<trackbot> Date: 22 October 2015

<richardschwerdtfeger> meeting: W3C WAI-PF ARIA Caucus

<scribe> scribenick: LJWatson

TPAC

<richardschwerdtfeger> https://www.w3.org/WAI/PF/wiki/Meetings/TPAC2015/ARIA

RS: Would like to look at 1.1, but because we're in Japan and because the Web Platform people will be there, I think we should look ahead at 2.0.
... The web has changed a lot since 1.1. Web Components for example.
... IndieUI didn't become a formal spec.
... Cynthia will lead this discussion as head of the CG.
... We have a limited number of people that can work on this.
... So 9.30am through 12.30am on Monday there will be an accessibility API discussion.
... Question is can we create a cross-browser API?
... Monday afternoon we'll meet with Web Platform.
... Want to hear their thoughts and concerns.
... Also issues like querySelector - how does that fit into the APIs?
... ETS will be there with experience of using Web Components in education.
... Tuesday ARIA 2.0 again the morning, then 1.1 and the test harness and exit criteria.
... The agenda is adjustable.
... Later a discussion with COGA from 3.30pm to 5pm.
... There is also a special session on Wednesday to follow-up on the acc API discussion.

MC: in APA we want to get organised on CSS.
... Meetings with DPub, Web Payments and Web Platform.

RS: We need to know how much SVG 2 to include in the abstract of the 1.1 update.

<clown> issue-707

<trackbot> issue-707 -- Update ARIA 1.1 Abstract and Introduction -- open

<trackbot> https://www.w3.org/WAI/PF/Group/track/issues/707

RS: ARIA 1.1 is going to synchronise gaps in HTML5 plus issues left over from 1.0.
... A common ARIA module that's synchronised with both HTML and SVG.

<MichaelC> Draft ARIA 1.1 Exit Criteria

MC: Drafted potential exit criteria for 1.1.
... Proposing we state explicitly we're only testing new or changed things since 1.0.
... Also that for ARIA itself we only need to find two test cases amongst tools using any of the mappings.
... Think we can argue this is sufficient for the ARIA spec, but the mappings will have a much higher bar.
... So the timelines for the mapping docs is likely to be different.
... Suggest we don't require that the mappings have reached PR to be used in the ARIA testing.

RS: Don't want us to duplicate effort. W ehave HTML and SVG mappings.
... Where either of those specs point to our spec, we could do the test cases for that.
... For the things specific to the host languages we have them do the mappings.

MC: Agree no duplication.
... In practice it doesn't matter who writes/runs the tests.
... Sounds like you want to prioritise ARIA mappings to pass 1.1?
... We'll need help from other WGs to test the rest, and that's reasonable.

RS: For example in SVG the name/desc computation are dramatically different to that for HTML.
... Because of descendent <title> and <desc> elements for example.

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

RS: Don't think ARIA WG needs to deal with those particular things.

MC: Yes.

RS: It'll allow us to get 1.1 done quickly.

Issue 741

<clown> action-741

<trackbot> action-741 -- Gregory Rosmaita to Ensure that our old comments are resubmitted, but not do any new review. -- due 2010-12-15 -- CLOSED

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/741

<richardschwerdtfeger> https://www.w3.org/WAI/PF/Group/track/issues/741

RS: We made changes for listitems.

<clown> +1

RS: My understanding is that we still want a name computation on links, the way we're doing it now.
... Anyone disagree?

<joanie> +1 to explicitly agree with naming links

RESOLUTION: Close issue 741 - No change to the spec.

ARIA 1.1

RS: Want to go through some outstanding actions.

<clown> action-601

<trackbot> action-601 -- Michael Cooper to Make property default value overrides machine readable -- due 2015-10-30 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/601

RS: Should we close or move to 2.0?

MC: Suggest move to 2.0.

RS: Ok, I'll create an issue.

action-1380?

<trackbot> action-1380 -- Joanmarie Diggs to #presentation should mention aria-hidden vs presentation role on raster and vector images in relation to ACTION-1379 -- due 2015-09-25 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1380

JMD: Haven't done anything.

RS: Can we get a rvised date?

