W3C

HTML Media Task Force Teleconference

13 Sep 2016

Agenda

See also: IRC log

Attendees

Present
paulc, Kaz, ddorwin, markw, MattWolenetzx, jdsmith, plh
Regrets
Philippe, Mark_Vickers, Francois
Chair
paulc
Scribe
paulc

Contents


HME WG F2F meeting, Lisbon, Mon-Tue Sep 19-20

https://lists.w3.org/Archives/Public/public-html-media/2016Sep/0043.html

Suggestion is to hold a joint Web & TV IG and HME WG meeting to discuss future use cases on Monday morning.

So attendees should find the Web & TV IG meeting space for that joint meeting

paulc: does the Web & TV IG have used cases?

Kaz: The Web & TV IG's task force may have use cases
... I want to talk to Mark Vickers about older use cases from Sapporo TPAC

paulc: I believe the WAVE community plans to submit new use cases ASAP
... This will be "unconference like" session with the agenda determined at the start of the joint meeting

paulc will coordinate with Web & TV IG chairs on this session

Matt: Please ensure there are good notes for non-attendees

Paulc: Will do

MSE Editorial issues, MSE Editors

https://github.com/w3c/media-source/milestones/V1Editorial

https://github.com/w3c/media-source/issues/159 will be triaged at V1Editorial and will be done as before we transition to PR

Jerry will prepare a pull request and Matt will review so we have this change ready.

https://github.com/w3c/media-source/issues/157 is also untriaged

ISSUE-157 is marked as V1NonBlocking

<scribe> Chair: How do we handle this?

Matt: If the registry can be updated separate from MSE then I can do a pull request for this.

<scribe> Chair: Matt should coordinate with Philippe since the MSE Registry work is still pending

Jerry: I have information now and will respond today

Matt: I will assign 157 to Jerry and will close with no action my PR 158

MSE Registry publication as WG notes, Philippe

This is still pending and plh told me he would guarantee that it does not hold up MSE's transition

See https://github.com/w3c/media-source/issues/74

CfC for MSE Proposed Recommendation closes Fri Sep 16

There has been support but no dissent yet.

https://lists.w3.org/Archives/Public/public-html-media/2016Sep/0053.html

Due to the AC meeting next week and the publication moratorium I am expecting the PR transition to occur either in Lisbon or after the TPAC week. Publication of the PR to the AC would likely occur in late Sep or early Oct.

FTR the usual PR review period for the AC is 6 weeks

MSE test suite and results

plh tells me the WebIDL new results are integrated as are recent Firefox results

<scribe> Chair: Any outstanding test results?

Matt: Do we need more text track testing?

<scribe> Chair: Can you send an email about text track testing to indicate what tests exist and which part of the spec is not tested.

Matt: I will start a new thread on the public-html-media email list

see Text tracks: https://github.com/w3c/web-platform-tests/issues/3673

MAtt: There are some WebIDL differences but these are outside of MSE and are more general problems

David: There are diffences on how Blink handles some semantic tests that actually test the webidl generator

MAtt: The implementation report is slightly out of date. Matt will identify the work that still needs to be added.

EME Editorial issues, EME Editors

ISsue-317 and 320 were outstanding as of yesterday

David: there is PR for 320 that needs to be reviewed
... issue-317 needs to be cleaned up
... issue-183 - we need to discuss his with Philippe

<scribe> Chair: plh told me yesterday that 183 is certainlly not going to blocked EME

<scribe> Chair: We need to simply decide if any changes to the registry SOTD are required.

EME NonBlocking issues, EME Editors

Issue-242 has a PR that needs to be reviewed

David: I opend Issue-312 in response to one of the objection threads but no one has responded.
... We are done the Editorial and Nonblocking items once the PRs are handled.

Other EME issues

ISSUE-85: https://github.com/w3c/encrypted-media/issues/85

<trackbot> Notes added to ISSUE-85 .

<scribe> Chair: The last time we touched on ISSUE-85 we said we were waiting for interop testing

EME at risk features

David: persistent usage record - pending testing results
... is the current status that Edge is the only browser that can pass?

