HTML Accessibility Task Force Teleconference

24 Feb 2011


See also: IRC log


Cynthia_Shelly, Eric_Carlson, Gregory_Rosmaita, John_Foliot, Léonie_Watson, Marco_Ranon, Martin_Kliehm, Michael_Cooper, MikeSmith, Rich, Steve_Faulkner, paulc
Laura_Carlson, Janina_Sajka


Canvas Subteam Report

RS: canvas subteam proposal -- number of facets to canvas a11y -- 1. provide accessible interfaces to AT that can be accessed to represent what is seen in canvas

<trackbot> Date: 24 February 2011

RS: support in IE9 RC beta and in chrome (in the last week)
... hardest part about canvas is getting fallback content loaded -- hidden from user, hidden content usually not mapped; now can map and can focus on other needs

<MikeSmith> http://dev.w3.org/html5/status/issue-status.html#ISSUE-131

RS: 2) need to be able to have a FocusRing follow system settings (color, background, font size, etc.) -- support for author to draw FocusRing while reporting focus position to AT -- simplified drawFocusRing -- had an issue where if author wantede to draw himself,would drop out of FocusRing and couldn't be reported -- now, forcing FocusRing to follow system conventions -- if system has...
... FocusRing drawing mechanism, uses that -- outlined algorithm into how can compute bounding rectangle for object and object info DOM -- if no drawing path, will be handled by UA

<MichaelC> scribe: oedipus

RS: old drawFocusRing took x,y coordinates -- not how you draw carets for text boxes; always moving focus -- on every platform, selection and carets a11y APIs separate from what need for FocusRing
... created setCaretSelectionRect -- forget if have RTE or not in Canvas, gives author ability to give position whether user selecting content or using cursor -- defines what caret position and selectoin position is and exposes that to AT
... need contestual info (element) and actual caret position so can center zoom point on caret based on magnification level -- may want caret to left of zoom, etc.
... 3rd part: provide ability to get system setting for blink rate -- for those who suffer from seizures -- author can detect if blink rate set to custom level in system
... 4th: put in for metric for baseline for text --FocusRing or caret or selection position within content associated with text can get baseline to compute actyual bounding rectangle
... hixie ok with plan, question is when?
... testing the proposal with actual tools
... 1 more requirement for magnification -- need position and bounding rectangle for any part of the canvas -- challange will be to introduce this before last call -- something we can't rush through -- some optoins --could do what imagemap does to provide x,y and bounding rectangle or can use CSS styling -- querying devs which would you prefer

MS: process in HTML WG -- issue state: waiting for alternate or counter proposals -- due date 22 March 2011
... if objections, objector should write counter-proposal before 22 March 2011

RS: would like to hear thoughts on positioning of content

MS: take to discussion on public-html list
... great to hear canvas a11y progress




RS: assuming we don't get pushback like to try to get ua devs to implement it -- FF and IE trying to get out new versions, so not much flexibility there

MS: squeaky wheel gets the grease -- discrete amount of time to accomodate HTML a11y

Media Subteam Report

JF: single largest issue remaining is multitrak API -- how to provide support to supplementary binary assets -- lot of discussion -- looked at 8 different ways forward -- polled people on call, still some divergence especially amongst UA devs
... this is new territory -- not repairing bugs, but building in and extending capabilities
... 1 change proposal against HTML WG Issue 152

<MikeSmith> http://dev.w3.org/html5/status/issue-status.html#ISSUE-152

JF: consensus on call to continue in fashion we have been using -- dialogue on list -- much common ground but still some divergence
... for f2f in march 2001, media subteam would like break-out session -- ideally on saturday -- so we can get together in same room and hash out details -- most of ateendees from weekly calls will be there -- silviaP trying to get funding to attend
... should have another change proposal in by March with internal consensus behind it

MS: HTML WG issue 152 is waiting for counter proposals -- change proposal submitted from silvia?

