- Monday, 14 November 2011

clue@jabber.ietf.org

Monday, 14 November 2011< ^ >

stpeter has set the subject to: CLUE WG | http://tools.ietf.org/wg/clue/ | IETF 81 | audio at http://ietf81streaming.dnsalias.net/ietf/ietf804.m3u | slides at https://datatracker.ietf.org/meeting/81/materials.html#wg-clue

Room Configuration

GMT+0
[00:21:26] stpeter joins the room
[00:39:51] burn joins the room
[00:49:35] pm joins the room
[00:49:39] tsuichi joins the room
[00:52:15] Jonathan Lennox joins the room
[00:52:38] suzukisn joins the room
[00:54:59] A. Jean Mahoney joins the room
[00:55:40] <Jonathan Lennox> Anyone here who isn't in Taipei?
[00:58:18] MeetechoAudio joins the room
[01:00:06] Simon Romano joins the room
[01:01:03] Lorenzo Miniero joins the room
[01:01:58] <Jonathan Lennox> Now that we're going, asking again: anyone here who isn't in Taipei?
[01:02:13] <Lorenzo Miniero> I'm in Italy
[01:02:27] <Simon Romano> Presentation stopped
[01:02:56] <Jonathan Lennox> Okay, Lorenzo: let me know if you want any comments relayed to the mic
[01:03:18] <Lorenzo Miniero> ok, tnx!
[01:04:16] victor.pascual.avila joins the room
[01:04:22] Xavier Marjou joins the room
[01:04:41] <Jonathan Lennox> Starting Use Case slides
[01:04:44] Raj joins the room
[01:04:47] Roy joins the room
[01:05:08] <Jonathan Lennox> Or "slide", actually
[01:05:19] Raj leaves the room
[01:06:58] <Simon Romano> Presentation stopped
[01:07:11] <Simon Romano> Slide 1: CLUE Requirements
[01:07:30] <Simon Romano> Slide 2: Dynamic
[01:08:22] <Simon Romano> Slide 3: 3 dimensions
[01:10:50] <Simon Romano> Slide 4: 3 dimensions continued
[01:11:02] <Simon Romano> Slide 5: Multi-view
[01:11:43] Cullen Jennings joins the room
[01:12:17] claude joins the room
[01:15:24] claude leaves the room
[01:16:43] jlcJohn joins the room
[01:17:07] <Simon Romano> Presentation stopped
[01:17:18] <Simon Romano> Slide 1: CLUE Framework
[01:17:37] <Simon Romano> Slide 2: Contents
[01:18:16] <Simon Romano> Slide 3: Changes since interim meeting
[01:18:16] Mike Bannion joins the room
[01:19:45] RjS joins the room
[01:19:57] <RjS> who's doing the typing in the etherpad?
[01:20:04] victor.pascual.avila leaves the room
[01:20:24] Yong Cui joins the room
[01:20:43] victor.pascual.avila joins the room
[01:20:49] <Simon Romano> Slide 4: Issues
[01:21:07] <Jonathan Lennox> I think it was Marshall who was going to be notetaker?
[01:21:16] Xiaolan Wan joins the room
[01:22:04] Gonzalo Camarillo joins the room
[01:22:14] <Simon Romano> Slide 5: Issue 1 Coordinate System
[01:22:16] <Simon Romano> Slide 6: What is this addressing
[01:22:18] Yong Cui leaves the room
[01:22:24] Mike Bannion leaves the room
[01:22:29] <Simon Romano> Slide 16: Issue 2 - Source Selection
[01:22:50] hta joins the room
[01:29:13] <hta> what was the statement about the adhoc? I was a bit late.
[01:29:15] pm leaves the room
[01:29:39] Xiaolan Wan leaves the room
[01:30:56] <hta> hmmm... the subject line of this chatroom points to IETF-81. We're currently in IETF-82.
[01:32:20] <Jonathan Lennox> I think only StPeter can change it since he's the group moderator?
[01:33:43] RjS2 joins the room
[01:33:44] <Jonathan Lennox> There's an RTCWeb/Clue commonalities adhoc meetup for Thursday lunch.
