16:53:59 <RRSAgent> RRSAgent has joined #aria
16:54:03 <RRSAgent> logging to https://www.w3.org/2025/07/10-aria-irc
16:54:03 <Zakim> RRSAgent, make logs Public
16:54:04 <Zakim> Meeting: ARIA WG
16:54:09 <jamesn> agendabot, find agenda
16:54:09 <agendabot> jamesn, OK. This may take a minute...
16:54:10 <agendabot> agenda: https://www.w3.org/events/meetings/5a155237-d896-464b-9c5f-6dd1654293ae/20250710T130000/
16:54:10 <agendabot> clear agenda
16:54:10 <agendabot> agenda+ -> New PR Triage https://github.com/search?l=&q=is%3Aopen+is%3Apr+created%3A%3E%3D2025-07-03+repo%3Aw3c%2Faria&type=Issues
16:54:10 <agendabot> agenda+ -> New Issue Triage https://tinyurl.com/2s3ah8u9
16:54:13 <agendabot> agenda+ -> WPT Open PRs https://bit.ly/wpt_a11y
16:54:15 <agendabot> agenda+ -> Deep Dive planning https://bit.ly/aria-meaty-topic-candidates
16:54:18 <agendabot> agenda+ Registration open for TPAC 2024 in mid July
16:59:09 <Francis_Storr> Francis_Storr has joined #aria
17:00:26 <Adam_Page> Adam_Page has joined #aria
17:00:50 <filippo-zorzi> filippo-zorzi has joined #aria
17:01:34 <aardrian> aardrian has joined #aria
17:01:34 <aardrian> present+
17:01:45 <hdv> present+
17:01:55 <hdv> scribe: hdv
17:01:58 <Adam_Page> scribe: Adam_Page
17:02:03 <Adam_Page> zakim, next item
17:02:03 <Zakim> agendum 1 -- -> New PR Triage https://github.com/search?l=&q=is%3Aopen+is%3Apr+created%3A%3E%3D2025-07-03+repo%3Aw3c%2Faria&type=Issues -- taken up [from agendabot]
17:02:07 <Adam_Page> present+
17:02:15 <spectranaut_> spectranaut_ has joined #aria
17:02:26 <Adam_Page> jamesn: no issues to triage?
17:02:34 <front-endian-jane> front-endian-jane has joined #aria
17:02:35 <Adam_Page> ... anyone?
17:02:42 <spectranaut_> present+
17:02:50 <front-endian-jane> present+
17:02:52 <spectranaut_> agenda?
17:03:04 <Adam_Page> zakim, close this item
17:03:04 <Zakim> agendum 1 closed
17:03:05 <Zakim> I see 4 items remaining on the agenda; the next one is
17:03:05 <Zakim> 2. -> New Issue Triage https://tinyurl.com/2s3ah8u9 [from agendabot]
17:03:07 <Adam_Page> zakim, next item
17:03:07 <Zakim> agendum 2 -- -> New Issue Triage https://tinyurl.com/2s3ah8u9 -- taken up [from agendabot]
17:03:12 <Francis_Storr> present+
17:03:49 <smockle> present+
17:04:03 <Adam_Page> jamesn: starting with June 26th
17:04:05 <Adam_Page> ... 3 new issues
17:04:09 <Adam_Page> ... first is editorial
17:04:27 <Adam_Page> ... allowed accessibility child roles on performance
17:04:36 <Adam_Page> ... needs someone to look at it
17:04:37 <Adam_Page> ... any volunteers?
