Minutes: March 17, 2016 ARIA Working Group Meeting

https://www.w3.org/2016/03/17-aria-minutes.html

Text:

- DRAFT -

Accessible Rich Internet Applications Working Group Teleconference

17 Mar 2016

See also: IRC log <http://www.w3.org/2016/03/17-aria-irc>
Attendees <>
Present
Michiel_Bijl, Fred_Esch, Ja_Eun_Ku, Rich_Schwerdtfeger, Joanmarie_Diggs, Joseph_Scheuhammer, Bryan_Garaventa, janina
Regrets
Chair
Rich
Scribe
jongund, matt_king
Contents

Topics <https://www.w3.org/2016/03/17-aria-minutes.html#agenda>
Technical Plenary <https://www.w3.org/2016/03/17-aria-minutes.html#item01>
ARIA 1.1 test effort <https://www.w3.org/2016/03/17-aria-minutes.html#item02>
Combobox <https://www.w3.org/2016/03/17-aria-minutes.html#item03>
Action 1743 <https://www.w3.org/2016/03/17-aria-minutes.html#item04>
Password Role <https://www.w3.org/2016/03/17-aria-minutes.html#item05>
ACTION 1489 <https://www.w3.org/2016/03/17-aria-minutes.html#item06>
Computed name and role <https://www.w3.org/2016/03/17-aria-minutes.html#item07>
ACTION 1624 <https://www.w3.org/2016/03/17-aria-minutes.html#item08>
ACTION 2023 Role=Text <https://www.w3.org/2016/03/17-aria-minutes.html#item09>
Summary of Action Items <https://www.w3.org/2016/03/17-aria-minutes.html#ActionSummary>
Summary of Resolutions <https://www.w3.org/2016/03/17-aria-minutes.html#ResolutionSummary>


<Rich> https://lists.w3.org/Archives/Public/public-aria/2016Mar/0124.html <https://lists.w3.org/Archives/Public/public-aria/2016Mar/0124.html>
<Rich> https://lists.w3.org/Archives/Public/public-aria/2016Mar/0124.html <https://lists.w3.org/Archives/Public/public-aria/2016Mar/0124.html>
<Rich> scribe: jongund

Technical Plenary

<Rich> Technical Plenary - 2016 (September 19-23; Lisbon, Portugal). Who will be able to attend?

<joanie> I will be there.

RS: WHois coming?

CS: I can come

MB: I can come

<Rich> I will try

FE: I can

RS: I am

MK: Maybe

JG: no

MC: Yes

<jamesn> JN: I will try - if i can get approvals

RS: We have enough to do real work

<joanie> JD: I will definitely be there

MC: Definiateley, I was assuming, the main question is who we want to meet with

RS: I am hoping september we will be well into testing
... We need to hook up with web apps
... CS what is going on with the community group

CS: Not a whole lot

RS: What to do something joint?

CS: Let me get back to you on that

RS: What are we going to do on personalization, if media queries are off the table, then we need an API

CS: What type?

RS: In order for us to address cognitive im[airments we need to know what the user needs, this can be done other ways but we need something from the browser
... We need to detect things like S/N are low and captions are needed
... Smarter apps need more information
... Doesn't mean that the API comes out of this group

CS: I am not sure rosen is the right person

RS: I need to setup a page

MC: There is a survey of chairs what they are planning for there working group

RS: We can split things with the ....

MC: You and JS coordinate on meeting schedule

RS: JS needs to be there the whole week

MK: If ARIA can be Thursday and Friday it would be better for me

RS: I don't have a problem

CS: I will be their the whole week

JN: It is in September

RS: RS: I will ask for thursday and friday

MC: You need to fill out the survey

RS: There will be no ARIA working group meeting March 24th

<clown> https://www.w3.org/2016/09/TPAC/ <https://www.w3.org/2016/09/TPAC/>
RS: Where is JG on the testing

<mck> scribe: matt_king

ARIA 1.1 test effort

<mck> rs: Michael has created exit criteria identifiying aria 1.1 changes

<mck> rs: defines what is needed for test statements

<mck> rs: We need people who have volunteered to write test statements

<mck> rs: It is a bit of an art to do it.

<MichaelC> draft status for CR with exit criteria: https://rawgit.com/w3c/aria/CR-pub/aria/aria.html#sotd <https://rawgit.com/w3c/aria/CR-pub/aria/aria.html#sotd>
<mck> jg: In this round, things that have not changed from 1.0, will they be tested?