Mark: This is the same for many tests so we have a general discussion item here
... ... mentioned use of various polyfills for some of these tests
... keystatus is not working the same on more than one browser
... We need to agree on how to handle features that only pass on only one browser

David: If we have features that don't pass testing then we either have to wait for changing implementations or take the features out and repeat CR

<scribe> Chair: Didn't we agree to have this discussion AFTER we completed the conversion of the Google tests?

Mark: Yes
... There are quite a few of things that are not passing and this could either be bad tests, bugs in implementations, etc.
... There will likely be some features that don't have two implementations

David: Have we reorganized the test suite so that the tests don't distinguish on key system?

Mark: I will check into whether the most recent results are merged

David: Chrome55 has better WebIDL results
... if you want to get those results

Jerry: Which version of Chrome should be test with?

David: Canary

rrsagent

<scribe> Chair: We need to follow up on the previous actions and update the test results

Jerry: We need to evaluate the results eventually

Mark: We need to merge the persistent license tests results in merger
... This depends DRMToday bug fix

Jerry: We are still waiting for the test cases to be converted

Jerry's status report: https://lists.w3.org/Archives/Public/public-hme-editors/2016Sep/0014.html

The above link indicates what tests are being done by Netflix and we are stll waiting for a volunteer for the 14 outstanding tests

Mark: I will report tomorrow on the current work being done by Netflix

Mark and Paul: Maybe we will have an EME test hacking session at TPAC

David: I understand Jerry's desire to minimize his work

Jerry: I will do an update with Mark's change to not distinguish on key system

<scribe> Chair: Paul provided an oral summary for late arriving plh

plh: We need to review the Netflix tests
... I know how long we need to finish MSE but it is NOT clear how to handle our charter for EME
... We need to know our timeline for EME before the end of TPAC
... We run a risk of not getting EME done if we don't have a clearer plan for EME testing
... We need someone to step up for reviewing the Netflix migration tests

Jerry: MSFT will review Netflix converted tests but we don't have the resources to do the conversion themselves

plh: we need to have the timeline at TPAC
... And if mark is right that we don't have all features implemented then the problem is bigger - the timetable is very fuzzy
... It is NOT clear how the Director would react to a request for 6 month extension to charter to get better test results
... I am not picking on today's attendees that are doing the work. We need other to step up to do the testing work.

Matt: Will the uncertainty for EME impact MSE?

plh: No. It appears that we can get a MSE recommendation by the end of October.
... We need to also discuss how to do any future work on MSE.
... We need to figure out how the uncertainty of EME does not impact MSE.
... The policitical pressure on EME is increasing. Remember that I warned that we needed to get done by the end of Sept.

Mark: Do you ahve comment on where EME V2 would be done?

plh: That topic needs to wait until we have a clearer story on EME V1

Mark: Maybe the story for V2 will influence people's views on how to handle lack of interop on EME V1

<scribe> Chair: IS there anyone on the call that can help with converting the outstanding 14 Google tests?

Mark: Netflix will help but we need time to get thru them all

<scribe> Chair: What is the latest date you can use to collect data for a charter extension?

plh: Sep 27
... If the Director does not give a charter extension he might tell the WG to publish EME as a WG Note

paulc: If EME was published as a WG Note could it be published with a license that would permit others to use the content?

plh: That could be considered.
... Patent commitments only apply to Recommendations and therefore any patent obligations on EME would be released

David: We may need to engage Firefox on the Key Status tests, especially for Clear Key since Edge does not support Key Status
... The Keystatus tests were working on Chrome

<scribe> Chair: We need to have a clear list of work tiems for EME testing

plh: It is possible to add people to the owner file for just EME. Just let plh know. You will get pinged by GitHub when there is a new PR.
... You may need to manually tell Github that you want futher updates on the PR

David: You will be added as a reviewer to new PRs if you are in the owners file.

Jerry: I will publish the updated results today

David: Then I will start a thread with the Editors the gap list for implementations based on a shorter less than two list.

Mark: I will give the conversion of tests updates as well

Summary of Action Items

Summary of Resolutions

    [End of minutes]

    Minutes formatted by David Booth's scribe.perl version 1.147 (CVS log)
    $Date: 2016/09/13 20:22:49 $