{minutes} HTML WG media telecon 2014-05-06 - MSE and EME status and bug discussion

http://www.w3.org/2014/05/06-html-media-minutes.html

Joe Steele



HTML Media Task Force Teleconference

06 May 2014

Agenda

See also: IRC log

Attendees

Present
+1.425.936.aaaa, Aaron_Colwell, +1.781.221.aabb, pal, davide, Cyril, adrianba, paulc, markw, ddorwin, BobLund, [Microsoft]
Regrets
Chair
paulc
Scribe
joesteele
Contents

Topics
Role Call
action items
MSE status and bugs
bug 24370
ACTION 65
bug 25518
MPEG audio byte stream format spec
MSE CR tests
EME status and bugs
EME use of Promises
EME bugs new and reopened
bug 25663
bug 25537
bug 25506
Summary of Action Items
Role Call

trackbot, start the meeting
<trackbot> Meeting: HTML Media Task Force Teleconference
<trackbot> Date: 06 May 2014
<paulc> ACTION-65?
<trackbot> ACTION-65 -- Paul Cotton to Make sure html5 bugs filed for item #3 related to MSE -- due 2014-04-16 -- OPEN
<trackbot> http://www.w3.org/html/wg/media/track/actions/65
<paulc> Agenda: http://lists.w3.org/Archives/Public/public-html-media/2014May/0005.html
action items

MSE status and bugs

paulc: acolwell sent out a draft with some closed bugs, any feedback?
acolwell: no but was not expecting
paulc: two open bugs
<paulc> http://tinyurl.com/kfeptqy
bug 24370

<paulc> https://www.w3.org/Bugs/Public/show_bug.cgi?id=24370
paulc: couple of outstanding items
acolwell: I need to implement the spec changes I proposed awhile ago (part C)
paulc: you updated the proposal?
acolwell: yes
<scribe> ... pending
acolwell: should be integrated in next week or so
<paulc> Revised proposal to be implemented: http://lists.w3.org/Archives/Public/public-html-media/2014Apr/0087.html
acolwell: that is proposal (i)
acolwell: not the most recent one though
... no that is the beginning of the thread -- there were later updates
<acolwell> http://lists.w3.org/Archives/Public/public-html-media/2014Apr/0095.html
<acolwell> that is the most recent IDL
paulc: looks like last message is #95
<paulc> Last April message on this thread is http://lists.w3.org/Archives/Public/public-html-media/2014Apr/0095.html
cyril: comment from Mark said you would only have 1 track per file but would need to set the kind -- is that correct?
acolwell: alternate proposal does solve that problem -- initial suggestion was commented on and I integrated those changes
... track ID is optional parameter on the constructor now
cyril: what happens if you don't put track defaults before?
acolwell: values will be unset
cyril: will they be derived from the initialization segment?
acolwell: yes -- if in the byte stream will derive that
markw: yes - what is proposed is the comments I made
... I guess that parameters could be nullable -- was thinking of them as integers
acolwell: I did not make it that in case someone put the GUID as the track ID
... will make a note to make that explicit
paulc: sounds like you are going to add this to the front of the queue -- for next couple of weeks
ACTION 65

action-65?
<trackbot> action-65 -- Paul Cotton to Make sure html5 bugs filed for item #3 related to MSE -- due 2014-04-16 -- OPEN
<trackbot> http://www.w3.org/html/wg/media/track/actions/65
paulc: #3 I believe that refers to and was a comment about the HTML5 spec being made clearer in this area -- can you refresh my memory?
acolwell: believe this has to do with kind and language expcted to be immutable
... if there is a change -- what happens?
... sounds like what happens is the track is removed and a new track is created - but nothing in the spec says that
... clarification is needed
<paulc> ACTION-65 refers to "Sylvia and/or Hixie update the HTML specs to reflect the expected behavior for language & kind changes."
acolwell: people have said that it is possible for them to change during a presentation so spec needs to say what happens
<paulc> See http://lists.w3.org/Archives/Public/public-html-media/2014Apr/0043.html
paulc: bullet items at the end -- item 3 seems to refer to me
... I believe I need to take this meesage and reply with Sylvia in the "to" field to see if any bugs apply to this message
... any other comments?
bug 25518

