W3C

– DRAFT –
ARIA Authoring Practices Task Force

25 March 2025

Attendees

Present
CurtBellew, howard-e, Jem, jemma, jongund, jugglinmike, Matt_King, Siri
Regrets
-
Chair
-
Scribe
jugglinmike

Meeting minutes

Setup and Review Agenda

https://github.com/w3c/aria-practices/wiki/March-25%2C-2025-Agenda

Matt_King: Any requests for change to agenda?

Hearing none, we'll stick with what we've planned

Matt_King: No meeting April 1

<Jem> I will be out for April 8 too

Matt_King: Next meeting: April 8

Publication planning

Matt_King: We published last week, so I marked the milestone for March as "complete". It ended up having just two things in it.

Matt_King: Since it was published, has anyone verified the changes in the live site?

Jem: No, but I can do that this afternoon

Matt_King: Thank you! It should be pretty quick

Matt_King: I set up a milestone for April https://github.com/w3c/aria-practices/milestone/38

Matt_King: We'll play the publication date by ear now that the W3C is more flexible. For the time being, I've set a tentative date for April 29

howard-e: No problem with that target on our end

Matt_King: Right now, this only has the "color settings" pull request. If possible, we'll add the "Expandable region" pull request that Adam is working on

Matt_King: I'll put that on the agenda for next week when Adam returns

howard-e: If we add a pattern which doesn't follow the expected structure (e.g. that it doesn't end in "-pattern"), then we need to modify a JSON file.

howard-e: It doesn't look like those conventions are properly documented, so I will look into documenting them

Matt_King: There's no reason not to follow those conventions, so we can conform to those

Matt_King: The conventions have been intuitively obvious to me, but you're right: they aren't properly documented, and they should be

howard-e: I created an issue to track that and assigned it to myself

Matt_King: Thank you!

Matt_King: I don't know if this "expandable region" thing should be a new pattern or not. We can discuss when Adam returns

Matt_King: There are also some other minor pull requests that could end up in the next publication

Matt_King: Any questions about the future publication?

Matt_King: Hearing none, we'll move on

PR 3252 - Editorial change on combobox

github: w3c/aria-practices#3252

Matt_King: This looks fairly simple

Matt_King: When I see this kind of pull request, I wonder if the problem exists elsewhere because we use similar phrasing in other places

Matt_King: I'd like reviewers to verify whether or not we have the same problem in other places

Matt_King: This problem is related to the accessibility features and the amount of padding/border (the way we indicate selected items)

jongund: I think I worked on this combobox. I can take a look if you'd like

Matt_King: I know you have a big pull request in progress, but that would be great

Siri: I can review it along with jongund

PR 2991 - Next steps on color settings practice

github: w3c/aria-practices#2991

jongund: I made some updates based on the feedback in recent meeting minutes

jongund: It seems like Safari is the only browser that supports one requested feature

jongund: I'm concerned that including it may suggest to readers to force the default color theme

Matt_King: There's a media query for this, and it's only supported in WebKit

jongund: So what should an author do with it?

jongund: it looks like it inverts images on Macs

jongund: If this were supported on all browsers, what would you tell authors to do?

Matt_King: I can't recall the utility

jongund: I did update the table to say that there is a media query which is supported in some browsers

Matt_King: So someone can tell that this is going on, but should they do anything with that information?

Matt_King: Or, put differently: if someone is running with "invert colors" turned on, is there some benefit to the author knowing that?

jongund: There are already so many things that we're telling authors to do (or at least to consider)

Matt_King: We don't want to say that authors cannot detect this (as we previously stated) because that's incorrect.

Matt_King: We can say that we don't recommend doing anything in response to it

CurtBellew: I've never played with "invert colors". The thing with "high contrast" is that things will tend to disappear if you're not careful. I don't know if that's the same with "invert colors", but could that be what it's for...?

Matt_King: Theoretically, it's impossible for something to disappear because of invert colors because everything is inverted (there is no exception)

jongund: On the Mac, it's implemented globally on a low-level of the display, so there's little-to-no room for inconsistency. Even the DOM doesn't recognize the change--when the user inverts colors, the DOM still reports the colors as authored

Matt_King: It's very useful to tell authors NOT to do something, and it's often quite welcome...

Matt_King: What if we told authors (essentially) "don't use this media query; we don't see any benefit to it", would you be in support of that?

jongund: I think this is a non-issue. No one on this call has experience with it, and it's only supported on one browser

Matt_King: Because it's out there and because people are writing about it, I think taking a clear position is beneficial.

<CurtBellew> +1

jongund: I support the APG discouraging its use

Matt_King: Okay, so we're aligned on that

Matt_King: It looks like you've simplified the structure of the page, generally

Matt_King: Are you ready for additional review?

jongund: I'm still making changes in response to some of your feedback. I should finish that later today

jongund: I will let you know when it's ready for additional review

Matt_King: Thank you very much!

PR 3249 - Add HTML search to landmark practice

github: w3c/aria-practices#3249

Matt_King: jongund, I saw you pushed some additional commits

Matt_King: After we're done with the "color settings" patch, I think we should talk about what we need to do in the "landmark" section overall

Matt_King: That said, I think the direction #3216 is consistent with the way we've structured the rest of APG

Matt_King: Instead of pulling stuff from the examples into the "patterns" page, if there is additional information, it should go to the "practices" page

Matt_King: The "patterns" page was intentionally very simple and referred to the "practices" page

Matt_King: I'm thinking that we can keep #3216 "on hold" for now, and we can talk about it in a later meeting

Matt_King: In the mean time, I did want to land #3249 because it's a very simple change and it came from a first-time contributor

Matt_King: CurtBellew was going to add another simple commit to that to adjust the example page for the "search" landmark

Matt_King: I'm kind of wondering: if you already did some of the work that CurtBellew was going to do (I saw you push a commit that was similar)--did you make a change to the "search example" landmark page?

jongund: Yes

Matt_King: Okay, so CurtBellew, you might want to look at that commit

jongund: I added a demonstration of an "HTML way" to create a "search" landmark

CurtBellew: That sounds like exactly what I had done

Matt_King: I didn't see you push a commit to #3249, yet

CurtBellew: I've never pushed a commit to an existing pull request, so I had questions about how to do that

<Jem> https://github.com/w3c/aria-practices/wiki/Submitting-Pull-Requests

Matt_King: [explains the process]

CurtBellew: That makes sense to me!

Matt_King: Great

Matt_King: Now, I pushed some more changes to this branch last Wednesday, and my changes still aren't showing up in the preview...

howard-e: It says it was last built on Wednesday

Matt_King: I have commit eb14fc6. That commit is not showing up in the preview

howard-e: I will check that out in the pull request

Matt_King: Ok, great

Matt_King: That's it for today. Thanks, everyone!

Minutes manually created (not a transcript), formatted by scribe.perl version 244 (Thu Feb 27 01:23:09 2025 UTC).

Diagnostics

Succeeded: s/color as/colors as/

Succeeded: s/ed14/eb14/

All speakers: CurtBellew, howard-e, Jem, jongund, Matt_King, Siri

Active on IRC: CurtBellew, Jem, jongund, jugglinmike, Siri