HTML Accessibility Task Force Teleconference

02 Feb 2012


See also: IRC log


MichaelC, LeonieW


<trackbot> Date: 02 February 2012

<janina> Meeting: HTML-A11Y Task Force Teleconference

<scribe> scribenick: LeonieW

Issue-164 Group http://www.w3.org/html/wg/tracker/issues/164

SF: When you have multiple h1 in an hgroup, only the first goes into the document outline.
... Within the ARIA spec it says that what hgroup does is to take the headings, and the hgroup itself becomes the level of the highest heading contained within it.
... In the document outline it's removed, it remains in the document itself though.
... You end up with a heading and a sub heading visually, but semantically you lose that information about the sub heading.

JS: Steve, you raised a change proposal, and since then other proposals have been submitted.

SF: Yes, several change proposals out there.
... When reviewing the real world examples I noticed that quite often hgroup wasn't used properly.
... My change proposal proposes the idea of a sub line.
... I have another proposal out there that recommends an attribute for removing the sub heading from the document outline.

JS: You have two proposals?

<MichaelC> scribe: MichaelC

jf: more important to pick up relationship

don't care about element

js: to recap, care about semantic value, not about the mechanism to provide

cs: info still in the spec, right?

sf: yes

cs: so address in API mapping?

sf: semantics of hgroup defined in HTML 5, including appropriate role

it removes the "headingness" of the two headings and makes them one

if the mapping document contradicts, need to update HTML

but prefer to take advantage of the new element

in a semantically useful manner

rather than hack around it

js: one CP could be to remove element

sf: it is already present

js: seems sense of TF is we want to preserve semantics; hgroup is problematic in that regard; don't care about mechanism

cs: while we want to preserve semantics, I don't want to die on sword for this one

jf: +1

<LeonieW> scribenick: LeonieW

<MichaelC> scribe: LeonieW

SF: We'll find there is momentum there, and if there's going to be change, we should have a position on that.

JB: If we could agree whether we think this change is relevant, that would seem straightforward and sufficient.

CS@ Not going to fight over it, but think we need to look at our priorities.

JS: Hearing no objections.

Issue-193 Bad CSS Example http://www.w3.org/html/wg/tracker/issues/193

CS: Suggest a reorg of the agenda to get to canvas?

<Stevef> http://www.w3.org/html/wg/wiki/ChangeProposals/outlinenone

SF: Change proposal is out there. Has plenty of evidence.

JS: Would be good if everyone could look at this CP over the next week so we can move to a resolution on it next week.
... Issue 193, worth noting it contradicts WCAG.

SF: Thanks. That's already in there.

Issue-194 Media Full Transcript http://www.w3.org/html/wg/tracker/issues/194

MC: Think it's fairly straight forward. The section on ARIA needs expanding.
... Planning on doing the substantial part of this tomorrow.

RS: Did you put in my info about role?

SF: Not yet.

RS: Did you see the post Michael?

MC: Yes.

SF: Talking about bringing info from V1 and V2 together.

MC: Believe this will address 11893. 11892 seems to be a duplicate though.
... 11891 will be caught.
... Think one CP will do here.

JS: Questions?

SF: What we don't want to have is a situation where HTML5 is locked into not allowing new roles as ARIA evolves.

MC: Yes. That's something I hope defining the role attribute within the HTML5 spec will help with.

RS: We need to be able to update ARIA without changing the HTML5 spec.

MC: The issue with saying that roles are recognised from the ARIA spec, it's dependent on user agents processing them.

RS: Could we talk about patterns?

MC: Trying to follow the ARIA spec. If you don't recognise the token, you go on until you find one you do. The API mapping doc will then be the source for the tokens you recognise.

JS: Sounds like we're on track. Please look out for the CP to review when it's ready.

Issue-199 ARIA Attributes Processing http://www.w3.org/html/wg/tracker/issues/199

Issue-201 Canvas Fallback http://www.w3.org/html/wg/tracker/issues/201

FO: If you could send comments or proposed edits on the CP it would be useful

JB: Any issues to discuss on the call?

FO: No, just to make everyone aware of it.

JF: Could we get an executive summary of the proposal?

FO: Proposal is to complete the sub dom idea.

<richardschwerdtfe> http://www.w3.org/wiki/Canvas_hit_testing

FO: The change defines regions on the canvas that relate to the shadow dom. If you interact with the bitmap it should fire the event through. So the bitmap is a proxy for the element underneath it.