17:05:05 <Adam_Page> jcraig: I’ll take it and add context
17:05:41 <Adam_Page> ... it’s not that an author should avoid too many elements, the point is that if we allow infinite nesting then we’re forcing the web engine to do that computation all the time
17:05:51 <Adam_Page> ... the note should just be for authors to be reasonable
17:06:04 <Adam_Page> jamesn: next is aria#2569
17:06:14 <Adam_Page> ... should label override content
17:06:18 <Adam_Page> ... briefly discussed in WG call
17:06:26 <Adam_Page> jcraig: I can also give context to this
17:06:34 <Rahim> present+
17:06:56 <Mario> Mario has joined #aria
17:07:11 <Adam_Page> ... what if there were something deeper in, like image title attribute, etc. what happens if you put an aria-label at the top, does it cascade and cause children to become presentational?
17:07:17 <Adam_Page> ... or could there be a smarter hybrid
17:07:28 <scott> scott has joined #aria
17:07:37 <scott> present+
17:07:42 <Adam_Page> aardrian: isn’t it limited to certain roles?
17:07:50 <Mario> https://github.com/w3c/aria/issues/2567
17:07:57 <Adam_Page> ... my kneejerk is that this is about scoping model and other things correctly
17:08:13 <Adam_Page> jcraig: stevef’s article isn’t about model, that’s just my secondary consideration
17:08:17 <Adam_Page> ... please read and comment on the issue
17:08:27 <Adam_Page> jamesn: this sounds like a great TPAC candidate
17:08:28 <filippo-zorzi> present+
17:08:31 <scott> present+
17:08:42 <Adam_Page> ... F2F
17:09:30 <Adam_Page> jamesn: last one is aria#2567
17:09:42 <Adam_Page> ... should toolbar role be widget instead of doc structure
17:09:54 <Adam_Page> Mario: recommend changing category
17:10:00 <Adam_Page> ... toolbar is really a region, a shortcut for menu
17:10:04 <jamesn> q+
17:10:17 <Adam_Page> ... a toolbar has to manage the focus for its sub elements
17:10:42 <Adam_Page> ... toolbar should be region
17:10:52 <Adam_Page> ... when you put focus on a button in a toolbar, then NVDA switches to reading mode — not good
17:11:03 <Adam_Page> ... I normally can’t use arrow keys to navigate to other parts of the toolbar
17:11:10 <Adam_Page> ... manually have to change back to navigation mode
17:11:15 <jamesn> ack me
17:11:16 <Adam_Page> ... maybe we should change the category in the spec
17:11:41 <Adam_Page> jamesn: I seem to remember the reason that toolbars don’t state that authors MUST manage focus is because Apple does this differently
17:11:54 <Adam_Page> ... and that’s why this is categorized the way it is
17:12:06 <Adam_Page> jcraig: I don’t recall that
17:12:10 <Adam_Page> ... but it’s very possible
17:12:32 <spectranaut_> found this old issue: https://github.com/w3c/aria/issues/713
17:12:41 <spectranaut_> Why is toolbar as a document structure role? #713
17:12:42 <Adam_Page> jamesn: maybe we should do some digging to determine why it’s categorized this way
17:13:02 <sarah> sarah has joined #aria
17:13:07 <Adam_Page> jcraig: toolbars in iOS did get some special handling
17:13:25 <Adam_Page> ... oh toolbar, yes I remember — I was thinking scrollbar
17:13:32 <Adam_Page> ... at the time Windows toolbars were very focused on keyboard navigation
17:13:39 <Adam_Page> ... kind of an exclusionary set like radio buttons
17:13:46 <Adam_Page> ... so managed focus was almost a requirement for this
17:14:02 <Adam_Page> ... but we chose MAY because no toolbars in macOS were managed focus
17:14:20 <Adam_Page> ... we have “segmented controls”, which are kind of like radio buttons
17:14:24 <Adam_Page> ... there’s “pop up button”
17:14:42 <Adam_Page> ... a few other things that would conflict with the idea of a toolbar always having managed focus
17:14:42 <Adam_Page> ... so I think we need to keep this as is
17:15:03 <Adam_Page> jamesn: maybe we can investigate it more, since it definitely causes a problem with NVDA switching modes
17:15:09 <Adam_Page> ... but NVDA could decide to change their behavior?
