Media Fragments Working Group Teleconference

<trackbot> Date: 20 October 2010

<scribe> Scribe: raphael

<scribe> Scribenick: raphael

1. Admin

PROPOSED to accept the minutes of the 06 October 2010 telecon

http://www.w3.org/2010/10/06-mediafrag-minutes.html

<foolip> +1

<davy> +1

minutes are accepted

ACTION-185?

<trackbot> ACTION-185 -- Yves Lafon to book zakim for the 2 days meeting of the group on Mon 1st and Tue 2nd Nov -- due 2010-09-29 -- OPEN

<trackbot> http://www.w3.org/2008/WebVideo/Fragments/tracker/actions/185

<foolip> November 1 and 2, any time between 0800 and 1800 in UTC+2

Raphael: for Silvia and Philip, could you write down time slot where we can phone you

<silvia> Monday is better for me than Tuesday - it's both in the afternoon/evening for me, I think

<foolip> in other words, any time is fine, since I'm in the same timezone as TPAC

<silvia> if you start early, I can join in the mornings

2. MEDIA FRAGMENT EXTENSION

ACTION-189?

<trackbot> ACTION-189 -- Raphaël Troncy to put the top-level production rules back into the document (in section 4.3.5 or 4.1) -- due 2010-10-06 -- OPEN

<trackbot> http://www.w3.org/2008/WebVideo/Fragments/tracker/actions/189

<silvia> ok

<foolip> can't say anything until I see the actual suggestion for the spec

Raphael: I will implement the ACTION-189 in the spec and ask all the group if we can close this issue

3. Implementation experiment

Raphael: problem with pixels, are we talking about physical pixels or css pixels?

Silvia: pixels in video are always a problem

Raphael: I'm looking for a reference that we could use so that any developers understand what we are talking about

Philip: this definition is provided by HTML5 and not CSS

<foolip> http://www.whatwg.org/specs/web-apps/current-work/multipage/video.html#concept-video-intrinsic-width

<foolip> http://dev.w3.org/html5/spec/video.html#concept-video-intrinsic-width

<scribe> ACTION: troncy to update our spec to talk about video intrinsic width [recorded in http://www.w3.org/2010/10/20-mediafrag-minutes.html#action01]

<trackbot> Created ACTION-190 - Update our spec to talk about video intrinsic width [on Raphaël Troncy - due 2010-10-27].

<foolip> well, there's an issue we discussed at FOMS

Davy: this issue is solved

<foolip> ah :)

Davy: there is indeed too many byte ranges to send to the UA

<foolip> do you always redirect, or just when number of ranges > n ?

Davy: therefore, we send a redirect pointing to a media fragment with the ?

<foolip> always seems simpler :)

Davy: currently, always redirect

Raphael: the UA issues a new request with the query parameter

Davy: yes, and we had a Link header pointing to the master resource

Philip: we could implement that but I really wonder if it is useful
... we will always need a specialized server
... so the query parameter migt be enough

Davy: there are other scenarios where the # is useful
... we should have both
... even if in the retrieval scenario, the # will always be converted into the ?

Philip: what should we do with the name dimension

Davy: in MP4, refer to the cue points; in Matroska, refer to the chapters names

Philip: OK, yes the chapters are part of WebM

<silvia> and Ogg can have chapters through a srt chapter track in Ogg Kate

Raphael: ok, so it seems important that we cover the name dimension even if it is limited to chapter names!
... last issue from Philip, make the hour optional in specifying the time dimension
... currently, we need one or two digits
... but most clips have duration less than 1 hour
... but the grammar will be more complicated
... because we could then have just one digit for the minutes

Philip: in WebSRT, the hours are optional

Raphael: I suggest we discuss this at the F2F

<silvia> I would support that if you can find a grammar to describe it correctly - it's also more complicated to implement then, btw

Look at the demos at http://www.w3.org/2008/WebVideo/Fragments/code/plugin/demos/

Davy: yes, Ninsuna understands range request with SMPTE time codes
... and answers with smpte time codes
... though I need to check

Yes Silvia, that's also our conclusion, re the grammar

scribe: the questio is who is willing to work out this more complex grammar? I suggest to hack this during the f2f but if you want to work on it before, please feel free

<silvia> do it with Yves - he is good at that stuff :)

4. AOB

no other topic to discuss

meeting adjourned

<foolip> ok, about the syntax then

<foolip> it's be simplest to just say that all components are 1 or 2 digits

<foolip> ugly, but simpler, and simple is good :)

:-)

<foolip> but let's discuss that further later

we will see what the grammar should loook like, indeed, Yves might be able to sort that out elegantly