HTML Media Task Force Teleconference -- 20 Oct 2015

<trackbot> Date: 20 October 2015

<scribe> scribenick: joesteele

ISSUE-85 TAG Discussion

<markw> The TAG issue on this as no updates (https://github.com/w3ctag/spec-reviews/issues/73)

paulc: sent a note to Travis yesterday and he said he would come today, but nothing written yet so we asked him to update their issue-73 or our issue-85 instead

<paulc> https://github.com/w3c/encrypted-media/issues/85

… will put this on the F2F agenda if no progress

… not sure whether they will be in Sapporo

… but something written would be preferable

+1 from the crowd

ddorwin: +1

paulc: ok will keep pushing

… see what we can do before then

Media TF F2F

<paulc> See http://lists.w3.org/Archives/Public/public-html-media/2015Oct/0052.html

paulc: as co-chair — I sent a message to all folks who had registered

… over 75 observers

… some groups have been re-labeled

effectively only Media TF is meeting not HTML WG

… I will put the archive in the minute

plh: for folks not attending in person, we are not registered

<paulc> Draft agenda: http://www.w3.org/wiki/HTML/wg/2015-10-Agenda

paulc: working on draft Agenda for F2F but no progress as yet

… spending time working on MSE and EME issues (they are doing the work)

<paulc> (under construction)

… lots of stuff being done via email

… lots of progress being made

<paulc> Proposal: Draft plan is for MSE to be discussed on Thu Oct 29 and EME on Fri Oct 30.

… one of the main items to confirm with EME folks — this proposal will be discussed on Thursday and Friday

s/Froday/Friday/

paulc: 16 hours time difference — so remote attendance on Friday is difficult

ddorwin: I can try to participate on Friday — see what the agenda is

paulc: Matt asked previously what the distributed meeting logistics are — don’t know yet

… have moved away from Zakim to CIsco/Polycom and will be done via the web

… have not seen the arrangements yet

paulc: stay tuned. I will email folks I expect to be remote directly

… any other questions on this?

updating MSE/EME issue tags and milestones

<paulc> http://lists.w3.org/Archives/Public/public-html-media/2015Oct/0056.html

paulc: email from Matt is our best guide

MattWolenetz: look at the email I sent

… for MSE it was pretty simple, not too many bugs so may not have used all the labels

… some mis-interpretation last week, but moving past that

… needs clarification can be a bug filer not the web author

… e.g. providing justification etc.

scribe: “needs followup” is a msg to the editors (from themselves) to investigate further

… “needs implementation” means editors have agreed and just need to make the pull request

<paulc> Are the Editors planning to remove the tag “needs implementation” off of issues when they are actually implemented and closed?

MattWolenetz: if bug is closed that would seem appropriate — David?

ddorwin: yes we can take it off. Should show up in another list though

jdsmith: whether or not we strip it off will not show up in the normal view

paulc: when looking for all issues, some closed issues had “needs implementation” set — that seems wierd

… would seem good to take this off

pal: maybe that should be a separate issue

jdsmith: I have no objection to clearing those irrelevant flags

ddorwin: we would need to clean up 22 of them.

… “needs implementation” refers to spec and not the agent implementation

… need another tag here?

paulc: so editors can use that as their work queue

MattWolenetz: agree with the discussion and the point about the agent implementation

… “blocked” means an external issue or another github issue

… “feature request” refers to a new use case not covered by the current spec

… this may be assigned to a future milestone

… “interoperability” refers to a lot of issues but meant as a severity flag — e.g. known incompatibilities, breaking stuff

… spec needs works here

… “wont fix”, “invalid”, “dup” these are things that should be closed

… Milestones should be followed with a digit — had some discussion re: v.Next as a version

… alphabetical sort could be an issue

… means issue is not in-scope of the current spec but good to track

paulc: david - anything to add?

ddorwin: no - did a lot of changing of tags and labels to match this

joesteele: was everything moved over?

paulc: 100% of MSE were moved over from bugzilla

… yesterday some of the EME issues started to be moved over

ddorwin: moved the clearly actionable ones over

<MattWolenetz> correction: s/100%/100% of MSE/

… there was one about distinctive identifiers I have not had time yet

… the rest I need to talk to the filers again

jdsmith: how many of those are there?