[01:34:48] RjS2 has set the subject to: CLUE WG | http://tools.ietf.org/wg/clue/ | IETF 82 | slides at https://datatracker.ietf.org/meeting/82/materials.html#wg-clue
[01:34:59] <RjS2> apparently anyone can
[01:35:05] suzukisn leaves the room
[01:35:06] <hta> that seemed to help.
[01:35:21] RjS leaves the room
[01:35:23] <Simon Romano> Slide 17: Issue 3 - Describing composed (including
[01:35:55] RjS2 leaves the room
[01:36:00] RjS joins the room
[01:37:49] <Jonathan Lennox> RjS2: apparently I haven't figured out how to do it in PSI.
[01:37:57] <Simon Romano> Slide 18: Issue 4 - selecting composed (including
[01:38:10] <Simon Romano> Slide 17: Issue 3 - Describing composed (including
[01:40:00] marshall joins the room
[01:40:10] <marshall> Hello
[01:40:16] <marshall> is there a jabber scribe
[01:40:18] <marshall> ?
[01:40:32] <Jonathan Lennox> I'm jabber scribing, but a backup isn't a bad idea
[01:40:37] <Jonathan Lennox> Simon is relaying slide titles
[01:40:46] rifaat.sy joins the room
[01:41:12] <marshall> Stephan Wenger : I believe that it is useless to think that MCUs would accurately relay RTCP
[01:41:16] <Simon Romano> rtsp://taipei1.conf.meetecho.com/3330001.sdp
[01:41:23] <Simon Romano> Have a listen...
[01:41:56] <Cullen Jennings> what's the audio like ?
[01:42:12] <marshall> I am in the room, so I am not going to listen
[01:42:29] <Cullen Jennings> same problem for me
[01:42:53] Shinnosuke Suzuki joins the room
[01:43:22] <marshall> Roni Evan : THis is also a point to point issue, not just a MCU issue
[01:43:30] <marshall> this is relevant for P2P
[01:43:46] <Jonathan Lennox> Lorenzo is remote -- can you say what the audio's like?
[01:43:58] <marshall> As for the mechanism, we need to be clear how to provide the right informatiion
[01:44:00] <Jonathan Lennox> If it's good we probably don't need to relay comments at the mic
[01:44:21] <Simon Romano> Slide 1: CLUE Framework
[01:44:34] <marshall> Allyn : What Roni was talking about about the mapping is what's in Jonathan's and my draft and will be discussed at length
[01:45:14] <Cullen Jennings> Hmm - just tried the RTSP and real player could not connect
[01:45:18] <marshall> BUT, we need to think deeply as to whether there is other information, the content of that information that needs to be conveyed to say
[01:45:23] <Lorenzo Miniero> Jonathan the audio is fine for me
[01:45:37] Shinnosuke Suzuki leaves the room
[01:45:40] <marshall> what is in a picture. That is different from the mapping of SSRCs
[01:45:45] Shinnosuke Suzuki joins the room
[01:46:08] <marshall> Roni : This is about mixing of original's . I think it iis a separate issue from Jonathan's draft
[01:46:09] <Simon Romano> Sorry...I thought they were done!
[01:46:37] <marshall> Mark : It seemed to us this might not be in the scope of clue
[01:46:56] <stpeter> Simon Romano: at http://tools.ietf.org/wg/appsawg/agenda?item=agenda82.html we are on draft-kucherawy-greylisting-bcp, probably we will get to the DCON BoF preview in about 10-15 minutes
[01:47:14] <marshall> Stephan : I think that this is much more in the scope of clue than thinking about mapping of rtcp streams
[01:47:40] suzukisn joins the room
[01:47:45] <Cullen Jennings> sorry need to correct prev thing I said. RTSP is working - just took awhile to connect. Seems to be abou 15 second delay from live.
[01:48:04] <marshall> Steve B. : I disagree with Stephan. Composed pictures have been done for a long time
[01:48:33] <marshall> Mary : If you can get this done this fall that would be good.
[01:49:01] <marshall> Issue 4 : Is very similar to previous discussion about source selection
[01:49:03] rifaat.sy leaves the room
[01:49:37] <marshall> but this not about selecting left or right cameras, but about algorithms for mixing sources
[01:49:50] <marshall> should the consumer be able to select something
[01:49:50] <Simon Romano> stpeter: I'm afraid I cannot make it...I'm the only Meetecho guy in clue right now :-(
[01:49:51] <marshall> ?
