Minutes from HTML-A11Y Media Subteam, February 9

Minutes from the HTML-A11Y Task Force's Media Subteam teleconference are
provided below as text. They're also available as html at:
http://www.w3.org/2011/02/09-html-a11y-minutes.html

   W3C

                                                           - DRAFT -

                                                       HTML-A11Y telecon

09 Feb 2011

   See also: IRC log

Attendees

   Present
          Silvia, John, janina, Eric, Judy

   Regrets
          Sean, Geoff

   Chair
          John_Foliot

   Scribe
          janina

Contents

     * Topics
         1. Identify Scribe
         2. Actions Review http://www.w3.org/WAI/PF/HTML/track/actions/open
         3. Issue-152 Multitrack API http://lists.w3.org/Archives/Public/public-html-a11y/2011Feb/0079.html
         4. Time Tracks Feedback from Google http://lists.w3.org/Archives/Public/public-html-a11y/2011Jan/0152.html
         5. Are we done with Time Tracks?
         6. Poster Issue http://lists.w3.org/Archives/Public/public-html-a11y/2010Dec/0054.html
     * Summary of Action Items
     __________________________________________________________________________________________________________________

   <scribe> agenda: this

Identify Scribe

   <scribe> scribe: janina

Actions Review http://www.w3.org/WAI/PF/HTML/track/actions/open

   <silvia> action-88?

   <trackbot> ACTION-88 -- Sean Hayes to review Media Fragment URI 1.0 http://www.w3.org/TR/2010/WD-media-frags-20100624/
   -- due 2010-11-24 -- OPEN

   <trackbot> http://www.w3.org/WAI/PF/HTML/track/actions/88

   <JF> action-96

   <silvia> clost action-88

   <silvia> close action-88

   <trackbot> ACTION-88 Review Media Fragment URI 1.0 http://www.w3.org/TR/2010/WD-media-frags-20100624/ closed

   <silvia> action-96?

   <trackbot> ACTION-96 -- Eric Carlson to media Sub Team to revisit bug 11395 (Use media queries to select appropriate
   <track> elements) -- due 2011-01-06 -- OPEN

   <trackbot> http://www.w3.org/WAI/PF/HTML/track/actions/96

   close action-96

   <trackbot> ACTION-96 Media Sub Team to revisit bug 11395 (Use media queries to select appropriate <track> elements)
   closed

   <silvia> Dave and Eric decided it would be too complex to extend media queries for this purpose

   <silvia> action-99?

   <trackbot> ACTION-99 -- Janina Sajka to annotate 9452 with clear audio discovery and selection, as well as independent
   control of multiple playback tracks -- due 2011-01-19 -- OPEN

   <trackbot> http://www.w3.org/WAI/PF/HTML/track/actions/99

Issue-152 Multitrack API http://lists.w3.org/Archives/Public/public-html-a11y/2011Feb/0079.html

   Silvia: WG wants change proposals by Feb 21
   ... Has a proposal, asking for feedback

   <silvia> http://www.w3.org/WAI/PF/HTML/wiki/Media_Multitrack_Media_API

   Silvia: I prefer solution #1
   ... People should read the wiki and indicate their preferences
   ... Also a proposal that defines a position on screen for the element (which can be changed with CSS)
   ... Microsoft had been for option #2
   ... We need continued discussion, as different people prefer different solutions

   Eric: Will do so later today
   ... Favoring option #2 with mod of having src element inside track to accomodate different encodings
   ... I've pinged Frank, but not heard back yet

   John: Asking about #7
   ... is it correct that sign video track would be positioned using css?
   ... PIP might be too smal on handhel, no?
   ... Wereas, if independent, could do better sizing

   Silvia: Yes, that's an advantage of #7
   ... These are some of the points it would be good to see on list

   Eric: Why can't that also via track element?

   Silvia: Would imply too fundamental changes, currently track only renders on viewport and nowhere else on page
   ... Track currently can't have children
   ... Don't know if that's open to mod in the WG?

   Eric: Suspect we'll discuss any of these in the WG, even though it's late in the timeline, it's just been postponed

   John: Didn't we identify a user req to position anywhere?