JF: no, i think FrankO of Microsoft
... silvia thinking of writing a CP; apple reps might write one, maybe a third; if stopped today would have to work through 3 or 4 -- thought best if could coalesce around CP that meets everyone's approval

MS: chairs have to evaluate each CP -- if can merge into common CP, that is preferable

JF: right, that is our goal
... a lot of issues coming to the attention of people -- more than a11y -- i18n, "widestreaming" -- complex topic
... very aware of 22 March 2011 deadline, but don't want to ruch -- we need to get this right

MS: should devote F2F time to this issue -- if need breakout session, then could do that

JF: not sure of agenda and logistics for F2F -- will discuss with Cooper
... engineers need to discuss low-level engineering as well

MC: should have time in agenda for breakouts -- media subgroup meet an addtitional day?

JF: not feasible

MC: ok, just checking

MS: should accomodate breakouts as needed

JF: issue of time stamp formats still quite open -- hearings in U.S. at FCC about mandating captioning -- keeping eye on that -- U.S. may mandate specific technology which will impact development

Bug Triage Subteam Report

MK: down to almost no bugs to which to apply a11ytf keyword -- looked at all marked a11y to see if needed a11ytf attention and keyword
... looking at what appears each week

<kliehm> http://www.w3.org/Bugs/Public/show_bug.cgi?id=10525

<MikeSmith> "Please try to improve Bugzilla's accessibility/usability problems before Last Call"

MK: would be good if someone in TF could handle HTML WG bug 10525 -- needs owner, needs to be done

GJR: i will do a bugzilla review of w3c version versus current version (version 4, recently released)

<scribe> ACTION: Gregory - review w3c bugzilla installation versus newest version 4, coordinate with those who volunteered, present options and needs [recorded in http://www.w3.org/2011/02/24-html-a11y-minutes.html#action01]

<trackbot> Created ACTION-105 - - review w3c bugzilla installation versus newest version 4, coordinate with those who volunteered, present options and needs [on Gregory Rosmaita - due 2011-03-03].

ARIA Mapping Subteam

SF: haven't had meeting lately -- waiting on the chairs to address issues outlined in CPs -- timeline?

MS: chairs haven't yet scheduled date -- perhaps paulc can comment

PC: issue number?

SF: 129

<MikeSmith> http://dev.w3.org/html5/status/issue-status.html#ISSUE-129

PC: not in current queue
... did receive request about this via public-html
... reason why wanted decision on this issue?

RS: because we have a lot of work to do with UA devs

PC: has impact outside of WG, right?

RS: yes

PC: not clear in original message

MS: does resolution of this issue have implementation impact on UAs

SF: yes, definitely

RS: replaced ARIA section to define where ARIA can and cannot be applied

<paulc> Other than the intro and signature the request only said: "When will the chairs have a decision on the straw poll vote on issue 129?"

SF: part about doc conformance; default roles defined in ARIA user agents must implement -- default role on element, has to apply that role -- one of the major issues

MS: should be prioritizing talk with chairs

PC: RS provided no rationale as to why chairs should consider earlier

RS: don't know what your queue looks like -- understand need for deadlines -- we rushed to meet it, and then it sits there, which is frustrating --

<JF> +1 to Rich

PC: only making the point that if the chairs get 5 emailis requesting review need to know why so important to requesting entities
... suggest RS respond to own message as to why an earlier decision on ARIA in HTML5 is needed
... are moving forward with change proposal review
... some issues may moot others -- chairs prioritize on ratinoale given to us by WG memberts, suggest RS do the same

MS: will follow up on this too to ensure gets into charis' queue
... default roles is higher priority than document conformance

SF: people are implementing HTML5 now, basing decisions on what is in spec, hence urgency

RS: have product teams implementing HTML5 now so need to get addressed

<Stevef> http://lists.w3.org/Archives/Public/public-html/2011Feb/0399.html

SF: there was a call for change in regards issue 161 (A11y API Mapping) -- expired yesterday -- sam sent out notice that "closed without prejudice" -- responded to him saying would provide CP today -- working on document now including advice as to how to procede

