- Monday, 5 November 2012

clue@jabber.ietf.org

Monday, 5 November 2012< ^ >

RjS has set the subject to: Clue WG

Room Configuration

GMT+0
[13:26:45] Tobia Castaldi joins the room
[13:27:47] Simon Romano joins the room
[13:28:13] Audio CLUE joins the room
[13:34:34] Mark Duckworth joins the room
[13:39:10] Lorenzo Miniero joins the room
[13:45:00] roberta presta joins the room
[13:50:26] Alex Amirante joins the room
[13:50:41] John Leslie joins the room
[13:52:47] Martin Thomson joins the room
[13:58:39] Andy Pepperell joins the room
[14:00:18] <Simon Romano> Hello everybody: meetecho is @ www.meetecho.com/ietf85/clue
[14:01:05] Mary Barnes joins the room
[14:01:33] <Simon Romano> If you're just interested in slides and audio/video, grab your popcorn and open this link: http://atlanta.conf.meetecho.com/WebLite/event.jsp?w=clue
[14:01:40] Enda joins the room
[14:02:45] Jonathan Lennox joins the room
[14:02:45] <Tobia Castaldi> Slide 1: CLUE WG IETF-85
[14:02:46] <Tobia Castaldi> Slide 2: Note Well
[14:02:48] <Tobia Castaldi> Slide 3: Agenda – Monday (09:00-11:30)
[14:03:08] Lorenzo Miniero leaves the room
[14:05:31] tsuichi joins the room
[14:06:03] <roberta presta> hi mary!
[14:06:04] <Martin Thomson> prefix comments with "mic:" if you want me to relay something
[14:06:36] <Tobia Castaldi> Slide 4: Agenda - Tuesday (13:00-15:00)
[14:06:41] <Tobia Castaldi> Slide 5: Current WG deliverables
[14:06:59] <Tobia Castaldi> Slide 6: Working Towards a Solution
[14:07:15] tomoyuki tsuichihara joins the room
[14:07:33] <Tobia Castaldi> Slide 7: Proposed Revised WG deliverables
[14:08:00] <Tobia Castaldi> Presentation stopped
[14:08:18] tomoyuki tsuichihara leaves the room
[14:08:27] <Martin Thomson> bernard aboba at mic
[14:08:33] Andy Pepperell leaves the room
[14:08:38] roberta presta leaves the room
[14:09:21] <Tobia Castaldi> http://datatracker.ietf.org/doc/draft-ietf-clue-framework/
[14:09:30] Andy Pepperell joins the room
[14:09:43] <Tobia Castaldi> Slide 1: CLUE framework updates
[14:10:20] Lorenzo Miniero joins the room
[14:10:24] <Martin Thomson> the last speaker was Gonzalo
[14:10:50] <John Leslie> audio?
[14:10:59] I lost audio too
[14:10:59] <Martin Thomson> are you getting no audio?
[14:11:08] Andy Pepperell leaves the room
[14:11:13] Audio CLUE leaves the room
[14:11:27] <Martin Thomson> the chairs will run through the charts
[14:11:29] <Simon Romano> Working on this..
[14:11:32] Andy Pepperell joins the room
[14:11:35] <Martin Thomson> hopefully we will regain audio soon
[14:11:58] <Martin Thomson> simulcast concerns
[14:12:06] <Andy Pepperell> My connection via Meetecho Java applet seems to keep timing out
[14:12:11] <Tobia Castaldi> Slide 2: “Capture encoding”
[14:12:13] <Tobia Castaldi> Slide 3: “Capture encoding” changes
[14:12:15] <Tobia Castaldi> Slide 4: Simulcast concerns
[14:12:26] audio joins the room
[14:12:28] <Tobia Castaldi> Slide 5: Maximum capture encodings
[14:12:28] tomoyuki tsuichihara joins the room
[14:13:01] <Martin Thomson> audio?