17:15:12 <aardrian> Q+
17:15:34 <Adam_Page> jcraig: here’s a strawman: could we give some explicit flag for managed focus true/false?
17:15:45 <Adam_Page> ... I think we may have considered and dismissed that in the past
17:15:51 <Adam_Page> jamesn: yes, we have
17:16:42 <Adam_Page> ... could be worthwhile to investigate
17:16:42 <Adam_Page> ... there was an `aria-interactive` idea that might be related
17:16:47 <Adam_Page> ... proposed as a more generic thing
17:16:54 <Adam_Page> ... we should start out small, and expand later on
17:17:21 <jamesn> ack aa
17:17:31 <Adam_Page> aardrian: my chaos approach would be let’s just call it a toolbar
17:17:36 <Adam_Page> ... aligns with many expectations
17:17:39 <Adam_Page> ... and aligns with Windows
17:17:54 <Adam_Page> ... and since Apple has different UI, it just gets documented
17:18:05 <Adam_Page> ... if enough implementations want it to be a toolbar, then we just document exceptions
17:18:16 <cyns> cyns has joined #aria
17:18:23 <Adam_Page> jcraig: the problem there is that we’re depending on is it a Windows user expectation or a Mac user expectation
17:18:42 <Adam_Page> ... the author MAY or may not manage focus
17:18:42 <Adam_Page> ... the technology would’t do that by default
17:19:11 <Adam_Page> aardrian: I was surprised to see that it’s a `document`
17:19:15 <Adam_Page> ... but yes, that’s a good point
17:19:28 <Adam_Page> Mario: could we do something more?
17:19:37 <Adam_Page> ... some more discussion
17:19:49 <Adam_Page> jamesn: if someone comes up with a proposal, then we can discuss that
17:19:54 <Adam_Page> jcraig: I have a proposal
17:20:03 <Adam_Page> ... consider moving this to authoring practices repo
17:20:12 <Adam_Page> ... and documenting that it’s reasonable for an author to manage focus on a toolbar
17:20:19 <sarah> q+
17:20:20 <Adam_Page> ... and VO on Mac can handle that if they do
17:20:42 <Adam_Page> ... but it’s also reasonable to not manage focus, to simulate Mac-like experience
17:20:49 <Adam_Page> ... then the question is what do we do for Windows screen readers?
17:20:57 <Adam_Page> ... or Linux
17:21:08 <Adam_Page> ... either way, reasonable to have a separate documented page about this platform difference in authoring practices
17:21:12 <jamesn> ack sa
17:21:18 <Adam_Page> sarah: quick question about cross-platform difference
17:21:28 <Adam_Page> ... I’ve done a lot of cross-platform user studies
17:21:38 <Adam_Page> ... never seen VO users have an issue with managed focus
17:21:46 <Adam_Page> ... it doesn’t matter since they’re using VO commands
17:21:49 <StefanS> StefanS has joined #aria
17:22:05 <Adam_Page> ... how big is the actual impact on Mac VO users
17:22:26 <Adam_Page> ... how does it’s categorization matter?
17:22:26 <StefanS> present+
17:22:42 <Adam_Page> ... to me, `widget` does make sense
17:22:48 <spectranaut_> +1 to try to make the spec make authoring sense
17:22:48 <Adam_Page> ... does moving it to a `widget` role have any bearing on what authors should do?
17:23:05 <Adam_Page> jcraig: good point; no objection to the taxonomy change as long as it doesn’t change the authoring requirement
17:23:10 <Adam_Page> sarah: another example
17:23:16 <Adam_Page> ... in Windows, you tab between menu items
17:23:26 <Adam_Page> ... even in Fluent UI, we had this discussion
17:23:29 <Adam_Page> ... do you tab between items
17:23:33 <Adam_Page> ... role had no bearing
17:24:00 <Adam_Page> jcraig: we would have to change the def of widget, if we moved it there
17:24:12 <Adam_Page> ... this would not be a discrete UI object
17:24:19 <sarah> q?
