W3C

– DRAFT –
AGWG Teleconference

03 March 2026

Attendees

Present
Adam_Page, alastairc, AWK, CClaire, cgrimley, Charles, Daniel, eloisa, filippo-zorzi, Francis_Storr, Frankie, Gez, graham, GreggVan, hdv, Heather, Helen, Jen_g, Jennie_Delisi, Jon_Avila, jtoles, julierawe, kevin, kirkwood_, laura, Laura_Carlson, LenB, LoriO, Makoto_U, maryjom, Patrick_H_Lauke, Rachael, shadi, stevef, SydneyColeman, tayef
Regrets
Brian, Bruce, John K
Chair
-
Scribe
laura

Meeting minutes

rm: Any new members?

Dirk Ginader: I just joined. 20-something years at Google for 13 at this point, Yahoo before that.

<ginader> hey all 🙂

<eloisa> presen+

Charles Hall: I know many of you, many of you know me. spent five and a half years with CVS Health.
… l was active in the Silver Task Force and AG between November 2017 and March of 2021.

<Rachael> https://www.w3.org/TR/wcag-3.0/

RM: on March 8th, the United States will do a time change.
… the draft was published.
… feel free to post and share that on social media.

WCAG 2 proposed changes (review by 9 March) https://lists.w3.org/Archives/Public/w3c-wai-gl/2026JanMar/0108.html

Kevin: there will be an announcement coming from, W3C too.

Patrick: Reminder to people that if they haven't had a chance to look at the email that was sent that lists the proposed changes, please do.

Continued discussion of and work on ACT Rules for Requirements https://docs.google.com/presentation/d/1klORYoNQq3oDMJLc52q6Qf5gCTIcGHsm2kBbroj7bMM/edit?usp=sharing

Patrick: if you can, just to look over that we didn't overlook anything.

Slideset: https://docs.google.com/presentation/d/1klORYoNQq3oDMJLc52q6Qf5gCTIcGHsm2kBbroj7bMM/edit?slide=id.p#slide=id.p and archived PDF copy

RM: We're going to continue the discussion of and work on ACT rules.
… We believe it will help us to create higher quality requirements.
… We've got this set of existing ACT rules, but ACT also has a rules format that's been documented. We've, signed off on it, it's published.
… there was a conversation at the chair's level,.
… I wanr to bring this understanding here and clarify it.
… In WCAG 3, what we were hoping to do is start here and create enough rules with enough coverage that when put together, that set of rules that ties to a requirement.
… Rules mostly prove the absence of a failure rather than 100% success.

Gregg: ACT test little bits of a provision. They don't test the whole provision, and so passing them doesn't mean that you pass.
… my kind of concern, is that passing an ACt rule doesn't actually mean you either pass or fail, it doesn't mean anything at all. It just means you passed the ACT.
… mean anything at all. It just means you passed the ACT.

RM: (Show example)

[Slide 5]

RM: we're looking for something, I believe, in order to pass page view title available, all of the following must be true.
… So, long-term, we expect that we will have multiple types of rules within WCAG3 methods.
… covers Text Color Adjustable

https://docs.google.com/document/d/1N2rWpUTAN_u6fP53k9V1q5EjI5drMY7fXFu6i5xfmws/edit?tab=t.0

rm: process of pulling examples, finding what worked, what didn't, led us to a really good conversation around, well, should this be text color adjustable and non-text color adjustable
… examples drove good conversations.
… this process has really helped refine and get to a point where we have better examples, and there's a working draft.

ac: covers https://docs.google.com/document/d/1ucdwQdwGCuTPzKXAfF8tTe5RRXsWVFE2bB--6zFRjiQ/edit?tab=t.4ts7mfo3ejq1#heading=h.2c4uehj7negn

<kirkwood_> Was there any discussion about surfacing the flagging frequency for the user through meta data such that they could have personalized needs for flagging frequency filtered?

