W3C

– DRAFT –
APA Weekly Teleconference

19 Aug 2026

Attendees

Present
Fazio, janina, JeffAdams-UN, JenStrickland, JonCohn, marcelo, matatk, Neha, PaulG, Vanessa
Regrets
-
Chair
Janina
Scribe
JeffAdams-UN

Meeting minutes

Agenda Review & Announcements

<matatk> AUR docs list: https://www.w3.org/WAI/research/user-requirements/ (and other resources too)

janina: Call for Consensus forthcoming from Research Questions Task force for gaming accessibility user requirements

TPAC 2026

janina: Reminding people to register for room since host hotel rooms tend to book up quickly

janina: Continuing to work on sessions and breakouts

New charters review

matatk: No new charters this week

matatk: Talking with the Immersive Web Working Group about their chater and had some architectural questions for them. They have made some clarifications

matatk: Spatial CSS is going to be intersting to watch

New on TR

Spec review requests

Issue tracking

WebMCP: accTree question

<matatk> w3c/a11y-tracking#317

matatk: They closed thread on their end and said it's not anything to do with the accessibility tree, which is the answer that we wanted.

janina: We should acknowledge that we were getting ready to make a comment in that direction and we're in agreement. It should be clear we didn't ignore it.

janina: Will draft the comment

WebMCP (as a whole)

<matatk> w3c/a11y-tracking#318

<matatk> That tracking thread links to the google doc we started to work on the comment: https://docs.google.com/document/d/1_Y5X5npxGePkU7andL1ibvi7-JX25frel6wFUNcAizg

matatk: Dr Keith added comment on financial processes. Marcelo added expanded questions about the browser to make an argument for browser being required to provide definite UI to humans.

matatk: Rare in standards to specify a tight level of what UI should be so people can experiment as well as pragmatic reasons. Most of questions are in as they were but open ended to WebMCP team can come back with how they'd solve the problem

matatk: More expansion on how assistive tech perceives the page in the first place.

matatk: Section called deployment/web content authoring question also from Marcelo. Propsing we take that section out for possibly later.

marcelo: Doesn't object to removal. Main reason thought important was to ensure there's equitable experience for assistive technology users. Wanted to engage on that early on

matatk: Thinks its covered by high level question with how will people who use assistive technology get the same information as people who do not.

janina: IT's a key point we'll likely keep coming back to

PaulG: Want to make sure we're respecting the boundary of what WebMCP is proposing. If we're discussing what an agent needs to do, we've crossed the boundray. If there's advertising that WebMCP capabilities are available, that should be human readable.

<Zakim> matatk, you wanted to talk about WebMCP vs MCP, and developer use cases

marcelo: MCP protocol has an extension called MCP apps. Idea is that LLM agent can access API via the protocol. What they get from backend, agent decides how to display that data to user. That transition may have different tools. The way we've been expericing the MCP today will become more complex

<marcelo> MCP Apps extension documentation: https://modelcontextprotocol.io/extensions/apps/overview

matatk: It might not have been best idea to call this WebMCP because it's similar but not an exact mirror of it. This info is good context for us. Some discussion with proponents of WebMCP and it's mostly around what developers want to be able to have agents be able to do. This isn't in the documentation yet and we've asked for it. Seeing that

would probably help answer questions.

matatk: Proposed to do 1) See if we can include Dr. Keith's mention of the WCAG success criterion on financial transactions. 2) Will post the contents of the google doc at the end of this week unless there are objections. Also, asking Paul to review use case coverage and extending for actions.

[css-forms-1] Handle prefers-contrast and forced-colors by default

<matatk> w3c/a11y-tracking#341

matatk: Suggesting default style sheets

PaulG: Hard to know without seeing it. If it lands and there's a problem, they'll get bug reports. If they produce the calculations and say this reaches AA, this reaches AAA, that would be helpful.

<marcelo> +1 on suggesting the conformance levels AA, AAA

PaulG: Web platform test (WPT) is fantastic idea because if it doesn't meet its own criteria if it's not provable that's the whole point.

matatk: Invitation: if anything in the a11y tracking is of interest and you have thoughts, please add comments because its the APA space for discussing things.

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

Diagnostics

All speakers: janina, marcelo, matatk, PaulG

Active on IRC: Fazio, janina, JeffAdams-UN, JenStrickland, JonCohn, marcelo, matatk, Neha, PaulG, Vanessa4