ddorwin: 8-10

paulc: there were 16 before

ddorwin: 2 will be moved still and 6 we need to discuss

paulc: lets get the link in there

<paulc> EME remaining Bugzilla bugs: https://www.w3.org/Bugs/Public/buglist.cgi?bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&component=Encrypted%20Media%20Extensions&list_id=60287&product=HTML%20WG&query_format=advanced

ddorwin: I will move 27168 and 27268

<paulc> https://www.w3.org/Bugs/Public/show_bug.cgi?id=27168

<paulc> https://www.w3.org/Bugs/Public/show_bug.cgi?id=27268

paulc: for the other 6, plan is to discuss?

ddorwin: jerry and I can talk about these

paulc: like to make progress on those by the F2F

… some of these you might want to go back to the original correspondent

… need help?

jdsmith: some might get closed instead of moved

ddorwin: e.g. the big “needs interop” bug — we need to track somewhere but not sure how to move

paulc: any advice?

<paulc> What to do about the scope related bug: https://www.w3.org/Bugs/Public/show_bug.cgi?id=20944 about CDM interop?

plh: I agree with David, we have to track this somewhere. Better to track as a bug but maybe with a flag saying not a blocker

… that flag may be controversial

paulc: david you are making great progress

Review of tags on 19 EME issues (open and NOT "needs implementation")

paulc: that number could be wrong now

<paulc> https://github.com/w3c/encrypted-media/issues?utf8=%E2%9C%93&q=is%3Aopen+is%3Aissue+-label%3A%22needs+implementation%22

paulc: link takes you to issue search

… these do not have “needs implementation” yet

… some have

… “help wanted” or “needs followup”

… how are we communicating this out to folks on the list?

… or the appropriate people on the list

ddorwin: some folks we do not have an owner

… might not need to be an editor

… almost like “to be implemented”

paulc: I have some on the agenda, since we are clear on what the status is now — lets drop into that part of the agenda
... still a huge backlog of “to be implemented” — 18 last night

… link coming

<paulc> https://github.com/w3c/encrypted-media/labels/needs%20implementation

paulc: think this shows only open issues by default

ddorwin: searches sometimes include pull requests as well — watch for that

ISSUE-98 - Decide on ideal "waitingforkey" event behavior when update() doesn't resume playback

https://github.com/w3c/encrypted-media/issues/98

paulc: agenda has short status

… needs author input and feedback

… David, who should we be targeting

ddorwin: I use this when it affects the authors — they need to pay attention to this

… could send an email on this

… this is about how they want the UA to behave

… I have a suggestion in there already

<paulc> David's change proposal is here: https://github.com/w3c/encrypted-media/issues/98#issuecomment-147801649

paulc: think we need the feedback on this — possibly resolve before F2F is not put on the agenda there

ISSUE-99 - Remove note recommending setMediaKeys() be called before providing media data

https://github.com/w3c/encrypted-media/issues/99

paulc: marked as “needs review”

ddorwin: I have a comment in 8 on this — not sure why this was added

… think we should remove this

… will assign to Jerry to comment

jdsmith: I will make sure we have feedback

ISSUE-100 - Is "running the Encrypted Block Encountered algorithm" the correct way to Attempt to Resume Playback If Necessary?

https://github.com/w3c/encrypted-media/issues/100

ddorwin: this is really spec input — please review

… probably does not impact interop — discovered via another bug

paulc: anyone want to take this on?

joesteele: I will put this to our internal team as well as anyone else

… try to get some feedback

markw: I am happy to look at it

ISSUE-101 - Normatively require distinctive identifiers to be different by top-level and EME-using origin

https://github.com/w3c/encrypted-media/issues/101

https://www.w3.org/Bugs/Public/show_bug.cgi?id=27269

ddorwin: next 6 or so are brought over from bugzilla

… this one is on me

ISSUE-102 - Define what to do when CDM becomes unavailable

https://github.com/w3c/encrypted-media/issues/101

https://www.w3.org/Bugs/Public/show_bug.cgi?id=27067

really this is : https://github.com/w3c/encrypted-media/issues/102

ddorwin: not assigned yet

… probably low priority

joesteele: I will ask Chris Pearce for more feedback

paulc: make sure he is aware where where the conversation has moved

EME Initialization Data Correlation

https://lists.w3.org/Archives/Public/public-html-media/2015Oct/0020.html

https://lists.w3.org/Archives/Public/public-html-media/2015Oct/0022.html

paulc: original emails are linked — raised by DASH.JS work

… somebody responded, but no responses from the editors yet

… how should we process?

