Media Pipeline Task Force Teleconference

10 May 2012


See also: IRC log


clarke, kaz, bryan, bob, joe, kevin, Juhani, eric, john, ??P32, 443-030-40xxxx, 425-214-xxxx, 858-735-xxxx, 617-761-xxxx, 415-832-xxxx, 617-761-xxxx, ??P58, 206-218-xxxx, 425-882-xxxx, 617-761-xxxx
kstreeter, kaz


<nisc> this passcode is not valid ? :(

<inserted> scribenick: kstreeter

face to face last week

Bob: Summary of last week:

aaron presented the media source proposal

[mark?] presented the encrypted media proposal

discussed the formation of a media task force

it was decided to keep the work in the group

discussion on whether to keep the two proposals together

John: none of the acrimonious comments on the enc media proposal really came up in the f2f

there is a transition happening between HTML5 and HTML next

Aaron: there is a sense that the HTML5 ship has sailed, and there is an opp for cooperation moving forwad in the HTML WG


Aaron: Paul Cotton indicated that he would moderate the calls regarding the media proposals

John: Google presented a couple of good demos, are these available online?

Aaron: yes, online will find a link


some of the metrics being discussed by webperf could be utilized by the media source elements

this group should familiarize themselves; we will discuss next week

bug tracking

many messages this week, means work is progressing


no new bugs

requirements documents

<Clarke> ABR Requirements: http://dvcs.w3.org/hg/webtv/raw-file/tip/mpreq/MPTF-ADB-Requirements.html

updated the terminology

group should review the terminology changes as well as the reqs

added use cases

reviewing use case 5.1

the terminology for ABR "media" or ABR "system" is ill-defined, we could clarify this

reviewing use case 5.2 U2

Kevin: could add the need to handle trick modes across time discontinuities

Joe: content protection can also impact trick modes, we should call that out as well

Aaron: what are the expectations around consistency between different systems?

Clarke: the primary thing is that from a user perspective dealing with trick modes in ABR should be similar to that of non-ABR content

Aaron: some adaptive systems might handle trick modes slightly differently, for example switch to lower bitrate

Bob: even in non-adaptive there are often differences on how different media handle trick play
... different user agents will often do trick modes differently to enhance user experience

Kevin: perhaps we could define this more as a req for the mechanisms to initiate trick plays, rather than the end experience

Bob: could express this in terms of the media element attributes that manage trick modes
... this could be more succinct

reviewing use case 5.3 U3. Search Adaptive Bit Rate Content

Aaron: we should use "seek" rather than "search"

<inserted> scribenick: kaz

Clarke: U5. Continuous Adaptive Bit Rate Content
... time reference?

bob: who made that comment?

kstreeter: kevin from Adobe

bob: we had some discussion before
... are you aware how systems today handle that?
... what should for adaptive streaming?

kstreeter: not aware
... I imagine system like Tivo should handle this

bob: ok

<nisc> kaz, who is here

<nisc> kaz, who is here?

<nisc> er

<nisc> sorry kaz ;-)

bob: granularity of time

Clarke: I guess we could leave it in the requirements, and reconsider it later
... any other comments for the UC 5?


<inserted> scribenick: kstreeter

reviewing use case 5.6 U6. Timed Tracks and Adaptive Bit Rate Content

Clarke: this is related to the time base requirements

DASH supports this in the MPD

Bob: this might boil down to that the implementation must support timed text tracks if present

Aaron: we are going to support inline text tracks

Kevin: how would this work with external caption file?

Aaron: yes, you would just use the existing track mechanism
... since this is built on existing HTML5, i don't think we need to call this out specifically as a requirement

Clarke: given the above requirements on timeline, this use case might not be required

Aaron: this req does allow a timed text track to be merged into content that did not already have one
... even if this is redundant timed text is important it is worth leaving in

next week we will talk about the enc content doc

Aaron: link points to old document

<Clarke> Thanks for scribing, Kevin

<kaz> [ adjourned ]

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.136 (CVS log)
$Date: 2012/05/10 16:02:30 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
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/Agenda:/scribenick: kstreeter
Succeeded: s/[Bob]/Bob:/
Succeeded: i/Summary of last week:/topic: face to face last week
FAILED: s/fowad/forward/
Succeeded: s/int eh/in the/
Succeeded: s/5.3 U3/5.3 U3. Search Adaptive Bit Rate Content/
Succeeded: s/time track/timed text track/
Succeeded: i/Continuous Adaptive/scribenick: kaz
Succeeded: i/reviewing use case 5.6/scribenick: kstreeter
Found ScribeNick: kstreeter
Found ScribeNick: kaz
Found ScribeNick: kstreeter
Inferring Scribes: kstreeter, kaz
Scribes: kstreeter, kaz
ScribeNicks: kstreeter, kaz
Present: clarke kaz bryan bob joe kevin Juhani eric john ??P32 443-030-40xxxx 425-214-xxxx 858-735-xxxx 617-761-xxxx 415-832-xxxx 617-761-xxxx ??P58 206-218-xxxx 425-882-xxxx 617-761-xxxx
Agenda: http://www.w3.org/2011/webtv/wiki/MPTF/Agenda_Telco_10th_May_2012
Got date from IRC log name: 10 May 2012
Guessing minutes URL: http://www.w3.org/2012/05/10-webtv-minutes.html
People with action items: 

[End of scribe.perl diagnostic output]