   Eric: Yes, but not possible with spec as it is now

   Silvia: Would be through js
   ... So, possible but not simple
   ... Proposing to widden the discussion

   Judy: Proposing it should be on the W3C list
   ... It's a critical piece of getting a11y addressed in W3C, so wouldn't make sense to not have it on W3C

   Silvia: Problem is I've had no response on the W3C list

   Judy: So, we should figure how to get the discussion going

   Silvia: No reason to take it off, but should not be a problem widdening the discussion
   ... Just wonder if it makes sense to post to the WG list

   Judy: Maybe that's why there's no response yet.

   John: My concern is to avoid multiple discussions, too many gotchya possibilities.

   Silvia: I'll keep track and report, but I want more opinions.

   Eric: Agree, this topic is too important
   ... We had this discussion sometime ago, and it's not progressing.

   Judy: Thought the reason it's been silent is that more work was anticipated? Not so?

   Silvia: The "More work part" is more discussion.
   ... Don't want a solution that's had too little vetting.
   ... I don't mean anything official--just to communicate what we're considering.

   Judy: Important point is that we need to move forward and have a wider discussion

   John: Do we revisit this next week? With the sense of a decision from the TF?

   Silvia: Makes sense.

   John: OK, any more on this?

   Judy: One question: Given discussion is being raised in the broader group, has there been any feedback that we should
   chase down?

   Janina: Yes, we should solicit feedback from a11y people with experience on this, NCAM, DAISY, etc

   John: More on this?

Time Tracks Feedback from Google http://lists.w3.org/Archives/Public/public-html-a11y/2011Jan/0152.html

   John: An executive overview?

   Silvia: So, I'm currently contracted to Google and have been working on this ...
   ... Teams from Chrome, YouTube, a11y, looking at WebSRT, track element
   ... Checked out 608 and 708 (used in U.S.)
   ... We're happy about a name change from WebSRT to WebVTT, in good part because SRT has negative reputation around
   copyright infringement

   Judy: Question: When looking at U.S. reqs, were you looking at the FCC reqs?

   Silvia: At Google's reqs
   ... The two existing standards for captions on TV
   ... These have limited set of features, insufficient for our reqs
   ... But, we want to replicate everything in 608 and 708

   Judy: Yes, but my question is different ...
   ... Since you're looking at U.S. reqs, did you consider the FCC VPAAC reqs. You're aware of VPAAC?

   Silvia: Unaware they've produced a req doc?
   ... Understand only a general req

   Eric: General req to carry captions in current broadcast video when broadcast over Internet
   ... Sounds like they have basis to believe it will meet reqs from VPAAC

   Judy: There are also issues about emergency crawls, voicing of those, etc. including on the web
   ... The reqs aren't mapped in the statute, but my impression is that there will be more reqs than previous practice

   Silvia: So, I guess the answer is: "No, we didn't look at that."

   Eric: Google is on the committee, yes?

   Judy: Yes, but the committee's barely started. Only one meeting so far, and there's not yet been an opportunity to get
   a fuller understanding
   ... Geoff did post his understanding that FCC would not mandate a protocol for this
   ... There may be some clarification to that, as the charge for VPAAC is to produce guidance

   Silvia: So, the email summarizes our results ...
   ... Discusses gaps on WebVTT -- also what we want to see improved
   ... I'm currently working on a js implementation for all this
   ... Think we're currently converging on changes needed to WebVTT, and they're not very large

   John: Anything we need to consider?

   Silvia: Don't think so--if any questions, happy to involve everyone in a discussion

   John: Duplicate track? Not sure what the answer should be? Is there?

   Silvia: Philip responded duplicate is same lang and same type format; answer just display both in the menu

   Janina: Because independent alternate media authors may have produced a second version, same lang, same type

   Eric: Correct, but spec says only one, and that's guidance for authors
   ... Text is just guidance for authors, so from that perspective it's reasonable

   John: Should we seek better spec?
   ... To specifically say that both should be made available in the menu?

   Silvia: Sounds like a reasonable bug, and makes it easier to conform across browsers

   John: Just thinking of making things as robust as possible
   ... Might as well get it into the spec, rather than by precedent, because there could be not so great precedent

   Janina: Agree

   John: I'll file

   <JF> silvia, are you still on IRC?