<MikeSmith> http://dev.w3.org/html5/status/issue-status.html#ISSUE-161

MS: not going to do anything about it but leave to discretion of chairs -- not going to second guess chairs' decisons on

SF: would lilke a reasonably timely reply -- don't want to wait 2 weeks to find out won't be opened

PC: if submit change proposal today, chairs will at least consider it -- 4 due yesterday -- chairs immediately requested counter proposals to be completed by 22 March 2011 --schedule to LC needs to give people writing coutner proposals as much time to compose as given to thiose writing CPs
... i speak for Paul and not for all 3 co-chairs

SF: the issue says create a reference -- creating document -- wo'n't be ready for LC, asking to have reference to docuement in HTML5
... issue is to have a reference to it

PC: don't see why HTML5 in LC can't reference an editor's draft -- no logistical or administrative barrier to that

<richardschwerdtfe> gotta drop off folks

SF: only thing been told is hixie stating "special status" for referenced documents in HTML5, but hasn't articulated what constitutes a "special status"

PC: i don't know what he means by that either
... why did you miss this deadlijne?
... been there for a moth -- on last weeks' WG agenda as closing next week -- TF needs to pay more attention to these dates

SF: human error -- previously with human error hixie has been given extentions

PC: concewrned we are speaking in circles here

SF: will get CP in today, wanted clarification on what CP should say so would be accepted

MS: SF please try to join HTML WG call at noon to discuss

SF: can't make those calls because have to pick up kids from school

RS: chairs pretty swamped -- what is big deal if don't get into today

PC: sam took action item to call for counterporopsals if submitted by today -- immediately ready to take action -- longer it takes to file CP, less time people have to do counter proposal, which may sway chairs to leave issue closed because unfair to counter proposals because original propoasal deadline extended
... want to ensure that TF undertstands pressure on the chairs

Change Proposal for Issue 134

HTML A11y TF Action-87 "Create change proposal for ISSUE-134 http://www.w3.org/html/wg/tracker/issues/134": http://www.w3.org/WAI/PF/HTML/track/actions/87

<MikeSmith> oedipus: allow menu and command to represent a tablist

HTML WG ISSUE-134: "Provide tablist and tab states for menu and command elements respectively" http://www.w3.org/html/wg/tracker/issues/134

bound to HTML Bug 10831: http://www.w3.org/Bugs/Public/show_bug.cgi?id=10831

change proposal for HTML WG ISSUE 134: http://www.w3.org/html/wg/wiki/ChangeProposals/tablist_and_tab_states_for_menu_and_command_elements (thansk to everett zufelt and jason kiss)

<inserted> scribenick: MikeSmith

oedipus: the state of tab controls is normally only visually indicated

… and this makes cases for tabs are unfocusable

… current state is that this creates much more work for Web devlopers

… but there is a solution for this, which is to provide support natively

<inserted> scribenick: oedipus

"The HTML5 menu element, along with the command element, can currently represent a menu, context menu, or toolbar. Since a group of tab controls effectively acts as a menu of command elements for showing and hiding associated content tab panels, it is, therefore, proposed that the menu and command elements are well-suited to serve the purposes of a tabbed interface, and that their functionality be

extended to represent a tablist of tab controls and to identify which of those tabs is selected. "

" * Add a tablist state to the menu element.

* Add a tab state to the command element.

* Add a tabpanel state to command element.

* Add a tabgroup attribute to the command element."

<richardschwerdtfe> those are role values

"If no action is taken on this issue, then, without developers doing rather significant work to make a tabbed interface accessible using WAI-ARIA and additional scripting, only the selected tab in each tablist will be focusable, meaning that the remaining tabs from each tablist will not be available or accessible to assistive technologies."

