Protocols and Formats Working Group Teleconference

01 Dec 2015

See also: IRC log


Rich_Schwerdtfeger, Joseph_Scheuhammer, Bryan_Garaventa, Joanmarie_Diggs, Cynthia_Shelly


<scribe> agenda: this

<scribe> agenda: ISSUE-682/ACTION-1745 (Joseph) What happens to required owned elements when the @role is different that the element's default? E.g. <table role="dialog"> … </table>

<joanie> scribe: joanie

ISSUE-723 (Cynthia) Check that UIA control type for "application" role is Pane.

<clown> issue-723?

<trackbot> issue-723 -- UIA mapping for application incorrect -- raised

<trackbot> https://www.w3.org/WAI/PF/Group/track/issues/723

JS: I did a bunch of modifictations to UIA mapping. You, Cynthia, wanted to double-check a couple.
... One was application.

<clown> http://rawgit.com/w3c/aria/master/core-aam/core-aam.html#role-map-application

JS: Link above.
... (Reads from spec)

CS: I think that's ok.
... It's either pane or group. And neither one is ideal.

JS: I will take the TBD off and sometime this week commit the change and close the issue.

ISSUE-730 (Cynthia) Check that UIA localized control type for "article" role is article.

<clown> issue-730?

<trackbot> issue-730 -- UIA mapping for article incorrect -- raised

<trackbot> https://www.w3.org/WAI/PF/Group/track/issues/730

JS: This one is article.

CS: So article is not a landmark anymore.
... But I'm not sure what to do with them as a non-landmark.

<clown> http://rawgit.com/w3c/aria/master/core-aam/core-aam.html#role-map-article

JS: You said localized control type is article.
... And control type is document.

CS: I think we should change it to group; it's not a document.

JS: I'll point out that it's a document in IA2 and ATK.

JD: This is why I'm in the queue. An article is not a document.

RS: If you want to make it a document, you might want to tell....

JS: Actually I confirmed it in Firefox for IA2.

CS: A document is not a smaller thing; it's more of a top-level thing.

JS: The superclass role of article is document in ARIA.
... The mapping spec for ATK/AT-SPI2 has ROLE_ARTICLE.

JD: Aha, so Firefox is just not doing that.

JS: I will change UIA control type from document to group and remove TBD for the localized control type.

(All) Reminder to check the role mappings for Dpub-aam for FPWD. - http://rawgit.com/w3c/aria/master/dpub-aam/dpub-aam.html

JS: This is just a quick, friendly reminder.

RS: It's probably a little late since it's going out the door now.
... But feedback would be great.
... I need to look at the subroles on Mac as they don't have everything we need.

<clown> Issue-682

<trackbot> Issue-682 -- Clarify what happens to required owned childrens' roles when native host language parent is overridden with a WAI-ARIA role. -- open

<trackbot> https://www.w3.org/WAI/PF/Group/track/issues/682

JS: This is something that came up at the last ARIA meeting.
... : It got transferred from the spec to the mapping spec.

<clown> <table role="dialog" > … </table>

JS: If you do the above....
... What happens to the required owned elements of the table?

RS: Why would you do that?

JS: Maybe you wouldn't it.

BG: Setting dialog on the table element?

JS: Yes.

BG: I've actually done that.

CS: One possibility is do the same thing as role presentation; the other is author error.

RS: You could say role dialog is not allowed on a table.
... I'm ok with making them presentational.
... You could do this with a list.

JS: I've seen this done with an unordered list with role menu

RS: Is this the spec or mapping?

JS: I thought the spec, but during the ARIA meeting everyone seemed to think it belonged in the mapping spec.

RS: Make them none or presentation?

JS: I will do both as we're still in transition.

(Group agrees)

Action/Issue/Bugzilla triage for ARIA 1.1

JS: Triage time!

<clown> https://www.w3.org/WAI/PF/Group/track/products/23

JS: We're trying to hit ARIA 1.1 soon.
... The list of issues and actions is at the URL above.


<trackbot> action-1309 -- Cynthia Shelly to Determine the best place to map aria-colindex and aria-rowindex for UIA. -- due 2015-12-01 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1309

<clown> action-1309?

<trackbot> action-1309 -- Cynthia Shelly to Determine the best place to map aria-colindex and aria-rowindex for UIA. -- due 2015-12-01 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1309

JS: Action 1309 is due today.

CS: I still need to figure this out.
... Please give me another week.
... Make that two weeks.

RS: We're going to start shutting things down in a week or two.
... Sometimes people use navigate by row in email clients.
... You might want to have rowindex there as well.

JS: It's that setsize thing again (where does the property belong, on the children or parent container?)

CS: Repeating the property is a performance issue.

<clown> action-1373

<trackbot> action-1373 -- Cynthia Shelly to Create a test case for when aria-posinset and aria-setsize are provided explicitly on some but not all elements. -- due 2015-12-01 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1373

CS: I've not done this one yet.
... When do test cases need to be done?

RS: We have to create the testable statements.
... The week I'm out, Michael will start coordinating the testable statement creation.
... So we'll start this discussion text week.

CS: When do you expect to have the tests done?

RS: You need testable statements first. :)
... We're probably going to go to CR in March.
... We should have tests done by February.

CS: Ok, let's go with February 1st for this action item.

JS: I would think you'd write the test cases and statements at the same time.
... Because it's in your mind then.

