MS=0A= (h)2612 1935 MS (a)2734 1935 MS (v)2845 1935 MS (i)2957 1935 MS (o)3013 = 1935 MS (r)3135 1935 MS ( )3212 1935 MS (o)3268 1935 MS (f)3390 1935 MS = ( )3456 1935 MS (m)3512 1935 MS (u)3690 1935 MS (l)3811 1935 MS (t)3867 = 1935 MS (i)3933 1935 MS (c)3989 1935 MS=0A= (a)4100 1935 MS (s)4211 1935 MS (t)4322 1935 MS ( )4389 1935 MS (p)4445 = 1935 MS (r)4567 1935 MS (o)4645 1935 MS (t)4767 1935 MS (o)4834 1935 MS = (c)4955 1935 MS (o)5066 1935 MS (l)5188 1935 MS ( )5244 1935 MS (c)5300 = 1935 MS (a)5411 1935 MS (n)5522 1935 MS=0A= ( )5644 1935 MS=0A= (a)979 2146 MS (d)1090 2146 MS (v)1212 2146 MS (e)1323 2146 MS (r)1434 = 2146 MS (s)1513 2146 MS (e)1624 2146 MS (l)1735 2146 MS (y)1792 2146 MS = ( )1901 2146 MS (a)1958 2146 MS (f)2069 2146 MS (f)2136 2146 MS (e)2203 = 2146 MS (c)2314 2146 MS (t)2425 2146 MS=0A= ( )2492 2146 MS (t)2548 2146 MS (h)2615 2146 MS (r)2737 2146 MS (o)2815 = 2146 MS (u)2936 2146 MS (g)3058 2146 MS (h)3180 2146 MS (p)3302 2146 MS = (u)3424 2146 MS (t)3546 2146 MS=0A= [167 0 0 -167 0 0]/Helvetica MF=0A= /IsChar{exch/CharStrings get exch known}bd/MapCh{3 -1 roll/Encoding get = 3 1 roll put}bd/MapDegree{dup 16#b0 exch/degree IsChar{/degree}{/ring}ifelse = MapCh} bd/MapBB{dup 16#a6 exch/brokenbar IsChar{/brokenbar}{/bar}ifelse = MapCh}bd /reencode{findfont begin currentdict dup length dict begin{1 index/FID = ne{def} {pop pop}ifelse}forall/FontName exch def dup length 0 ne{/Encoding = Encoding 256 array copy def 0 exch{dup type/nametype eq{Encoding 2 index 2 index put = pop 1 add}{exch pop}ifelse}forall}if pop currentdict dup end end/FontName get = exch definefont dup MapDegree = MapBB}bd/LATENC[0/grave/acute/circumflex/tilde/macron /breve/dotaccent/dieresis/ring/cedilla/hungarumlaut/ogonek/caron/dotlessi= /fi/fl /Lslash/lslash/Zcaron/zcaron/minus/.notdef/.notdef/.notdef/.notdef/.notde= f /.notdef/.notdef/.notdef/.notdef/.notdef/.notdef/space/exclam/quotedbl /numbersign/dollar/percent/ampersand/quotesingle/parenleft/parenright/ast= erisk /plus/comma/hyphen/period/slash/zero/one/two/three/four/five/six/seven/ei= ght /nine/colon/semicolon/less/equal/greater/question/at/A/B/C/D/E/F/G/H/I/J/= K/L/M /N/O/P/Q/R/S/T/U/V/W/X/Y/Z/bracketleft/backslash/bracketright/asciicircum= /underscore/grave/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/bra= celeft /bar/braceright/asciitilde/.notdef/.notdef/.notdef/quotesinglbase/florin /quotedblbase/ellipsis/dagger/daggerdbl/circumflex/perthousand/Scaron /guilsinglleft/OE/.notdef/.notdef/.notdef/.notdef/quoteleft/quoteright /quotedblleft/quotedblright/bullet/endash/emdash/tilde/trademark/scaron /guilsinglright/oe/.notdef/.notdef/Ydieresis/.notdef/exclamdown/cent/ster= ling /currency/yen/brokenbar/section/dieresis/copyright/ordfeminine/guillemotl= eft /logicalnot/hyphen/registered/macron/degree/plusminus/twosuperior/threesu= perior /acute/mu/paragraph/periodcentered/cedilla/onesuperior/ordmasculine /guillemotright/onequarter/onehalf/threequarters/questiondown/Agrave/Aacu= te /Acircumflex/Atilde/Adieresis/Aring/AE/Ccedilla/Egrave/Eacute/Ecircumflex= /Edieresis/Igrave/Iacute/Icircumflex/Idieresis/Eth/Ntilde/Ograve/Oacute /Ocircumflex/Otilde/Odieresis/multiply/Oslash/Ugrave/Uacute/Ucircumflex /Udieresis/Yacute/Thorn/germandbls/agrave/aacute/acircumflex/atilde/adier= esis /aring/ae/ccedilla/egrave/eacute/ecircumflex/edieresis/igrave/iacute /icircumflex/idieresis/eth/ntilde/ograve/oacute/ocircumflex/otilde/odiere= sis /divide/oslash/ugrave/uacute/ucircumflex/udieresis/yacute/thorn/ydieresis= ]def LATENC /_Helvetica /Helvetica reencode=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 2389 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (R)1242 2389 MS (P)1363 2389 MS (s)1474 2389 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 2626 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (S)1242 2626 MS (w)1351 2626 MS (i)1484 2626 MS (t)1529 2626 MS (c)1584 = 2626 MS (h)1677 2626 MS (o)1779 2626 MS (v)1881 2626 MS (e)1972 2626 MS = (r)2065 2626 MS ( )2130 2626 MS (f)2176 2626 MS (r)2232 2626 MS (o)2297 = 2626 MS (m)2399 2626 MS ( )2547 2626 MS=0A= (\()2593 2626 MS (*)2648 2626 MS (,)2713 2626 MS (G)2758 2626 MS = (\))2887 2626 MS ( )2943 2626 MS (t)2988 2626 MS (o)3044 2626 MS ( )3146 = 2626 MS (\()3192 2626 MS (S)3247 2626 MS (,)3358 2626 MS (G)3404 2626 MS = (\))3534 2626 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 2862 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (M)1242 2862 MS (S)1381 2862 MS (D)1492 2862 MS (P)1613 2862 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 3137 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (R)979 3137 MS (e)1123 3137 MS (c)1235 3137 MS (o)1346 3137 MS (m)1468 = 3137 MS (m)1646 3137 MS (e)1824 3137 MS (n)1936 3137 MS (d)2058 3137 MS = (a)2180 3137 MS (t)2291 3137 MS (i)2358 3137 MS (o)2414 3137 MS (n)2536 = 3137 MS (s)2658 3137 MS (?)2769 3137 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 3381 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (U)1242 3381 MS (s)1363 3381 MS (e)1456 3381 MS ( )1549 3381 MS (S)1595 = 3381 MS (S)1706 3381 MS (M)1817 3381 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 3617 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (I)1242 3617 MS (f)1288 3617 MS ( )1344 3617 MS (u)1390 3617 MS (s)1492 = 3617 MS (i)1584 3617 MS (n)1630 3617 MS (g)1732 3617 MS ( )1834 3617 MS = (A)1880 3617 MS (S)2001 3617 MS (M)2112 3617 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 3854 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (P)1504 3854 MS (l)1615 3854 MS (a)1661 3854 MS (c)1754 3854 MS (e)1847 = 3854 MS ( )1940 3854 MS (R)1986 3854 MS (P)2107 3854 MS ( )2218 3854 MS = (n)2264 3854 MS (e)2366 3854 MS (a)2458 3854 MS (r)2551 3854 MS ( )2616 = 3854 MS (s)2662 3854 MS (e)2755 3854 MS=0A= (n)2848 3854 MS (d)2950 3854 MS (e)3051 3854 MS (r)3144 3854 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\226)1354 4090 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (D)1504 4090 MS (o)1625 4090 MS ( )1727 4090 MS (n)1773 4090 MS (o)1875 = 4090 MS (t)1977 4090 MS ( )2033 4090 MS (u)2079 4090 MS (s)2181 4090 MS = (e)2273 4090 MS ( )2366 4090 MS (M)2411 4090 MS (S)2550 4090 