W3C

– DRAFT –
APA Weekly Teleconference: Special Presentation

20 Aug 2025

Attendees

Present
Ananya, Demelza, Fazio, Fredrik, janina, Lionel_Wolberger, lola, matatk, mike_beganyi, Neha, niklasegger, PaulG, Roy_Ruoxi, stevef, tsaulter
Regrets
-
Chair
Janina
Scribe
Lionel_Wolberger, janina, matatk

Meeting minutes

Welcome & Announcements

janina: Maturity Model, we expect progress this week

Guest Introduction -- Matthew

janina: Please start the recording

janina: Introduced this special presentation 20 August 2025

matatk: Introduced Lola, who is active in W3C as well as accessibility, including ARIA and TAG

Presentation

Lionel_Wolberger: Lola presented, we direct people to the recording link that will be shared on the list.

<lola> https://github.com/lola-s-lab/accessibility-compat-data/tree/master

Lionel_Wolberger: Can you tell us about how you see your audience and their relationship to WCAG as opposed/contrasted with ACD

lola: We see WCAG as giving criteria on how content should be presented on the web
… think about a text. WCAG says, be this size, meet these requirements, etc
… ACD is saying, if the paragraph element doesn't work, this dataset will tell you that
… when ACD is complete, a matrix or cross-reference could be made with WCAG

matatk: The Web Platform offers building blocks to developers. They can look up the implementation of these building blocks in different browsers
… but participants never had a lookup of how accessibility is implemented, particularly the accessibility tree

lola: The information is quite skewed currently as the accessibility tree is overlooked in these developer resources
… it may be in the DOM and not in the accessibility tree

matatk: Asks after ultimate goal of the work

matatk: would like to see it part of baseline

matatk: There's general greement about this, but do you agree?

lola: Yes

lola: But there are nuances

lola: Integrating into baseline is problematic for several reasons

lola: it itself is a simplification of a simplification

lola: backward compatat also a concern

lola: I've also expressed this sentiment--but again there are nuances

lola: For now adding acb to baseline is tabled

lola: data is more for understanding the support of the feature

matatk: demonstrating that acb works and provides reliable results might take us to the correct integration

janina: Baseline and BCD relatively new terms to me. The outcome defined in those terms is to me not as interesting as to if this is really successful, there will be some kind of integration relied upon by the wider web accessibility community.
… Very much of use to people developing content on the web. If people are relying on semaphores saying green, yellow, red, and that doesn't incorporate accessibility considerations, that adds an extra layer of checking.
… However the data get presented ultimately, when accessibility is included it'll be of great help to content authors.
… Another area we're working in is supporting people incorporating accessibility into the platform.
… I think we've got some compelling solutions in some areas. We've expanded the list of groups who benefit from accessibility considerations. I think when you have ACD and BCD working well, we will know in which areas we need to focus.
… APA tends to worry about which areas we have not covered in W3C specs, via our horizontal review functions.
… Is that a fair understanding of the two areas where this could help?

lola: Yes, good summary. On the relationship between this dataset and APA, I've been thinking about this, having worked with matatk on TAG.
… I'm confident APA could use the dataset in terms of horizontal review, but it's a bidirectional relationship.
… Only part of this can be tested automatically with browsers. The AT side of this still needs human testers. If we have a core data set and want to test new things, we may impact how we test.
… Starting small, but I have blue sky ideas for this project. A multitude of people providing accessibility expertise for the data, so that it meets more diverse needs. Particularly people who use ATs.

janina: Thank you; exciting!

janina: How might this integrate for people creating websites?

lola: one of the main goals is to integrate this into the tools web devs use.

lola: BCD is very effective because it's deeply integrated. E.g. GitHub uses Baseline to ensure new features are safe to use across browsers.

lola: We want to go to where they are - integrate into tools, MDN (the premier place for web docs).
… CanIUse etc.
… IDEs as well. IntelliJ suite includes an HTML IDE that looks up BCD on the fly.
… Mozilla are positive on including accessibility data on MDN.
… I think it's feasible, but lots of work ahead.

lola: Half of this work relies on manual testing.
… So much of the testing for web features is automated. E.g. James Craig from Apple is leading the effort on WPT accessibility testing.

matatk: Asks about wider integration of pwd supporting at?

lola: phps in the future we can expand on what we already have in aria-at driver

lola: even it isn't yet release ready yet

<lola> Thank you all for having me!

janina: Thank you!

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

Diagnostics

Succeeded: s/agend?/agenda?/

All speakers: janina, Lionel_Wolberger, lola, matatk

Active on IRC: Ananya, Demelza, Fazio, Fredrik, janina, Lionel_Wolberger, lola, matatk, mike_beganyi, Neha, niklasegger, PaulG, Roy_Ruoxi, stevef, tsaulter