{minutes} HTML WG media telecom 2014-08-19 EME status and bugs

http://www.w3.org/2014/08/19-html-media-minutes.html

Joe Steele



HTML Media Task Force Teleconference

19 Aug 2014

Agenda

See also: IRC log

Attendees

Present
paulc, glenn, jdsmith, Aaron_Colwell, +1.408.536.aaaa, joesteele, markw, davide, +1.425.936.aabb, ddorwin, ReimundoGarcia, adrianba, +1.303.661.aacc, BobLund, pal, +1.425.605.aadd
Regrets
Chair
paulc
Scribe
joesteele
Contents

Topics
MSE Status and Bugs
Review ACTION items
MSE Test Suite Status
EME status and bugs
Bug 26575 - Separate creation of the MediaKeySession from "message" event generation
Bug 26600 - Text is confused between persistent session vs persistent licenses
EME bugs with a proposal
Bug 25268 - Reduce the burden on applications to deduplicate initData from many needkey events
Bug 24771 - Provide guidance on object and CDM lifetime (including when events are guaranteed to be fired)
EME bugs awaiting input from Task Force or actions
Bug 26372 - Revisit the need for EME-specific DOMException names and the "error" attribute and event
[Bug 26332] Applications should only use EME APIs on secure origins (e.g. HTTPS)
[Bug 26401] Key message destinationURL usage is not reflected in examples
Bug 25923 - isTypeSupported should be asynchronous
Do we need LoadSession?
EME Use cases Wiki
Any other business?
Summary of Action Items
<trackbot> Date: 19 August 2014
<scribe> Scribe: joesteele
<paulc_> Agenda: http://lists.w3.org/Archives/Public/public-html-media/2014Aug/0013.html
MSE Status and Bugs

Review ACTION items

<paulc_> ACTION-59?
<trackbot> ACTION-59 -- Paul Cotton to Coordinate with web apps on the streams api, give them mse requirements and timeline -- due 2013-12-17 -- OPEN
<trackbot> http://www.w3.org/html/wg/media/track/actions/59
paulc: dug up this old action because we have a dependency
<paulc_> See best status: http://lists.w3.org/Archives/Public/public-webapps/2014JulSep/0305.html
paulc: sent the other chairs a note and got this back
... just asking the editors what their plan of action is
... if no response from Art or Charles then will followup in next week
MSE Test Suite Status

paulc: 2 items -- set of 7 pull requests, important ones are the open ones
<paulc_> Open pull requests: https://github.com/w3c/web-platform-tests/pulls?q=is%3Aopen+label%3Amedia-source+is%3Apr
paulc: in particular last time we discussed was end of June, original pull request for blink MediaSource test was still outstanding
<paulc_> Aaron's original pull request is not closed: https://github.com/w3c/web-platform-tests/pull/725
paulc: still not closed
... did not get many volunteers for this
... sent MarkW a note about this, he had promised to try to find someone to look at this
<paulc_> See Paul's request to Mark: http://lists.w3.org/Archives/Public/public-html-media/2014Aug/0017.html
markw: I did that, but no results as yet
... folks were on vacation last time
paulc: we have a fair number of open pull requests many from Shishimaru
... we need to make progress
<markw> s/promised to find someone/promised to try to find someone/
paulc: one on getting us out of CR, and one on test suites and one on test results
... no progress on getting these resolved
paulc: can stay in CR forever if not addressed
... what are the next steps here?
... would like a company to step forward to promise to make progress on this
acolwell: believe I have made all the changes needed for my initial import
... remaining are comments that UI suggests I don't close because they were raised by others
... needs more comments from folks who raised them
... believe I have resolved the issues despite what the UI says
paulc: Jerry or Adrian -- can you unblock this?
acolwell: I can close, but it warns me not to
jdsmith: I will circle back with Chris on that
paulc: please send us an email on that so we have a record
jdsmith: ok
paulc: what about volunteers for other pull requests?
acolwell: others look like they may be dups -- is that OK?
paulc: maybe make comments to that effect -- otherwise folks might submit additional tests that are dups
acolwell: ok
paulc: is there anyone who can volunteer themselves or someone else?
acolwell: I can look at the posted stuff
adrianba: I think we are running into a common situation, which is that folks on the call are concerned with the spec, and folks who do testing are not here
... maybe we need to ask the testing folks for this? Testing Task Force?
paulc: they have been MIA for this Task Force -- wiki is dormant and they have not been attending WG meetings
... I think you are suggesting we are asking the wrong audience?
adrianba: Testing interest group might be the right group
paulc: would have thought this would be the right group -- we have the vested interest
paulc: I am really asking the member companies how we will get through this
adrianba: agree with the point about this task force, but I disagree that folks in the testing task force are just interested in testing
paulc: here is the link
<paulc_> Testing IG: http://lists.w3.org/Archives/Public/public-test-infra/
paulc: sending to this list might be one tack we can take
joesteele: will poke around and see if I can find someone
markw: I will try again as well
paulc: will try to close Aarons original request, Aaron will close additional dups if possible, I will send email for guidance from other editors, Joe and Mark will look for additional resources
EME status and bugs