[14:13:14] <Jonathan Lennox> We just heard "you are now unmuted" … can you hear Mary now?
[14:13:18] John Leslie leaves the room
[14:13:23] <Simon Romano> Audio is back...
[14:13:28] <Andy Pepperell> I cannot hear Mary at the moment, no
[14:13:28] <Simon Romano> Sorry about that :-(
[14:13:41] <Tobia Castaldi> Slide 6: “Point on Line of Capture”
[14:13:52] Enda Mannion joins the room
[14:13:59] Enda leaves the room
[14:14:38] <Tobia Castaldi> Slide 6: “Point on Line of Capture”
[14:14:41] Cullen Jennings joins the room
[14:14:54] <Tobia Castaldi> Slide 5: Maximum capture encodings
[14:15:02] Alexandru Petrescu joins the room
[14:15:12] <Jonathan Lennox> Roni Even at the mic — asking a question about the max capture encodings
[14:15:22] Alexandru Petrescu leaves the room
[14:15:32] <Martin Thomson> point is about what streams can be simulcast
[14:15:59] <Martin Thomson> when a sender says that they can send 2, which two can they send?
[14:16:13] <Simon Romano> @andy: unfortunately, the audio netbook connection seems to be intermittent. We're changing the eth cable now...
[14:16:22] <Simon Romano> Do you get audio from the room now?
[14:16:35] <Andy Pepperell> No, am hearing no audio still
[14:16:55] <Simon Romano> Launch the applet again..it should be ok now
[14:17:02] <Andy Pepperell> to answer that specific point, when a sender says they can send "a maximum of 2" that would be a per-media capture attribute
[14:17:13] <Martin Thomson> mic?
[14:17:52] <Martin Thomson> roni is observing that the problem exists with a dynamic mapping, where a static mapping would be easier to solve
[14:18:07] roberta presta joins the room
[14:18:14] <Martin Thomson> there is some discussion about separate streams as opposed to simulcast
[14:18:49] Andy Pepperell leaves the room
[14:19:42] <Martin Thomson> jonathan is saying that if each capture only had a single ssrc, then it becomes easier
[14:19:52] Andy Pepperell joins the room
[14:20:05] <Martin Thomson> all this seems to be based on speculation
[14:20:33] <Andy Pepperell> relaunching the Java applet crashed my machine
[14:20:38] <Andy Pepperell> I do seem to have audio now
[14:20:41] <Tobia Castaldi> Presentation stopped
[14:20:48] <Martin Thomson> speculation that it might be possible to drop encoding groups
[14:21:24] <Martin Thomson> do you have audio now?
[14:21:31] I have audio
[14:21:32] <Andy Pepperell> I do, yes
[14:21:38] <Martin Thomson> great :)
[14:21:50] <Martin Thomson> we're on the data model scheme draft
[14:22:00] <Martin Thomson> XML!
[14:22:09] <Tobia Castaldi> Slide 2: outline
[14:23:10] <Tobia Castaldi> Slide 3: Overview
[14:24:11] John Leslie joins the room
[14:25:58] <Martin Thomson> a picture from xmlspy or eclipse xml schema editor
[14:26:39] Enda Mannion leaves the room
[14:28:48] <Andy Pepperell> The audio seems to have died again for me...