17:24:30 <scott> a row is a widget.
17:24:42 <Adam_Page> jamesn: `widget` doesn’t sound appropriate
17:24:42 <Adam_Page> ... for this multiple use
17:24:48 <Adam_Page> jcraig: especially nested managed focus widgets
17:25:00 <Adam_Page> ... managed focus toolbar that contains a managed focus radio buttobn
17:25:07 <jamesn> q?
17:25:14 <jamesn> agenda?
17:25:14 <Adam_Page> Mario: JAWS understands it as a managed focus widget
17:25:18 <Adam_Page> ... and you can automatically use arrow keys
17:25:22 <Adam_Page> ... but NVDA doesn’t
17:25:31 <Adam_Page> sarah: yes, I think both JAWS and Narrator agree on that
17:25:41 <Adam_Page> ... and I think NVDA probably should too
17:25:49 <Adam_Page> ... maybe should bring this up with them independently
17:26:00 <scott> magic.  magic is in narrator
17:26:38 <jcraig> q+
17:26:42 <Adam_Page> jamesn: plainly this is not a simple change
17:26:44 <Adam_Page> ... if someone wants to come up with a proposal
17:26:48 <Adam_Page> ... then we can review and discuss
17:26:55 <Adam_Page> ... just remember platform differences
17:27:18 <Adam_Page> jcraig: Mario, the scenario you’re talking about, is it simple toolbar buttons?
17:27:35 <Adam_Page> ... because we’re considering more complex patterns
17:28:09 <Adam_Page> ... is the concern just for a one-level deep flat toolbar?
17:28:23 <Adam_Page> Mario: for example, in a horizontal toolbar, you need to be able to use arrow left and arrow right to go to every child
17:28:28 <Adam_Page> ... or in a vertical, to use up and down
17:28:42 <Adam_Page> ... it’s managed
17:29:12 <scott> agreed.  even changing the spec would still mean nvda would need to change
17:29:16 <scott> just changing the spec wouldn't do anything
17:29:19 <Adam_Page> jcraig: I agree with sarah that the issue should be raised with NVDA
17:29:29 <Adam_Page> Mario: I’ll file an issue with NVDA
17:29:33 <Adam_Page> zakim, next item
17:29:33 <Zakim> I see a speaker queue remaining and respectfully decline to close this agendum, Adam_Page
17:29:35 <jamesn> ack jcraig
17:29:41 <Adam_Page> zakim, next item
17:29:41 <Zakim> agendum 3 -- -> WPT Open PRs https://bit.ly/wpt_a11y -- taken up [from agendabot]
17:29:52 <Adam_Page> jamesn: anything new?
17:30:05 <Adam_Page> jamesn: wpt#53676
17:30:09 <Adam_Page> Rahim: I had a question on this
17:30:17 <Adam_Page> ... keithamus, you raised this
17:30:22 <Adam_Page> ... are these crash tests reliable?
17:30:42 <Adam_Page> ... questions around availability of a11y tree relative to DOM load event
17:30:51 <Adam_Page> keithamus: yeah, it was jteh who raised this initially
17:30:53 <Adam_Page> ... and you’re right
17:30:57 <Adam_Page> ... the crash tests just wait for the load event
17:31:06 <Adam_Page> ... and if the page is still running, then it shuts down and assumes a pass
17:31:10 <Adam_Page> ... but the way Gecko works
17:31:17 <Adam_Page> ... we have an out-of-process a11y manager that handles all the a11y logic
17:31:25 <Adam_Page> ... and communicates via RIC
17:31:29 <Adam_Page> ... fully asynchronous
17:31:42 <Adam_Page> ... so it’s problematic
17:31:48 <Adam_Page> ... so crash tests aren‘t conclusive for Gecko
17:32:00 <Adam_Page> jcraig: right, is it crashing the main process, or the a11y process off the thread
17:32:10 <Adam_Page> ... in the results, Firefox is passing even though there is presumably a crash
17:32:16 <Adam_Page> ... is there harm in checking this in?