Bug 26575 - Separate creation of the MediaKeySession from "message" event generation

https://www.w3.org/Bugs/Public/show_bug.cgi?id=26575
paulc: some very specific changes in here
... is this related to the persistence discussion?
ddorwin: not related to that
paulc: anyone else reviewed yet?
joesteele: not me
jdsmith: I would like to look at the problem more
paulc: will put on the agenda for next week
jdsmith: changes seems reasonable, but want to look at the underlying issue
Bug 26600 - Text is confused between persistent session vs persistent licenses

https://www.w3.org/Bugs/Public/show_bug.cgi?id=26600
paulc: Mark you want to talk about this?
markw: trying to make a proposal everyone could agree to, but this is not likely to be controversial
... can work this out in the bug
paulc: can close in the next week or at least come closer
ddorwin: agree that text is not exactly clear, was trying to fix existing text which is totally broken
... problem with persistent license in temporary session
EME bugs with a proposal

Bug 25268 - Reduce the burden on applications to deduplicate initData from many needkey events

https://www.w3.org/Bugs/Public/show_bug.cgi?id=25268#c7
<paulc_> https://www.w3.org/Bugs/Public/show_bug.cgi?id=25268#c7 contains Glenn's proposal
paulc: last week we thought we could resolve with no change
... can we close?
ddorwin: he thought we could move on, didn't see him say we could close this. WORKSFORME might not be right at this point
<adrianba> i agree with david
paulc: you are right -- he said delay for later version
<paulc_> Last meeting discussion: http://www.w3.org/2014/08/12-html-media-minutes.html#item09
paulc: will just move this to the bottom of the queue
Bug 24771 - Provide guidance on object and CDM lifetime (including when events are guaranteed to be fired)

https://www.w3.org/Bugs/Public/show_bug.cgi?id=24771
paulc: David you said you will make the change after the re-org
<paulc_> https://www.w3.org/Bugs/Public/show_bug.cgi?id=24771#c4
paulc: just make sure folks are aware this is coming
... you were going to start the re-org right after the heartbeat
ddorwin: would like to deal with the createSession/loadSession bug
paulc: probably 25% of outstanding bugs are in that group
... record as dependent
EME bugs awaiting input from Task Force or actions

Bug 26372 - Revisit the need for EME-specific DOMException names and the "error" attribute and event

https://www.w3.org/Bugs/Public/show_bug.cgi?id=26372
paulc: this is the one about the diagnostic error code, last week Jerry was going to open a new bug
ddorwin: this is not directly related to the system code bug, have since renamed this bug
... we have Promises and those deal with that type of errors, some issues we don't have ways to report yet
... seems like there are a few issues, maybe we don't need generic solution
<ddorwin> The new summary is "Report issues/events not related to a specific method call"
jdsmith: commenting on the system code bug, have not prioritized this yet as working on loadSession issues.
... think the DOMException bug this was tied to is already resolved
paulc: David, what will happen to bug 26372? most recent change of the summary
... you are saying we don't have a proposal for this
ddorwin: yes
<paulc_> New summary is "Report issues/events not related to a specific method call"
paulc: you made a change to the doc last week to remove the error related stuff
... then you changed this bug
... who is on point to make a proposal to report issues not related to a specific message call?
... want to know who to go to
glenn: the reporter is the default assignee right?
paulc: David has lot on his plate -- can we get someone else in parallel to Jerrys work?
... not hearing volunteers, going to move on
jdsmith: I will take this bug
[Bug 26332] Applications should only use EME APIs on secure origins (e.g. HTTPS)