<Patrick_H_Lauke> a cursor would also not cover more than a few pixels, so likely wouldn't trigger anything?

<kirkwood_> Flagging/flashing

Gregg: a transition is a shift in in luminosity, in the opposite direction.
… If you go down, and then go down again, that doesn't... second one doesn't count as a transition, just a tweak for you there.
… I think it would be helpful if we labeled all the ACT rules as being comprehensive.
… The other flag is the technology
… agnostic.
… if it's only for HTML, then it really falls into the category of being a technique.
… l'm not sure what the purpose of examples are.
… if it's not a comprehensive rule ,Then the examples of passing and failing don't even tell me if it passes or fails.

Helen: we have examples of code in HTML, but a lot of the actual pass-fails can be agnostic.
… the point of the examples are not about, being device specific in some cases, the code examples are for the automation, but the generic what happens that is to help the manual tester, to understand how to do the testing effectively.
… As we've grown, we've been doing more manual rules and there's ones that are very subjective.

<Zakim> Rachael, you wanted to say we are not only writing html rules

Helen: They can't be implemented by automation. So if you have a look at the proposed rules, a lot of those are the manual testing rules that cannot be automated as easily, so it's not about automation as such, it's just automation tools can use the rules.

to help make sure that they're being agnostic
… of one person's ideas of what meets the success criterion.

<Rachael> https://www.w3.org/WAI/standards-guidelines/act/rules/

<Zakim> Charles, you wanted to discuss process clarity to comment on demonstrated doc examples.

RM: we aren't saying that you should only be writing HTML rules. We should be writing generic rules that, when put together, become a comprehensive or, um, compound role.

Charles Hall: What's the preferred method for adding comment?

AC: you can either comment in the survey or comment on the document.

<Charles> where might I find a list of active subgroups, scope, and how to join?

Patrick: How big is the flashing surface? Because taking it to the extreme, we shouldn't really fail a page for a single pixel that flashes, even if the frequency is very high, because, you know, yeah, it comes down to the actual size of it.

AC: But the frequency test is separate from the size test.
… So, you would have to fail both.
… In order to actually fail, and then the size test is looking for

ac: in the document, we've got top-down procedures like non-decorative images.
… they aim to help people understand the provision.

<alastairc> acl lori

ac: The ACT rules kind of a harder thing to do, and involves finding more examples on things..

Lori: ACT rules states that they are technology specific.

<Helen> There is an FAQ section: https://docs.google.com/document/d/1i_O5J7D4BWMZAB7NaD4iplDRuqOjCvCGAaHdAirLUhE/edit?usp=sharing

RM: But I hear that concern, and let me see if I can get a better clarification on that.

AC: we're gonna need rules for things that are word-based. Doesn't matter what, uh, technology you're on.

<Patrick_H_Lauke> ACT rules WILL have to have high-level, non-technological, user-acceptance criteria style rules. which i believe the ACT *format* allows

RM: It is an iterative process, and that ACT rules may update what they have as their form.
… if you're most comfortable with the HTML, write the HTML first, and then we can work towards generic.

ac: we're going to be going into our breakout rooms.

Minutes manually created (not a transcript), formatted by scribe.perl version 248 (Mon Oct 27 20:04:16 2025 UTC).

Diagnostics

Succeeded: s/sReminder/Reminder

Maybe present: ac, Gregg, Lori, Patrick, rm

All speakers: ac, Gregg, Helen, Kevin, Lori, Patrick, rm

Active on IRC: Adam_Page, alastairc, AWK, CClaire, cgrimley, Charles, Daniel, eloisa, filippo-zorzi, Francis_Storr, Frankie, Gez, ginader, graham, GreggVan, hdv, Heather, Helen, Jen_g, Jennie_Delisi, Jon_Avila, jtoles, julierawe, kevin, kirkwood_, laura, LenB, LoriO, Makoto_U, maryjom, Patrick_H_Lauke, Rachael, shadi, stevef, SydneyColeman, tayef