See also: IRC log
<trackbot> Date: 23 June 2014
<MichaelC> meeting: ARIA
<MichaelC> Element.computedRole? https://www.w3.org/WAI/PF/Group/track/issues/427
<MichaelC> value https://www.w3.org/WAI/PF/Group/track/issues/493
<MichaelC> item 2 = Issue 493 Support undefined without having to have an undefined token value https://www.w3.org/WAI/PF/Group/track/issues/493
<MichaelC> item 2 = Issue 427 Do we want to go for a Element.computedName and/or Element.computedRole? https://www.w3.org/WAI/PF/Group/track/issues/427
<clown> http://lists.w3.org/Archives/Public/public-pfwg/2014Jun/0091.html
<MichaelC> scribe: cgallello
Joanie: Somewhere in the history of the bug there's a long discussion on why this should be done. We may want to do something different with states.
<MichaelC> Discussion in bugzilla https://www.w3.org/Bugs/Public/show_bug.cgi?id=23350
joanie: If objects are visible on the screen but not active, taking it out of the tree doesn't make sense.
MichaelC: jcraig' proposal is that aria-hidden and aria-inert wouldn't be pruned from the tree
<clown> https://www.w3.org/Bugs/Public/show_bug.cgi?id=23350#c14
<clown> "In short SENSITIVE state should work for inert subtrees (when subtree is inert then SENSITIVE state is absent). Note, ATK has ENABLED state for disabled controls."
clown: Proposal is to use ATK's sensitive state
MichaelC: Difficulty is deciding
what inert should be mapped to in HTML, then separate
discussion on if we need to create an ARIA property
... We need to familiarize ourselves with the background on
this topic, then get a clear idea on what the HTML mapping
should be. Bug was last discussed Oct 2013.
<Stefan> +q
<Stefan> +q
jcraig: Two places where this can be used. One is the dialog API - they're working on the dialog element in HTML. There are some javascript methods for dialogs like hiding/showing. All background contents are still focusable. Work around this in ARIA using aria-hidden="true". We either need to make the background no longer focusable or intercept the tab/shift tab. Concept of inert is 'displayed content' but like it's disabled.
<Stefan> +q
jcraig: Everything in the inert view would be inert. Even if the dialog was included in the body, you could put inert="false" on the dialog
Stefan: Why don't we put a dialog bound modal attribute - I can't see any other good uses for inert.
jcraig: There are other uses. The
presentation of carousels is such that you can interact with
one at a time but you can still see these other panes even
though they're not active.
... Even if you mark up with aria-hidden, you'd still need to
handle the focusability
... when you pop up menus, sometimes the background might be
inert. Depends on the platform.
... Support having an overridable attribute on menus and
dialogs to indicate that it's modal
Stefan: You don't necessarily need an explicit inert property though
jcraig: We could make inert the
default for dialogs/menus - then it's up to the user agent to
determine if the dialog/menu has appeared
... The APIs may want to update in order to handle some of
this.
<Zakim> MichaelC, you wanted to ask about the UA implementation issues of temporarily overriding keyboard handling
MichaelC: Is this too much on the user agent to override keyboard handling?
<jcraig> summary: this is probably more fitting as a host lang attr
<jcraig> html@inert
<jcraig> as opposed to @aria-inert
Stefan: Seems exceptional that you need to use a mix of html and aria markup
<Zakim> jcraig, you wanted to say it looks like @inert is no longer in either html spec
<jcraig> http://developers.whatwg.org/editing.html#editing
<jcraig> http://dev.w3.org/html5/spec-preview/editing.html
jcraig: Inert attribute has been taken out (from above links), we may want to work with these WGs and request that inert be brought back in
<Stefan> The dialog element's showModal() method makes
<Stefan> use of this mechanism.