… long responses

… David or Jerry have you looked?

paulc: want to make sure someone is on point to answer

… and whether the answer makes sense

jdsmith: discussed with our media team

ddorwin: raises some issues we have discussed before and avoided

… causes some architectural issues — Jerry can take this on

… long discussion

ddorwin: looks to be about non-identical initData and deducing what to do

… we have discussed

… other issues also

jdsmith: are those issues related?

paulc: one is a reply to the other

ISSUES 103 through 110

paulc: look at these at a high level

plh: there is a general issue on dependencies in the specification — this is touching on the WebIDL

… there are two versions of WebIDL today

… the Web Platform WG has not decided what they want to do with theirs

… and there is a new one — with a reduced set of features which they believe can be moved to recommendation

<paulc> ISSUE-107: https://github.com/w3c/encrypted-media/issues/107

<trackbot> Notes added to ISSUE-107 .

… Web IDL Level 1

… they mentioned a few issues but think they are in WebIDL Level 1 — when we move to CR can we trust these dependencies

… if we move too early they may have moved and we need to recreate dependencies again and again

… my recommendation is to not worry about this yet, until we know one will be a problem

paulc: at what point should we do this? we have a LC-CR bug already

plh: I am talking about the move to PR

… this is even later

… we should not ait to the last minute, but if EME draft is not stable enough yet we should wait, maybe David can tell us?

ddorwin: other than “iterable” these are pretty stable, we can correct the issues now or wait until PR

plh: I would not worry for intermediate drafts, but for the move to rec and v.Next we need to do this

… we only burn the references once we move to PR

paulc: If iterable is the only item and add a note that this is the only known feature not in WebIDL Level 1

ddorwin: I was saying whether we use is still up in the air

plh: think this is not a blocked for moving to CR
... just wanted to mark as something we need to handle in the next 6 months
... ISSUE-107 could be assigned to me

https://github.com/w3c/encrypted-media/issues/107

paulc: leave this in your hands for now — change is pretty obvious

plh: I added a comment

paulc: 105, 106, 108 are all tagged for needs followup

ddorwin: 105 is mine — I need to draft a proposal but input would be good

https://github.com/w3c/encrypted-media/issues/105

paulc: 108 is marked also

https://github.com/w3c/encrypted-media/issues/108

paulc: looks like another reference to the JSON spec

… is this covered by the PLH? ISSUE-108?

plh: yes — another way is to have a local biblio in the draft

… feel free to assign to me

paulc: 106 has a help wanted

… as entry to TS Common Encryption

ddorwin: this is blocked on the re-org bug

… I think Bob has a draft on this?

… would be good to have that since none of the editors are experts in this

BobLund: I submitted a couple of drafts that split the spec and added a new verion for the CENC and MPEG2 transport but would like feedback on what is already proposed

paulc: you may be the principal expert here

jdsmith: we should have someone from our media team look at the draft as well

paulc: if we make progress on this in parallel with the bugs Philipe is working on we can figure out what to publish

jdsmith: is there a link to that draft?

paulc: link in the bug

… linked from comment #21

<paulc> For Bob's proposal: https://www.w3.org/Bugs/Public/show_bug.cgi?id=26738#c21

paulc: out of time — lots of other stuff to cover but I will carry forward on some of those

… Jerry you promised to implement 7-8 bugs but asked questions and you need to revisit them again

… input is available

… otherwise they will be on the F2F agenda

paulc: Thanks to Matt, Jerry and David for all the work they have done in the last few week on this

… getting everything organized and out of bugzilla

… this makes things much easier for me

plh: for making progress on 105 who is best to help me?

paulc: david

markw: if anyone want to pass “needs implementation” issues to me I can help

paulc: please self-assign as needed
... some pull requests still waiting for editor review as well
... if there are new to be implemented ones that needs disucssion let us know

markw: some of those pull requests are blocked

paulc: thanks everyone!
... will decide what the next meeting schedule is in Sapporo

s/shoud not ait/should not wait/

- DRAFT -

HTML Media Task Force Teleconference

20 Oct 2015

Attendees

Contents