<clown> action-1379

<trackbot> action-1379 -- Richard Schwerdtfeger to TAC should cover rendered descendants from external sources: svg, iframe, etc. (Compose text and get final review form ARIA subteam) -- due 2015-09-25 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1379

JMDRS: This might have to do with presentaitonal and/or hidden images?
... Is it critical to 1.1?

JS: I wrote an action to say that since the elements would either be HTML or SVG, the guidance belongs in those respective mapping docs.

RS: Don't have a problem with that.

JS: One of the first example is an iframe. Does the HTML AAM already cover this?

JN: Does it tell you where to get the acc name if one hasn't been provided?

<clown> http://w3c.github.io/aria/accname-aam/accname-aam.html

RS: Think this says if you put -hidden in an SVG element that has descendents, does that state apply to the child descendents?

JS: That's an SVG question.

RS: No, could be HTML too - an image in an iframe for example.

LW: IAccessible 2 and MSAA seem to map <title> to acc name and <desc> to acc desc, if role="img" is applied to the parent <svg>.

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

JS: SVG name calculation.
... It replaces chunks of the original computation with SVG specifics.

JN: So if my <svg> has both <title> child and title attribute, the <title> child will win out?

RS: There is no title attrib in SVG.

JS: Think it can have an XML title attribute.

RS: For SVG we're covered then, right?

JS: Guess so. The changes don't refer to iframe and img.

JN: Isn't the source image in this issue an SVG?

<clown> <img src="foo.svg"> …

FE: It's when the SVG is a child element I think.

<richardschwerdtfeger> https://svgwg.org/svg2-draft/embedded.html#ImageElement

RS: Core attributes... no mention of title in there.
... It's in the XML attributes list though.
... So SVG is covered, leaving only HTML.
... We could turn that over to HTML to look at?

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

JS: If you want to keep HTML elements out of the acc name, we have to turn it over to HTML AMM.
... What about a <img> that's source is a .svg file fall?

RS: That would be HTML.

LW: If you pull .svg as src to an <img> it's treated like any other img src file.

FE: I don't think that's the case. Will check again.

RS: Ok, think this belongs in HTML.

<mck> back ... was unexpectedly dropped

<clown> acton-1380

<clown> action-1380

<trackbot> action-1380 -- Joanmarie Diggs to #presentation should mention aria-hidden vs presentation role on raster and vector images in relation to ACTION-1379 -- due 2015-09-25 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1380

RS: so this is about the presentation role.
... if you put presentation on an iframe should it descend to its children?

RESOLUTION: Close action 1379, moved to bug 29230 in HTML AAM.

<richardschwerdtfeger> https://www.w3.org/Bugs/Public/show_bug.cgi?id=29230

RS: I think 1380 is talking about HTML?

JS: Can't tell.
... It talks about presentation and -hidden.

LW: I think it is HTML, yes.

RS: If you put aria-hidden on an image, it should go away.
... It's a single entity.

JDM: Presentation doesn't mean it's a single entity.

RS: No, but if it's <img src="bla.png"> it is.

JS: So -hidden and role="presentation" do the same thing?

RS: So the type of img src doesn't matter.

LW: No.

RS: Perhaps we need something for clarity?

JS: Is the issue focusing on the difference between -hidden and presentation?

RS: If you have role="presentation" on a <img> it gets one node in the DOM.

JN: Yes.

<clown> <img role="presentation" src="foo.svg"> -> no mapped in the a11y tree.

RS: Right, so it doesn't matter whether you use -hidden or presentation. It won't be mapped.

LW: Yes.

RS: So we could clarifiy the fact the two do the same thing in this context.

<clown> <img aria-hidden="true" src="foo.svg"> -> not mpaped in the a11y tree.

FE: If you put role="group" on the <img> it doesn't mean you'll get access to the inner SVG does it.

JS: No.

FE: SVG 2.0 doesn't make the distinction between embeded and stand-alone content.

JN: But how the host language treats the source file is down to the host language - in this case HTML.

RS: Don't think we need anyhting in ARIA for this. Anyone disagree?
... Unless we want Joanie to include a clarification sentence?

