W3C

- DRAFT -

ACT Rules Community Group Teleconference

31 Oct 2024

Attendees

Present
CarlosD__, Helen, Sage, Dan_Tripp, giacomo-petri, thbrunet, wilco, rachael
Regrets
Chair
SV_MEETING_CHAIR
Scribe
dmontalvo

Contents


<CarlosD__> scribe+ Dan_Tripp

<CarlosD__> scribe+ dmontalvo

scribe- Dan_Tripp

ACT stand-up

Carloss: Did some reviews during last week at the Office Hours

<Wilco> https://github.com/act-rules/act-rules.github.io/pull/2220/

Wilco: Working on a PR to update all rules to the ACT-Rules Format v1.1

Dan: Reviewed Wilco's request, and also updated my PR

Sage: Doing surveys

Helen: Chatted with a user of voice recognition software. He's agreed to create a google survey with the examples that Dan built

Calros: Would the alternative be to host this on the W3C survey tool?

Daniel: Ideally we can create it on the W3C space, I'll create one and try to give you write access to it

Carlos: Are you collecting personal tuff?

Helen: Just thoughts on the examples, email addresses
... I'd like to avoid people doing it multiple times

Carlos: Still that is personal information, What is the procedure in this instances?

Daniel: If we are not rendering code, I strongly recommend using the W3C Balloting system

<CarlosD_> Daniel: Taken care of redirectts, Opened a page on the CG blog and opened a PR on security and privacy

Daniel: Cleared up some misplaced redirects for rules

giacomo: I opened a PR based on discussions with Wilco, reviewers welcome

ACT TF update

Wilco: No TF updates, working through the rules annual review

Carlos: Was there a deadline for the rules submission to the AG?

Wilco: We were tarlking about when the ideal time would be for the changes to the Format. We need to get this done before the end of the Charter. There are minor changes to the Format that Daniel put together. We need to do HR first and then we can move to CR

ACT F2F meeting announcement https://github.com/act-rules/act-rules.github.io/issues/2201

Carlos: We have decided two weeks ago to host our F2F meeting in Edinburgh hosted by Mark on the 6 and 7 February
... We would like to confirm who is going to attend in person.
... Wilco, Jean-Yves, Daniel, Mark, and myself

Helen: I plan to be there

Tom: I won't be able to attend in person

giacomo-petri: I don't know yet. I can attend remotely, just not sure if I'll be able to make it in person

Dan: Not in person this time, probably remotely

Sage: Same here, I don't think i'll be able to attend in person, probably remotely

Tom: I could attend remotely on Friday, not on Thursday

Carlos: Are we putting a survey to confirm people's attendance?

<Wilco> https://github.com/act-rules/act-rules.github.io/wiki/2025-F2F-meeting-edinburgh

Wilco: It was just a mention. Now we know that there are more than five people, I don't think the survey is needed
... I am working on updating the wiki page with event details

Applicability in color contrast rules https://github.com/act-rules/act-rules.github.io/issues/2215

Carlos: The contrast rules have exceptions on the applicability, mainly for disabled widgets
... But there are anchor elements without href that are not covered by the exceptions
... We are proposing to be more conservative, basically targetting at whether the element has aria-disabled

Dan: I've been recommending clients to add href = javascript void. That seems pretty consistent

Wilco: You can't put aria-disabled on an a element without a href. We should not encourage people to do this to get around this problem
... As soon as you remove the href that's no longer an implicit role of link

Tom: href="#"

Wilco: That is a clever idea. There should be other ways.
... I dont see this that often

Dan: I see calendars that have a link and then they remove the href
... Would href="#" jump somewhere?

Tom: Since it does not have a href value it won't jump anywhere.

Dan: Role="link", that was mentioned by Tom, would it improve anything?

Wilco: The most common scenario is that the href gets removed

Tom: If you put role="link" it becomes a widget

Wilco: And the rule passes it

Carlos: The rule becomes inapplicable because that is an exception in the applicability.
... I read we don't want to encourage aria-disabled

Wilco: The real fix maybe is to add a subjective part to the applicability that the target is not interactive

Carlos: Removing the "disabled" and replacing with a subjective part?

Wilco: I think the "disabiled" can stay

Carlos: And then add another list item to the exception?

Helen: I'd prefer Inactive items that do not need to be read

Wilco: That does not seem part of WCAG

Helen: The problem with the use of disabled items is that you can no longer read personal data, and that is a UX issue

Wilco: I think the best we can do is to add a note

Rachael: For 3.0, please email me the request and I'll get it added to the correct location

Wilco: If it is an incative user interface component then it does not have to meet contrast

<Helen> https://web-a11y.slack.com/archives/C050D9NHZ/p1730222960406989 - for the discussion I was mentioning :)

Tom: Without the href it has no role and it's not a user interface component

Wilco: Or it is a user interface component, it needs a role, and it requires to meet contrast

giacomo: Browsers expose the disabled state even though the tag is not a widget

Tom: I can't distinguish this from a jump target

Wilco: Screen readers can't either
... If that is an inactive component that should be communicated to the screen reader
... I think the position we should take is if it's disabled it should have a role. Even if it is a false possitive, it's coming from another accessibility error. This we allow in the assumptions

Tom: That code is always going to trigger issues

Carlos: Don't we have a rule that checks aria-disabled on non-widget elements?

Wilco: No
... Sorry, we do. You can disable a group or a widget

Carlos: That should be flagged already

Dan: Not all implementations honor this, not sure if it's because aria-disabled became deprecated for generic
... The tool implementations may or may not match the rule these days

Wilco: We should then say something about allowing aria-disabled in non-widgets

Carlos: I agree, because we have a rule that flags this elsewhere, so it is a false possitive. I think the assumptions could take care of it
... Does everyone agree/

Wilco: We could change the applicability to say "the ancestor is disabled"

<giacomo-petri> +1

Wilco: if it's a disabled interactive control, then it needs to have a role, you can't just remove an href from an anchor and claim that it's disabled

Giacomo: Agree with Wilco

Wilco: I'll leave a comment on the issue with our conclusions

Dan: I am still processing it

Helen: JavaScript is not always the solution. If we replace the href value with a fuzzy JavaScript function we should raise an issue

Dan: The combination of these two things is confusing. Maybe these should be mutually exclusive

giacomo: Having an a tag without an href value is also invalid
... At the end having a link with aria-disabled or having no href doesn't change for some users, for example low vision
... I would require the author to set role="link" based on how the SC is written

Tom: The a without a href is not a link in HTML, according to HTML

Dan: That can be seen as true at the same time as what Giacomo said

Tom: The fact that HTML uses the same element for links and text is an old artifact

Dan: Would you go to developers and explain it?

Tom: They should change their code because what they specified is not a link

Carlos: What Wilco is proposing is that these rules shouldn't check if aria-disabled is used or not because we have another rules for that

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version 1.200 (CVS log)
$Date: 2024/10/31 15:01:42 $

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)

Succeeded: s/the ACT/all rules to the ACT/
Default Present: CarlosD__, Helen, Sage, Dan_Tripp, giacomo-petri, thbrunet, wilco, rachael
Present: CarlosD__, Helen, Sage, Dan_Tripp, giacomo-petri, thbrunet, wilco, rachael
No ScribeNick specified.  Guessing ScribeNick: dmontalvo
Inferring Scribes: dmontalvo

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) Succeeded: s/the ACT/all rules to the ACT/ Default Present: CarlosD__, Helen, Sage, Dan_Tripp, giacomo-petri, thbrunet, wilco, rachael Present: CarlosD__, Helen, Sage, Dan_Tripp, giacomo-petri, thbrunet, wilco, rachael No ScribeNick specified. Guessing ScribeNick: dmontalvo Inferring Scribes: dmontalvo 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.) line 412 column 1 - Warning: trimming empty <ol> Info: Document content looks like HTML Proprietary Tidy found 1 warning and 0 errors! One or more empty elements were present in the source document but dropped on output. If these elements are necessary or you don't want this behavior, then consider setting the opt