Minutes: [aapi] UAI TF Meeting Tue Mar 11 2014

Link: http://www.w3.org/2014/03/11-aapi-minutes.html

Plain text follows:

    [1]W3C

       [1] http://www.w3.org/

                                - DRAFT -

            Protocols and Formats Working Group Teleconference
                               11 Mar 2014

    See also: [2]IRC log

       [2] http://www.w3.org/2014/03/11-aapi-irc

Attendees

    Present
           Joseph_Scheuhammer, +1.415.624.aaaa, Joanie_Diggs,
           [Microsoft], Bryan_Garaventa

    Regrets
           David_Bolter

    Chair
           Joseph_Scheuhammer

    Scribe
           cyns, cyns_

Contents

      * [3]Topics
      * [4]Summary of Action Items
      __________________________________________________________

    <trackbot> Date: 11 March 2014

    <clown> agenda: this

    <cyns> scribe: cyns

    <clown> take up item 1

    JS: I have created a version 1.1 UAIG that's a full copy of 1.0
    and incorporated jason kiss's javascript for table display
    ... still working on build issues. I will publish when I've
    figured that out

    <clown> take up item 2

    <clown> issue-481?

    <trackbot> issue-481 -- aria-live removal events should or must
    happen before object removal -- open

    <trackbot> [5]https://www.w3.org/WAI/PF/Group/track/issues/481

       [5] https://www.w3.org/WAI/PF/Group/track/issues/481

    <cyns_> scribe: cyns_

    <clown> action-1368?

    <trackbot> action-1368 -- Cynthia Shelly to Investigate
    implementation of aria-live region events in ie apropos
    issue-481. -- due 2014-02-11 -- OPEN

    <trackbot>
    [6]https://www.w3.org/WAI/PF/Group/track/actions/1368

       [6] https://www.w3.org/WAI/PF/Group/track/actions/1368

    <clown> take up next item

    <clown> take up next item

    <clown> take up item 3

    <clown> issue-441?

    <trackbot> issue-441 -- Normative UAIG requirements for what
    UAs do when aria-posinset or aria-setsize is provided
    explicitly on some, but not all, elements within a set. -- open

    <trackbot> [7]https://www.w3.org/WAI/PF/Group/track/issues/441

       [7] https://www.w3.org/WAI/PF/Group/track/issues/441

    <clown> take up item 4

    <clown> issueu-618?

    <clown> issue-618?

    <trackbot> issue-618 -- MSAA selection event mapping for single
    selection case when item becomes unselected -- open

    <trackbot> [8]https://www.w3.org/WAI/PF/Group/track/issues/618

       [8] https://www.w3.org/WAI/PF/Group/track/issues/618

    <clown> take up item 5

    <clown> issue-635?

    <trackbot> issue-635 -- Determine if uaig mappings for rowgroup
    are correct -- open

    <trackbot> [9]https://www.w3.org/WAI/PF/Group/track/issues/635

       [9] https://www.w3.org/WAI/PF/Group/track/issues/635

    <clown> take up item 6

    <clown> issue-644?

    <trackbot> issue-644 -- Update ATK/AT-SPI mapping for
    aria-hidden="true" to reflect elements should not be
    exposed/mapped -- open

    <trackbot> [10]https://www.w3.org/WAI/PF/Group/track/issues/644

      [10] https://www.w3.org/WAI/PF/Group/track/issues/644

    changed due dates for cynthia's actions to 4/8

    <clown> "Element SHOULD NOT be exposed, unless it is focused or
    fires an accessibility event. If the object is exposed in the
    accessibility tree, map all attributes are normal. In addition
    expose object attribute hidden:true."

    JS: david said mozilla ok with this wording.

    JD: it's better than current language, but not as strong as
    she'd like

    JS: David said he'd like to do this, but doens't think he can
    get developers to do it.
    ... can't have a MUST because of the 'unless' clause according
    to RFC

    JD: this seems like an author error to me

    JS: if you use display:none with labeledby doesn't get
    calculated
    ... it does get put in the label text, but no pointer to hidden
    accessible.
    ... I think the same thing probably happens with aria-hidden

    JD: I'm ok with that. But it is a lot of work for Orca to pull
    this info out of elements.

    JS: possible use case: tooltip. You might use aria-describedby
    to point to tooltip, but the tooltip isn't in the tree until
    you mouse over it. once you mouse over it, the tooltip is
    displayed, the accessible appears in the tree, and the
    relationship exists. when the tooltip goes away so does the
    accessible

    <bgaraventa1979> a quick question, any ideas how this would
    work in touch screen devices?

    JD: Orca doesn't want to care if this is aria or some other way
    to get accessibiltiy info.

    CS: AT shouldn't have to care if accessibiltiy tree info comes
    from aria or some other technology. Narrator, for example, does
    not know where the aapi values came from

    JD: could do a children changed at instead? Should this thing
    that doesn't exist be in the tree?

