W3C

– DRAFT –
ARIA WG

16 December 2021

Attendees

Present
jamesn, Jaunita_George, MarkMcCarthy, sarah_higley, scottono, spectranaut, StefanS
Regrets
Curt, Harris, Joanie
Chair
JamesNurthen
Scribe
Jaunita_George, jcraig, Juanita_George, MarkMcCarthy

Meeting minutes

<jamesn> agendabot this is aria

New Issue Triage

[we started off with some intros for Frank - welcome!]

<jamesn> https://github.com/search?l=&q=is:open+is:issue+created:%3E%3D2021-12-09+repo:w3c/aria+repo:w3c/accname+repo:w3c/core-aam+repo:w3c/dpub-aam+repo:w3c/dpub-aria&type=Issues

jamesn: 5 new issues - starting with 1662

jcraig: no strong preference on what they _should_ be, and not certain if we need a new type of list,

jcraig: if we can remove the longer names, we could use listitemkey and listitemvalue

jcraig: just some suggestions, nothing concrete!

jamesn: 1.3?

scottono: yeah

jamesn: i'm also pretty sympathetic to this

sarah_higley: +1

pkra: i also linked this to Joanie's issue of DLs vs OLs

jamesn: thanks!

jamesn: maybe this would be a good deep dive conversation

<Jaunita_George> +1

jamesn: next issue, dpub, issue 40

jamesn: i don't have a milestone for this, we'll just leave it in dpub and take care of it later on - objections?

[none]

jamesn: issue 1661, Joanie opened this. we have some comments - scott, what should the milestone be?

scottono: joanie's looking at things a bit, also pointing out that native selects and aria selects behave differently. i'm not sure how important this is, but it's a good example of HTML/ARIA differences (and joanie rightly pointing out it's confusing)

Jaunita_George: maybe throw this to APG, they can help explain the differences in behaviours?

spectranaut: not sure it's that simple though

jamesn: can you do some more research on this Valerie?

spectranaut: sure, but I'm not sure what more I can provide than Joanie already did. I just think it needs more discussion and more eyes

jamesn: okay - I'll 1.3 it and agenda it for a later meeting

spectranaut: good idea

mck: yep

New PR Triage

<jamesn> https://github.com/search?l=&amp;q=is:open+is:pr+created:%3E%3D2021-12-09+repo:w3c/aria+repo:w3c/accname+repo:w3c/core-aam+repo:w3c/dpub-aam+repo:w3c/dpub-aria+repo:w3c/html-aam&amp;type=Issues

<jamesn> https://github.com/w3c/aria/pull/1663

jamesn: 1663 - we have 3 approving reviews, will be merged soon. look at it quick if you want to make comments!

jamesn: 1660 - jcraig, this needs reveiwers. joanie is signed up and scott signed off.

jcraig: in a nutshell, this mostly removed unecessary specificity and adds helpful vagueness

jcraig: this is mostly editorial, and removes some unnecessary limiting language

jamesn: carmacleod is down to review it, so we should be good there then

jamesn: 1658 pkra is now an editor!

group: congrats, well done, sympathies, rock on!

CR Published

jamesn: we should have announced something to some people, and should hold an announcement until after the holidays...give people all of January to review it

jamesn: objections?

carmacleod: don't disagree, but wonder if there's a timeline?

jamesn: we need implementations done first

carmacleod: gotcha

jamesn: valerie, do you know where the implementations are?

<pkra> sorry have to drop off.

spectranaut: yes, but I don't know the status of it, Joanie probably does

jamesn: i need to have a meeting with her about this anyway. we only have one main implementer which is bottlenecking, would love others!

jamesn: and of course aaronlev as well. :)

jamesn: unless there's objections, i think it's fair with holidays to extend things out a bit.

Activation point or sub-element click target (maybe aria-hitpoint?)

<jamesn> https://github.com/w3c/aria/issues/788

jamesn: jcraig put some comments into the issue following our deep dive last week. what are the next steps on this?

jcraig: quick recap - there's a feature in VoiceOver that allows direct manipulation to sliders. basically can simulate drag events

jcraig: ...We can use a sub-level element and the AT can decide if it wants to mimic tap behavior
… We might not need a role for that, we could use an ARIA attribute for it
… I had a few proposals, one being "hit point" and is that the right name? How should that be exposed?

siri: So the slider will rotate the product?

jcraig: Let me give a demo

sarah_higley: I don't know if screen reader users need track clicking. Is this only for screen reader users or are other use cases?

jcraig: With most AT besides screen readers, we do use pointer events

sarah_higley: Another interesting example is charts

frank: There are use cases with selection ranges

jcraig: Let me demo this

<sarah_higley> I guess some of my questions are circling the idea that a JS implementation might be more flexible, but I don't know what roadblocks were encountered with that 😓

jamesn: I can see this used for two D controls as well as charts

jcraig: We can implement this as aria-slider but could cause people to bypass track

jamesn: If you had it the way you are proposing where you have the hit point on the thumb itself, would this allow multiple thumbs?

jcraig: No

jamesn: If you have a line graph where you could manipulate the lines ...

jcraig: The container itself would have a role in that case

frank: What would an audio graph do?

siri: How is that different from regular sliders?

matt: Sliders don't work properly in iOS

correction: Mark: sliders don't work like native iOS

jcraig: This is just a proposal to control standard drag points

sarah_higley: We could treat this like a state

jcraig: We do this in other cases
… do you want this to be a role? Do you want this to point inward or outward? I think it should be outward

jcraig: This is closer to aria-controls

jamesn: Do we need roles for this?

Zakim: next item

Need an attr to indicate element activation triggers audio, video, etc.

jcraig: We have two APIs on iOS -- one is used for longer sound and the play sound trait triggers audio, even if it's short

Minutes manually created (not a transcript), formatted by scribe.perl version 185 (Thu Dec 2 18:51:55 2021 UTC).

Diagnostics

Succeeded: s/next/jamesn: next

Maybe present: carmacleod, correction, frank, group, jcraig, matt, mck, pkra, siri, Zakim