W3C

– DRAFT –
APA Weekly Teleconference

22 Jul 2026

Attendees

Present
chiace, Fazio, janina, JonCohn, matatk, Neha, PaulG, Roy_Ruoxi, stevef
Regrets
-
Chair
Janina
Scribe
chiace

Meeting minutes

<PaulG> present

zakim

zakim next item

Agenda Review & Announcements

David Fazio and Janina were in Czech Republic last week and it went really well

TPAC 2026

TPAC Registration is open, register or request the support

matatk will participate to 8 breakouts, scheduling is really important

<matatk> Propose breakouts on this repo: w3c/tpac2026-breakouts

matatk the support to attend is in 2 parts as invited expert: 1) registration fee 2) travel and support . See from the IRC session, it is in the registration page with all the info, You can attend in remote or in person. The deadline is the end of next week the 31st. Moving on to breakouts there were three APA breakouts.

TPAC Breakout Proposal: Accessibility and Privacy

matatk The first breakout proposal discussed was Accessibility and Privacy. The session would explore the tension between protecting users of assistive technologies from fingerprinting and enabling useful accessibility adaptations. The group wants a broader discussion on how these trade offs should be assessed, rather than automatically excluding

potentially valuable features because of privacy risks.

matatk suggests to add the comments on the Google Doc

matatk provide feedback before the end of the week

TPAC Breakout Proposal: What does good AI training look like from an accessibility perspective?

<matatk> Relevant Web & AI IG minutes: https://www.w3.org/2026/07/14-webai-minutes.html#b564

matatk This came from the recent discussion from the Web and AI Interest Group. There is the tracking issue and the Google Doc to provide comments. The proposed topic is what good AI training should look like from an accessibility perspective. The current proposal is still very preliminary, and APA members were invited to review the Google Doc

and provide feedback through either the document or the related tracking issue. The aim is to refine it enough to share outside APA by the end of the week.

janina emphasised that AI systems should not reproduce the poor practices already present on the web, but should instead help improve them.

This issue extends beyond accessibility and will likely require the involvement of those developing models and agentic systems. However, APA should still contribute to the discussion to help ensure that AI produces more accessible content and solutions, rather than creating additional problems that later need to be fixed.

matatk the deadline is like it is no even the 8th of October, trying to get a good spot in the timetable

New charters review

Roy_Ruoxi no new charters, charter 1 is a web security.

<Roy_Ruoxi> w3c/strategy#552

matatk we choose to sign up on the web app SEC one. The reply from ATAG for a working draft by the end of the charter.

New on TR

Strings on the Web: Language and Direction Metadata

<Roy_Ruoxi> - tracker: w3c/a11y-longitudinal-tracking#52

<Roy_Ruoxi> - spec: https://www.w3.org/TR/string-meta/

Roy_Ruoxi we got 3 new. Strings on the web is to add best practices from metadata to indicate the directions of this document

<matatk> FAST checklist tracking issue: w3c/a11y-tracking#335

matatk There is an issue that was filed called Fast Checklist.

matatk The section that is most of interest is if the technology that document semantics is relevantm that section is highlited. As well internationalisation support serviced by naking them more clear. Check on how we can be involved or could give him some extra use cases or what have you/ why wpuold we proposing to do. A11Y tracking 335

PaulG will help in reading them

Character Model for the World Wide Web: String Matching

<Roy_Ruoxi> - tracker: w3c/a11y-longitudinal-tracking#53

<Roy_Ruoxi> - spec: https://www.w3.org/TR/charmod-norm/

janina COGA, is a nice to have level for people with cognitive learning disabilities. Like their text that is typically in their language presented verbal vertically

<matatk> FAST checklist issue for this one too :-) w3c/a11y-tracking#336

CSS Linked Parameters Module Level 1

<Roy_Ruoxi> - tracker: w3c/a11y-longitudinal-tracking#54

<Roy_Ruoxi> - spec: https://www.w3.org/TR/css-link-params-1/

matatk A second Internationalisation specification was introduced, together with its FAST accessibility checklist. As the specification affects the visual rendering of content, several accessibility considerations may apply.

matatk Links to both the checklist and the specification will be added to the tracking issue. The group then asked for a volunteer to review the character model and identify any potential accessibility risks, while cautioning reviewers not to take on too much work.

matatk This spec introducs a way to pass CSS values into resources such as SVG images so that they can be used as CSS custom environment variables in the destination resource

PaulG We are applying the thing that we already know we use to SVGs. He doesn't see any accessibility issues here

Neha They already take care of making a contrast or adapting to the system setting of accessibility. And the user who is now giving CSS properties and changing the settings is already present in the switching

PaulG If you do not change any of the code that's an IMG tag, so if you are doing IMG and you are sourcing an SVG then you can add the param to it. Than taht frame needs to be linked and then you would need to update. You would have to opt in by editing your SVG files and adding the environment. Variable function

matatk We should add an accessibility consideration section

PaulG if there is time they will be added

Spec review requests

o 2026-07-14: Digital Credentials

<matatk> w3c/a11y-tracking#325

o 2026-07-29: Attribution Level 1

<matatk> w3c/a11y-tracking#334

matatk The group briefly reviewed the Digital Credentials specification, which concerns an API allowing browsers to communicate with a digital wallet to retrieve, issue, or present government style digital credentials.

matatk Although the specification had likely been discussed previously, no notes were found in the tracking issue. As the review deadline was approaching and time was limited, the group identified the need to confirm the earlier discussion and update the tracking record.

o 2026-07-31: CSS Image Animation Module Level 1

<matatk> w3c/a11y-tracking#333

PaulG UI scope, this API doesn't even define a standard schema, so I think that also mean that it is even too low level for that question . There is no human readable anything in that

<matatk> See Paul's comment: w3c/a11y-tracking#333 (comment)

matatk it is actually an ISO Standard rather than W3C, so ther is not very much that we could do about it.

<matatk> Shameless plug for my 'gaps' issue, too: w3ctag/gaps#15

PaulG CSS image animation. The discussion on CSS Image Animation highlighted several accessibility issues with the proposed approach. Simply adding tabindex to an animated image does not provide appropriate semantics, state information, or accessible controls.

PaulG Interactive animated images may require a new accessibility mapping so that assistive technologies can recognise them correctly. They should also include controls such as play and pause, and animations triggered by hover or focus should respect the user’s reduced motion preferences through the prefers-reduced-motion media query.

Issue tracking

WebMCP (continued)

<matatk> Tracking issue: w3c/a11y-tracking#317

matatk the accessibility tree should not be used for agent specific information. WebMCP appears to cover broader agent interactions than simply completing forms, raising concerns about the increasing “botification” of web content.

matatk concern that adding agent only information to the accessibility tree could pollute it and create architectural problems. Also it felt that embedding agent specific mechanisms directly into the web platform requires significantly more experimentation before being standardised.

matatk The declarative version of the API, mainly focused on form completion, was considered a safer and potentially more valuable area for experimentation.

matatk From an accessibility perspective, it could support useful outcomes, particularly if browsers are required to keep users informed or allow them to choose whether they remain involved in the process.

matatk A consolidated comment is still being prepared, drawing together the discussions and feedback collected over recent weeks.

Due to time constraints, the remaining items were deferred to the next meeting. Members were asked to add comments to the TPAC breakout proposals in the shared Google Docs so they can be developed further.

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, Fazio, janina, janina1, JonCohn, matatk, Neha, PaulG, Roy_Ruoxi, stevef