MS (D)2661 = 4090 MS (P)2782 4090 MS ( )2893 4090 MS=0A= (i)2939 4090 MS (f)2985 4090 MS ( )3041 4090 MS (i)3087 4090 MS (t)3132 = 4090 MS ( )3188 4090 MS (c)3234 4090 MS (a)3326 4090 MS (n)3418 4090 MS = ( )3520 4090 MS (b)3566 4090 MS (e)3668 4090 MS ( )3761 4090 MS (a)3807 = 4090 MS (v)3900 4090 MS (o)3991 4090 MS=0A= (i)4093 4090 MS (d)4139 4090 MS (e)4241 4090 MS (d)4334 4090 MS=0A= showpage=0A= %%Page: 11 11=0A= 90 rotate 0 -612 translate=0A= 13.059 598.762 translate 72 600 div dup neg scale=0A= 0 0 transform .25 add round .25 sub exch .25 add round .25 sub exch = itransform translate=0A= n=0A= 5999 4499 195 193 B=0A= 1 g=0A= f=0A= [83 0 0 -83 0 0]/Helvetica-Bold MF=0A= 0 g=0A= (1)5751 4495 MS (1)5797 4495 MS=0A= [300 0 0 -300 0 0]/Helvetica-Bold MF=0A= (W)2002 741 MS (E)2285 741 MS (B)2485 741 MS (R)2702 741 MS (C)2919 741 = MS ( )3136 741 MS (r)3219 741 MS (o)3336 741 MS (a)3519 741 MS (d)3686 = 741 MS (m)3869 741 MS (a)4136 741 MS (p)4303 741 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 1443 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (F)979 1443 MS (u)1101 1443 MS (l)1223 1443 MS (l)1278 1443 MS ( )1333 = 1443 MS (T)1389 1443 MS (e)1511 1443 MS (c)1622 1443 MS (h)1733 1443 MS = (n)1855 1443 MS (i)1977 1443 MS (c)2033 1443 MS (a)2144 1443 MS (l)2255 = 1443 MS ( )2311 1443 MS (p)2366 1443 MS=0A= (a)2488 1443 MS (p)2599 1443 MS (e)2721 1443 MS (r)2832 1443 MS=0A= [167 0 0 -167 0 0]/Helvetica MF=0A= /IsChar{exch/CharStrings get exch known}bd/MapCh{3 -1 roll/Encoding get = 3 1 roll put}bd/MapDegree{dup 16#b0 exch/degree IsChar{/degree}{/ring}ifelse = MapCh} bd/MapBB{dup 16#a6 exch/brokenbar IsChar{/brokenbar}{/bar}ifelse = MapCh}bd /reencode{findfont begin currentdict dup length dict begin{1 index/FID = ne{def} {pop pop}ifelse}forall/FontName exch def dup length 0 ne{/Encoding = Encoding 256 array copy def 0 exch{dup type/nametype eq{Encoding 2 index 2 index put = pop 1 add}{exch pop}ifelse}forall}if pop currentdict dup end end/FontName get = exch definefont dup MapDegree = MapBB}bd/LATENC[0/grave/acute/circumflex/tilde/macron /breve/dotaccent/dieresis/ring/cedilla/hungarumlaut/ogonek/caron/dotlessi= /fi/fl /Lslash/lslash/Zcaron/zcaron/minus/.notdef/.notdef/.notdef/.notdef/.notde= f /.notdef/.notdef/.notdef/.notdef/.notdef/.notdef/space/exclam/quotedbl /numbersign/dollar/percent/ampersand/quotesingle/parenleft/parenright/ast= erisk /plus/comma/hyphen/period/slash/zero/one/two/three/four/five/six/seven/ei= ght /nine/colon/semicolon/less/equal/greater/question/at/A/B/C/D/E/F/G/H/I/J/= K/L/M /N/O/P/Q/R/S/T/U/V/W/X/Y/Z/bracketleft/backslash/bracketright/asciicircum= /underscore/grave/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/bra= celeft /bar/braceright/asciitilde/.notdef/.notdef/.notdef/quotesinglbase/florin /quotedblbase/ellipsis/dagger/daggerdbl/circumflex/perthousand/Scaron /guilsinglleft/OE/.notdef/.notdef/.notdef/.notdef/quoteleft/quoteright /quotedblleft/quotedblright/bullet/endash/emdash/tilde/trademark/scaron /guilsinglright/oe/.notdef/.notdef/Ydieresis/.notdef/exclamdown/cent/ster= ling /currency/yen/brokenbar/section/dieresis/copyright/ordfeminine/guillemotl= eft /logicalnot/hyphen/registered/macron/degree/plusminus/twosuperior/threesu= perior /acute/mu/paragraph/periodcentered/cedilla/onesuperior/ordmasculine /guillemotright/onequarter/onehalf/threequarters/questiondown/Agrave/Aacu= te /Acircumflex/Atilde/Adieresis/Aring/AE/Ccedilla/Egrave/Eacute/Ecircumflex= /Edieresis/Igrave/Iacute/Icircumflex/Idieresis/Eth/Ntilde/Ograve/Oacute /Ocircumflex/Otilde/Odieresis/multiply/Oslash/Ugrave/Uacute/Ucircumflex /Udieresis/Yacute/Thorn/germandbls/agrave/aacute/acircumflex/atilde/adier= esis /aring/ae/ccedilla/egrave/eacute/ecircumflex/edieresis/igrave/iacute /icircumflex/idieresis/eth/ntilde/ograve/oacute/ocircumflex/otilde/odiere= sis /divide/oslash/ugrave/uacute/ucircumflex/udieresis/yacute/thorn/ydieresis= ]def LATENC /_Helvetica /Helvetica reencode=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 1686 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (T)1242 1686 MS (a)1344 1686 MS (r)1437 1686 MS (g)1502 1686 MS (e)1604 = 1686 MS (t)1696 1686 MS ( )1752 1686 MS (d)1798 1686 MS (a)1900 1686 MS = (t)1992 1686 MS (e)2048 1686 MS ( )2141 1686 MS (A)2186 1686 MS (p)2307 = 1686 MS (r)2409 1686 MS (i)2474 1686 MS=0A= (l)2520 1686 MS ( )2566 1686 MS (1)2612 1686 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 1961 MS=0A= %%IncludeFont: Helvetica-BoldOblique=0A= [200 0 0 -200 0 0]/Helvetica-BoldOblique MF=0A= (n)979 1961 MS (s)1101 1961 MS ( )1212 1961 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (c)1268 1961 MS (o)1379 1961 MS (d)1501 1961 MS (e)1623 1961 MS ( )1734 = 1961 MS (a)1790 1961 MS (n)1901 1961 MS (d)2023 1961 MS ( )2145 1961 MS = (w)2199 1961 MS (e)2357 1961 MS (b)2467 1961 MS (s)2589 1961 MS (i)2700 = 1961 MS (t)2756 1961 MS (e)2822 1961 MS=0A= ( )2933 1961 MS (o)2989 1961 MS (f)3111 1961 MS ( )3177 1961 MS (r)3233 = 1961 MS (e)3311 1961 MS (s)3422 1961 MS (u)3533 1961 MS (l)3655 1961 MS = (t)3711 1961 MS (s)3777 1961 MS ( )3888 1961 MS (a)3944 1961 MS (v)4055 = 1961 MS (a)4166 1961 MS (i)4277 1961 MS=0A= (l)4333 1961 MS (a)4389 1961 MS (b)4500 1961 MS (l)4622 1961 MS (e)4678 = 1961 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 2205 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (T)1242 2205 MS (a)1344 2205 MS (r)1437 2205 MS (g)1502 2205 MS (e)1604 = 2205 MS (t)1696 2205 MS ( )1752 2205 MS (d)1798 2205 MS (a)1900 2205 MS = (t)1992 2205 MS (e)2048 2205 MS ( )2141 2205 MS (A)2186 2205 MS (p)2307 = 2205 MS (r)2409 2205 MS (i)2474 2205 MS=0A= (l)2520 2205 MS ( )2566 2205 MS (1)2612 2205 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 2480 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (D)979 2480 MS (u)1123 2480 MS (m)1245 2480 MS (m)1423 2480 MS (y)1602 = 2480 MS ( )1712 2480 MS (n)1768 2480 MS (e)1890 2480 MS (t)2002 2480 MS = ( )2069 2480 MS (s)2125 2480 MS (i)2236 2480 MS (m)2292 2480 MS (u)2470 = 2480 MS (l)2591 2480 MS (a)2647 2480 MS=0A= (t)2758 2480 MS (i)2824 2480 MS (o)2880 2480 MS (n)3001 2480 MS (s)3123 = 2480 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 