<paulc> https://www.w3.org/Bugs/Public/show_bug.cgi?id=25518
https://www.w3.org/Bugs/Public/show_bug.cgi?id=25518
acolwell: something we found in implementation -- need positive infinity
... allow removing everything
... trivial change -- adds functionality
paulc: any comments on that?
MPEG audio byte stream format spec

https://dvcs.w3.org/hg/html-media/raw-file/default/media-source/mpeg-audio-byte-stream-format.html
paulc: original proposal put out after the F2F and was revied by Jerry
s/reviewedreviewed/
<paulc> Arron's response to Jerry's review: http://lists.w3.org/Archives/Public/public-html-media/2014Apr/0085.html
paulc: question is -- what is the next step?
acolwell: consider it done -- last update contains what I proposed
paulc: so this just sits as an editors draft?
acolwell: when it goes to CR we would snapshot the specs as we did before
paulc: you remind me that when we published before, we published some of the other bytestream specs with it?
acolwell: yes we had to make copies for some reason
adrianba: feedback from the director was that we should have something that was more traceable than the version control system
... so they were copied into the W3C space -- some discussion about the long term future for this
... i.e. was a wiki an appropriate method -- wanted something more formal
<paulc> Example of what was published before: http://www.w3.org/2013/12/byte-stream-format-registry/webm-byte-stream-format.html
adrianba: need to discuss this more specifically -- because not a rec-track document?
paulc: I gave an example above
<paulc> Registry: http://www.w3.org/2013/12/byte-stream-format-registry/
paulc: think there are two steps here (1) update the registry (2) move into W3C space
... are those the two steps?
acolwell: right now the editors draft should be pointing to the editors draft of the registry and the MP3 spec is in there
adrianba: should we stop doing that?
... is the registry done? we should look at the byte stream format stable enough to call done as well?
... should determine how to publish them
... as a complete doc that the WG has published -- should we wish to add/update in the future
... update in the one stable registry
paulc: the actual CR draft has a section A.2 "Informative References" which covers the registry -- but editors draft is missing that
<paulc> CR draft: http://www.w3.org/TR/2014/CR-media-source-20140109/
paulc: open that document and go to the end -- you will see that there is a reference to the registry in A.2
<paulc> A.2 Aaron Colwell Media Source Extensions Byte Stream Format Registry. 02 December 2013 URL: http://www.w3.org/2013/12/byte-stream-format-registry/
paulc: that is not in the editors draft
acolwell: yes
paulc: ok -- where is the updated registry?
<acolwell> https://dvcs.w3.org/hg/html-media/raw-file/tip/media-source/byte-stream-format-registry.html
acolwell: if you look in the editors draft there is a pointer to it -- not an informative reference -- did not update to include it
paulc: that is a bug
acolwell: it is in the introduction
paulc: that is a bug
... need to fix the reference
... Adrian - is your suggestion that we should make a statement that we do not want a dated version in the registry? i.e. just a floating reference?
adrianba: not sure what you mean
acolwell: currently the MPEG spec depends on the current editors draft
... if we put the registry in a static location and update what the CR points to, the MPEG spec won't make sense
... the algorithm is not there
adrianba: we will need a plan for how to update the CR draft -- whether it requires a call -- then we can tackle the issue of where we keep the registry
acolwell: fine with that -- did not intend to fork at CR
paulc: can you open an MSE bug for this please?
... we should have a CR bug that raises these questions and proposes how we solve it so we can point to this bug in the future
MSE CR tests