<Stefan> this is what I meant with modal
<Stefan> attributte
<Stefan> OK
<Stefan> got it
jcraig: We should at least have a modal attribute in ARIA, not sure we'd gain anything by having an inert attribute in ARIA
<Stefan> +1 to te aria-modal for role=dialog
<jcraig> issue-433?
<trackbot> issue-433 -- Vague statement about modal dialog states and properties in APG. -- closed
<trackbot> https://www.w3.org/WAI/PF/Group/track/issues/433
<Stefan> thanks james
<joanie> +1 for aria-modal from me as well
<clown> issue-478?
<trackbot> issue-478 -- Dialog (Non-Modal) design pattern conflicts with browser keys and lacks example -- open
<trackbot> https://www.w3.org/WAI/PF/Group/track/issues/478
<clown> issue-568?
<trackbot> issue-568 -- Introduce system modal attributes for dialogs -- closed
<trackbot> https://www.w3.org/WAI/PF/Group/track/issues/568
<clown> action-1347?
<trackbot> action-1347 -- jcraig Craig to Add an attribute for differentiating modal vs non-modal dialogs/menus/etc. (possibly aria-modal) -- due 2014-01-31 -- OPEN
<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1347
<Stefan> yes
<MichaelC> action-1347: address this in the context of issue-623 as well
<trackbot> Notes added to action-1347 Add an attribute for differentiating modal vs non-modal dialogs/menus/etc. (possibly aria-modal).
<MichaelC> issue-623: also impacted by action-1347
<trackbot> Notes added to issue-623 Consider aria-inert for disambiguation of modal states (re: html|*@inert).
<Stefan> I think we really need that in ARIA 1.1
<Stefan> yes it is due to issues we have with Jaws
<Stefan> and modality
<Stefan> good :)
jcraig: Action for us is to raise this in the HTML accessibility task force and the SVG accessibility task force. No way we can make an HTML attribute that can pull event listeners off of nodes. We explicitly kept tabindex out of aria for a specific purpose.
<Stefan> +1 to ask Steve Faulkner
<jcraig> or jason kiss
<jcraig> concerned that @inert was dropped from HTML5; there are implications and potential needs for accessibility that aren't met without it
<MichaelC> ACTION: cooper to ask an HTML-AAPI editor to bring question of @inert to HTML A11Y TF [recorded in http://www.w3.org/2014/06/23-aria-minutes.html#action01]
<trackbot> Created ACTION-1461 - Ask an html-aapi editor to bring question of @inert to html a11y tf [on Michael Cooper - due 2014-06-30].
<MichaelC> trackbot, associate action-1461 with issue-623
<trackbot> action-1461 (Ask an html-aapi editor to bring question of @inert to html a11y tf) associated with issue-623.
<MichaelC> actioin-1461: see http://www.w3.org/2014/06/23-aria-minutes.html
<MichaelC> action-1461: see http://www.w3.org/2014/06/23-aria-minutes.html
<trackbot> Notes added to action-1461 Ask an html-aapi editor to bring question of @inert to html a11y tf.
<jamesn> http://www.w3.org/html/wg/drafts/html/master/editing.html
jamesn: inert's not in HTML due to lack of implementation
<jamesn> http://www.whatwg.org/specs/web-apps/current-work/multipage/editing.html#inert-subtrees
jcraig: that's the concept of inert subtrees but not the inert attribute
note for the minutes - the previously referred to 'jcraig' will now be 'jcraig'
jcraig: Before ARIA group decides
on this we need buy in from accessibility teams from each of
the major browsers
... Opera isn't a major player in the accessibility space so we
don't need them on board. I can work on the webkit side. We
need buy in from David's team at Mozilla, Dominic at
blink
... accessibility code doesn't get run as much on the open web.
Screen readers aren't exercising a lot of this extra code. If
the user's using additional technology, they can execute the
additional codepath.
