HTML Accessibility Task Force Teleconference -- 15 Apr 2010

<trackbot> Date: 15 April 2010

<janina> Meeting: HTML-A11Y telecon

<janina> Chair: Janina_Sajka

<janina> agenda: this

<Marco_Ranon> hi, using skype on a slow connection though. audio a bit jumpy at my end...

<oedipus> if you didn't fill out http://www.w3.org/2002/09/wbs/44061/200404_ftf-proposals/ DO SO NOW!

<scribe> scribe: kliehm

ACTION-2

<MichaelC> action-20 due 30 May

<trackbot> ACTION-20 Dig up history on column element due date now 30 May

<oedipus> http://www.w3.org/WAI/PF/HTML/track/

<MichaelC> close action-23

<trackbot> ACTION-23 Write proposal for imagemaps on canvas closed

<oedipus> http://www.w3.org/WAI/PF/HTML/track/23

<oedipus> doh! http://www.w3.org/WAI/PF/HTML/track/actions/23

<MichaelC> close action-24

<trackbot> ACTION-24 Raise bug on ARIA being restricted to a11y closed

<MichaelC> action-25 due 22 April

<trackbot> ACTION-25 Post notice to a11y TF about http://dev.w3.org/html5/alt-techniques/ and how to make comments due date now 22 April

<oedipus> http://www.w3.org/WAI/PF/HTML/wiki/Work_Topics

<richardschwerdtfe> manana

Setting Deadlines for Open Issues

Janina: We should be setting reasonable deadlines for various issues next Thursday.

Canvas Subteam Change Proposals

One proposal has been forwarded, Rich is working on another. At the Birmingham face-2-face meeting Charles committed to write another proposal, too.

Media Subteam Update

John Foliot will set up a telcon date early next week.

<MikeSmith> I'm here

Dick Bulterman: I sent out a complementary proposal about text requirements that should be sent to the list and added in the Wiki.

<oedipus> RichS' most recent Canvas 2D Draft with a11y: http://u.nu/2u6d8 (cover letter: http://u.nu/2u6d8)

<oedipus> http://www.w3.org/WAI/PF/HTML/wiki/Media

Calls for Consensus

<oedipus> http://www.w3.org/2002/09/wbs/44061/200404_ftf-proposals/

<oedipus> http://www.w3.org/2002/09/wbs/44061/200404_ftf-proposals/results

Janina: please vote on the CfC (deadline tonight)

Janina giving an overview about the current state of votes on http://www.w3.org/2002/09/wbs/44061/200404_ftf-proposals/results

<oedipus> Candidate TF Resolution: ISSUE-30 longdesc vote total as of time of typing: 14 yes, 6 no

Steve Faulkner: Maciej had questions on the title attribute that Steve cleared.

<oedipus> Candidate TF Resolution: ISSUE-66 image-analysis vote total as of time of typing: 17 yes 2 no

<oedipus> Candidate TF Resolution: ISSUE-80 Title vote total as of time of typing: 18 yes 1 no

<oedipus> Candidate TF Resolution: ISSUES-90, 91, 93, 95, 96, & 97 vote total as of time of typing: 16 yes, 5 no

Jon Gunderson: What is the process? Advance votes where there is a consensus to the WG?

Janina: Yes.

Jon: We are dependend on browser developers to implements the elements. The spec is huge, we should consider making it less complicated by removing obsolete elements.

Cynthia: To my knowledge browser manufacturers do not oppose the elements.

Janina: Only one individual in the WG opposed them.
... The voting is open until midnight Boston time (4 AM UTC), you can still change votes.

@Alt Discussion

Janina: We looked at Laura's change proposal and it expresses with one minor change what the majority in Birmingham agreed on.

<oedipus> http://www.w3.org/2010/04/06-html-a11y-minutes.html#item08

Janina: The discussion then focused whether a missing @alt should trigger a warning or an error message in a validator.

<oedipus> GJR notes that disagreement strongest amongst those who wanted equal priority for missing @alt and missing @src in regards error/warning

<JF> mikes link: http://www.w3.org/html/check

<cyns> Is there a test file to try it with?

Cynthia: Is wondering if there is a test file that could be submitted to the validator?