[01:50:08] <marshall> options discussed
[01:50:25] <marshall> do nothing (my (Mark's) opinion)
[01:50:37] <marshall> or define some policies
[01:50:42] Shinnosuke Suzuki leaves the room
[01:51:00] <marshall> let the consumer pick one
[01:51:14] <stpeter> Simon Romano: ok
[01:51:28] <stpeter> Simon Romano: I will say a few words about the DCON BoF, then
[01:51:39] <Simon Romano> Great, thanx!
[01:51:46] <marshall> Charles : There is something in the framework about selecting "hints"
[01:52:07] <marshall> Mark : It didn't occur to me to use that - I thought it would have to be more provider driven
[01:52:41] <marshall> Charles : One of my concerns it might be hard to say all of te many options a provider can do
[01:52:46] <Lorenzo Miniero> ops Cullen you were right sorry -- I have fixed this, RTSP should be real time now
[01:53:00] <marshall> I thought that that was one of things that 3 way would handle
[01:53:24] <marshall> Mark : How would you enumerate some policies that were limited enough to be useful
[01:53:49] <marshall> Jonathan : Do we have some examples ? Loudest speaker ?
[01:54:00] <Roy> who is on the mic just now
[01:54:16] <marshall> Jonathan
[01:54:24] <Roy> thanks!
[01:54:30] MeetechoAudio leaves the room
[01:54:58] <marshall> I know one thing that has come up is, "when if ever does the loudest speaker want to see himself"
[01:55:15] <marshall> Steve B : Not just What kind of composition, but who is in it
[01:55:55] <marshall> I think that there are a huge number of composition modes in existing MCUs and it would hard to capture them all
[01:56:01] <marshall> Mark : Solving his doesn't seem to be part of the CLUE charter
[01:56:34] <marshall> Allyn " What is important about CLUE is that it is extensible. It is nopt clear why this is crucial right now
[01:56:59] <marshall> Roni Even :: There is a difference between select and request
[01:57:48] <marshall> Originally I looked at this from the P2P case - the description said, just compose. The provider can provide different composed schemes
[01:58:17] <marshall> Mary : So you want this functionality in here ?
[01:58:34] navii joins the room
[01:58:43] <marshall> Roni I want the provider to be able to specify what's in the mix when he offers a mix
[01:58:50] <marshall> Steve B : Layout ?
[01:59:22] <marshall> Roni : It may be layout, but - I can send you the 2 active speakers, or the center and the active speaker ?
[01:59:30] <marshall> Mary : Can you provide use cases
[01:59:37] <marshall> Roni : OK
[02:00:03] MeetechoAudio2 joins the room
[02:00:19] <Simon Romano> Presentation stopped
[02:00:39] <Simon Romano> Slide 1: ??
[02:00:40] <marshall> Mary : There was a lot of policy in XCON early on and it lead us down a deep rat hole.
[02:00:43] <marshall> NEXT
[02:00:48] <Simon Romano> Slide 2: ??
[02:00:50] <marshall> Coordinate systems : Brian B
[02:01:06] <marshall> What is this addressing ?
[02:01:17] <marshall> describe devices in a 3-D space
[02:01:35] <marshall> enable receivers to rrender correctly
[02:01:51] <marshall> enable implementers to give a sense of real world dimensions
[02:02:00] <Simon Romano> Slide 3: ??
[02:02:08] <Simon Romano> Proposal
[02:02:10] <marshall> slide 3
[02:02:23] <marshall> coordinate system is cartesian
[02:02:26] <Cullen Jennings> Are the meetecho slides updating for others ?