... allowing author to request accessibility property will
allow for new codepaths to be run - so potential sec issues for
all rendering engine - that's the discussion that we'd want to
have everybody on board
... We need accessibility teams to buy in enough to discuss
with the sec teams on the browser side.
clown: What if we keep to a read only state? Lessens the sec concern?
james: we still need to spin up the accessibility code just to figure out what the role is - if you don't use a screen reader then none of that code gets used at all
jcraig: we'll spin up accessibility specific code in the web inspector - but that's something that developers have to explicitly enable and run on their own.
<MichaelC> ACTION: cgallello to discuss computed role with IE team [recorded in http://www.w3.org/2014/06/23-aria-minutes.html#action02]
<trackbot> Created ACTION-1462 - Discuss computed role with ie team [on Christopher Gallello - due 2014-06-30].
<jcraig> s/issues/issues/gi
<MichaelC> trackbot, associate action-1462 with issue-427
<trackbot> action-1462 (Discuss computed role with ie team) associated with issue-427.
<jcraig> s/sec/sec/gi
<MichaelC> ACTION: clown to keep issue-427 on radar for AAPI meetings [recorded in http://www.w3.org/2014/06/23-aria-minutes.html#action03]
<trackbot> Created ACTION-1463 - Keep issue-427 on radar for aapi meetings [on Joseph Scheuhammer - due 2014-06-30].
<MichaelC> trackbot, associate action-1463 with issue-427
<trackbot> action-1463 (Keep issue-427 on radar for aapi meetings) associated with issue-427.
<jcraig> s/concern/concern/gi
<MichaelC> action-427: jcraigC discussing webkit implementation; DavidB discussing Firefox implementation
<trackbot> Notes added to action-427 work with Matt on process to update CAPTCHA note.
<MichaelC> ACTION: dominic to discuss computed role with blink team [recorded in http://www.w3.org/2014/06/23-aria-minutes.html#action04]
<trackbot> Created ACTION-1464 - Discuss computed role with blink team [on Dominic Mazzoni - due 2014-06-30].
<MichaelC> trackbot, associate action-1464 with issue-427
<trackbot> action-1464 (Discuss computed role with blink team) associated with issue-427.
<jcraig> action-1464?
<trackbot> action-1464 -- Dominic Mazzoni to Discuss computed role with blink team -- due 2014-06-30 -- OPEN
<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1464
<MichaelC> trackbot, associate action-1464 with issue-427
<trackbot> action-1464 (Discuss computed role with blink team) associated with issue-427.
<jcraig> action-1464?
<trackbot> action-1464 -- Dominic Mazzoni to Discuss implications of Element.computedRole() with Blink team -- due 2014-06-30 -- OPEN
<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1464
<jcraig> action-1463
<trackbot> action-1463 -- Joseph Scheuhammer to Keep issue-427 on radar for aapi meetings -- due 2014-06-30 -- OPEN
<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1463
<jcraig> action-1335
<trackbot> action-1335 -- jcraig Craig to Propose edit for issue-493 to explicitly disallow strings matching "" or "undefined" in this sentence: content authors must provide values for required states and properties. -- due 2014-01-30 -- OPEN
<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1335
MichaelC: Nothing to discuss until action 1335 is completed.
<jcraig> Content authors must provide a non-empty value for required attributes; this value may not match the format "undefined".
jcraig: It's an editorial issue - just recrafting one sentence. More for the validator than anything else. This will allow the validator to give more explicit error messages.
<jcraig> ISSUE-493: When states and properties are required, authors must provide a value; but the explicit value of "undefined" is not prohibited. The empty "" or "undefine" values should prohibited.
<trackbot> Notes added to ISSUE-493 When states and properties are required, authors must provide a value; but the explicit value of "undefined" is not prohibited. The empty "" or "undefine" values should prohibited..
jcraig: True/false/undefined has gotten us into a corner. Is there ever a case where undefined is not the default value?