Steve: Will post a URL.

<cyns> +1 to john that we need to talk about which is more important hard-error or pointer to WCAG

<Stevef> image alt test page http://www.paciellogroup.com/blog/misc/HTML5/test.html

<richardschwerdtfe> we are rehashing the face to face

John: I have been talking about it on the weekend. We should embrace Mike's proposal. The @alt is the holy grail for all of us, I'm proposing a pragmatic compromise. It's an error, but that doesn't solve the problem. So "warning" is a compromise we should live with.

<JF> not all were at the f2f rich

<richardschwerdtfe> i know

<richardschwerdtfe> frustrating

David Singer: People are unlikely to notice a new error because usually there are hundreds of them. We should categorize our error and warning notices in the spec.

Cyntha: Agrees surpressing or categorizing errors would be useful. We should find a consensus. Although error messages get more attention from developers than warnings.

<oedipus> GJR just wants missing @alt and @src accorded the same priority

<Zakim> cyns, you wanted to say that I think the hard-error is more important, because developers pay attention to errors and largely ignore warnings

Rich: Most applications will never see a validator.

<oedipus> cyns, MikeSmith checker results for mrtweet.com http://u.nu/7zpd8

Rich: Maybe we whould have a classification of accessibility validator errors in HTML5.

<dsinger> classification would mean that we could run validator on sites and ask *only* for accessibility errors and we'd get a useful report...

<richardschwerdtfe> agreed

BTW, Mike's missing @alt mail: http://lists.w3.org/Archives/Public/public-html-a11y/2010Apr/0165.html

<oedipus> are we talking about http://www.w3.org/html/wg/wiki/ChangeProposals/ImgElement20090126#With_Suggested_Text

<cyns> The most importan thing is for developers to know that missing alt *is wrong*, not just *might be wrong*.

<cyns> +1 to classifying errors

<JF> +1 to that Rich

Rich: We could adopt Open AJAX error classifications for HTML5.

<oedipus> plus 1 to reusing Open AJAX error classifications

<cyns> url for open ajax classifications?

<oedipus> http://www.openajax.org/index.php

Rich: Also Open AJAX classification catch dynamic content that a normal validator never catches.

HTML to ARIA mapping

<cyns> HTML to ARIA mapping

oops

<oedipus> s/TOPIC: ARIA to HTML5 mappings/TOPIC: HTML to ARIA mapping/

<oedipus> steveF, do you have the google docs spreadsheet URI handy to paste into IRC?

Cynthia: The HTML5 spec allow very little overwriting of HTML elements by ARIA. Mapping is straightforward, deciding what is allowed to overwrite is more difficult. Grouping elements now. We made significant progress, need a little more discussion.

<paulc> I have to drop off to get ready to scribe the HTML WG meeting.

<Stevef> draft aria conformance http://www.paciellogroup.com/blog/misc/HTML5/aria-html5.html

[discussion whether headings and @alt should be overwritable]

<oedipus> discussion of ARIA mapping from PFWG 2010-04-14 (member-only) http://www.w3.org/2010/04/14-pf-minutes.html#item09

Discussion continued on a telcon with Steve, Cynthia, Rich and David, schedule tbc.

<davidb> 2015

<oedipus> s/Canvas accessibility mappings/ARIA to HTML5 mapping/

They will form a new subteam.

New and open bugs

http://lists.w3.org/Archives/Public/public-html-a11y/2010Mar/0168.html

<oedipus> http://lists.w3.org/Archives/Public/public-html-a11y/2010Apr/0116.html

<oedipus> MikeSmith, didn't laura prepare a proposed CAPTCHA survey on the wiki?

John: All new bugs seem to be related to a fundamental question what we do when images are incomplete.
... 7 new bugs, most relate to text alternatives.

<oedipus> draft CAPTCHA survey (laura) http://lists.w3.org/Archives/Public/public-html-a11y/2010Mar/0436.html

Janina: vote on the CfC, check out Mike Smith's test

<oedipus> http://www.w3.org/2002/09/wbs/44061/200404_ftf-proposals/

<JF> Repeat - mikes link: http://www.w3.org/html/check

<Marco_Ranon> bye

<davidb> ciao