W3C

HTML Accessibility Task Force Teleconference

12 May 2011

See also: IRC log

Attendees

Present
Michael_Cooper, JF, Janina_Sajka, Steve_Faulkner, Judy, MikeSmith, paulc, Marco_Ranon, Cynthia_Shelly, Eric, Rich_Schwerdtfeger
Regrets
Chair
Janina_Sajka
Scribe
MichaelC

Contents


<trackbot> Date: 12 May 2011

<janina> Meeting: HTML-A11Y telecon

<janina> agenda: this

<janina> http://www.w3.org/WAI/PF/HTML/wiki/index.php?title=Scribe_List

<janina> http://www.w3.org/html/wg/wiki/ChangeProposals/EnTitle

<mranon> zackim, P7 is me

<scribe> scribe: MichaelC

<scribe> chair: Janina_Sajka

jf: multitrack mostly wrapped up

Subteam Reports: Media; Canvas; Bug Triage

== media ==

discussion ongoing about text alternatives for video element

js: request for stable URI for <missed>

also need to work on navigational concerns

== canvas ==

<nobody present>

== bug triage ==

mr: bug triage processed pending tf decision bugs this week

gave a few the a11ytf keyword, not many

gave rationales for each

no major highlights

== ARIA mapping ==

cs: first discussion about HTML API map document just before this call

lots of work to do

have identified starter points

js: that meets just before this call?

cs: yes, for now

js: need help?

cs: yes

splitting time between HTML API map, and ARIA API map

js: good, looking to publish HTML implementation guide in June

Text Subteam Issues http://www.w3.org/2011/05/09-text-minutes.html

<judy> Text Alternatives Sub-Team -- recent meeting: http://www.w3.org/2011/05/09-text-minutes.html

jb: 7-8 issues on the table

<lists too fast to scribe>

jf: re meta name = generator

<JF> http://www.w3.org/html/wg/wiki/ChangeProposal/meta_name=generator_does_not_make_missing_alt_conforming

leif has done a lot of work

creating a change proposal in wiki

still work to do

<Stevef> rich will try to join later

jb: expect work in near future?

jf: yes, but more to do in wiki

jb: role=presentation

being dealt with by bugs

cs: talked with Rich, think ok to allow null alt as equivalent to role=presentation

concern about case in which there is non-null alt plus role=presentation

author doesn't know what they're doing in this case

talked with david singer

<JF> +q

since this doesn't truly break anything, think it's a validation warning / accessibility evaluation issue

not a validator error

jb: need to check support for this?
... over to Steve for title

Leif making a case

sf: he has made a change proposal, discussion going on

jb: ready for discussion within tf?

sf: yes

jb: figcaption

would like to take off urgent track

recognize not everyone wants to do that, but many do

cs: taking off urgent track is good

jb: longdesc

were looking for a stablized proposal from Laura

kept getting additional suggestions

want to do consensus check on proposal as it stands at the moment

jb: poster alt

jf: being discussed in media subteam

alternative text for paused video

same problem from different angle

active discussion, strong divide in opinion

there is [candidate?] formal objection to chair decision on poster alt

not pursuing further at moment

jb: table summary

gregory working on, but out this week

probably taking off urgent path

jb: location of alt guidance

clear disagreement with HTML WG decision

probably will address on non-urgent basis

jb: for today: 1) alt validation title 2) longdesc

but first, bug support issue

cs: role=presentation we meant not to impact alt requirements

but after discussion we view as not a big deal

do we need to push more?

jf: if <img src=foo alt=important role=presentation> which wins?

cs: role=presentation wins

but it's author error

from a validation perspective it's allowed

might be a warning

note validator doesn't know if alt is important

changing requirements won't affect prevalence of this kind of error

topic under discussion was expanding things could be given name, not restricting alt

js: leif noted this could be used to hide ascii art

cs: not sure ascii art in alt would work...

can see cases where having both alt and role makes sense

can't flag every case, if it makes sense sometimes

so would have to look for non-null string, which makes validation quite complex

therefore moves into warning territory

jf: agree sounds like warning

but important to capture the point

js: proposal to file bug

cs: having warning gives a hook

jb: sounds like this is execution of action for CS and RS to discuss

still needs to come back into text sub-team

not consensus there yet, needs further discussion

cs: worried about differing from WAI CG consensus document, because of perception of flip-flopping

jb: need further conversation on that

group yesterday considered new info

Monday seemed pretty strong consensus on a proposal except for Leif

jb: moving on to title

cs: ok with either there

<LeifHSilli> Just want to note the question I consider most important: Does e.g. @aria-labelledby=foo win over empty @alt=""? If empty @alt equals role=presentation, then empty @alt should win. However, ARIA says that aria-labelledby wins. See http://lists.w3.org/Archives/Public/public-html-a11y/2011May/0290

sf: Leif and I disagreed on what was meant by device independence

<something about input devices>

asked vendors, got inconclusive responses

looked at Webkit, when @title and not(@alt) it falls back to title

but title shouldn't be a repair mechanism, it should be a visible caption

or at least available to all, not to some people some times

<JF> +1 to Steve

no requirement in HTML 5 that content be available in input device independent manner

filed a bug a year ago

was rejected

<re needing mouse to display title>

jb: questions?

objections to supporting proposal?