https://www.w3.org/Bugs/Public/show_bug.cgi?id=26332
paulc: seem to have folks strongly on both sides
... not sure how we will get consensus on this item -- just want to bring it up again
... this is a "polarizing" issue
[Bug 26401] Key message destinationURL usage is not reflected in examples

https://www.w3.org/Bugs/Public/show_bug.cgi?id=26401
paulc: Joe offered to update this bug
... traffic back and forth between David and Joe
<paulc_> See start of dialogue at https://www.w3.org/Bugs/Public/show_bug.cgi?id=26401#c2
<paulc_> Joe: Discussion brought out at least two issues
<paulc_> ... message versus needKey
<paulc_> ... Joe had an opinion that David's response has clarified and there might be a needed change
<paulc_> ... In last set of comments, David was indicating that he did not think the URL should come from the InitData
<ddorwin> i was dropped
<paulc_> ... Joe wants discussion about whether PSSH boxes should not contain URLs or if they do they should not be used by CDM
<paulc_> ... Original design of PSSH boxes was if the information is in the boxes then the CDM should be able to use it
<ddorwin> PSSH boxes can contain URLs, but EME implementations may not expose them as discussed in https://www.w3.org/Bugs/Public/show_bug.cgi?id=25920
<paulc_> ... Side issue about where else the URL could come from
jdsmith: it is historically common for the PSSH to contain the URL and to use it -- we (Microsoft) think it should be there and should be used as well
ddorwin: essentially "used" and "exposed" mean the same thing here
<ddorwin> Also, PSSH boxes are a problem for interoperability. We should not encourage more data in PSSH boxes.
paulc: should we continue in the bug? Jerry can you add that to the bug?
ddorwin: should we close the other as a dup?
<paulc_> https://www.w3.org/Bugs/Public/show_bug.cgi?id=25920 was RESOLVED FIXED
<ddorwin> Joe wants to extract URL(s) from PSSH data. That is what https://www.w3.org/Bugs/Public/show_bug.cgi?id=25920 was about
jdsmith: that bug was specifically about default URL's and did not cover the initData containing a URL
<ddorwin> i'm disconnected again
jdsmith: we had a discussion that said the CDM could provide the destinationURL -- that includes any source the CDM wishes to use
paulc: so this is not a dup?
jdsmith: this is a previous issue
joesteele: do not believe this is a dup
paulc: Jerry believes that bug is not a true dup
... given the time let's go broad
... leave that item for more discussion
Bug 25923 - isTypeSupported should be asynchronous

https://www.w3.org/Bugs/Public/show_bug.cgi?id=25923
paulc: this bug looks like it is dependent on bug 26207 -- what should we do with this now?
ddorwin: sounded like they had a lot in common
paulc: think we are waiting to see if other changes will emerge -- can we make this change now?
ddorwin: there are good reasons for and against this change
... could be ok but difficult
... input from Steve (a developer) that would rather not do this
... don't think we have a clear path
... this makes the model much more complicated for canPlay types
paulc: concerns on both sides then?
ddorwin: yes
paulc: seems like a call for input from other task force members on the path to follow -- Mark do you have an opinion?
markw: not really
... as long as it is there
paulc: Jerry would help if you can look at it
jdsmith: agreed. Have viewed this as a mixed rational change -- not the most pressing thing to resolve
Do we need LoadSession?

paulc: Joe offered to follow up -- he did. Email has generated a response from David
<paulc_> Update from Joe: http://lists.w3.org/Archives/Public/public-html-media/2014Aug/0015.html
<paulc_> David has responded and Joe will review
EME Use cases Wiki

paulc: Joe offered to update the wiki with outstanding bugs
<paulc_> https://www.w3.org/wiki/HTML/Media_Task_Force/EME_Use_Cases now contains links to bugs
<paulc_> We need TF members to review
joesteele: yes -- not sure completely accurate -- please review
Any other business?

paulc: propose that we meet again next week, push MSE out two weeks, just EME next week
... David when will re-org happen?
ddorwin: planning to do that this week
paulc: if you get that done, just let the TF know so they can review
ddorwin: ok
paulc: Thanks everyone!
Summary of Action Items

[End of minutes]
Minutes formatted by David Booth's scribe.perl version 1.138 (CVS log)
$Date: 2014-08-19 16:14:26 $