   John: Anything else?

   Silvia: Would ask people to read through our results and respond with their thoughts.

Are we done with Time Tracks?

   John: Anything more we should say?

   Judy: Let me try ... On the broader question
   ... The question remains a concern. There may be no other way, and considering the impace downstream is important ...
   ... Is there a point for this group to comment?

   John: That's the question.
   ... Market forces will decide what each browser does ...

   Judy: But, there's also continued discussion re our reqs

   Eric: I have concerns with SMPTE-TT, now that I've read the spec, from an a11y perspective, specifically with
   background image handling

   Judy: Agree there are things to look at there
   ... My understanding is that VPAAC won't mandate, but will comment on appropriatness of various options

   <silvia> +q

   Judy: Understand there are strong leanings on the part of some stakeholders
   ... We're also looking at this in W3C
   ... Curious to learn more about the background image issue
   ... W3C needs to be responsive to the entire field--all stakeholders

   Eric: Not sure that background is necessarily harmful to a11y, but analgous to CSS background-image--and we should have
   a discussion

   Judy: Agree we should understand it better. Let's do

   Eric: Happy to do so

   Silvia: Done a prelim on SYMPTE TT; agree with Eric, unclear what we get if we use it
   ... If I understand correctly, one key purpose is to get legacy content onto the web with captions, using Internet as a
   transport, not necessarily as web content
   ... It's an exchange format, so makes sense to use for encapsulating and transporting; But that doesn't necessarily
   imply presentation

   Judy: Curious to explore something on this ...
   ... Understand that's the basis of their approach, have heard this elsewhere as well
   ... Not as convinced that some of the broadcast people aren't also looking at using it on the web
   ... Are you certain that there aren't already entities forseeing use of TT for web delivery?
   ... Don't think this answers what W3C should do, just trying to clarify our understanding of where people are coming
   from

   Silvia: I was trying to understand how the SYMPTE standard came to be
   ... e.g. there aqre binary captions in legacy content
   ... Transformational and transport formats aren't necessarily the best choices for web presentation
   ... We have to deal with the converging world, and we have these two groups coming from different perspectives
   ... Don't know if we can consolidate the two
   ... May be delivered across the net, but not delivered via a browser

   Judy: Very helpful, Silvia. I appreciate your perspective.

   Janina: Suspect browser will emerge because there are also a11y reqs on the user interface now

   John: Or plugins, which I expect

   Judy: Several people are wondering about convergence possibilities

   John: With a few minutes left, I want to recap ...
   ... Judy, you mentioned a two week timeline?

   Judy: W3C is participating on the VPAAC, we're there to be helpful and explain what we're working on, but we don't have
   a position
   ... We'd like to bring info as up to date and as useful as possible, so very interested in our analysis of SYMPTE TT vs
   W3C TTML
   ... Then the a11y relevance
   ... So, the possibility of convergence before many years go by with people working in different formats, that's an
   important question.

   John: Specifically we've a pressing timeline for multitrack api, just thinking in terms of time alltoment here

   Judy: Don't have a good answer right now.
   ... Perhaps 20 minutes on the SMPTE next week? Three of us have looked at it already?

   <Judy> s/SYMPI/SMPT/

   Silvia: Suggesting Eric and I look more closely at SMPTE-TT, and also Judy, so we should discuss it

   John: Just concerned we conclude on multitrack

   akim, next item

Poster Issue http://lists.w3.org/Archives/Public/public-html-a11y/2010Dec/0054.html

   John: Nothing to add ...

   Janina: Pf interested that we have screenshots to mark up for the two proposals, that we not leave this to a handwave

   rrsagent make minutes

Summary of Action Items

   [End of minutes]
     __________________________________________________________________________________________________________________
Found Scribe: janina
Regrets: Sean, Geoff


-- 

Janina Sajka,	Phone:	+1.443.300.2200
		sip:janina@asterisk.rednote.net

Chair, Open Accessibility	janina@a11y.org	
Linux Foundation		http://a11y.org

Chair, Protocols & Formats
Web Accessibility Initiative	http://www.w3.org/wai/pf
World Wide Web Consortium (W3C)

Received on Thursday, 10 February 2011 00:09:46 UTC