<MikeSmith> http://dev.w3.org/html5/status/issue-status.html#ISSUE-134

<richardschwerdtfe> dropping now folks. sorry

<JF> http://dev.w3.org/html5/status/issue-status.html#ISSUE-134

richard correct tablist, tab, and tabpanel would be @role values

tabgroup would be a new attribute on command element

Face2Face Meeting

<MikeSmith> http://lists.w3.org/Archives/Public/public-html-a11y/2011Feb/0210.html

MS: list of proposed topics
... have some cases of miscommunication about process -- would like to prevent going forward -- useful to have process discussion and decision policy document and some of the contentious issues around communication
... 2nd) session on how to write a good change proposal -- how to avoid accusations of incomplete
... tuturial on change proposal writing would be helpful -- volunteers?
... 3rd) discussion over technologies: media, canvas, how to procede with longdesc, ARIA integration
... ambiguity about keyboard access

GJR: can prepar keyboard access discussion starter

MS: bugs and issues that need to be brought to the WG's attention

qck oed

<inserted> scribenick: MikeSmith

oedipus: I can prepare a keyboard-access overview and current-status focusing on high-level and discrete issues

… I will attend remotely

… I would like to go through it all

… so I can present what are issues, what has changed

<Stevef> chane proposal for issue 161 http://lists.w3.org/Archives/Public/public-html-a11y/2011Feb/0212.html

Stevef, thanks

<Stevef> np

<oedipus> ACTION: Gregory - prepare keyboard access high-level issues discussion for F2F [recorded in http://www.w3.org/2011/02/24-html-a11y-minutes.html#action02]

<trackbot> Created ACTION-106 - - prepare keyboard access high-level issues discussion for F2F [on Gregory Rosmaita - due 2011-03-03].

<inserted> scribenick: oedipus

JF: need to plan timeframe for 2 days -- even a skeleton at this point would be helpful

MS: narrow down to day and part of day you will have critical mass

JF: want to do breakout in afternoon to accomodate austrialian time zoners

MS: can discuss on mailing list -- will kick off discussion with list of topics -- try to send out tomorrow


Summary of Action Items

[NEW] ACTION: Gregory - prepare keyboard access high-level issues discussion for F2F [recorded in http://www.w3.org/2011/02/24-html-a11y-minutes.html#action02]
[NEW] ACTION: Gregory - review w3c bugzilla installation versus newest version 4, coordinate with those who volunteered, present options and needs [recorded in http://www.w3.org/2011/02/24-html-a11y-minutes.html#action01]
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2011/02/24 17:09:28 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.135  of Date: 2009/03/02 03:52:20  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/tab list/tablist/
Succeeded: s/starter discussion/overview and current-status focusing on high-level and discrete issues/
Succeeded: i/canvas subteam proposal /TOPIC: Canvas Subteam Report
Succeeded: i/oedipus: the state of tab controls/scribenick: MikeSmith
Succeeded: i/"The HTML5 menu element/scribenick: oedipus
Succeeded: i/oedipus: I can prepare a keyboard-access/scribenick: MikeSmith
Succeeded: i/JF: need to plan timeframe/scribenick: oedipus
Found Scribe: oedipus
Inferring ScribeNick: oedipus
Found ScribeNick: MikeSmith
Found ScribeNick: oedipus
Found ScribeNick: MikeSmith
Found ScribeNick: oedipus
ScribeNicks: oedipus, MikeSmith
Default Present: John_Foliot, Rich, Gregory_Rosmaita, Eric_Carlson, MikeSmith, Michael_Cooper, Marco_Ranon, Steve_Faulkner, paulc, Cynthia_Shelly, Microsoft, Léonie_Watson, kliehm
Present: Cynthia_Shelly Eric_Carlson Gregory_Rosmaita John_Foliot Léonie_Watson Marco_Ranon Martin_Kliehm Michael_Cooper MikeSmith Rich Steve_Faulkner paulc
Regrets: Laura_Carlson Janina_Sajka
Agenda: http://lists.w3.org/Archives/Public/public-html-a11y/2011Feb/0210.html
Found Date: 24 Feb 2011
Guessing minutes URL: http://www.w3.org/2011/02/24-html-a11y-minutes.html
People with action items: gregory

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

[End of scribe.perl diagnostic output]