RS: It depends on who is writing the tests.

JS: Action item due date updated.


<trackbot> action-1528 -- Alexander Surkov to Bolter to investigate the proper ia2 mappings for aria-current -- due 2014-11-18 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1528


<trackbot> action-1529 -- Joseph Scheuhammer to Investigate the proper atk/at-spi mappings for aria-current -- due 2014-11-18 -- PENDINGREVIEW

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1529


<trackbot> action-1535 -- James Craig to Investigate the axapi mappings for aria-current. -- due 2015-08-18 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1535

JS: We need to create an action for UIA, unless it's done.

<clown> http://rawgit.com/w3c/aria/master/core-aam/core-aam.html#ariaCurrent

RS: So Alex has never sent you a mapping?

JS: There's a long discussion on the IA2 mapping in which Alex objects because it's different from ATK.

<clown> http://lists.linux-foundation.org/pipermail/accessibility-ia2/2015-September/001939.html

JS: The above is the last message in the list.

RS: What happens when you have a navigation list with links?

<clown> scribenick: clown

JD: the locus of focus for a list item is 'selected'
... the locus of focus for a focussed widget is "focussed"
... for the current link in a list of links, we use "active"

RS: If you tab through the links, the one with focus is not the active one, necessarily

JD: The "thing" you are on...
... There are different states for the thing you are on.
... Putting active on something not focusable nor selectable is not a problem.

RS: Alex says that not possible?

JD I'd have to go back to list.

<joanie> scribe: joanie

RS: (Reads list message)

JS: Alex is saying if you have a widget that say is focused(?), it will be active in IA2 and ATK.
... Joanie says that it's not active in that case; it's STATE_FOCUSED.
... Alex is agreeing with James Teh.
... James Teh is responding to Joanie.
... You might need to read the entire thread.

RS: Ok, I'll do that.

JS: What I was going to do is respond to this one and say, "Progress? What's happening?"
... I know that Joanie wants the current item to map to STATE_ACTIVE.

RS: I think that's fine. I don't understand Alex's concern.

JS: I mentioned they are working on it for AX API.
... It looks like they finished and closed the bug.

<clown> https://bugs.webkit.org/show_bug.cgi?id=146012

JS: From the bug above, they have values in AX API corresponding to the different values for aria-current.
... AXCurrent is what they are going to use.

RS: So how are you going to do the active state for that?

JS: ATK is going to use the object attribute

RS: That makes sense.

JS: The only thing about this bug. There's a wrinkle in ATK/AT-SPI that when you change the state it spawns an event.
... And an AT can listen to that event.
... Apple doesn't have such a state.

RS: I'll follow up via email.

<clown> http://lists.linux-foundation.org/pipermail/accessibility-ia2/

<clown> https://lists.linuxfoundation.org/mailman/listinfo/accessibility-ia2

<clown> http://rawgit.com/w3c/aria/master/core-aam/core-aam.html#mapping_events_selection

JS: I will ask James Craig if there are any notifications for AX API.

<clown> Expose current=<value> in AriaProperties.

CS: Confirmed.

JS: Do you want an event?

CS: We get a property-change notification.
... So I don't think we need to do anything additional.

JS: Anything else on aria-current?


<trackbot> action-1530 -- Cynthia Shelly to Investigate uia mapping of role="text" -- due 2014-11-18 -- PENDINGREVIEW

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1530

JS: You made some suggestion. This is for role="text".
... I added for UIA use the text control pattern.

CS: Some things have chanaged in Edge. We no longer have separate text patterns per element.
... So I think role="text" is going to be the same as role="presentation"
... So not mapped. We'd expose the string as part of the page's text pattern.

<clown> <img role-"text" alt="hi">

RS: So how do you get things like lines if you have things in the middle of the content.

CS: The text pattern does that.
... We have one big text pattern for the entire document.
... Paragraphs are a thing inside the text pattern.
... Headings are not elements; the are an attribute of a range within the document text.
... There's not a separate node in the tree.
... It's similar to what's done with spans.
... For this one, I'm not yet sure how to map it.
... I think it's going to do the same thing as presentation, but I need to verify that.

JS: What due date?

CS: Give me a week from Tuesday (the 15th).

Meeting next week

JD: I won't be around next week.
... I think Rich won't be around next week. Cynthia just indicated the same.

Conclusion of group: No meeting next week.

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.144 (CVS log)
$Date: 2015/12/01 21:02:33 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.144  of Date: 2015/11/17 08:39:34  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/role/control type/
Succeeded: s/localized role/localized control type/
Succeeded: s/Cynthis/Cynthia/
Found embedded ScribeOptions:  -final


Found Scribe: joanie
Inferring ScribeNick: joanie
Found ScribeNick: clown
Found Scribe: joanie
Inferring ScribeNick: joanie
ScribeNicks: clown, joanie
Default Present: Rich_Schwerdtfeger, Joseph_Scheuhammer, Bryan_Garaventa, Joanmarie_Diggs, Cynthia_Shelly
Present: Rich_Schwerdtfeger Joseph_Scheuhammer Bryan_Garaventa Joanmarie_Diggs Cynthia_Shelly
Found Date: 01 Dec 2015
Guessing minutes URL: http://www.w3.org/2015/12/01-aapi-minutes.html
People with action items: 

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

[End of scribe.perl diagnostic output]