<mck> rs: Right, however, we have a new col in AAM for UIA

<Zakim> clown, you wanted to ask if the exit criteria delimits what needs to be tested.

<mck> So, we need to test all aria features for Edge , including 1.0

<mck> CS: We are automating that and making good progress.

<mck> Joanie: I have added about 200 assertions to webkit

<mck> Fred and I need to talk for SVG

<mck> RS: Joanie, do you we can automate IA2 or Ma?

<mck> CS: Should be easy to port the UIA automated testing to other platforms; will be able to say more in a couple weeks.

<mck> Joanie: I will look into Mac

<mck> rs: Has MS created 1.1 test cases, or just 1.0?

<mck> CS: we have done some of 1.1

<mck> RS: can you share with JF?

<mck> rs: Have we heard more from Alex about MSAA UIA Express?

<mck> CS: No

<mck> RS: I don't think we are going to need that if we go to common mappings in the future; not good use of time.

<mck> rs: The exit criteria section includes everything we need testable statements for

<mck> RS: Where is CRPub? Can we get it with git pull?

<mck> MC: Us git sync

<mck> rs: John circle back with cynthia and JF's team.

<mck> CS: can we talk about it at CSUN?

Combobox

<Rich> http://rawgit.com/w3c/aria/action1490-combobox/aria/aria.html#combobox <http://rawgit.com/w3c/aria/action1490-combobox/aria/aria.html#combobox>
MK: Jaws versions causing crashing problems
... Action 1490 branch
... To make this time most useful, I would like to go through the changes from discussion at least weeks meeting
... I want to go through this section of without discussing everything anout tree, gird and dialog
... The first think is that you didn't like listbox being a popup

RS: I think that's good

MK: Second paragraph is all about the test box element
... I think every body agrees it should be a single line text box

RS: A textbox by default is single line

MK: We need to explicitly define it is single line

JS: Or you could say line terminators

MK: I think it is OK the way it is
... Next is about expand collapse
... We elevated this from should to must, put on textbox element
... When a combobox is expanded, authors MUST ensure it contains or owns an element that has a role of listbox, tree, grid, or dialog. This element is the combobox popup. Elements with the role combobox have an implicit aria-haspopup value of true.

RS: exapnd goes on the combobox

MK Correct

MK: The fourth one is about focus...
... Authors should put focus on the textbox, authors should move focus....
... To be keyboard accessible, authors SHOULD manage focus of descendants for all instances of this role, as described in Managing Focus. When a combobox receives focus, authors SHOULD ensure focus is placed on the textbox element. Authors SHOULD provide keyboard mechanisms for moving focus to and from the textbox element and the pop-up. For example, one common convention is that Down Arrow...
... moves focus from the text input to the first focusable descendant of the pop-up element. If the popup element supports aria-activedescendant, authors MAY specify a value for aria-activedescendant on the textbox element that refers to a descendant of the popup element.
... ARIA activedescendant can point to the element in the list box that it owns

RS: They could put it on the combobox?

MK: The listbox does not have focus

RS: When you say it maybe on the textbox or the list

MK: If you put DOM focus on the listbox it meets the current requirement
... They can put DOM focus on the UL element

RS: You don't think authors will get confused

MK: In this particular spec language

RS: I understand it, but new users might get confused

CS: I have another question

<Zakim> MichaelC, you wanted to note DPub also starting to think test cases

MK: Maybe in the last sentence I can add
... It starts with if the element the author wants to keep focus on the textbox .....

RS: We have people on the cue

MC: Deque is testing accessibility API mappings

CS: So would you expect autocomplete on Google or bing would use this pattern

<bgaraventa1979> +q

MK: I am not familiar with bing, but they should work with facebook and google

CS: You think of them as comboboxes?

MK: If you want to talk about the definition of combobox

CS: That is not where I was going

MK: I wrote it in broad terms about down arrow....

CS: There is a feature in Narrator that if can a textbox or autocompelete, but if there could be an aria-controls

MK: The AT vendors wanted some siblings relationship, they didn't care about controls, if it is an authoring requirement..

