W3C

- DRAFT -

AAPI

04 Oct 2011

See also: IRC log

Attendees

Present
Andi_Snow_Weaver, David_Bolter, Cynthia_Shelly
Regrets
Chair
Andi_Snow-Weaver
Scribe
Andi_Snow-Weaver

Contents


<scribe> Scribe: Andi_Snow-Weaver

<scribe> ScribeNick: Andi

http://www.w3.org/Bugs/Public/show_bug.cgi?id=11811

http://www.w3.org/WAI/PF/aria-implementation/#document-handling_frames

bug 11811

DB: when building a page using frames, you specify the source and it gets pulled into the document - that's a sub-document
... example is a chat application
... if you put aria-attributes on the frame, you might be trying to override the attributes of the sub-document
... don't think this is a common case
... Chrome and Firefox might be the only browsers that would do this.

AS: summarizes the text in Section 6.1
... in most cases, the attributes on a sub-document override the attributes on a document
... but in the case of aria-live, -atomic, -relevant, and -busy the attributes of the sub-document override the attributes on the frame in the document

CS: sounds complicated but if the browsers do it differently, it's even more complicated
... if have a test file, I can find out what IE is doing
... okay with this being normative but need to check with some other people

DB: think this applies to all APIs

AS: but we need to find out how they do it because they don't have the container- object attributes

<scribe> ACTION: David to find out if Mozilla has a test file to send to Cynthia [recorded in http://www.w3.org/2011/10/04-aapi-minutes.html#action01]

bug - David - to find a test file

bug 13205 http://www.w3.org/Bugs/Public/show_bug.cgi?id=13205

DB: ATK is probably similar to IA2
... similar wording, just need the ATK equivalent to IAccessibleAction

13208 http://www.w3.org/Bugs/Public/show_bug.cgi?id=13208

CS: will do this week

<davidb> for atk: "If an AT calls an action on an accessible object, the user agent SHOULD simulate a click on the DOM element which is mapped to the that accessible object." (interface is AtkAction)

http://www.w3.org/Bugs/Public/show_bug.cgi?id=13288

DB: checkbox example, don't expose the default attributes in any particular way as if the author had put aria attributes on it

AS: aria-busy - default value is "false"

DB: FF doesn't add an object attribute for aria-busy="false"

<scribe> ACTION: Andi to fix the aria-grabbed row - false=false, undefined=not mapped [recorded in http://www.w3.org/2011/10/04-aapi-minutes.html#action02]

DB: for aria-live, if the author specifies "off", we expose it, but if not, we don't expose the object attribute
... not really an expectation of an AT needing to see the default value specified
... explicitly

CS: think IE only puts in the aria properties what the author has specified
... whether or not they are valid
... in AriaProperties field in UIA
... it's just a string

DB: where the aria state or property is expected to be supplied by the author, there might be a case for exposing a default value, but otherwise not

AS: can we define where it is expected?

DB: not any of the global properties except where defaults are unique based on the role. e.g. aria-live has different defaults depending on the role
... aria-checked is undefined is supposed to mean the widget can't be checked

but for role="checkbox", if aria-checked is not specified, are we supposed to expose "undefined"?

CS: whatever the author provides, gets put into the properties field and the UA is not validating in any way
... wrt the question, think a "MAY" would be okay.
... UAs SHOULD expose all of the properties in object attributes or properties as provided by the author and SHOULD NOT expose any that the author didn't provide.

AS: but for some we do provide the object attribute (aria-live) so we need to add "unless specified in the following table"

bug 14295 http://www.w3.org/Bugs/Public/show_bug.cgi?id=14295

DB: if the element is in the tree, everything is exposed as normal

http://www.w3.org/WAI/PF/aria-implementation/#mapping_state-property

http://www.w3.org/WAI/PF/aria-implementation/#mapping_events_state-change

http://www.w3.org/WAI/PF/aria-implementation/#mapping_events_visibility

<scribe> ACTION: David to review all sections of 5.8 (Events) for correct specification of aria-hidden behavior [recorded in http://www.w3.org/2011/10/04-aapi-minutes.html#action03]

s/[Microsoft],//

Summary of Action Items

[NEW] ACTION: Andi to fix the aria-grabbed row - false=false, undefined=not mapped [recorded in http://www.w3.org/2011/10/04-aapi-minutes.html#action02]
[NEW] ACTION: David to find out if Mozilla has a test file to send to Cynthia [recorded in http://www.w3.org/2011/10/04-aapi-minutes.html#action01]
[NEW] ACTION: David to review all sections of 5.8 (Events) for correct specification of aria-hidden behavior [recorded in http://www.w3.org/2011/10/04-aapi-minutes.html#action03]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.136 (CVS log)
$Date: 2011/10/04 17:04:27 $

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/rrsagend, make minutes//
FAILED: s/[Microsoft],//
Found Scribe: Andi_Snow-Weaver
Found ScribeNick: Andi
Default Present: Andi_Snow_Weaver, David_Bolter, [Microsoft]
Present: Andi_Snow_Weaver David_Bolter Cynthia_Shelly
Got date from IRC log name: 04 Oct 2011
Guessing minutes URL: http://www.w3.org/2011/10/04-aapi-minutes.html
People with action items: andi david

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


[End of scribe.perl diagnostic output]