W3C

– DRAFT –
MATF 3 December 2025

03 December 2025

Attendees

Present
Carol, Detlev, hdv, Joe_Humbert, julianmka, pauljadam, Tanya
Regrets
qbalsdon
Chair
JJ
Scribe
Tanya

Meeting minutes

Updates to EN 301 549 v4 and WCAG2ICT

<JJ> Zakim nominate a scribe

<JJ> nominate a scribe

<JJ> https://www.etsi.org/deliver/etsi_en/301500_301599/301549/04.01.00_20/en_301549v040100ev.pdf

<JJ> https://services.w3.org/htmldiff?doc1=https%3A%2F%2Fwww.w3.org%2FTR%2Fwcag2ict-22%2F&doc2=https%3A%2F%2Fw3c.github.io%2Fwcag2ict%2F

<JJ> ack @hdv

JJ: Summarises the status of the v4.1.0 EN standard and WCAG2ICT. As a group, we probably need to check whether our guidance should be aligned.

<Zakim> hdv, you wanted to mention new TAG document on web user agents https://www.w3.org/TR/2025/DNOTE-web-user-agents-20251127/

User Agent

JJ: Summarises the previous work and the status of WebView, and emphasises the complexity for mobile apps.

<pauljadam> is they because they have more exceptions for apps vs web pages?

Joe_Humbert: Confused about why a WebView embedded in the app does not meet the definition of a web page, even though it is still a web page.

<Detlev> We always audited web views under the app audit acc. to clause 11

<pauljadam> I've always audited web views and apps under same WCAG SCs, I just would do things like not apply Parsing to apps but yes apply parsing to web views, but parsing is obsolete now 🤷‍♂️

hdv: Summarises that the Technical Architecture Board of W3C published a document on the User Agent, which we might want to review. A more philosophical perspective.

<JJ> Tanya: explains reasoning behind comment w3c/matf#63 (comment)

JJ: How should we move forward - the “user agent” concept that only applies to web content, or the broader definition that also includes operating systems, web browsers etc

pauljadam: Trying to understand why EU legislation seems to lean towards applying the User Agent concept only to HTML content.

<Detlev> But it says somewhere that what the user agent is for web, is the patform (OS) for native apps.

Joe_Humbert: UA discussion is closely related to discussion of exceptions, like "except when it is determined by teh UA.." and a couple of other examples (e.g. 2.5.1)

ACTION: If user agent is only applied to web content, consider extending success criteria exception clauses with a term that covers native operating system components / behavior

<pauljadam> If an app is not a user agent and user agents only view HTML and WCAG only applies to web then folks may say that WCAG does not apply to apps and that apps need their own separate guidelines. I prefer to say that apps are web and WCAG does apply to apps and users don't know the difference between an app view or a web view.

Detlev: Our line is that when something might fall under “determined by the user agent,” it should rather be interpreted as "determined by the underlying layer or software" and be considered as an exception.

Set of (Web)pages

JJ: Summarises what has been done and discussed before. It seems that we can use the “page” definition, but we might want to include some examples.

<JJ> WCAG-EM 2.0: https://w3c.github.io/wai-wcag-em/ Sampleset: https://w3c.github.io/wai-wcag-em/#sampleset

hdv: In WCAG-EM we don’t often talk about multiple sets of pages; it’s not very common. In the next version of WCAG-EM, we are updating that. We are talking more about “sample sets.”

pauljadam: I worry about multiple sets of pages. There might be issues with consistent identification and navigation. Within an app, I would consider the app pages to be one set of pages.

JJ: set of pages = one flow? How do we deal with different flows

ACTION: Finetune definition proposal by shoobe01

Underlaying layer

<shoobe01> On screen keyboard

JJ: Summarises previous comments.

shoobe01: what is the point of this definition: e.g. who is in control?

JJ: To allow people who are auditing to understand when exceptions apply.

<JJ> julianmka: Example Apple Pay add card flow.

<JJ> Also JJ - Example 2: NFC scanning sheet.

julianmka: Apple Pay onboarding example, keyboard alerts and permissions - will probably make it clearer to present what each layer means

JJ: on iOS - NFC scanning sheet where you cannot adjust anything

ACTION: Collect examples of mobile app interactions where you cannot access the layer that is showing, e.g. Apple Pay flow or NFC scanning flow

JJ: we have open drafts on some definitions and SC's.

Summary of action items

  1. If user agent is only applied to web content, consider extending success criteria exception clauses with a term that covers native operating system components / behavior
  2. Finetune definition proposal by shoobe01
  3. Collect examples of mobile app interactions where you cannot access the layer that is showing, e.g. Apple Pay flow or NFC scanning flow
Minutes manually created (not a transcript), formatted by scribe.perl version 248 (Mon Oct 27 20:04:16 2025 UTC).

Diagnostics

Maybe present: JJ, shoobe01

All speakers: Detlev, hdv, JJ, Joe_Humbert, julianmka, pauljadam, shoobe01

Active on IRC: Carol, Detlev, hdv, JJ, Joe_Humbert, julianmka, pauljadam, shoobe01, Tanya