W3C

– DRAFT –
APA Weekly Teleconference

20 May 2026

Attendees

Present
chiace, Dr_Keith, Fazio, janina, JonCohn, matatk, mpaiva, PaulG
Regrets
Gottfried_Zimmerman, Jennifer_Strickland
Chair
Matthew
Scribe
chiace

Meeting minutes

Agenda Review & Announcements

TPAC 2026

TPAC 2026 we need to think about what we are going to do, It hasn't been request yet, but we need to start to think about it: meeting, breakout, etc, let us know

matatk TPAC 2026 we need to think about what we are going to do, It hasn't been request yet, but we need to start to think about it: meeting, breakout, etc, let us know

matatk W3C program for Invited Expert it will happen soon

APA Members' Spec Review Guide

matatk about the spec review please file issues or pull requests on it as required, please use the guide as well and provide feedback

New charters review

Browser Testing and Tools Working Group Charter

<Roy_Ruoxi> - charter: https://w3c.github.io/charter-drafts/2026/charter-wg-btt.html

<Roy_Ruoxi> - issue: w3c/strategy#558

Roy_Ruoxi has been closed

janina we tangled our way into understanding it

matatk We are talking about browser testing and tools working group

matatk Web driver and the new version which is a webdriver which is bidirectional. Defining the protocol that allows you to remote control browser instances. So you can headless browsing. Probably lots of us have used this sort of thing

<PaulG> https://aria-at.w3.org/reports

JonCohn believe they already did NVDA support for it too, has been working on this for over a year

JonCohn collect the info that gets returned and there by off what you expect to get returned

matatk Aria link from Paul getting back the screen reader

PaulG it has in these reports they have the test, The tests have asseruibs like must convey roll button and so or must convert, convey heading level 3

matatk the differences are here, it has been going on for a little while, so not a totally new deliverable. the normative specs, they are just saying they finished webdriver. But the bi directional version they are intending to

mpaiva Will they consider all the AI automation and QA? AI automation that the progression that's been happening

matatk good point, and he wonders if it is specific to the AT driver stuff or in general

<matatk> https://www.w3.org/groups/wg/browser-tools-testing/tools/ - repos owned by the group

<PaulG> w3c/webdriver-bidi

<matatk> w3c/at-driver

<matatk> w3c/webdriver

PaulG we've got too many concepts with the same word, Obviously the web driver project started way before LMS became a mainstay of

matatk webdriver, baidu stuff, intresting

PaulG they just wanto to normalize the protocol for testing various browsers

WebTransport

<Roy_Ruoxi> charter: w3c/strategy#537

matatk we talk about it last week, he made a note on the tracking issue

<matatk> w3c/a11y-tracking#310

Fazio wcag dev - framework of testing it can make sense , to keep it in line with other groups

New on TR

CSS-AAM

<matatk> w3c/css-aam

matatk Paul did give us a heads up about this. So there is a CFS AAM or A11y APA API mapping CSS. We need to be aware of this and follow significant changes to it. We have had a number of issues over the last 6-9-12 months where this type of thing has come up in the issue

matatk ARIA and CSS is working on it

PaulG last week there was a discussion and they said that they pivoted and they are doing AAM mapping and sending it. But they cannot style it.

<matatk> Paul is referring to w3c/a11y-tracking#247 wrt making the mapping directly, rather than going long way 'round via actually setting ARIA attrs

matatk Target current pseudo class should also imply Aria current on active scroll marker, is that is a11y tracking 247

PaulG Steve said last week was the right move and but ut does create this, dual mapping. You are making CSS that is sort of resilient against implantation choices or you have progressive enhancement, or you have fallbacks for problems that do not support certain features

matatk Were wondering how to expose that semantically to ATS, but if they ve already got the target current pseudo class, presumably they can style it

matatk Current scroll marker, CSS rule. Thta API it is a different way around, it is just gonna go in the opposite direction

Spec review requests

<matatk> w3c/a11y-tracking#322

matatk we maybe need a Google Doc, we need to collate the threads about HTML and canvas, understand the right people are talking about it . So we want to do this much more in depth

<PaulG> w3c/aria#2758

matatk please feel free to edit this

matatk post on the a11y tracking with any questions you have that you do not think are answered

Extended Descriptions (DPUB)

<matatk> w3c/a11y-tracking#311

<matatk> Please look and put feedback initially on that tracking thread

matatk Explainer for the digital publishing people for extended descriptions for publications. Everyone should look at it

Issue tracking

Long Animation Frames

<matatk> w3c/a11y-tracking#316

PaulG Long Animation Frames API has the ability to prevent that blocking. Blocks UI interaction, API to keep the UI interactive.Google is already shoved and small language model into a browser thread

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

Diagnostics

Active on IRC: chiace, Dr_Keith, Fazio, janina, JonCohn, matatk, mpaiva, PaulG, Roy_Ruoxi