CS: If it is an authoring practices and examples it will most likely be there

MK: We don;'t have aria-controls on grids for sorting

RS: I don't think it is alot of work

MK: It is a decision related to making is easier for authors or for ATs

CS: It this author doesn't use autocomplete then there is DOM inspection that is fragile
... autocomplete is a may

MK: There are plenty of examples where autocomplete is not needed
... The main thing for the browser is that it is in the expanded state that there are two controls, that is a must

CS: I need a controller relationship

MK: Your advocating an author should for aria-controls

CS: On the listbox and listbox

RS: Or anything that is open
... The textbox could be a search box

MK: Good point I forgot the search role
... The way it is written it is OK, searchbox is a descendant of textbox

RS: I think we need to make that explicit
... We are running out of time
... I don't have a problem with this, but I don't think you want a controls for a switch

MK There is an author must...

MK: Is everything is OK with the example, we have consensus on anything else

JS: I sent an editorial comment

<clown> mck: my editorial suggestions are here: https://lists.w3.org/Archives/Public/public-aria/2016Mar/0191.html <https://lists.w3.org/Archives/Public/public-aria/2016Mar/0191.html>
BG: COuld I get an answer to the questions I asked earlier today

MK: I have not got to the email yet

RS: I think we need to tightened

MK: This is not wide open
... We have an author must statement, on exactly on how it should be constructed

RS: Authors must have owns...., I guess it is in two separate sentences
... Is everyone OK with the dialog box

MK: CSUN is a good place to talk about this

RS: I am not a big fan of dialog box, there are some people who are not comfortable with it

MK: We want to limit unnecessary role proliferation, I am happy we have made this much progress, ....

<jamesn> Date picker is an example of dialog right?

Action 1743

<clown> action-1743

<trackbot> action-1743 -- Joanmarie Diggs to Put aria-activedescendant on application and request wg review -- due 2016-02-17 -- PENDINGREVIEW

<trackbot> http://www.w3.org/WAI/ARIA/track/actions/1743 <http://www.w3.org/WAI/ARIA/track/actions/1743>
RS: Putting aria-activedescendant on role=application

JOANIE: I ahve not done anything with this

MK: Can we put that on the agenda for after CSUN

RS: Joanie has a branch
... Can we assign these to you MK

JOANIE: Fell free to make changes in my brand

MK: ONly in application and activedesendant

RS: What date can you come back to the group

MK: 2 weeks

RS: Need to find the agenda

Password Role

<joanie> https://rawgit.com/w3c/aria/password-role/aria/aria.html#password <https://rawgit.com/w3c/aria/password-role/aria/aria.html#password>
<clown> action-2004

<trackbot> action-2004 -- Joanmarie Diggs to Create proposal for password role -- due 2016-03-10 -- PENDINGREVIEW

<trackbot> http://www.w3.org/WAI/ARIA/track/actions/2004 <http://www.w3.org/WAI/ARIA/track/actions/2004>
RS: Joanie you want to put int he branch

<joanie> https://rawgit.com/w3c/aria/password-role/aria/aria.html#password <https://rawgit.com/w3c/aria/password-role/aria/aria.html#password>
RS: Issues with supported properties and states
... This looks ok to me

JS: What would you put in for placeholder

Joanie: That would change the how ATs would want to see the role exposed

RS: HTML5 type=password support placeholder

JN: Yes people use it all the time

RS: I can see where you would use it

JN: Make and model of a car

RS: Goven that do people have issues with the new Password role?

MK: I am still reading the text

JOANIE: I thought we said haspopup should not be supported

RS: It could be help information

MK: Looks good to me

RESOLUTION: Accept joanie's proposal, and close action
... Accept joanie's proposal for new ROLE=PASSWORD, and close action 2004

JOANIE: I'll close it when its merged

MC: They are not at the top of the page
... But they are there
... These are no W3C announcements on working draft changes
... But the groups can publicize them and send them to people

JOANIE: Did you what me to remove ...
... Input for entering the password

MK: yes

ACTION 1489

https://www.w3.org/WAI/ARIA/track/actions/1489 <https://www.w3.org/WAI/ARIA/track/actions/1489>
RS: Move one

https://www.w3.org/WAI/ARIA/track/actions/2040 <https://www.w3.org/WAI/ARIA/track/actions/2040>
<clown> issue-686

<trackbot> issue-686 -- need a getComputedLabel or getComputedName -- open

<trackbot> http://www.w3.org/WAI/ARIA/track/issues/686 <http://www.w3.org/WAI/ARIA/track/issues/686>
Computed name and role

