W3C

– DRAFT –
ARIA WG

29 June 2023

Attendees

Present
Matt_King, StefanS
Regrets
-
Chair
ValerieYoung
Scribe
jamesn, jcraig

Meeting minutes

[New Issue Triage](https://tinyurl.com/bdurz7mn)

w3c/core-aam#183

UIA issue, Doug to review for aria-errormessage

w3c/html-aam#494

scotto__: jamesn called me out on underexplaining....

jamesn: needs constraints

scotto__: could be an easier pattern in the UA so authors don't have to

spectranaut_: add agenda+

<jcraig> s/jn/jamesn: needs constraints/

[New PR Triage](https://tinyurl.com/bdeb2e63)

[Deep Dive planning](https://bit.ly/aria-meaty-topic-candidates)

<spectranaut_> w3c/aria#939

spectranaut_: role parity summary issue

week of July 27th

spectranaut_: consider use cases for aria-modal

scotto__: need aleventhal for that one

[Consider supporting aria-checked on gridcells](https://github.com/w3c/aria/issues/1960)

sarah_h_: from probs with aria-selected and mixed state selection

esp if you have a selectable grid... checked column. select all rows via checkbox

when arrowing through the grid, this would help to keep focus on the check box gridcell itself

would need AAM mapping

jamesn: question: do we have examples if current usage and how it works? needs testing standard patterns across platforms

sarah_h_: not great support.

jamesn: would help progress if test cases provided

<sarah_h_> example use case: https://react.fluentui.dev/?path=/docs/components-datagrid--default

MarioBatusic: why should it be possible to uncheck a row if you already have a checkbox there

understand need for selection, but not unchecking

sarah_h_: added example to the chat.. changes windows screen reader modes

Matt_King: i think these are screen reader bugs... maybe not cause for an ARIA change

would we have to write something similar to what we did for option and treeitem so that authors would not use both in conflicting ways

changed and provided guidance for authors and UAs to not provide both... with UA MUST error handling

with grid, there aren't well established patterns... Not sure I want to hear "selected" on every cell... That may be tangential

sarah_h_: currently standard practice for WIndows SRs to switch modes... would be a big change

Matt_King: not in JAWS, excepting one known bug with some links

sarah_h_: have tested recently, but will verify the version I was using

helpful to have that info exposed

gives example about file selection... filename including selection state was helpful... even though no checkbox provided

re: checked vs selected.... Agree with you to avoid that anti-pattern

* discussing some very specific user testing results

Matt_King: mentions another example in Google Sheets selection

sarah_h_: clarifying checked vs selected may need to be a prerequisite for this

Matt_King: in gridcell role, could add the same language in authoring guide from treeitem role

sarah_h_: will add PR and also talk to browser folks

Matt_King: need to coalesce on rows in tree vs treegrid

may need to add the same language on the row role too

[Discussion tracking for ARIA Notification proposal](https://github.com/w3c/aria/issues/1957)

Doug: at F2F broke down into 7 different discussion groups

Doug: based on that we created individual discussion groups.

Doug: wanted to verify hit everyones issues

Doug: also wanted to go through individually

Doug: also want to go back to API doc and go back and change based on feedback

Doug: want to look into what is best course of action to move this forward

spectranaut_: might be good to go through some in the meetings maybe a few at a time until we get through them

Doug: some easier than others

Matt_King: there is a google doc with a spec

Matt_King: this API would have to be implemented at the platform level by both IA2 and UIA to work (on windows)? Who is the implementor here?

Doug: the browser would implement based on the best platform available

cyns: chrome will imeplement on IA2

Matt_King: does it require changes to IA2?

Doug: don't believe IA2 has what is needed. Believe UIA is close

Matt_King: wish the spec could be in comething other than a google doc

Matt_King: could it move into a repo?

jcraig: WICG?

cyns: I think Travis set up a repo

scotto__: based on the google doc it doesn't look like there are any comments

scotto__: the GH thread was to allow the discussion to be open

scotto__: shouldn't let that stop us talking about it now

scotto__: we could even put it in a wiki on ARIA

scotto__: changes could go there on a discussion thread

Matt_King: this spec needs to live somewhere

Matt_King: where does this go?

Doug: don't know the answer

jamesn: WICG seems best to me

<Zakim> jcraig, you wanted to mention the WICG move should not necessarily be tied to a solution or spec

jcraig: would call this something like Accessibility Notifications. Do not tie to proposed solution

cyns: is there a good reason not to add it to AOM

cyns: seems easier

jcraig: reasonable to be a topic in that meeting but would prefer its own repo

jcraig: the primary reason to incubation groups is to keep small groups of people very task-focused

spectranaut_: lets go through discussion points

Doug: Does the notification API need to support Braille?

Doug: do we need to do this up front - do we do it later?

jcraig: absolutely we need it

jcraig: hearing screen reader users, deaf-blind screen reader users and hybrid

jcraig: shorthand braille notifications is very useful. Some extension to that notification is useful

jcraig: absolutely consider up front

Doug: there could be folks that don't understand the difference and may use incorrectly

jcraig: we put a lot of through into braillelabel wording - its a tool and folks can do the wrong thing

Matt_King: do you want to move deeper and find what the braille requirements might be

"Would the notification API need to provide support for separate braille specific notification strings? What are people's thoughts on this being a v1 release feature or if this is maybe something that could be rolled out as part of future updates?"

Mario: screen readers control braille and web authors may not know the right thing for the braille

jcraig: I also missed the comment that this is about v1. should be considered for v1 but not be on the critical path for v1

cyns: what james just said. need a path forward but doesn't have to be in v1

Doug: has helped a lot to move forward

spectranaut_: can continue to discuss in the meeting

Minutes manually created (not a transcript), formatted by scribe.perl version 210 (Wed Jan 11 19:21:32 2023 UTC).

Diagnostics

Succeeded: s/jn/jamesn: needs constraints

Failed: s/jn/jamesn: needs constraints/

Succeeded: s/@@@/option and treeitem/

Succeeded: s/the reason for incubation groups is to keep small tasks focussed/the primary reason to incubation groups is to keep small groups of people very task-focused/

Maybe present: cyns, Doug, jamesn, jcraig, Mario, MarioBatusic, re, sarah_h_, scotto__, spectranaut_

All speakers: cyns, Doug, jamesn, jcraig, Mario, MarioBatusic, Matt_King, re, sarah_h_, scotto__, spectranaut_

Active on IRC: dmontalvo, jamesn, jcraig, Matt_King, sarah_h_, spectranaut_, StefanS