See also: IRC log
<Clarke> Agenda: http://www.w3.org/2011/webtv/wiki/MPTF/Agenda_Telco_18th_August_2011
<inserted> scribe: francois
Kaz: checking with Philippe le Hégaret.
Clarke: Bob, do you know?
Bob: to respond to a comment? Just hit Bugzilla and add a comment. Pretty straightforward.
Jan: I didn't want to make the bug look alive when I'm simply looking for clarification on proposed solution.
[I confirm, simply hitting Bugzilla with comment should be good]
Kaz: you already submitted a bug, right? You need an account
Bob: That's correct.
Bob: I think I sent that out.
Jan: Yes, sorry I didn't
... We can take it over by email.
Clarke: It looks like we'll have
two to four hours for Media Pipeline TF.
... We'll ensure no overlap with HNTF meeting.
... Agenda for today's call includes a list of possible agenda items, let me know if you want to add or refine them.
... [going through the list]
... Pulling out the requirements from the use cases as we do in the HNTF. Third item to get a final recommendation for adaptive bit rate media as there's been discussion here.
Bob: Back on the mapping table, I
don't expect there will be many resolutions on that. I would
expect to discuss what needs to be done, and hopefully get some
insight from chairs as to whether we continue that in the MPTF
or move to another group in W3C.
... Most of the table, we don't have specifications yet.
Clarke: ok. Next item is DRM, goal is to start discussion here.
Bob: I have a number of isolated
proposals for the workshop.
... A number of people have identified DRM as a problem. No use cases so far.
... I can propose use cases withing MPTF if you think that would be good material for the F2F discussion.
Clarke: yes, I think it would be
good to set some context for the discussion.
... Anybody disagrees with that?
Bob: We could also reach out to the authors of the proposals talking about DRM for the workshop and ask them to contribute that as use case for MPTF.
Clarke: Makes sense, yes.
Bob: I'll contact people who mentioned DRM, then.
Clarke: OK. Final item is
consolidation of SVG video and video tags. A bit hesitant
because the topic is large, but it's been discussed here.
... If we can come up with useful recommendations, that could be useful.
... Worst case, we end up with diverging solutions.
... Any other items?
Clarke: ok, feel free to suggest additional topics.
Narm: Got a lot of discussion about HTTP Live Streaming in Berlin. Is the third item about that?
Clarke: yes, that's exactly meant
... The only agenda item I have for today is to review issues. We've started to discuss them, but right now there are in the "raised" status.
... I suggest we go through them and see if we're satisfied with discussion and can accept use case(s) or need more time.
<Clarke> issue 31: http://www.w3.org/2011/webtv/track/issues/31
<trackbot> ISSUE-31 -- Client Ad Insertion -- raised
Clarke: we can move it to various status and close it or say it's accepted.
Jan: Looking at the ability to put in the video a cue to play an ad at specific time.
<kaz> Client Ad Insertion
Bob: Targeted use case is
advertising. One of the media stream would carry the ad at
specific time. You'd like an API so that an application can
react to the ad.
... To replace the ad in the initial stream.
... Another use case is to stop the primary stream, play the ad, and restart the primary one immediately after.
[scribe missed last exchanges because of echo]
Bob: Two fundamental concepts. You have an existing media stream and you want to insert something in that. The other idea is that you just provide a playlist to the video tag to start with, and the browser just plays the playlist.
Jan: trying to create a very basic type of playlist. Not having to do a long playlist.
Bob: First use case that's
described is meant to address this. You can easily say: play at
the end of the first one.
... Then you can create a chain.
Clarke: Jan, if you can make sure your comment is captured in the use case, feel free to do that.
Jan: will do.
francois: just to clarify status really mean whatever you want them to mean.
Clarke: suggest to move it to
... I don't think we should close it right now.
<Clarke> Issue 32: http://www.w3.org/2011/webtv/track/issues/32
<trackbot> ISSUE-32 -- Media Synchronized Web Content -- raised
Clarke: Adequately covered? Do you want to move it to accepted or do you think more discussion is needed?
Bob: I'd like to see this accepted.
[looks good to me]
Clarke: OK, let's accept it. If you have further comments, feel free to pass them on.
RESOLUTION: Accept and close ISSUE-32
Clarke: Two important use cases: closed captions and subtitles. There may not be anything new to do, probably covered
Bob: Yes, this one and ISSUE-32 are pretty much handled but worthwhile being explicit, I think as still under development.
