W3C

Accessible Rich Internet Applications Working Group Teleconference

18 Jan 2018

Attendees

Present
Joanmarie_Diggs, MichaelC, Stefan, janina, matt_king, jongund
Regrets
JamesNurthen
Chair
Joanmarie_Diggs
Scribe
Stefan

Contents


<joanie> agenda: be done

<scribe> Scribe: Stefan

Publishing new Working Drafts of Graphics ARIA and Graphics AAM

<joanie> https://rawgit.com/w3c/graphics-aria/master/

<joanie> https://rawgit.com/w3c/graphics-aam/master/

joanie: here are links to latest editor drafts
... there were recent changes
... everybody reviewed, submitted changes etc.
... we shoul do cfc, but we need new WD out of the door
... any comments?
... I will send later out today a CFC
... we also need CFC about Transistion to CR

Tri-state toggle button issue: https://github.com/w3c/aria/issues/684

Joanie: what about tristate isssue?

Matt: (looking it up)
... the question is: what happens for aria-pressed for tristate?
... spec table seems to be ambiguous here
... do we allow toggle buttons? if so then I propose new wording for aria-pressed

<joanie> A value of mixed means that the values of more than one item controlled by the button do not all share the same value.

Joanie: state is not binary

Matt: we have example in the authoring practices

<joanie> Toggle buttons require a full press-and-release cycle to change their value.

Joanie: pushbutton is different compared to tri-state button

Matt: sentence above explains what it meant

Joanie: do you agree that this limitation should't be there in the wording?

Matt: next sentences are explaining and specifiying it better
... I don't think whole button down can be separately handled in ARIA without extra events

Joanie: proposed change that fixes it?

Matt: I will take ownership taking text changes

Joanie: tha's editorial

Matt: maybe we have checks for aria-pressed

Joanie: was tested in 1.0 cycle

Matt: then it is editorial

Sanity checking roles supporting aria-expanded: https://github.com/w3c/aria/issues/681

Matt: was not supported on menuitems in ARIA 1.1
... no state signaling was possible
... on touch devices
... in APG we have used aria-expanded=true on menuitems
... simple spec change allows it now to be on menuitems, sanity check reveals whole bunch of other cases that are maybe inappropriate
... all patterns using aria-expanded puts it on controlling element
... disclosure expand/collapsed not in AT tree when it is collapsed
... what if there is NO controlling element ????
... I've never actually seen that
... but maybe we disallow it .. need discussion about this

Bryan: there may be hieraric effects using aria-expaned=true

Matt: I am thinking of other roles than menuitem
... I went through all roles supporting it and not
... we should remove for some and add support for it to others
... button, combo, listbox, treeitem, menuitem have use of it
... remove: alert dialog, cell, link, menuitemcheckbox, tooltip..
... and some unclear things
... for example alert role
... is this editorial change?
... annother one is role=article
... you can do this withoit controllable element, but even with would this be legal making the container itsself interactive?

STefan: we need a concept for interactive containers

Matt: do we need to require separate visual focusable items in the container?

Bryan: we should avoud focus to regions
... this will fail on touchscreen devices

Matt: you can focus by swipe, not directly
... interactive focusable widget or not was the line we drew
... regarding focusability
... can you apply an aria state to a static role for interoperability?

Bryan: inheritance may occur which destroys things
... aria-expanded on a region will cause also checkboxes to get expanded state

Matt: focusable container must have a node in the accessibility tree
... we don't actually prohibit this currently
... we coud test it and make it work with AT or can make spec more restrictive
... one idea is addig property only when element is focusable

Joanie: people with expertise what to do .. we need consensus

Matt: this discussion is extremly helpful

Joanie: put it to task force :)

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.152 (CVS log)
$Date: 2018/01/18 19:01:59 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.152  of Date: 2017/02/06 11:04:15  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Succeeded: s/trace state/tristate/
Succeeded: s/want expertise/with expertise/
Found embedded ScribeOptions:  -final

*** RESTARTING DUE TO EMBEDDED OPTIONS ***

Present: Joanmarie_Diggs MichaelC Stefan janina matt_king jongund
Regrets: JamesNurthen
Found Scribe: Stefan
Inferring ScribeNick: Stefan
Found Date: 18 Jan 2018
People with action items: 

WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)


[End of scribe.perl diagnostic output]