<janina> +1

<mranon> +1

<judy> please indicate in IRC who is supporting Steve's title proposal

<janina> +1

<mranon> +1

<judy> +1

<JF> +1 to steves proposal

rs: what are problems viewed with Leif's proposal?

sf: about accessible name property in AAPIs

<paulc> Where is Steve's proposal? Link please.

<janina> Referenced in the agenda for today ...

main diff from my proposal is that title used as fallback to alt

rs: that's just webkit, right?

<paulc> Title http://www.w3.org/html/wg/wiki/ChangeProposals/notitle

sf: don't want to promote that approach

rs: clarification?

sf: my proposal says title and alt should not be displayed in same way

rs: title used specifically for tooltip on mouseover

advisory text

<JF> +q to this topic

<Zakim> MichaelC, you wanted to say ARIA pretty explicitly allows title as last resort http://www.w3.org/WAI/PF/aria/roles#namecalculation

mc: ARIA allows title as last resort

sf: yes for accessible name

but shouldn't impact graphical rendering

jf: there's an issue of conventional behaviors vs specified behaviors

lots of screen readers by default ignore title

because of redundancy with alt in legacy content

because back in the day @alt wasn't rendered to tooltip so authors put in both places

jb: any change in support for Steve's proposal after this discussion?

rs: support

my understanding is Leif's is to have @title rendered when images turned off

that's not something I support

important to separate functionality of alt and title

jb: proposing change?

<JF> akc JF

rs: supporting Steve's

<Zakim> JF, you wanted to this topic

<richardschwerdtfe> I support Steve Faulkner's proposal

<richardschwerdtfe> on title

pc: note Leif not here

js: was on Monday, other discussions

pc: proposal has a number of "new information"

is there a companion rationale to reopen decision?

<judy> Leif, do you at this point object to Steve's proposal on title?

jb: embedded, think Sam's reviewed

sf: rationale is that there is new info

jb: pinging Leif, no response yet, but did discuss Monday that we'd look for consensus but not necessarily unanimity

at moment think we have that

suggest we declare TF consensus

js: think we have that now

were specific that the question would be called in this call

<paulc> Title http://www.w3.org/html/wg/wiki/ChangeProposals/notitle

RESOLUTION: Task Force supports Steve's reopened request and change proposal at http://www.w3.org/html/wg/wiki/ChangeProposals/notitle and recommends that it be forwarded to the HTML WG

jb: moving back to longdesc

pc: trying to understand goal at the moment

current proposal being constantly updated

<paulc> http://www.w3.org/html/wg/wiki/ChangeProposals/InstateLongdesc updated on may 12

<judy> ===== longdesc =====

<judy> http://www.w3.org/html/wg/wiki/ChangeProposals/InstateLongdesc

change proposals not necessarily expected to be static

it's before the WG now

<paulc> http://dev.w3.org/html5/status/issue-status.html#ISSUE-030

chairs track these things at above link

jb: we've requested Laura to stabilize this

she asked for consensus, got additional suggestions, so has continued to change

so put on agenda here to measure support from task force

who supports Laura's proposal as is right now?

<JF> +q

no responses, is the issue that it's still changing?

jf: conditionally I support, but need to finish fine-tuning

want to see final version

same

<JF> +1 to Rich's comments

think we're arriving at something that adds value to all users

jb: sounds like we can't call question on TF support until it stabilizes

to PC, what is process expectation?

pc: Sam & I will be at AC meeting next week, but both on travel so jetlag...

hope to respond within 3-4 days to reopen requests, but difficult next week

will enter our queue

jb: that's just regarding making decision whether or not to reopen?

pc: yes

though we have already started discussing this

and have meeting next Tuesday morning CEST

<JF> bye all

<MikeSmith> yep

<MikeSmith> wi

<MikeSmith> will be

Next week

Mike to chair, Steve to scribe

trackbot, end meeting

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.136 (CVS log)
$Date: 2011/05/12 16:05:32 $

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/conetnt/content/
Succeeded: s/tilte/title/
Succeeded: s/scrivbe/scribe/
Found embedded ScribeOptions:  -final

*** RESTARTING DUE TO EMBEDDED OPTIONS ***

Found Scribe: MichaelC
Inferring ScribeNick: MichaelC
Default Present: Michael_Cooper, JF, Janina_Sajka, Steve_Faulkner, Judy, MikeSmith, paulc, Marco_Ranon, Cynthia_Shelly, Eric, Rich_Schwerdtfeger
Present: Michael_Cooper JF Janina_Sajka Steve_Faulkner Judy MikeSmith paulc Marco_Ranon Cynthia_Shelly Eric Rich_Schwerdtfeger
Found Date: 12 May 2011
Guessing minutes URL: http://www.w3.org/2011/05/12-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]