HTML Media Task Force Teleconference -- 04 Feb 2014

<trackbot> Date: 04 February 2014

<scribe> scribenick: joesteele

Role Call

<paulc> Agenda: http://lists.w3.org/Archives/Public/public-html-media/2014Feb/0000.html

Heartbeat

paulc: Heartbeat CFC sent out

<paulc> Copy of CfC on public-html-admin@w3.org: http://lists.w3.org/Archives/Public/public-html-media/2014Feb/0002.html

paulc: please respond to the version on the admin list if you respond
... possible that CFC may take a week to get published even if it closes this week

EME Bug Status

paulc: in Adrians original order -- some that were not in the original list though
... some are in editors queue already, we will skip those
... need to check status on others
... some need discussion, would like to know how we will process.
... generate a deadline if possible

Bug 17673

<paulc> https://www.w3.org/Bugs/Public/show_bug.cgi?id=17673

paulc: minutes indicate David was going to split off isTypeSupported and talk to David Singer -- any progress?

ddorwin: no progress this week

adrianba: sent mail to David Singer, no reply yet. Will follow up

<paulc> ddorwin: will try to do this this week

Bug 17750

https://www.w3.org/Bugs/Public/show_bug.cgi?id=17750

paulc: don't think we touched on it last week
... last entry is a question from David on Jan 25th -- no responses since then

<paulc> David's question: https://www.w3.org/Bugs/Public/show_bug.cgi?id=17750#c25

paulc: David any thoughts?

ddorwin: reading ...
... believe the issue is that we have state checks but the checks are all done synchronously

paulc: any questions or comments?

adrianba: haven't looked at all the cases in detail, but suspect that David is right about needing to include some language in the check -- in the release case think it might be ok
... since the close can happen at any time without the app doing anything
... would be unfortunate if the app tried to call release but the CDM had already closed the session and generated an error because of this
... could generate a no-op in this case
... might need just moving those checks

paulc: more input needed?

ddorwin: yes -- more comments in the bug

Bug 21798

<adrianba> 17750 is a tricky one and will need careful analysis

https://www.w3.org/Bugs/Public/show_bug.cgi?id=21798

paulc: bugs says that the constants were reviewed, need to add them, how will that happen? what are you waiting for?

ddorwin: just a lower priority - need to go thtough whole thread and generate a response

Bug 23619

https://www.w3.org/Bugs/Public/show_bug.cgi?id=23619

paulc: last week you said you needed more time Adrian to consider comment #9
... no reply as yet

<adrianba> i need to discuss with travis

<paulc> https://www.w3.org/Bugs/Public/show_bug.cgi?id=23619#c9

paulc: can it be done in a week?

adrianba: think so

Bug 24025, 24082

https://www.w3.org/Bugs/Public/show_bug.cgi?id=24025, https://www.w3.org/Bugs/Public/show_bug.cgi?id=24082

paulc: linked in the agenda
... mark you mentioned these

markw: think it was use cases for extensibility

<paulc> Mark's use case for 24025: http://lists.w3.org/Archives/Public/public-html-media/2014Feb/0001.htmlhttp://lists.w3.org/Archives/Public/public-html-media/2014Feb/0001.html

markw: /msg adrianba thanks!

<paulc> http://lists.w3.org/Archives/Public/public-html-media/2014Feb/0001.html

paulc: please add your comments to 24025
... any other responses?

paulc: if you need time to review -- indicate you are not sure

Bug 24027

https://www.w3.org/Bugs/Public/show_bug.cgi?id=24027

paulc: Adrian suggested John could review Pavels proposal

johnsim: think I agree with the proposal -- general approach looks right
... as long as there is support for multiple key ids

<paulc> Pavel's support: https://www.w3.org/Bugs/Public/show_bug.cgi?id=24027#c2

<paulc> Mark's comments: https://www.w3.org/Bugs/Public/show_bug.cgi?id=24027#c4

<glenn> +1

markw: we have found it useful to include a hash of the key to allow the client to quickly determine it has the right key
... decoder should be robust, but might have lost the context by the time you hit a failure

ddorwin: when would you get the wrong key?

scribe: still need to come up with a format

markw: misconfiguration, confusion about which file goes with which content
... could be cases where an attacker supplies the wrong key

joesteele: we have seen misconfiguration lead to this issue

s/miconfirguration/misconfiguration/

paulc: closed the action
... heard David say we need more data in the proposal -- mark can you propose an extension

<markw_> yes - I will make a proposal

Bug 24216

https://www.w3.org/Bugs/Public/show_bug.cgi?id=24216

paulc: Last week we said Jerry was looking at this

adrianba: he has been delayed
... travelling
... delayed by the SuperBowl
... parade in downtown Seattle tomorrow so may be some delay there as well

paulc: on the agenda for next week

Bug 24323

https://www.w3.org/Bugs/Public/show_bug.cgi?id=24323

paulc: need feedback from TF -- have seen none?

ddorwin: boils down to whether we need to fire a needkey if initData was not provided

<ddorwin> "This allows something other than Initialization Data to trigger this algorithm."

joesteele: this would impact us -- I will take a look

Bug 24368

https://www.w3.org/Bugs/Public/show_bug.cgi?id=24368

paulc: Jerry was to review -- assume this moves to next week

Bug 24381

https://www.w3.org/Bugs/Public/show_bug.cgi?id=24381

paulc: Adrian you thought this had been reported before -- have you found the dup?

adrianba: no time last week

Bug 24419

https://www.w3.org/Bugs/Public/show_bug.cgi?id=24419

paulc: not discussed last week at all

ddorwin: currently keymessage is some undefined value extracted from initData, proposal is to make a JSON based license format
... this is a mirror of the license done in JSON WebKey

<markw_> you mean the keymessage, not needkey, right ?

paulc: on your queue to implement?

joesteele: that sounds right -- thanks mark

s/needKey is/keyMessage is/

paulc: any comments on this one?

<paulc> s/17682 is on your queue to implement/17682 is on queue to implement/

<ddorwin> yes, I meant the [key]message event.

paulc: what do you need David? is there a question?

ddorwin: if noone objects I will implement

<adrianba> we haven't discussed it

paulc: any objections?
... gone through the items on my list -- skipped a few
... scan through agenda item#5 for other items to discuss
... next week will break agenda item #5 into 2-3 parts
... low priority, items we have discussed, maybe one other category
... encourage folks to work on these items so we can make progress

adrianba: there is a lot of work to do at the moment -- long queue
... how do people feel about the cadence of our weekly meetings? is that effective? or move back to two weeks?

paulc: I was thinking about this also -- I tend to agree. Not seeing a lot of progress made

+1 -- I could use this hour for reviewing

<ddorwin> +1

paulc: might be tempted to assign very explicit action items to make sure folks know what they are working on
... motion from the floor for next meeting to be Feb 18th. Any objections?
... thinkg I have some notes that I can send out some prompts to people about actions resulting from todays discussion
... I will make sure there is something in the email archive about items waited on as well
... maybe this will help
... thanks for that suggestion Adrian
... any other comments?
... let's adjourn

<scribe> Agenda: http://lists.w3.org/Archives/Public/public-html-media/2014Feb/0000.html

s/\/msg //

- DRAFT -

HTML Media Task Force Teleconference

04 Feb 2014

Attendees

Contents