[14:28:54] <Tobia Castaldi> :S
[14:29:00] no audio for me either
[14:29:10] <Martin Thomson> we're just going through the XML schema in more detail
[14:29:31] <Martin Thomson> media capture type/spatial information type
[14:30:01] <Martin Thomson> yuck: IDREF
[14:30:14] audio leaves the room
[14:30:18] <Martin Thomson> (that was just me, making commentary :)
[14:30:37] audioclue joins the room
[14:31:03] <Martin Thomson> s/unsignedInt/nonNegativeInteger/
[14:31:13] <Martin Thomson> audio capture type example
[14:31:43] tomoyuki tsuichihara leaves the room
[14:31:59] <Martin Thomson> video should be similar
[14:32:15] audioss joins the room
[14:32:22] tomoyuki tsuichihara joins the room
[14:32:29] audioclue leaves the room
[14:32:50] <Andy Pepperell> The slides for me are still showing "Capture scene type" - dunno if that's a consequence / related to the audio no longer working
[14:33:03] same for me
[14:33:23] Martin Thomson leaves the room
[14:33:28] now slides are updating
[14:33:35] Martin Thomson joins the room
[14:33:42] <Martin Thomson> hmm, my last message didn't go to the room
[14:34:01] <Martin Thomson> our audio problems were more important than advancing the slides :)
[14:34:10] <Martin Thomson> we are up to "A mor ecomplex capture"
[14:36:45] ena2413 joins the room
[14:37:02] ena2413 leaves the room
[14:37:44] Zaheduzzaman Sarker joins the room
[14:39:36] <Tobia Castaldi> what about audio now?
[14:42:56] Andy Hutton joins the room
[14:43:10] audio is working for me
[14:43:37] <Martin Thomson> moving on to Roni's slides on SDP O/A and CLUE relationship
[14:43:55] <Martin Thomson> introduction
[14:43:57] <Tobia Castaldi> Slide 1: Relation between CLUE signaling and SDP offer answer
[14:44:02] <Tobia Castaldi> Slide 2: Introduction
[14:44:27] <Andy Pepperell> Audio is working for me, but is quite difficult to hear due to crackling
[14:44:42] <Lorenzo Miniero> which stream are you using Andy?
[14:44:56] <Andy Pepperell> I'm using the Java audio applet
[14:45:16] <John Leslie> audio working now, but only after restarting the stream -- is this something I'll have to do often?
[14:45:18] <Lorenzo Miniero> have you tried if any of the passive streams is better?
[14:45:32] <Lorenzo Miniero> John hopefully not :)
[14:45:37] <Lorenzo Miniero> not anymore at least
[14:45:41] <Tobia Castaldi> Slide 3: Capture Attributes
[14:45:43] <Andy Pepperell> I haven't tried one of the passive streams as I had hoped to participate
[14:45:48] <Lorenzo Miniero> ah ok
[14:45:50] I can hear Roni well, using VLC player with stream rtsp://atlanta.conf.meetecho.com/wlcoimh.sdp
[14:46:12] <Tobia Castaldi> John, we are sorry... we are having some problems
[14:47:46] <Tobia Castaldi> but all should work smoothly from now on
[14:48:13] <John Leslie> <fingers=crossed>
[14:48:48] <John Leslie> (was worried I might need to restart audio often...)
[14:59:20] <Tobia Castaldi> Slide 4: Encoding parameters (1).
[15:01:23] Cullen Jennings leaves the room
[15:02:15] <Martin Thomson> we're having some audio issues
[15:03:31] Cullen Jennings joins the room
[15:04:00] <Tobia Castaldi> Slide 5: Encoding parameters (2).
[15:04:40] <Lorenzo Miniero> Martin I'm listening to the audio RTMP and it seems fine
[15:04:55] <Lorenzo Miniero> which issue are you getting?
[15:05:14] <Martin Thomson> that was for the benefit of remote participants, I'm in the room
[15:05:40] <Lorenzo Miniero> ops sprry :)
[15:05:45] <Lorenzo Miniero> I'm not in the room
[15:13:18] Mathieu joins the room
[15:13:56] <Tobia Castaldi> Slide 6: Encoding parameters (3).
[15:15:41] <Tobia Castaldi> Slide 7: Observations.
[15:16:34] <Martin Thomson> observation: clue needs to decide whether configuration requests are able to make adjustments to the selection that is made from the advertisement, or whether any receiver constraints are applied based on what is negotiated in SDP. Example is for a stream that is one of many where the advertisement is 1080p and the receiver is only capable of handling 720p for that stream. SDP might not provide adequate constraints for this.