17:32:30 <Adam_Page> ... I think no, it’s just not effective
17:32:37 <spectranaut_> q+
17:32:42 <Adam_Page> keithamus: just the false sense of security
17:32:43 <Adam_Page> ... the a11y process may well still be crashing
17:32:53 <Adam_Page> jcraig: should we raise a new issue on Firefox?
17:33:02 <Adam_Page> ... if this out-of-process crash happens, it should crash the main thread?
17:33:11 <Adam_Page> keithamus: yes, there’s something more to pick at here
17:33:20 <Adam_Page> ... around WPT crash tests
17:33:26 <Adam_Page> ... there are others that do async operations
17:33:34 <Adam_Page> ... it may well not correctly flag
17:33:36 <Adam_Page> ... not just a11y code
17:33:44 <Adam_Page> ... the fact that it’s waiting on DOM load; an eager event
17:33:49 <Adam_Page> ... we need to address that in some capacity
17:33:56 <Adam_Page> ... maybe script can tell when a test should complete
17:34:02 <Adam_Page> ... or add a new mechanism for waiting
17:34:08 <Adam_Page> jcraig: I’m gonna add you as a reviewer
17:34:21 <Rahim> q+
17:34:23 <Adam_Page> ... would you reach out to other crash test writers?
17:34:42 <Adam_Page> ... that did exercise these crashes
17:34:42 <Adam_Page> ... in the a11y Slack
17:35:09 <Adam_Page> spectranaut_: do crash tests actually run with a11y turned on in the browser?
17:35:21 <Adam_Page> ... and I would also have to talk to those same people
17:35:28 <Adam_Page> jcraig: presumably, that’s what caused the crash
17:35:35 <Adam_Page> spectranaut_: but on the server
17:35:44 <Adam_Page> jcraig: understood, I don’t know
17:36:05 <Adam_Page> keithamus: there’s a blurred line; it could touch code in the DOM
17:36:09 <Adam_Page> ... or it could turn on the a11y engine
17:36:28 <Adam_Page> spectranaut_: with Acacia testing system, we *will* be able to write a crash test
17:36:42 <Adam_Page> ... or at least some kind of failure
17:36:49 <Adam_Page> jcraig: it’s possible some of these are so old that they were written before a11y was decoupled from the main thread
17:37:07 <Adam_Page> jamesn: do folks know what they’re doing with this one?
17:37:20 <Adam_Page> spectranaut_: is there an issue for it?
17:37:24 <Adam_Page> ... if so, assign me to that
17:37:26 <jamesn> ack spectranaut_
17:37:32 <jamesn> ack Rahim
17:37:45 <Adam_Page> Rahim: keithamus, is there a way we could run this crash test with a previous Firefox build?
17:37:50 <Adam_Page> ... to see if it’s detected
17:37:53 <Adam_Page> keithamus: I don’t know
17:38:02 <Adam_Page> ... I’ll dig up the bg
17:38:15 <Adam_Page> ... we might have a separate test for that
17:38:18 <Adam_Page> ... which might make this redundant
17:38:45 <Adam_Page> ... if we’re targeting a crash that happened in Gecko, I imagine there’s a test in Gecko’s suite
17:39:05 <Adam_Page> jcraig: keithamus, would you take an action to reach out to Aaron, Tyler, and spectranaut_
17:39:17 <Adam_Page> ... to see about improving reliability and validity of a11y crash tests?