Clarke: right, let's move it to accepted then.
RESOLUTION: Accept and close ISSUE-33
<trackbot> ISSUE-32 Media Synchronized Web Content closed
<trackbot> ISSUE-33 Timed Text closed
Clarke: covers DASH and other formats, how we handle them. I think we should continue the discussions online and try to resolve it during F2F. Big enough issue to benefit from F2F.
Clarke: one way to do streaming
is to do it as a playlist that gets continuously updated.
... another is an infinite segment
<trackbot> ISSUE-35 -- Continuous Streaming -- raised
<kaz> Continuous Streaming
Bob: There is nothing that
prevents the video tag from playing a stream that is created by
a recorder. The second, the ability to play a playlist, does
... The requirements would be quite different. I suggest splitting the two.
Clarke: Yes. I think it's a good idea.
+1 to splitting
Clarke: I'll split this into two
... The first one is covered, I think we can accept it.
Jan: Have you considered streaming from [??echo]
Bob: That's what I had in mind,
if you feed stream from a satellite or cable and take it to a
recorder and then stream it to the browser doesn't create any
problem as far as we could see.
... The play time will wrap around because it's a finite number of bits, but that resolves in about 100 years, so no problem.
... This is broader than adaptive streaming. Independent of the transport, the goal is to support the fact that the stream has not start and no end.
Jan: I think it should cover any
form of broadcast.
... You could have a dvb scheme and tune directly to a channel.
Bob: we agree with that scenario. You could have a gateway that makes broadcast content available as IP on local network.
Jan: well, let's take a TV, no
need for a gateway, you have a broadcast termination
... You could do VOD on cable.
... Understand the focus on video tag in Web browser, but not necessarily restricted to that.
Bob: so you're suggesting to include broadcasting use case here.
Jan: Yes, I just want to extend it.
David: Another idea, [linked to program boundaries but scribe missed]
Bob: related to parental
controls. I think we have a use case on that.
... Program boundaries with change of rating.
[discussion missed because of echo]
Bob: what I'm suggesting is support for time signals, including signals that could be used to identify program boundaries
David: I understand your comment here.
Clarke: on ISSUE-35, I'll create
a separate use case on playlist
... and extend the remaining part to broadcasting (?)
Jan: [comment on some form of playlist]. It's streaming from the network.
Jan: that's what I'm trying to
avoid: creating another video tag.
... I'd like to create a seamless user experience and reuse the existing video tag, also easier for developers.
Bob: Two concepts it seems. You
can provide a new source to the video tag as of today. It stops
playing the first and starts playing the second. The first is
stopped, not paused.
... Not sure what "pause" means for a continuous stream.
Jan: Not clear on the
... Tricky transition.
Bob: you want both of them displayed?
Bob: sounds you need two video tags, here.
[echo remains, scribe missed next steps on issue]
<trackbot> ISSUE-36 -- Parental Controls -- raised
Clarke: [going through use case]
<kaz> Parental Controls
David: independent parental controls, what does it mean?
Clarke: a bit fuzzy here. The
idea is that you would subscribe to parental control
... Perhaps already covered by existing technologies. Need to check.
David: One challenge here is to have a GUID to give to your provider.
Bob: why do we need W3C to do this?
David: Important to identify the level.
Bob: yes, need some mechanism for content rating to be extracted and provided by the application.
Clarke: Suggestion is to continue
the discussion here.
... Any other comment?
<Clarke> Francois, thanks for scribing
This is scribe.perl Revision: 1.136 of Date: 2011/05/12 12:01:43 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: i/Topic:/scribe: francois Succeeded: s/Streaming/Streaming in Berlin/ Succeeded: s/Clarke:/Jan:/ Found Scribe: francois Inferring ScribeNick: francois Default Present: Kazuyuki, rberkoff, JanL, +1.215.286.aaaa, Tatsuya_Igarashi, davidmays, francois, Narm, Bob, Clarke, Panu_Markkanen Present: Kazuyuki rberkoff JanL Tatsuya_Igarashi davidmays francois Narm Bob Clarke Panu_Markkanen Agenda: http://www.w3.org/2011/webtv/wiki/MPTF/Agenda_Telco_18th_August_2011 Got date from IRC log name: 18 Aug 2011 Guessing minutes URL: http://www.w3.org/2011/08/18-webtv-minutes.html People with action items:[End of scribe.perl diagnostic output]