FE: The weird part is that you can assign any role you want and it won't make any difference.

RS: We could say if the host language treats the img as a single entity... etc.

LW: I can write a clarification sentence and send it to Joanie for inclusion/wordsmithing.

JS: Question is where is this sentence going?

<clown> action-1465

<trackbot> action-1465 -- Shane McCarron to Or shane to fix default values in 1.1 respec -- due 2014-06-30 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1465

<clown> http://w3c.github.io/aria/aria/aria.html#presentation

JMD: This is a change to Respec?

RS: Shall I send an email to Shane?

JMD: I can send it.

action-1475?

<trackbot> action-1475 -- Cynthia Shelly to Work with michael c. and joseph s. to create up to 4 accessibility diagrams showing the application (before and after) of applying role=“presentation”/none -- due 2015-09-25 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1475

<clown> action-1475?

<trackbot> action-1475 -- Cynthia Shelly to Work with michael c. and joseph s. to create up to 4 accessibility diagrams showing the application (before and after) of applying role=“presentation”/none -- due 2015-09-25 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1475

RS: Think this should be ARIA 2.0?

MC: We might want to move it to the practices.

RS: Can we assign this to APG?

JN: We don't want it!

MK: No, it doesn't belong in the APG.
... Seems like it should be somewhere, but it isn't an authoring practice.

<jamesn> http://www.w3.org/TR/aria-in-html/#use-of-role-presentation

RS: Related to issue 522...
... if aria-label and aria-labelledby are undefined, and the element is not presentational, check for the presence of an equivalent host language method for applying a label.

<clown> issue-522

<trackbot> issue-522 -- Name Computation 2A - bullet 3 confusion -- open

<trackbot> https://www.w3.org/WAI/PF/Group/track/issues/522

JN: This is that people think role="presentation" doesn't get included in name computation.
... That's not what it says, but it seems to confuse a lot of people.

JS: We could take your suggestion and put it in the acc name calculation?

JN: We're done then?

<clown> http://w3c.github.io/aria/accname-aam/accname-aam.html#step2D

JS: Well, Rich left a note to say it has closed, but the issue is still open...

JN: Think the edits have been done, but not the suggested diagrams.

MK: If we have diagrams, they should go in the acc name/desc computation doc

<jamesn> http://www.w3.org/TR/aria-in-html/#use-of-role-presentation

JN: Did anyone look at the link to Steve's doc? Think it does a lot of what we're asking here.
... Has diagrams of acc trees etc.
... Maybe we need some extra info in here?

RS: Don't know that we need to worry about a diagram now, so suggest we move it to ARIA 2.0 name calculation.
... Michael can you create a product for the Acc Name Computation spec?

MC: Should be there.

<mck_> have to leave for another meeting ... see you in Japan! happy travels!

<clown> action-1489

<trackbot> action-1489 -- Michael Cooper to Propose spec text to limit what aria attributes can be overridden by strong native semantics (e.g., aria-label and aria-labelledby) -- due 2015-09-25 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1489

action-1849?

<trackbot> Sorry, but action-1849 does not exist.

MC: Need to put a clarification in the spec.

RS: Ok, new date?

MC: 12th Nov, but will try to get to it before then.

action-1490?

<trackbot> action-1490 -- James Nurthen to Propose spec text edit for issue-610: comboboxes should allow complex children elements -- due 2014-10-06 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1490

<clown> action-1490

<trackbot> action-1490 -- James Nurthen to Propose spec text edit for issue-610: comboboxes should allow complex children elements -- due 2014-10-06 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1490

JN: Won't get this done before TPAC!

This is scribe.perl Revision: 1.140  of Date: 2014-11-06 18:16:30  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/rbowser API/browser API/
Succeeded: s/since the images/since the elements/
Found ScribeNick: LJWatson
Inferring Scribes: LJWatson
Present: LJWatson Rich_Schwerdtfeger fesch Joseph_Scheuhammer Joanmarie_Diggs Bryan_Garaventa Matt_King
Found Date: 22 Oct 2015
Guessing minutes URL: http://www.w3.org/2015/10/22-aria-minutes.html
People with action items: