W3C

– DRAFT –
ARIA WG

20 July 2023

Attendees

Present
Adam_Page, alison, andrea, benb, benjamin, bryan, CurtBellew, Daniel, Francis_Storr, jaunita_george, Jem, mpaiva, scotto, siri
Regrets
-
Chair
spectranaut_
Scribe
Jemma

Meeting minutes

<spectranaut_> remove item 6

[New Issue Triage](https://tinyurl.com/2d2j7ssd)

<Jem> w3c/aria#1986

<Jem> val: Mario's comment is about clarifying the language about the confusion. It sounds like a minor issue

<Jem> scott: My PR will also clarify 1986

<Jem> scott: only pending thing is Matt's comment

<Jem> ... if you can just verify that, 1949, it would be great.

<Jem> matt: I can do review during this call.

<Jem> w3c/aria#1985

<mpaiva> I have two related issues 1956 and 1970

<Jem> Val: I see your comment, Scott. Do you have the comment to add?

<Jem> scott's comment "though presently the spec is pretty clear that the expectation is there would be interactive elements within the application role container... maybe that should/could change? But even if it does, that might be for nothing if you have to add nested interactive elements to address all the things i mentioned above, anyway."

<Jem> scott: if james Crag is in the meeting, he would say that there is nothing to prevent this.

<Jem> w3c/aria#1984

<Jem> curt: "The aria-selected state description seems to indicate that if one tab in a tablist is marked with aria-selected="true" and all other tabs have no aria-selected attribute then the tabs without aria-selected should return selected false. ...However in testing, browsers seem to return all of the tabs as selected. Simple sample code:"

<scotto> clarification to what i said "if james craig was in the meeting i would assume he might say we should not necessarily prevent this, as we dont want to prevent an author from doing something they might need to - which I would think is a good stance that we should consider, but not necessarily be in a rush to change, imo"

<scotto> np. i just wanted to be clear since i was 'fake speaking' for someone else :)

<Jem> matt: in that sentence it is vague which widget is referring to.

<Jem> matt: we may need to look at the language of "tab" since some tabs can not be selected. In APG, we add aria selected false for unselected tab.

<Jem> Val: since this is a triage session, I would like to set separate time.

<Jem> matt: To resolve this, someone can see three different sections and clarify the intents.

<Jem> curt: I will do that.

<Jem> w3c/aria#1983

<Jem> Why does role=none|presentation only prohibit aria-label and aria-labelledby #1983

<Jem> scott's comment on the issue - "If i recall correctly, it wasn't really considered something to worry about since if a none/presentation element gets a global ARIA attribute, the presentational role is dropped in favor of the implicit or fallback role."

<Jem> ..."it is a bit silly/confusing to say that naming them is prohibited, but all other global attributes are allowed/inherited... but also, if those attributes are used it negates the presentational role anyway... so in a very round about way, they're prohibited without saying as much."

<Jem> matt: we would need to update the script so that it can update the allowed attribute.

<Jem> Val: I can look into this.

[New PR Triage](https://tinyurl.com/2xky9pjp)

<Jem> w3c/aria#1982

<Jem> Daniel will review this with Valerie.

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

<spectranaut_> w3c/aria#939

<Jem> Val: We have a plan to discuss one hour before ARIA meeting.

<Jem> Matt: Scheduling the meeting earlier will be great.

<spectranaut_> to schedule: w3c/aria#1950

<spectranaut_> to schedule: w3c/aria#1951

<Jem> Val: We can invite Jaws and other screen developers to the aria-modal discussion

<Jem> we need Aaron for aria-modal discussion

<Jem> and screen reader developers for aria hidden

[Clarification about prohibited acc name and title attribute](https://github.com/w3c/aria/issues/1981)

<Jem> val: we discussed and briefly triaged this.

<Jem> scotto:

<Jem> we would need clarification for accessible name calculation and prohibited status

<Jem> scotto: we dont have a good answer to the accessible name to the generic roles like div and so on

<Jem> matt: I can say from my expereince

<Jem> ... we definitely don't want accessible name to generic because it hinders to access the actual contents. The computation should ignore/throw out the title attriute

<Jem> Matt:we don't support those naming for generic, do we?

<Jem> scott: yes.

<Jem> matt: if we have accessible name for strong, the screen reader user will be very confused which one is real accessible name.

<Jem> scott:title is global attribute so it is tricky..

<Jem> matt: just putting a title to div and "suffacing it to the screen reader" - being the part of name computation - is really problematic

<Jem> scott: I strongly agree with Matt and I can submit the PR to Html and AAM.

<Jem> Cory: I agree with Scott.

<Jem> ...SEO/marketing world tend to abuse those title attributes

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

<Jem> doug: aria live region is one example of ARIA notification

<Jem> ... goal is to take the string to pass out to the screen reader with simple javascript.

<Jem> .. MS set up some example and API was introduced to Face to face meeting.

<Jem> .. it has some problems...

<spectranaut_> https://github.com/w3c/aria/discussions/1958#discussioncomment-6223153

<Jem> matt; conclusion we got last time is that aria - live is not working for braille and speech...

<Jem> .. so that could be the priority 1

<Jem> Mario: may it be problematic to mix with braile or speech?

<Jem> matt: we will use braile role description for the braile notification.

<Jem> Cory and Dough are talking about interconnnection among browser, screen reader and braille version of notifictaions.

<Jem> there is synchronization need.

<Jem> matt: interesting thing for speech mark up is simply affecting the processing speech version of string. may be separate flag for each mark up....

<Jem> Doug: AT may not be recognize in timely manner. Therefore, it should be the separate string or it should be priority1.

<Jem> .. it is easier for braille to put it as the separate string.

<Jem> doug: when we did survey this, it is related to "map" so the shortend word can be pronounced fully like Dr reads as "doctor"

<Jem> presnt+ francis

<Jem> Daniel: to say accessibility platform WG asked joint meeting to work together notification or pronunciation ...

<Jem> .. the other working group can also help this dicussion.

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

Diagnostics

Active on IRC: Adam_Page, benb, CurtBellew, dmontalvo, Francis_Storr, jaunita_george, Jem, mpaiva, scotto, spectranaut_