W3C

- DRAFT -

Accessibility Conformance Testing Teleconference

26 Mar 2026

Attendees

Present
Wilco, giacomo-petri, Jean-Yves, Kathy, Godwin
Regrets
Chair
SV_MEETING_CHAIR
Scribe
giacomo-petri

Contents


scrive+

scribe+

RESOLUTION: We are now going to only use #wcag-act as our IRC channel

Liaison standup

<Wilco> scribe+

Jean-Yves: nothing much, group left to Godwin, I'll start tomorrow (Rachael's subgroup)

Godwin: Errors meeting on Monday (Hidde), putting things together well
... want to discuss something related to the meeting later today

Kathy: did some offline review of my sub-group work

<Wilco> Giacomo: We had some concerns about deciding what's meaningful title and what isn't

<Wilco> ... the rules aren't too clear about this. People can argue about what is too much and what's not enough. We tried to add more examples to clarify

<Wilco> ... We used some examples, like in a checkout process if every step is a different view the title should update.

Wilco: worth digging a little bit more, maybe not just examples, but adding also a note for clarification

giacomo-petri: there might be some repetition if we use both the understanding material and the notes within the ACT Rule

Wilco: potentially using a link to calrify what qualifies as sufficient vs not (from the ACT Rule to the Understanding/Methods)

Jean-Yves: possibly having a definition for meaningfulness

FAQ: general rules

Wilco: What about modal dialog that do not reload the page?

<Wilco> https://docs.google.com/document/d/1i_O5J7D4BWMZAB7NaD4iplDRuqOjCvCGAaHdAirLUhE/edit?tab=t.bo2e33svugxv#heading=h.8k3m19m21u2

giacomo-petri: we discussed it in the group but left out for now due to the different possible scenarios; e.g., shipping address popup in a shipping page, that asks you to validate your address; it's not so drastic to require a page update
... while a store locator opened in the homepage would require it

Kathy: we are first aiming to HTML Rules, so I've updated the WCAG3 guide to reflect even better this

Godwin: with the errors group, we are talking about how errors should be presented, the group was lost when applying HTML to this
... as they are getting into it, there are a bunch of ARIA to be added and tech aspects, and they get a bit lost
... if we write more generic manual test, it would be easier for reader

Wilco: it seems to me like knowing what's semantics require it's quite a critical piece... how can we do that without

Godwin: IMO it's critical for implementation, not for testing
... I believe it's harder for people to understand what they have to test if it's too technical

<Wilco> giacomo: I believe that we're perfectionists and that a11y isn't for everyone. When it comes to semantic structure I believe you need someone technical to validate if the implemtnation is correct.

<Wilco> .. If you don't know the structure is correct you need to test with 1000 assistive tech. One specific combination coudl be able to adapt.

<Wilco> ... For some of the requirements you really need some technical pass/fail example.

Godwin: I'm concerned with this group in producing something that is very hard to read for people that doesn't have technical knowledge

<Wilco> Gaicomo: I think that's why we need another section in the requirement such as understanding to clarify the intent to non-technical people

<Wilco> ... In WCAG 2 we have the intent section, and we have the ACT rules that clarify the violations.

<Wilco> Gaicomo: Passing examples say you are passing the provision, not that there are no other options to do something

Godwin: I do worry that if we are only technical and to precise, we might be too strict

Jean-Yves: I see the need of both, if too technical it's a problem for non-tech people (e.g., people didn't join the gorup due to too tech aspects); there is room for a need for audience that is not technical
... but if it's not technical, then it will lead to too much subjectivity. We need both the tech rule and the non-tech rule, for different people
... I can see the need for both, maybe in different place, but we want both kind of people involved
... maybe we can also try to be lesss tech for some ACT Rule if it works, let's see if it works
... it's up to the ACT people also to work on evolving technologies and ensure rules are matching them

Wilco: thanks for bringing this up, it's an important conversation to continue having
... if it's a compelling topic, please let us know

Godwin: if I'm testing for errors in forms, even with a strong dev background, I won't start looking at the code; so in this case, even audience is different, the test might be very similar
... ACT Rules in WCAG, who are the indended users of that?