RS: We met witht he group 2 years ago
... No action in 2 years, so move to ARIA 2.0

MK: We kind of have to

Proposed moving Issue 686 on APIs on roles and accessible name to ARIA 2.0

MC: We don't need a CFC for moving it

<Rich> https://www.w3.org/WAI/ARIA/track/issues/686 <https://www.w3.org/WAI/ARIA/track/issues/686>
RESOLUTION: MOve Issue 686 to ARIA 2.0

ACTION 1624

<clown> action-1624

<trackbot> action-1624 -- Michael Cooper to Work with shane to develop publication strategy and solution that would exclude abstract roles for authors -- due 2016-03-17 -- OPEN

<trackbot> http://www.w3.org/WAI/ARIA/track/actions/1624 <http://www.w3.org/WAI/ARIA/track/actions/1624>
MC: Can you put this higher on list after CSUN

MK: What do links do ....

JS: Point to the full spec

MC: We have to sort out in the process of doing the actions

<clown> action-2023

<trackbot> action-2023 -- Joanmarie Diggs to Write a proposal about how to modify the definition of role text to limit its use. -- due 2016-02-23 -- OPEN

<trackbot> http://www.w3.org/WAI/ARIA/track/actions/2023 <http://www.w3.org/WAI/ARIA/track/actions/2023>
ACTION 2023 Role=Text

<clown> issue-1011

<trackbot> issue-1011 -- For role="text", there needs to be requirements on user agents for overrides, such as controls, events, and other global aria properties, like there is for role="presentation". -- open

<trackbot> http://www.w3.org/WAI/ARIA/track/issues/1011 <http://www.w3.org/WAI/ARIA/track/issues/1011>
ACTION-2023: Write a proposal about how to modify the definition of role text to limit its use.

<trackbot> Notes added to ACTION-2023 Write a proposal about how to modify the definition of role text to limit its use..

Write a proposal about how to modify the definition of role text to limit its use.

RS: When you have text content a bit differently
... I know you can put this in button
... If you put a link inside a role=text, that that would be dangerous
... ***** role=text with aria-label="4 stars"
... IN the case of buttons the roles are presentations
... role=text can not have interactive elements

FE: I want to make sure it can work in SVG with a path used for text

RS: That would not be an issue

JN: I want to make sure we solve the problem for people who write good code, bad coders will always do bad things

MK: My understanding of the role=text is making content easier to listen to, so it is getting into the realm ...
... Realm of presentation, so the author is making a judgement that is different structure or roles

<cyns> +1

MK: I want to make sure that the original content is available tot he AT user

<Zakim> clown, you wanted to say it could be handled as exceptions to role presentation.

MK: FOr example if it is a graphic I can get to the original image

JS: If you put presentation on a link it the browser will ignore the role and present the link

RS: I would go with that

CS: I have a cautious approval, I have concerns about authoring

RS: The next question is this...., Joanie had the action
... Should Joanie come back with a proposal?

<clown> +1

<mck> +1

<cyns> +1

RS: Anything that is a descendant that is interactive will invalidate the role

<Rich> +1

JN: What it is has tabindex=-1 and aria-hidden?

CS: We need to look at a draft

JN: James put ...

Joanie: I don't like the role, so I need everyones comments

MK: The text role provides an alternative presentation, user agents must not hide the accessible

JS: What about title attribute?

MK: It is kind of like that

RS: We are out of time

JOANIE: What is my next step?

<clown> https://www.w3.org/WAI/ARIA/track/issues/1011 <https://www.w3.org/WAI/ARIA/track/issues/1011>
RS: JN can get you a link

JN: I can't because W3C is blocking Oracle

MC: This happens, there is page that explains it

<clown> joanie: inside that issue is the link to github where the discussion occurred (in a note)

<jamesn> i get directed here - https://www.w3.org/Help/abuse-info/re-reqs.html <https://www.w3.org/Help/abuse-info/re-reqs.html>
I published the minutes

Summary of Action Items <>
Summary of Resolutions <>
Accept joanie's proposal, and close action <https://www.w3.org/2016/03/17-aria-minutes.html#resolution01>
MOve Issue 686 to ARIA 2.0 <https://www.w3.org/2016/03/17-aria-minutes.html#resolution02>


Rich Schwerdtfeger