    JS: it should not be n the tree unless it's visible.
    ... tooltip starts out aria-hidden, author should chagne that
    when it's shown
    ... joanie wants it to say 'if aria-hidden=true' it shouldn't
    be in the tree

    JD: correct

    <clown>
    [11]http://www.w3.org/TR/wai-aria-implementation/#mapping_state
    -property_table

      [11] http://www.w3.org/TR/wai-aria-implementation/#mapping_state-property_table

    <clown> "Not exposed in AX API unless focused. If focused,
    expose as normal."

    JS: trying to get language for ATK that's closer to Apple's
    language

    JD: exactly

    JS: would like this new language across all columns. cynthia,
    are you ok with this for UIA and UIAExpress?

    <clown> "Element SHOULD NOT be exposed, unless it is focused or
    fires an accessibility event. If the object is exposed in the
    accessibility tree, map all attributes are normal. In addition
    expose aria-hidden is true in AriaProperties."

    <clown> *map all attributes AS normal.

    CS: looks good to me.

    <clown> ACTION: Joseph to add the above language to the UAIG
    for aria-hidden for UIA Express, UIA, IA2, and ATK/AT-SPI
    [recorded in
    [12]http://www.w3.org/2014/03/11-aapi-minutes.html#action01]

    <trackbot> Created ACTION-1399 - Add the above language to the
    uaig for aria-hidden for uia express, uia, ia2, and atk/at-spi
    [on Joseph Scheuhammer - due 2014-03-18].

    <clown> action-1399?

    <trackbot> action-1399 -- Joseph Scheuhammer to Add the above
    language to the uaig for aria-hidden for uia express, uia, ia2,
    and atk/at-spi -- due 2014-03-18 -- OPEN

    <trackbot>
    [13]https://www.w3.org/WAI/PF/Group/track/actions/1399

      [13] https://www.w3.org/WAI/PF/Group/track/actions/1399

    JD: aria-hidden is content that shouldn't be visible to AT
    ... non-current slide in carosel

    CS: modal dialog with greyed out background is another example.

    BG: voiceover doesn't make this distinction currently.

    CS: this area is buggy in all the browsers.

    JS: UAIG says in essence "if it's focused, it's in the
    accessibility tree"

    <clown>
    [14]http://www.w3.org/TR/wai-aria-implementation/#include_eleme
    nts

      [14] http://www.w3.org/TR/wai-aria-implementation/#include_elements

    JS: I don't like this wording

    CS: focusable rather than focused? That breaks the modal dialog
    use case, which is, afaik, the most common use case for
    aria-hidden

    JD: need to add some guidance about aria-hidden=false

    BG: div with display:none and aria-hidden=false, what happens?

    JS: @hidden + aria-hidden=false should be not visible but in
    tree.

    <joanie>
    [15]https://bugzilla.mozilla.org/show_bug.cgi?id=945194

      [15] https://bugzilla.mozilla.org/show_bug.cgi?id=945194

    CS: having something that is not visible but in accessibility
    tree is an anti-pattern, not a feature

    <joanie>
    [16]https://bugzilla.mozilla.org/show_bug.cgi?id=945194#c60

      [16] https://bugzilla.mozilla.org/show_bug.cgi?id=945194#c60

    BG: there are cases where offscreen is a useful technique.

    JD: this should be 2.0, because it's a complicated issue

    JS: what if html5.1 gets it in?

    CS: we should synch with html a11y tf on this
    ... +1 to Joanie

    issue: should aria-hidden=false put things in teh tree, or stay
    a no-op as it is now?

    <trackbot> Created ISSUE-647 - Should aria-hidden=false put
    things in teh tree, or stay a no-op as it is now?. Please
    complete additional details at
    <[17]https://www.w3.org/WAI/PF/Group/track/issues/647/edit>.

      [17] https://www.w3.org/WAI/PF/Group/track/issues/647/edit%3E.

    <clown> issue-647?

    <trackbot> issue-647 -- Should aria-hidden=false put things in
    teh tree, or stay a no-op as it is now? -- raised

    <trackbot> [18]https://www.w3.org/WAI/PF/Group/track/issues/647

      [18] https://www.w3.org/WAI/PF/Group/track/issues/647

    <joanie> or in between "false" and "put"

    <clown> :-)

    <joanie> they appeared in the past right?

Summary of Action Items

    [NEW] ACTION: Joseph to add the above language to the UAIG for
    aria-hidden for UIA Express, UIA, IA2, and ATK/AT-SPI [recorded
    in [19]http://www.w3.org/2014/03/11-aapi-minutes.html#action01]

    [End of minutes]
      __________________________________________________________


     Minutes formatted by David Booth's [20]scribe.perl version
     1.138 ([21]CVS log)
     $Date: 2014-03-11 20:08:47 $
      __________________________________________________________

      [20] http://dev.w3.org/cvsweb/~checkout~/2002/scribe/scribedoc.htm
      [21] http://dev.w3.org/cvsweb/2002/scribe/
      [24] http://dev.w3.org/cvsweb/~checkout~/2002/scribe/scribedoc.htm

-- 
;;;;joseph.


'A: After all, it isn't rocket science.'
'K: Right. It's merely computer science.'
              - J. D. Klaun -

Received on Tuesday, 11 March 2014 20:16:35 UTC