Kathy: from my understanding ACT involvement in helping to create the test instructions and ruiles for wcag 3 are to:
... 1. provide more details of what we have in wcag 2
... 2. help in refining the requirements and ensure they can be testable
... having ACT with existing rules, and liasons to help sub-groups develop the rules. I've seen improvements in adjusting requirements (or at least creating discussions within the group)
... the sub-groups realised sometime that applying ACT rules, maybe the requirement is not well written, covers more than it should be, or the opposit

Rachael: agree, improve the wording of the requirement notably
... also improve coverage for WCAG 3 from the beginning
... maybe not full coverage, but for sure more than what we have
... we appreciate your effort in helping us

Godwin: will them be published?

Rachael: yes

Godwin: who we intend to be reading and benefit when ACT are published in wcag 3

Kathy: I'm almost maintain manual methodology, and whenever these intructions and rules are aavailable, I would update these to cover as much as possible
... manual methodology initially covered the techniques in WCAG 2, and we discovered they were too generic, that we had a lot of stuff to do to cover scenarios

Wilco: IMO, technical aspects are the foundation and other things can be built on
... there are a variety of testing methodology (RGAA, etc.). People need additional information to test edge cases.
... and this is why is very technical; so it requires people with a tech degree to understand the rules.

<Rachael> This may be helpful: https://docs.google.com/presentation/d/1klORYoNQq3oDMJLc52q6Qf5gCTIcGHsm2kBbroj7bMM/edit?slide=id.g3cd4cba3573_0_0#slide=id.g3cd4cba3573_0_0

Godwin: procedures as a manual way to validate requirement, is it correct?

Wilco: yes, usually there is a layer on top
... for error messages for example I need to know when aria-labelledby or other tech solution is sufficient or not. If we don't specify it will be too geenric and won't qualify something as passed/failed
... do we want to take another look at what Kathy changed?
... I'd like to change the word "only" with "first"

Kathy: do you want to include ARIA, CSS?

Wilco: I think they are not separate from HTML

giacomo-petri: agree with Wilco, and maybe adding also javascript which might be required for specific requirements

Wilco: anything else?

Godwin: before publication we need to think about pdf, mobile, etc. Someone was confused and worried we are not covering other technologies

Wilco: Good rules for HTML should consider ARIA, CSS, Javascript will also make easier to build rules for other techs in the future
... question 10 in FAQ is new, any comment on this?
... about subjectivity of the rule

<Wilco> gaicomo: I think it may mix what we discussed about title. I think we can use examples to clarify intent. "Checkout" is fine on a single page step but not on multi-view steps. We should have additional content such as an understanding document to explain the background, maybe including real-world examples.

Wilco: this is a question for Rachael
... there might be a need for an understanding document
... this is not maybe an ACT conversation probably

Jean-Yves: if we have subjective rules in ACT we need some documentation somewhere with explanation

Summary of Action Items

Summary of Resolutions

  1. We are now going to only use #wcag-act as our IRC channel
[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version 1.200 (CVS log)
$Date: 2026/03/26 15:00:38 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision VERSION of 2020-12-31
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Default Present: Wilco, giacomo-petri, Jean-Yves, Kathy, Godwin
Present: Wilco, giacomo-petri, Jean-Yves, Kathy, Godwin
No ScribeNick specified.  Guessing ScribeNick: giacomo-petri
Inferring Scribes: giacomo-petri

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth


WARNING: No date found!  Assuming today.  (Hint: Specify
the W3C IRC log URL, and the date will be determined from that.)
Or specify the date like this:
<dbooth> Date: 12 Sep 2002

People with action items: 

WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)


[End of scribe.perl diagnostic output]
This is scribe.perl Revision VERSION of 2020-12-31 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00) Default Present: Wilco, giacomo-petri, Jean-Yves, Kathy, Godwin Present: Wilco, giacomo-petri, Jean-Yves, Kathy, Godwin No ScribeNick specified. Guessing ScribeNick: giacomo-petri Inferring Scribes: giacomo-petri WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth WARNING: No date found! Assuming today. (Hint: Specify the W3C IRC log URL, and the date will be determined from that.) Or specify the date like this: <dbooth> Date: 12 Sep 2002 People with action items: WARNING: IRC log location not specified! (You can ignore this warning if you do not want the generated minutes to contain a link to the original IRC log.) Info: Document content looks like HTML Proprietary No warnings or errors were found. About HTML Tidy: https://gi