<clown> http://www.w3.org/WAI/PF/aria-1.1/states_and_properties#aria-selected
<jcraig> http://www.w3.org/WAI/PF/aria/complete#aria-invalid
MichaelC: whenever false is allowed, false is default
<jcraig> ACTION: jcraig or shane to fix default values in 1.1 respec [recorded in http://www.w3.org/2014/06/23-aria-minutes.html#action05]
<trackbot> Created ACTION-1465 - Or shane to fix default values in 1.1 respec [on jcraig Craig - due 2014-06-30].
<jcraig> action jcraig to verify in 1.1 spec that aria-checked "used in or inherits into roles" should include checkbox
<trackbot> Created ACTION-1466 - Verify in 1.1 spec that aria-checked "used in or inherits into roles" should include checkbox [on jcraig Craig - due 2014-06-30].
MichaelC: We need to identify
what we mean by "undefined" on aria-checked
... To me it's clear that undefined should be the default
value
<jcraig> action-1466: this is an editorial or generator issue in 1.0 Rec. it is listed correctly on the checkbox role
<trackbot> Notes added to action-1466 Verify in 1.1 spec that aria-checked "used in or inherits into roles" should include checkbox.
MichaelC: If a developer uses undefined, should the validator flag it?
jcraig: On some roles, specific properties are required - should undefined be not allowed for those?
<jcraig> issue-522?
<trackbot> issue-522 -- Name Computation 2A - bullet 3 confusion -- open
<trackbot> https://www.w3.org/WAI/PF/Group/track/issues/522
jamesn: Don't have a clarification proposal but people are getting confused.
<jcraig> If the element is marked as hidden, skip it. If the element has role presentation or none, proceed to the child contents
<jcraig> Is that the general clarification you're asking for?
jamesn: The description is too
complicated right now. Having a flowchart would help.
... yes, if role="presentation" was set, it would skip using
alt, but it wouldn't necessarily skip using the child
content
<MichaelC> ACTION: jcraig to propose edit to issue-522 [recorded in http://www.w3.org/2014/06/23-aria-minutes.html#action06]
<trackbot> Created ACTION-1467 - Propose edit to issue-522 [on jcraig Craig - due 2014-06-30].
<MichaelC> trackbot, associate action-1467 with issue-522
<trackbot> action-1467 (Propose edit to issue-522) associated with issue-522.
<jcraig> action-1467?
<trackbot> action-1467 -- jcraig Craig to Propose edit to issue-522 -- due 2014-06-30 -- OPEN
<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1467
<jamesn> "If aria-labelledby and aria-label are both empty or undefined, check for the presence of an equivalent host language attribute or element for associating a label, and use those mechanisms to determine a text alternative unless the element is marked as presentational (role="presentation")."
<jcraig> : rrsagent, make minutes
This is scribe.perl Revision: 1.138 of Date: 2013-04-25 13:59:11 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) WARNING: Bad s/// command: s/vulnerabilities/issues/gi WARNING: Bad s/// command: s/security/sec/gi WARNING: Bad s/// command: s/risk/concern/gi Succeeded: s/vulnerabilities/issues/g Succeeded: s/security/sec/g Succeeded: s/risk/concern/g Succeeded: s/James/jcraig/g Found Scribe: cgallello Inferring ScribeNick: cgallello Default Present: Michael_Cooper, Chris_Gallello, Joseph_Scheuhammer, Susann_Keohane, +49.322.110.8.aaaa, Stefan_Schnabel, Joanie_Diggs Present: Bryan_Garaventa Chris_Gallello Joanie_Diggs Joseph_Scheuhammer Michael_Cooper Stefan_Schnabel Susann_Keohane James_Craig James_Nurthen Regrets: Rich_Schwerdtfeger Janina_Sajka Agenda: http://lists.w3.org/Archives/Public/public-pfwg/2014Jun/0112.html Found Date: 23 Jun 2014 Guessing minutes URL: http://www.w3.org/2014/06/23-aria-minutes.html People with action items: cgallello clown cooper dominic jcraig or shane WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option.[End of scribe.perl diagnostic output]