[02:03:11] <marshall> coordinates are cartesian, with arbitrary origin
[02:03:18] <marshall> coordinates are real or virtual
[02:03:40] <Lorenzo Miniero> I see slide3 right now
[02:03:55] <Lorenzo Miniero> but I'm in Italy so I can0t check if it's right :)
[02:03:58] <marshall> Culleen : It seems you have to be real to be life size
[02:04:05] <Simon Romano> We're there
[02:04:09] <Simon Romano> Slide 3
[02:04:13] <marshall> Having 2 coordinate system complicates things
[02:04:56] <marshall> Jonathan Lennox : I was thinking of things that were not real (virtual presenations)
[02:05:13] <marshall> like the screens right now
[02:05:24] <marshall> Steve B: I would agree with presentations
[02:05:59] <marshall> you need to know the distance to the person, and I am not sure how a mobile unit would get there
[02:06:04] <marshall> get that
[02:06:51] <marshall> Cullen : you have a cone or prism as a field of view. I don't get the mobile
[02:08:36] Gonzalo joins the room
[02:08:51] <marshall> Mark : I see it as one coordinate system, where the provider is either claiming its units are mm, or not
[02:09:07] <Simon Romano> Slide 4: ??
[02:09:16] <Simon Romano> Directionality
[02:09:40] Gonzalo Camarillo leaves the room
[02:12:45] <Simon Romano> Slide 5: relevant terms
[02:12:52] <marshall> Slide 5 : Relevant terms
[02:13:17] <Simon Romano> Slide 6: Point of Capture/Area of Capture
[02:17:59] <Simon Romano> Slide 7
[02:18:10] navii leaves the room
[02:18:43] <Simon Romano> Slide 6 again
[02:19:27] <Simon Romano> Slide 7
[02:20:41] hta leaves the room
[02:22:52] Magnus joins the room
[02:23:52] hta joins the room
[02:25:15] <Simon Romano> Slide 8: 3 Camera Example
[02:25:55] <Simon Romano> Slide 9: Curved Table Example
[02:26:19] <Simon Romano> Slide 10: 2 Camera 'Crossed' Example
[02:31:39] victor.pascual.avila leaves the room
[02:32:29] victor.pascual.avila joins the room
[02:33:59] <Simon Romano> Presentation stopped
[02:34:02] marshall leaves the room
[02:34:21] <Simon Romano> Slide 1: Options to Transport CLUE Messages draft
[02:34:30] <Simon Romano> Slide 2: Constraints
[02:34:56] RjS leaves the room: Computer went to sleep
[02:35:30] <Simon Romano> Slide 3: Conclusion #1
[02:35:43] <Simon Romano> Slide 4: Options for transporting CLUE exchange
[02:36:47] <Simon Romano> Slide 5: Conclusion #2
[02:37:49] <Simon Romano> Slide 2: Constraints
[02:52:00] <Simon Romano> Slide 6: CLUE negotiation over SIP-established �
[02:52:43] <Simon Romano> Slide 7: Options for CLUE “media” stream
[02:55:12] <Simon Romano> Slide 8: Conclusion #3
[02:55:44] <Simon Romano> Slide 9: CLUE message Content Representation
[02:56:44] yaovct joins the room
[02:57:07] <Simon Romano> Slide 10: Conclusion #4
[02:57:25] <Simon Romano> Slide 11: Conclusions Summary
[02:57:48] victor.pascual.avila leaves the room
[03:00:36] Lorenzo Miniero leaves the room
[03:02:16] yaovct leaves the room
[03:03:44] <Simon Romano> Presentation stopped
[03:04:08] <Simon Romano> Slide 1: RTP Usage for CLUE
[03:06:16] kenny wang joins the room
[03:06:27] <Simon Romano> Slide 2: Source Multiplexing: Motivation
[03:07:07] <Simon Romano> Slide 3: Source multiplexing
[03:08:16] Tobia Castaldi joins the room
[03:08:48] <Simon Romano> Slide 2: Source Multiplexing: Motivation
[03:11:01] <Simon Romano> Slide 4: Source multiplexing: exceptions
[03:11:45] <Simon Romano> Slide 5: Source multiplexing: complications
[03:14:16] victor.pascual.avila joins the room
[03:14:51] <Simon Romano> Slide 6: CLUE-specific Complications
[03:15:29] Cullen Jennings leaves the room
[03:19:49] <Simon Romano> Slide 7: Demultiplexing RTP Streams in Same Sessi
