W3C

- DRAFT -

AAPI

09 Feb 2010

See also: IRC log

Attendees

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

Contents


<scribe> scribe: Andi

<scribe> chair: Cynthia

goals and schedule for next draft

AS: 4.7 Actions, 4.8 Events, and 5.0 Special Document Processing need the most work

CS: is WAI-ARIA working to a target date?

AS: no - have to process comments
... got about 50 new comments or disagreement with disposition of comments on previous LC draft.

CS: April would be the soonest? Seems doable but do have SXSW in March.
... May seems more doable. Can work on these sections in a concentrated way the first two weeks of May.

AS: maybe we should split up the work into smaller pieces to get started before May.

<scribe> ACTION: all to read UAIG front to back [recorded in http://www.w3.org/2010/02/09-aapi-minutes.html#action01]

<scribe> ACTION: Andi to review all bugs and prioritize the ones that impact the Actions, Events, and Special Document sections [recorded in http://www.w3.org/2010/02/09-aapi-minutes.html#action02]

<scribe> ACTION: Andi to review all bugs and prioritize the ones that we have not dealt with yet at all in the UAIG. [recorded in http://www.w3.org/2010/02/09-aapi-minutes.html#action03]

Open issues: http://www.w3.org/WAI/PF/Group/track/products/10

http://www.w3.org/TR/2009/WD-wai-aria-implementation-20091215/#document-handling_author-errors

issue 97: http://www.w3.org/WAI/PF/Group/track/issues/97

AS: we added a table of defaults when required states/properties not provided

DB: at an abstract level, are we adding attributes to the DOM that aren't there.

CS: was thinking we were just adding attributes to the accessibility API.

DB: so we're saying the UA should act as if the ARIA attributes were provided but not modify the DOM

<scribe> ACTION: Andi to create a bug to clarify the text around supplying default values for required properties that were not provided [recorded in http://www.w3.org/2010/02/09-aapi-minutes.html#action04]

DB: something like "behave as if the following default values were provided"
... then issue 97 should be closable

http://www.w3.org/TR/2009/WD-wai-aria-implementation-20091215/#document-handling_author-errors

issue 98 (http://www.w3.org/WAI/PF/Group/track/issues/98) and 99 (http://www.w3.org/WAI/PF/Group/track/issues/99)

text at top of 5.3 addresses this - close these

issue 107: http://www.w3.org/WAI/PF/Group/track/issues/107

section 4.8.2 currently discusses node changes

DB: would make sense as a caution or alert but don't know if we do that in W3C documents

CS: could add paragraph after the two bullets that discuss node changes

issue 108: http://www.w3.org/WAI/PF/Group/track/issues/108

http://www.w3.org/TR/2009/WD-wai-aria-implementation-20091215/#document-handling_author-errors

http://www.w3.org/TR/2009/WD-wai-aria-implementation-20091215/#mapping_general

CS: this is about ID's on the accessible tree nodes
... could add a bullet to the list in section 4.1
... up to the user agent to get this right and authors should be free of having to worry about it

DB: not sure we need to say anything
... if others do, I'm fine with that.

CS: every node in the accessibility tree has an ID

DB: in MSAA, you have a bunch of data that specifies an accessible object - use ID to communicate with AT

CS: maybe something about managing IDs in the accessibility tree is the responsibility of the UA, not the author

<scribe> ACTION: David to talk to Rich for clarification on issue 108 [recorded in http://www.w3.org/2010/02/09-aapi-minutes.html#action05]

issue 115: http://www.w3.org/WAI/PF/Group/track/issues/115

http://www.w3.org/TR/2009/WD-wai-aria-implementation-20091215/#mapping_role

addressed by bullet 4 in the 4.4 Role Mapping

close issue 115 and refer to this section

issue 359: http://www.w3.org/WAI/PF/Group/track/issues/359

CS: in API mappings, it would go in description field
... valid issue - should deal with before our next draft

DB: think this is AT issue

CS: if aria-describedby points to a big chunk of HTML, how would you pass that in a description field

DB: don't know if we're going to be able to do this for ARIA 1.0

CS: maybe something like "UAs MAY provide a way for the user to navigate to an associated describedby and ATs SHOULD provide such a method."

DB: think this is a good idea but think it would require platform API changes

CS: it's the sort of thing ATs do.
... useful for anyone to be able to do

DB: example in e-mail - image that is described by a data grid

CS: AT could navigate it

DB: not sure how AT would know where to do

close this issue with text such as Cynthia suggested above.

issue 379: http://www.w3.org/WAI/PF/Group/track/issues/379

CS: don't think we can get IE to change implementation

DB: would rather change implementation than have diverging ones

AS: Apple would disagree with implementing this the way IE has done it

DB: do we even need it in the spec?

CS: the IE implementation is per the advice of the MSAA inventors
... conflict caused by IA2
... concept of relationships that are other than the tree relationships doesn't exist in MSAA
... IA2 has the concept of additional relationships
... doesn't map well to MSAA or UIA
... pretty sure there's no way in UIA to expose the relationship

DB: think maybe it shouldn't belong in the spec
... is anyone using this?

CS: would recommend restructuring HTML so you don't need aria-owns

DB: it's in jquery

CS: that means it's everywhere
... specific to one accessibility API - so may not make sense to have it in ARIA at all

DB: would have to argue that the accessibility forecast is worse if we continue with diverging implementations

CS: two of the major APIs don't have this

DB: ATK has a way of exposing any kind of relationship

CS: not sure about Apple

DB: sounds like they do

CS: exists already in IE8 - will be around for a long time
... can ask for feature in UIA but don't think would get it

<scribe> ACTION: David and Cynthia to take the issue of aria-owns to the ARIA TF [recorded in http://www.w3.org/2010/02/09-aapi-minutes.html#action06]

<davidb> atk relations: http://library.gnome.org/devel/atk/unstable/AtkRelation.html#AtkRelationType

s/issues/98 )/issues/98 )/

Summary of Action Items

[NEW] ACTION: all to read UAIG front to back [recorded in http://www.w3.org/2010/02/09-aapi-minutes.html#action01]
[NEW] ACTION: Andi to create a bug to clarify the text around supplying default values for required properties that were not provided [recorded in http://www.w3.org/2010/02/09-aapi-minutes.html#action04]
[NEW] ACTION: Andi to review all bugs and prioritize the ones that impact the Actions, Events, and Special Document sections [recorded in http://www.w3.org/2010/02/09-aapi-minutes.html#action02]
[NEW] ACTION: Andi to review all bugs and prioritize the ones that we have not dealt with yet at all in the UAIG. [recorded in http://www.w3.org/2010/02/09-aapi-minutes.html#action03]
[NEW] ACTION: David and Cynthia to take the issue of aria-owns to the ARIA TF [recorded in http://www.w3.org/2010/02/09-aapi-minutes.html#action06]
[NEW] ACTION: David to talk to Rich for clarification on issue 108 [recorded in http://www.w3.org/2010/02/09-aapi-minutes.html#action05]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2010/02/09 19:33:56 $

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)

WARNING: Bad s/// command: s/issues/98)/issues/98 )/
Succeeded: s/98)/98 )/
Found Scribe: Andi
Inferring ScribeNick: Andi
Default Present: Andi_Snow-Weaver, David_Bolter, Cynthia_Shelly
Present: Andi_Snow-Weaver David_Bolter Cynthia_Shelly
Got date from IRC log name: 09 Feb 2010
Guessing minutes URL: http://www.w3.org/2010/02/09-aapi-minutes.html
People with action items: all andi cynthia david

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


[End of scribe.perl diagnostic output]