2724 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (T)1242 2724 MS (a)1344 2724 MS (r)1437 2724 MS (g)1502 2724 MS (e)1604 = 2724 MS (t)1696 2724 MS ( )1752 2724 MS (d)1798 2724 MS (a)1900 2724 MS = (t)1992 2724 MS (e)2048 2724 MS ( )2141 2724 MS (J)2186 2724 MS (u)2279 = 2724 MS (n)2381 2724 MS (e)2483 2724 MS=0A= ( )2576 2724 MS (1)2622 2724 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 2999 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (R)979 2999 MS (e)1123 2999 MS (a)1235 2999 MS (l)1346 2999 MS ( )1402 = 2999 MS (w)1456 2999 MS (o)1613 2999 MS (r)1735 2999 MS (l)1813 2999 MS = (d)1868 2999 MS ( )1989 2999 MS (e)2045 2999 MS (x)2156 2999 MS (p)2267 = 2999 MS (e)2389 2999 MS (r)2500 2999 MS=0A= (i)2578 2999 MS (m)2634 2999 MS (e)2812 2999 MS (n)2923 2999 MS (t)3045 = 2999 MS (s)3112 2999 MS=0A= [167 0 0 -167 0 0]/_Helvetica MF=0A= (\225)1054 3242 MS=0A= [167 0 0 -167 0 0]/Helvetica-Bold MF=0A= (T)1242 3242 MS (a)1344 3242 MS (r)1437 3242 MS (g)1502 3242 MS (e)1604 = 3242 MS (t)1696 3242 MS ( )1752 3242 MS (d)1798 3242 MS (a)1900 3242 MS = (t)1992 3242 MS (e)2048 3242 MS ( )2141 3242 MS (A)2186 3242 MS (u)2307 = 3242 MS (g)2409 3242 MS (u)2511 3242 MS=0A= (s)2613 3242 MS (t)2706 3242 MS ( )2762 3242 MS (1)2808 3242 MS=0A= [200 0 0 -200 0 0]/Wingdings02009940 MF=0A= (\203)754 3518 MS=0A= [200 0 0 -200 0 0]/Helvetica-Bold MF=0A= (R)979 3518 MS (F)1123 3518 MS (C)1245 3518 MS ( )1389 3518 MS (t)1445 = 3518 MS (a)1512 3518 MS (r)1623 3518 MS (g)1701 3518 MS (e)1823 3518 MS = (t)1934 3518 MS ( )2002 3518 MS=0A= LATENC /_Helvetica-Bold /Helvetica-Bold reencode=0A= [200 0 0 -200 0 0]/_Helvetica-Bold MF=0A= (\226)2058 3518 MS=0A= (S)2225 3518 MS (e)2358 3518 MS (p)2469 3518 MS (t)2591 3518 MS (e)2658 = 3518 MS (m)2769 3518 MS (b)2947 3518 MS (e)3069 3518 MS (r)3180 3518 MS = ( )3258 3518 MS (2)3314 3518 MS (0)3425 3518 MS (0)3536 3518 MS (2)3647 = 3518 MS=0A= showpage=0A= FontSV restore=0A= PageSV restore=0A= %%Trailer=0A= %%DocumentNeededFonts:=0A= %%+ Helvetica=0A= %%+ Helvetica-Bold=0A= %%+ Helvetica-BoldOblique=0A= %%DocumentSuppliedFonts:=0A= %%+ Wingdings02009940=0A= end=0A= %%Pages: 11=0A= %%EOF=0A= =1B%-12345X@PJL EOJ=0A= ------=_NextPart_000_0000_01C1D0F1.9F8788C0-- _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt From daemon@optimus.ietf.org Mon Mar 25 08:23:01 2002 Received: from optimus.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA09176 for ; Mon, 25 Mar 2002 08:23:01 -0500 (EST) Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id IAA25286 for rmt-archive@odin.ietf.org; Mon, 25 Mar 2002 08:23:02 -0500 (EST) Received: from optimus.ietf.org (localhost [127.0.0.1]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA23806; Mon, 25 Mar 2002 08:17:37 -0500 (EST) Received: from ietf.org (odin [132.151.1.176]) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id IAA23771 for ; Mon, 25 Mar 2002 08:17:35 -0500 (EST) Received: from CNRI.Reston.VA.US (localhost [127.0.0.