paulc: we have tests submitted (Mike Smith) -- we need a reviewer
... any volunteers?
... this is message #81
<paulc> http://lists.w3.org/Archives/Public/public-html-media/2014Apr/0081.html
cyril: replied privately to Mike for clarification -- no reply as yet
paulc: more likely to get him on IRC -- if you have clarifications other might appreciate seeing them
... should consider responding on the list
... or CC me and I will try to get his attention
cyril: question was -- how many tests in total? would like to see what I can review
paulc: next item is pertinent
... clikc on the link -- see about 10 tests
cyril: yes but those are only some
<paulc> https://github.com/w3c/web-platform-tests/pulls/shishimaru
paulc: your question is where are the other tests?
cyril: yes - which tests apply to MSE?
<paulc> ACTION-64?
<trackbot> ACTION-64 -- Aaron Colwell to Review at the MSE test suite comments in https://critic.hoppipolla.co.uk/r/951 -- due 2014-04-22 -- OPEN
<trackbot> http://www.w3.org/html/wg/media/track/actions/64
<paulc> https://critic.hoppipolla.co.uk/r/951 to be reviewed
paulc: looking at the agenda -- next item was ACTION-64
action-64?
<trackbot> action-64 -- Aaron Colwell to Review at the MSE test suite comments in https://critic.hoppipolla.co.uk/r/951 -- due 2014-04-22 -- OPEN
<trackbot> http://www.w3.org/html/wg/media/track/actions/64
paulc: looking at that link -- comments that were submitted on Aarons tests
... submitted on Mar 7th
<paulc> Arron's tests: https://critic.hoppipolla.co.uk/c6841a09?review=951
paulc: that is the link
... here is the pull request
<paulc> See https://github.com/w3c/web-platform-tests/pull/725 for Arron's tests
paulc: Aaron tooks action 64 to respond to those comments at the F2F
... need someone to respond to Shishimarus tests
cyril: ok would like to review those tests and have help on the reviewing system
paulc: might be more folks to help you with that
... I can help you find someone (Kris Krueger)
jdsmith: I will also volunteer
<paulc> Kris Krueger is a good contact via jerry.
acolwell: I will also respond -- that is on my todo list
paulc: I will put this on the agenda for 2-3 weeks out
EME status and bugs

http://tinyurl.com/7tfambo
paulc: bugs have closed since last night
<paulc> 24 bugs open now
paulc: good news
... 2 agenda items here
EME use of Promises

https://www.w3.org/Bugs/Public/show_bug.cgi?id=25199
paulc: implemented after the F2F meeting
... message #68
... noticed a question from Jerry about that
... Jerry was your question answered?
<paulc> See David's response: http://lists.w3.org/Archives/Public/public-html-media/2014May/0002.html
jdsmith: thinking
... since I wrote this I had some internal discussions -- I think the solution of looking for an existing session and returning null makes sense
... I withdraw my comment
paulc: wanted to check that editors are looking through the now-unblocked bugs and see which ones can be handled
<paulc> 25199 Blocks: 17750 21798 24081 24216 24771
ddorwin: fixed 3 already -- more on my list. closed 10 bugs since the F2F
<ddorwin> Still open: 21798, 24771
ddorwin: 21798, and 24771 are still outstanding
<paulc> 21798 and 24771 are Open
paulc: previous agenda had a big batch of EME bugs
... believe I have dealt with the items except item #5
... for next week need assistance from the editors on bugs to resolve
EME bugs new and reopened

paulc: 6 bugs on this list
<ddorwin> https://www.w3.org/Bugs/Public/show_bug.cgi?id=17673
ddorwin: mention bug 17673 -- I have implemented that so it would be great to have folks review it
<markw> I will
paulc: should we have volunteers? any volunteers?
<jdsmith> I will
paulc: I will add that explicitly in the next agenda
bug 25663

<paulc> https://www.w3.org/Bugs/Public/show_bug.cgi?id=25536
<paulc> s/25563/25536/
ddorwin: we added an ambiguous release() which causes problem -- this shows how to make it non-ambiguous be adding another method
<paulc> Mark: look at https://www.w3.org/Bugs/Public/show_bug.cgi?id=25536#c2
paulc: Mark -- there is a question back to you -- have you seen it?
markw: yes I have
bug 25537

<paulc> https://www.w3.org/Bugs/Public/show_bug.cgi?id=25537
ddorwin: proposal to expose the license expiration time on the media key session
... assumes that a session is a license
paulc: joe?
joesteele: https://www.w3.org/Bugs/Public/show_bug.cgi?id=25218
joesteele: This is where discussion has been ongoing
joesteele: two outstanding issues -- what is a license versus a session? what kinds of keys can be delivered in a session?
markw: I will ask my question in the bug
bug 25506

paulc: we will deal with this with the editors I think
ddorwin: correct
<paulc> https://www.w3.org/Bugs/Public/show_bug.cgi?id=25506
paulc: ok -- other two bugs we will carry forward
<paulc> Note that 25434 and 25385 were recently reopened.
paulc: meet again next week
... I will organize the agenda
... Aaron let me know when we have made enough progress on MSE
... or send information to the list
... by default it will be on the agenda in 2-3 weeks
... we are adjourned.
Summary of Action Items

[End of minutes]
Minutes formatted by David Booth's scribe.perl version 1.138 (CVS log)
$Date: 2014-05-06 16:08:29 $