JF: This sounds functionally similar to a client side image map?

FO: Somewhat in concept. Imagemaps have other issues.

JF: Right. In the sense I'm clicking on a picture but activating something underneath it?

FO: Yes.
... There isn't a mechanism in canvas at the moment to deal with multiple controls. Anything more involved than a single interaction isn't catered for.

RS: It allows you to do hit testing on canvas, and also to provide a magnifier.

JF: What happens when your hit region is animated?

FO: This would solve that problem too. The hot spot and the animation can be dealt with separately, but brought together.

RS: The last thing drawn is the one that gets the hit.

JS: Sounds like there's general agreement here.

RS: Believe this has to be submitted by the 11th?
... Need to get a resolution on this next Thursday.

Issue-202 Fig-Caption Word Count http://www.w3.org/html/wg/tracker/issues/202

JS: Will we end up with a proposal?

JB: My aim is to have a proposal before the sub team meeting on Tuesday and review it then, before forwarding to the TF next Thursday.

Issue-203 Media Element Text Description http://www.w3.org/html/wg/tracker/issues/203

JS: We need to keep this alive, but we need issue 130 to be dealth with first?

JF: In issue 203, you're right. It's related to longdesc in some ways. If we could use longdesc on media elements it would be one solution. It's also related to issue 142 (poster alt).
... As we discussed at TPAC, the solution may lie outside of HTML in ARIA 1.1.

RS: It's in for ARIA 1.1.

JF: There was positive feedback from Frank at MS and Jonas from Mozilla.

Issue-199 ARIA Attributes Processing http://www.w3.org/html/wg/tracker/issues/199

Issue-201 Canvas Fallback http://www.w3.org/html/wg/tracker/issues/201

Issue 193.

JF: The proposal is to introduce an attribute.

CS: Why is an attribute needed? Wouldn't a link by the media element work?

JF: It isn't as clean if a link is used.

CS: There are pages with media and plenty of other links.

JF: What about if the video is viewed full screen?

CS: They close the window and follow the link.

JF: It isn't a hill to die for, but adding an attribute like this seems fairly innocuous.

<Judy> Text sub-team met Tuesday 20120131; minutes posted http://www.w3.org/2012/01/31-text-minutes

<Judy> - Meta name generator: confirmed edits to CP requested during the TF review last week. Poll to list is pending from Janina.

<Judy> - Title: discussed multiple edits to CP to clarify presentation of rationale and supporting evidence. Those edits pending from Steve. TF subteam will review them next Tuesday then bring to TF.

<Judy> - WG location of alt guidance: discussed several questions regarding primary focus of CP in progress by Michael, and main points for rationale. Updated version of that is pending from Michael.

s/Topic: Issue 193/Topic: Issue 194/

<Judy> - Timing is tight on some of these that need to go from Text subteam to TF to polls.

<Judy> - Our next scheduled meeting is next Tuesday, 1pm, US Eastern. All welcome.

CS: We were working on editorial stuff within the API mappings group.

RS: Nothing to report from the canvas sub team.

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.136 (CVS log)
$Date: 2012/02/02 17:01:56 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.136  of Date: 2011/05/12 12:01:43  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/that would be good/that would seem straightforward and sufficient/
FAILED: s/Topic: Issue 193/Topic: Issue 194/
Found ScribeNick: LeonieW
Found Scribe: MichaelC
Inferring ScribeNick: MichaelC
Found ScribeNick: LeonieW
Found Scribe: LeonieW
Inferring ScribeNick: LeonieW
Scribes: MichaelC, LeonieW
ScribeNicks: LeonieW, MichaelC

WARNING: No "Present: ... " found!
Possibly Present: CS Cooper Cynthia_Shelly FO Frank_Olivier IPcaller JB JS Janina_Sajka John_Foliot Judy LeonieW MC MichaelC Microsoft Mike MikeSmith P0 P7 RS Rich SF Steve_Faulkner Stevef aaaa aabb html-a11y inserted janina jf joined onie_Watson richardschwerdtfe scribenick trackbot
You can indicate people for the Present list like this:
        <dbooth> Present: dbooth jonathan mary
        <dbooth> Present+ amy

Regrets: Laura_Carlson
Agenda: http://lists.w3.org/Archives/Public/public-html-a11y/2012Feb/0020.html
Found Date: 02 Feb 2012
Guessing minutes URL: http://www.w3.org/2012/02/02-html-a11y-minutes.html
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.

[End of scribe.perl diagnostic output]