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA08914; Mon, 25 Mar 2002 08:17:33 -0500 (EST) Message-Id: <200203251317.IAA08914@ietf.org> Mime-Version: 1.0 Content-Type: Multipart/Mixed; Boundary="NextPart" To: IETF-Announce: ; Cc: rmt@ietf.org From: Internet-Drafts@ietf.org Reply-to: Internet-Drafts@ietf.org Date: Mon, 25 Mar 2002 08:17:33 -0500 Subject: [Rmt] I-D ACTION:draft-ietf-rmt-bb-webrc-01.txt,.ps Sender: rmt-admin@ietf.org Errors-To: rmt-admin@ietf.org X-Mailman-Version: 1.0 Precedence: bulk List-Id: Reliable Multicast Transport X-BeenThere: rmt@ietf.org --NextPart A New Internet-Draft is available from the on-line Internet-Drafts directories. This draft is a work item of the Reliable Multicast Transport Working Group of the IETF. Title : Wave and Equation Based Rate Control building block Author(s) : M. Luby, V. Goyal Filename : draft-ietf-rmt-bb-webrc-01.txt,.ps Pages : 28 Date : 20-Mar-02 Wave and Equation Based Rate Control (WEBRC) provides rate and congestion control for data delivery. WEBRC is specifically designed to support protocols using IP multicast, but could also be used to provide support to protocols that use unicast. A URL for this Internet-Draft is: http://www.ietf.org/internet-drafts/draft-ietf-rmt-bb-webrc-01.txt To remove yourself from the IETF Announcement list, send a message to ietf-announce-request with the word unsubscribe in the body of the message. Internet-Drafts are also available by anonymous FTP. Login with the username "anonymous" and a password of your e-mail address. After logging in, type "cd internet-drafts" and then "get draft-ietf-rmt-bb-webrc-01.txt". A list of Internet-Drafts directories can be found in http://www.ietf.org/shadow.html or ftp://ftp.ietf.org/ietf/1shadow-sites.txt Internet-Drafts can also be obtained by e-mail. Send a message to: mailserv@ietf.org. In the body type: "FILE /internet-drafts/draft-ietf-rmt-bb-webrc-01.txt". NOTE: The mail server at ietf.org can return the document in MIME-encoded form by using the "mpack" utility. To use this feature, insert the command "ENCODING mime" before the "FILE" command. To decode the response(s), you will need "munpack" or a MIME-compliant mail reader. Different MIME-compliant mail readers exhibit different behavior, especially when dealing with "multipart" MIME messages (i.e. documents which have been split up into multiple messages), so check your local documentation on how to manipulate these messages. 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: Multipart/Alternative; Boundary="OtherAccess" --OtherAccess Content-Type: Message/External-body; access-type="mail-server"; server="mailserv@ietf.org" Content-Type: text/plain Content-ID: <20020325080625.I-D@ietf.org> ENCODING mime FILE /internet-drafts/draft-ietf-rmt-bb-webrc-01.txt --OtherAccess Content-Type: Message/External-body; name="draft-ietf-rmt-bb-webrc-01.txt"; site="ftp.ietf.org"; access-type="anon-ftp"; directory="internet-drafts" Content-Type: text/plain Content-ID: <20020325080625.I-D@ietf.org> --OtherAccess-- --NextPart-- _______________________________________________ Rmt mailing list Rmt@ietf.org https://www1.ietf.org/mailman/listinfo/rmt