[15:18:48] <Martin Thomson> roni now has another presentation
[15:18:54] <Tobia Castaldi> Presentation stopped
[15:18:55] <Martin Thomson> rtp mapping
[15:19:19] <Tobia Castaldi> Slide 1: Mapping RTP streams to CLUE media captures
[15:19:26] <Tobia Castaldi> Slide 2: Introduction
[15:20:18] <Tobia Castaldi> Slide 3: Assumptions
[15:21:50] Tobia Castaldi leaves the room
[15:22:05] <Simon Romano> Slide 1: Mapping RTP streams to CLUE media captures
[15:22:09] <Simon Romano> Slide 4: SSRC behavior
[15:22:16] <Simon Romano> Relatore corrente: Roni Even
[15:22:17] <Simon Romano> Slide 4: SSRC behavior
[15:25:18] Roy joins the room
[15:25:21] <Andy Pepperell> Audio has become very bad quality for me now - even when I switched away from the Java applet and on to Passive audio (streaming) (flash)
[15:25:40] Yes. Audio got much louder and noisier a few minutes ago.
[15:26:25] <Martin Thomson> jonathan lennox at the mic
[15:27:32] <Simon Romano> @Mark: what about now?
[15:27:39] better now. Thanks
[15:27:53] <Simon Romano> ;-)
[15:28:21] <Simon Romano> It's our first day here...still fighting with the mixers, sorry.
[15:28:27] <Andy Pepperell> yes, much better - did you make a change?
[15:29:36] <Martin Thomson> on the Mapping options - SDP overview slide
[15:29:41] <Simon Romano> Lowered down the output channel...
[15:29:45] <Simon Romano> Slide 5: Mapping options – SDP overview
[15:30:25] Tobia Castaldi joins the room
[15:30:50] Cullen Jennings leaves the room
[15:31:17] <Simon Romano> Slide 6: Mapping options
[15:33:07] <Simon Romano> Slide 7: SDP examples – Static mapping
[15:33:35] Cullen Jennings joins the room
[15:34:38] <Simon Romano> Slide 8: SDP examples – Static mapping - simulc
[15:35:07] Cullen Jennings leaves the room
[15:35:14] <Simon Romano> Slide 7: SDP examples – Static mapping
[15:35:58] <Martin Thomson> bernard aboba at the mic
[15:39:03] <Simon Romano> Slide 6: Mapping options
[15:39:05] <Simon Romano> Slide 5: Mapping options – SDP overview
[15:39:41] <Simon Romano> Slide 7: SDP examples – Static mapping
[15:41:12] <Simon Romano> Slide 8: SDP examples – Static mapping - simulc
[15:43:58] <Martin Thomson> Roni likes Clapton
[15:44:05] <Simon Romano> :-)
[15:44:08] <Lorenzo Miniero> :D
[15:44:44] <Simon Romano> Slide 9: SDP examples – Static mapping – usin
[15:44:54] <Simon Romano> using src name for mapping
[15:46:08] <Martin Thomson> justin uberti at mic
[15:46:58] <Simon Romano> Slide 8: SDP examples – Static mapping - simulc
[15:47:33] roberta presta leaves the room
[15:51:20] hta joins the room
[15:51:35] <hta> if this is very complicated to explain, are we doing it wrong?