17:39:49 <Adam_Page> keithamus: yes
17:39:57 <Adam_Page> spectranaut_: we still have to solve this problem in Acacia
17:40:17 <Adam_Page> jcraig: you can do it if you call computedRole or computedLabel
17:40:42 <Adam_Page> ... in WPT if there were a way to show devtools, that could be an option too
17:40:42 <Adam_Page> keithamus: should look into webdriver as well
17:41:01 <Adam_Page> spectranaut_: webdriver is kind of in maintenance mode
17:41:07 <Adam_Page> ... but wouldn’t be opposed to changes
17:41:12 <Adam_Page> ... and this could be motivating
17:41:19 <Adam_Page> jamesn: do we have a direction?
17:41:28 <Adam_Page> keithamus: let’s say yes
17:41:32 <Adam_Page> ... we’re heading in *a* direction ;-)
17:41:54 <Adam_Page> Rahim: the broader issue of writing WPT tests
17:42:03 <Adam_Page> ... I’m still working on case sensitivity tests
17:42:06 <Adam_Page> sarah: also
17:42:15 <Adam_Page> ... NVDA does switch modes for toolbars, so no need to file an issue
17:42:19 <Adam_Page> zakim, next item
17:42:19 <Zakim> agendum 4 -- -> Deep Dive planning https://bit.ly/aria-meaty-topic-candidates -- taken up [from agendabot]
17:42:42 <Adam_Page> jamesn: any plans? any one want to plan one?
17:42:42 <Adam_Page> aardrian: next week, no?
17:42:44 <Adam_Page> jamesn: yes!
17:42:50 <Adam_Page> ... layout tables
17:42:57 <Adam_Page> ... great
17:43:05 <Adam_Page> ... if you want to plan another, please add it in GH and add the tracker
17:43:09 <Adam_Page> ... or email the chairs
17:43:11 <Adam_Page> zakim, next item
17:43:11 <Zakim> agendum 4 was just opened, Adam_Page
17:43:17 <Adam_Page> zakim, close this item
17:43:17 <Zakim> agendum 4 closed
17:43:18 <Zakim> I see 1 item remaining on the agenda:
17:43:18 <Zakim> 5. Registration open for TPAC 2024 in mid July [from agendabot]
17:43:18 <Adam_Page> zakim, next item
17:43:19 <Zakim> agendum 5 -- Registration open for TPAC 2024 in mid July -- taken up [from agendabot]
17:43:31 <Adam_Page> jamesn: registration opens in mid July
17:43:37 <Adam_Page> ... in Kobe, Japan 🇯🇵
17:43:42 <Adam_Page> ... tentatively scheduled meetings
17:43:47 <Adam_Page> ... still open to change
17:43:58 <Adam_Page> ... Thursday we have one cross-group meeting
17:44:03 <Adam_Page> ... we’ve managed to not clash with CSS so far
17:44:11 <aardrian> Q+
17:45:17 <Adam_Page> aardrian: I may not be able to attend in person, how difficult would it be to run sessions remotely?
17:45:21 <Adam_Page> keithamus: people have done it
17:45:42 <Adam_Page> jcraig: we can schedule them for convenient times
17:46:01 <Adam_Page> Daniel: we have plenty of time to shift our agenda
17:46:42 <Adam_Page> jamesn: realistically, we’ll schedule session for remote attendees in the morning
17:46:49 <Adam_Page> jcraig: and logistically, rooms are set up pretty well now
17:51:50 <Adam_Page> spectranaut_: everyone, go use the last 10 minutes to go review issues!
17:52:08 <Adam_Page> zakim, end meeting
17:52:08 <Zakim> As of this point the attendees have been aardrian, hdv, Adam_Page, spectranaut_, front-endian-jane, Francis_Storr, smockle, Rahim, scott, filippo-zorzi, keithamus, StefanS
17:52:11 <Zakim> RRSAgent, please draft minutes v2
17:52:13 <RRSAgent> I have made the request to generate https://www.w3.org/2025/07/10-aria-minutes.html Zakim
17:52:20 <Zakim> I am happy to have been of service, Adam_Page; please remember to excuse RRSAgent.  Goodbye
17:52:20 <Zakim> Zakim has left #aria