W3C

– DRAFT –
ARIA Editors

12 February 2025

Attendees

Present
Bryan, Daniel, James, James N, N, Peter, Rahim, scotto, Valerie
Regrets
-
Chair
-
Scribe
pkra

Meeting minutes

<Daniel> aagendabot, clear agenda

<Daniel> aagendabot, clear agenda

present jamesn, spectranaut, dmontalvo, bryan, rahim

[new] aria#2430 revisit PRs touching on posinset, setsize, level

pkra: this came from last week's discussion around aria-setize, posinset, level
… found a lot of related PRs
… and a lot of old ones
… feels like we're not getting to those old ones

jamesn: are those in different stages

pkra: yes.

jamesn: we should track that

spectranaut_: should we just track old PR review?

pkra: maybe. Not sure if it's editors' jobs (for each spec) or chairs (for group tasks)

spectranaut_: but it seems like we should do sometihng

Daniel: we should also think about communicating implemented changes. with everything being evergreen, even draft means it is already implemented.
… so other parts of the communities need to be aware.

jamesn: [earlier] we should try to nerd-snipe developers.

pkra: jamie had obviously some time last week, commented on older PRs.

<Daniel> Daniel: Maybe we need to better use labels to indicate where input is most required

spectranaut_: I do think it's worth figuring this out. We've discussed this a lot but didn't resolve enough.

jamesn: maybe deep dives

rahim: for old PRs?
… triaging issues with scotto now once week

pkra: good point.

scotto: I agree. We go through triage. We always agenda new stuff, rarely check back in on old thins.
… I've tried to do this more.

jamesn: that's our mode of operation.
… if you think something needs to be back on agenda, you agenda it

spectranaut_: but we need some kind of safety net

scotto: in openUI they use a stale bot.
… that might be too much.
… but for PRs perhaps.
… re-check then.

scotto: I've sent out reminders for PRs I had forgotten about.
… maybe "meets review"
… feed into some kind of queue to merge / check on bugs etc.
… I'm happy to poke but I think I don't scale.

pkra: the weekly github digests seem like they could be more useful

rahim: could we schedule regular time in the main meeting? Every few weeks.

spectranaut_: right, we talked about this before. It might be time to implement sometihng.
… "rally the group" effort.

rahim: is there a reporting by assignee perhaps?

spectranaut_: we sort of did that. Towards end of 1.3.
… but it would be nice.

pkra: let's track then?

spectranaut_: yes.

more generally: old PRs

[new] html-aam#576 Clarify attribute mapping tables

scotto: came up with rahim last week.
… old open issue on htm-aam "use aria mappings is not helpful"
… which is fair
… especially since html-aam refers to core-aam for many mappings.
… we obviously don't want to copy&paste around.
… with monorepo, could we do some scripting to take these core-aam mappings and pull them into other AAMs
… pulling in those mappings would help implementers.
… don't know how but sounds like we could do this now.

rahim: could this be respec territory?

pkra: this reminds me of the json blob that modules are importing for aria-related stuff
… but rahim reminds me of respec's export features which don't work right now

daniel: xref etc.

pkra: but this is would need moving away from custom pdef/rdef

spectranaut_: it's a good idea. Do we know who has issues? Does this help them?

scotto: https://github.com/w3c/html-aam/issues/122, marcos. Aaron also brought it up in the past.
… again, I think monorepo allows this now.

dmontalvo: we could try to put the mappings into separate file. then use data-include to pull that in.

spectranaut_: sounds interesting.

scotto: same. now we seem to be in a place to do this.

<Daniel> Respec data-include

spectranaut_: we spoke about having an editors hackathon. Perhaps that fits there.

scotto: sounds good. seems like a problem of finding shared time.

spectranaut_: last time, we blocked afternoons.
… maybe a board for it.

pkra: aside: should we start a board for agenda-editors ?

Daniel: yes. I'll create one if you can't do organizational boards

spectranaut_: I'll start planning.

[hasUpdate] aria#2411 aria.js: always add "Accessible Name Required" to characteristics tables

pkra: this was more with regard to the first item today

[revisit] html-aam#122

pkra: we discussed this earlier today.

[revisit] aria #2242 : Netlify to replace pr-review

pkra: I noticed that some relative links are re-written to point ot editors draft

Daniel: I'll take a look

spectranaut_: right. found an example now. makes sense we should fix it.

[revisit] aria#2407 aria.js: drop "deprecated on this role"

pkra: should I close this and file a follow up issue on the decision of the group?

spectranaut_: sounds good.

[hasUpdate] [peter] aria#2226 remove spec specific css or inline it

pkra: I pushed a lot of changes today.
… most for documentation (explaining why which rule is dropped)
… two main ones are
… 1 moving nested list design into just accname.
… 2 starting to switch table designs in aria as a test run

spectranaut_: for accname, I don't like the naming. Could we find a way to name things better? So that new lists are correctly numbered automatically?

pkra: right. I'd have to ask accname editors to take on naming that.

bryan: Mel should help since it's visual.

spectranaut_: right. Let's ping her.

pkra: ok.
… [screenshare aria spec preview]

spectranaut_: looks good.

pkra: fyi I'll remove the two unused css files unless anyone knows what they're good for.
… and continue the table changes in other specs.

Minutes manually created (not a transcript), formatted by scribe.perl version 242 (Fri Dec 20 18:32:17 2024 UTC).

Diagnostics

Succeeded: s/changes/implemented changes/

Maybe present: dmontalvo, jamesn, pkra, spectranaut_

All speakers: bryan, Daniel, dmontalvo, jamesn, pkra, rahim, scotto, spectranaut_

Active on IRC: Daniel, pkra