[15:52:01] <Martin Thomson> hta: yes
[15:52:09] <Simon Romano> Slide 9: SDP examples – Static mapping – usin
[15:53:11] <Jonathan Lennox> hta: that's why we were speculating earlier about dropping or simplifying the encoding concepts
[15:53:52] Mathieu Rene joins the room
[15:54:16] <Martin Thomson> as I understood it: ssrc = whatever; srcname = source of media (e.g., camera); cname = source of RTP, synchronized; msid = arbitrary grouping, synchronized
[15:55:40] <hta> martin, just to add to the confusion, msid = identification of the MediaStream and of the MediaStreamTrack inside that stream (the a=msid mechanism it carries two IDs)
[15:56:13] <Martin Thomson> right, the secondary "track id" part of msid is something I find to be uninteresting for the purposes of this discussion
[15:56:34] <Martin Thomson> let's call that the "msid track id"
[15:57:31] <Jonathan Lennox> The Clue capture ID — or maybe capture encoding — might be more like a track id than an msid grouping
[15:58:25] <Simon Romano> Slide 10: Proposal
[15:59:26] <hta> jonathan: I haven't seen it grouping yet, so yes, it seems like a track. But I'm not tracking very well.
[16:00:33] <Martin Thomson> capture id and track id seem analogous
[16:01:26] <Martin Thomson> can a single capture be simulcast? or does it get a new capture ID for each stream? what about FEC?
[16:02:42] <Simon Romano> Presentazione interrotta
[16:03:18] <Simon Romano> Slide 4: Agenda - Tuesday (13:00-15:00)
[16:03:27] <Simon Romano> Relatore corrente: Mary Barnes
[16:03:28] <Simon Romano> Slide 4: Agenda - Tuesday (13:00-15:00)
[16:05:31] <Simon Romano> Slide 6: Encoding parameters (3).
[16:06:38] <Simon Romano> Slide 6: Encoding parameters (3).
[16:09:35] <Simon Romano> Slide 3: Capture Attributes
[16:12:30] <Jonathan Lennox> Martin: A single capture can (in current mode) be simulcast, or FEC, with the same capture ID.
[16:13:26] <Jonathan Lennox> The capture ID indicates the logical media stream. When you're doing simulcast, a single capture can have multiple capture encodings; they'll have the same capture ID
[16:14:56] <hta> Jonathan: "logical media stream"? "single source, single media type media stream"? (video and audio from a single camera would be two logical media streams, not one?)
[16:16:10] <Jonathan Lennox> Right. "logical media stream" isn't well put — the point here is that this is the same media, encoded differently.
[16:17:35] <Martin Thomson> that is consistent with my view, I just wanted to confirm. In that case, it probably doesn't correspond to the msid track, as currently specified, but that doesn't mean that we can't close the gap
[16:19:34] <Martin Thomson> Paul: a la carte model
[16:21:21] Lorenzo Miniero leaves the room
[16:21:36] <Simon Romano> Goodbye e
[16:21:44] Martin Thomson leaves the room
[16:21:46] <Simon Romano> See you for tomorrow'
[16:21:51] <Simon Romano> s session
[16:21:58] <Simon Romano> Presentazione interrotta
[16:22:02] Mark Duckworth leaves the room
[16:22:25] Simon Romano leaves the room
[16:22:48] audioss leaves the room
[16:22:58] Mary Barnes leaves the room
[16:23:04] Andy Pepperell leaves the room
[16:23:26] John Leslie leaves the room
[16:23:26] hta leaves the room
[16:23:35] Mathieu Rene leaves the room
[16:23:40] tomoyuki tsuichihara leaves the room
[16:23:50] Andy Hutton leaves the room
[16:23:54] Alex Amirante leaves the room
[16:24:12] Tobia Castaldi leaves the room
[16:25:31] Ralph Giles joins the room
[16:25:41] tsuichi leaves the room
[16:27:28] Jonathan Lennox leaves the room
[16:31:03] Roy leaves the room
[16:31:33] Zaheduzzaman Sarker leaves the room
[16:32:38] Ralph Giles leaves the room
[17:32:08] Roy joins the room
[17:32:22] Roy leaves the room
[17:51:39] Jonathan Lennox joins the room
[17:51:42] ena2413 joins the room
[17:53:50] ena2413 leaves the room
[17:56:12] Jonathan Lennox leaves the room
[18:02:35] hta joins the room
[18:02:35] hta leaves the room
[18:02:36] hta joins the room
[19:45:01] hta leaves the room