[03:20:08] <Simon Romano> Slide 8: Case to Consider: Infinite Sources
[03:21:42] <Simon Romano> Slide 9: How to Demultiplex
[03:22:11] <Simon Romano> Slide 10: SSRCs Pros
[03:22:20] <Simon Romano> Slide 11: The issue with SSRC: 3 streams sent
[03:22:52] <Simon Romano> Slide 12: 2 of 3 streams stop
[03:23:07] <Simon Romano> Slide 13: VC2 and VC3 continue with different SSRC
[03:23:25] <Simon Romano> Slide 14: Timeline
[03:23:38] <Simon Romano> Slide 15: SSRCs Challenges
[03:23:41] <Simon Romano> Slide 14: Timeline
[03:24:25] <Simon Romano> Slide 15: SSRCs Challenges
[03:24:48] burn leaves the room
[03:25:20] <Simon Romano> Slide 16: Where to put metadata
[03:26:36] <Simon Romano> Slide 17: Sending metadata
[03:26:41] <Simon Romano> Slide 18: Multiplex ID Advantages
[03:27:08] Xavier Marjou leaves the room
[03:27:19] <Simon Romano> Slide 19: Multiplex ID Cons
[03:27:42] <Simon Romano> Slide 20: Sending Multiplex IDs
[03:27:45] <Simon Romano> Slide 21: Hybrid Scheme, Pros and Cons
[03:28:26] <Simon Romano> Slide 22: Hybrid Challenges
[03:28:44] Gonzalo leaves the room
[03:30:49] victor.pascual.avila leaves the room
[03:32:47] Magnus leaves the room
[03:34:10] hta leaves the room
[03:35:25] <Simon Romano> Slide 12: 2 of 3 streams stop
[03:35:58] <Simon Romano> Presentation stopped
[03:36:10] <Simon Romano> Bye bye!
[03:36:14] Simon Romano leaves the room
[03:36:23] MeetechoAudio2 leaves the room
[03:36:24] Jonathan Lennox leaves the room: Computer went to sleep
[03:38:07] kenny wang leaves the room
[03:38:11] A. Jean Mahoney leaves the room
[03:38:23] Tobia Castaldi leaves the room
[03:38:25] suzukisn leaves the room
[03:39:35] tsuichi leaves the room
[03:42:43] A. Jean Mahoney joins the room
[03:47:35] A. Jean Mahoney leaves the room
[03:53:29] Roy leaves the room
[04:15:52] stpeter leaves the room: Disconnected: connection closed
[04:19:12] hta joins the room
[04:19:27] hta leaves the room
[04:19:55] Roy joins the room
[04:29:03] burn joins the room
[04:38:58] tsuichi joins the room
[04:42:51] tsuichi leaves the room
[04:45:42] victor.pascual.avila joins the room
[04:46:42] Roy leaves the room
[05:04:51] stpeter joins the room
[05:05:09] stpeter leaves the room
[05:16:42] victor.pascual.avila leaves the room
[05:27:02] victor.pascual.avila joins the room
[05:33:05] RjS joins the room
[05:33:12] burn leaves the room
[05:33:34] RjS has set the subject to: Clue WG
[05:33:40] RjS leaves the room
[05:39:36] burn joins the room
[05:43:16] burn leaves the room
[05:47:34] victor.pascual.avila leaves the room
[05:48:21] victor.pascual.avila joins the room
[06:12:42] A. Jean Mahoney joins the room
[07:07:10] victor.pascual.avila leaves the room
[07:16:32] victor.pascual.avila joins the room
[07:23:40] A. Jean Mahoney leaves the room
[08:19:00] victor.pascual.avila leaves the room
[08:21:47] victor.pascual.avila joins the room
[08:40:37] victor.pascual.avila leaves the room
[08:41:47] victor.pascual.avila joins the room
[08:50:52] victor.pascual.avila leaves the room
[08:51:31] victor.pascual.avila joins the room
[09:06:15] victor.pascual.avila leaves the room
[09:06:45] victor.pascual.avila joins the room
[09:07:08] victor.pascual.avila leaves the room
[09:09:30] victor.pascual.avila joins the room
[09:09:49] victor.pascual.avila leaves the room
[09:17:21] victor.pascual.avila joins the room
[09:57:21] victor.pascual.avila leaves the room
[09:57:39] victor.pascual.avila joins the room
[11:06:00] victor.pascual.avila leaves the room
[12:04:40] victor.pascual.avila joins the room
[15:16:53] victor.pascual.avila leaves the room
[15:26:10] alfredh joins the room
[16:13:13] alfredh leaves the room