W3C

Accessibility Conformance Testing Teleconference

13 Oct 2022

Attendees

Present
Helen, ToddL, kathy, trevor, Will_C, ChrisLoiselle, thbrunet, Wilco, Daniel
Regrets
Chair
Trevor
Scribe
Wilco

Contents


Trevor: Wilco asked me to run things today.

Daniel: It was really good to meet people again. It had been a long time.
... We finished everything related to WAI-ARIA migration from 1.1 to 1.2.
... There was an issue opened whether aria-label triggers conflict resolution. I think ARIA 1.3 will be addressing that.
... Helen also raised an important point. Whether we should write more rules on manual testing.
... This will be discussed on the CG call today.
... There was some text spacing work as well.

Helen: I also agreed to do public relations for the group

Announcing ACT Implementation pages https://wai-wcag-act-rules.netlify.app/standards-guidelines/act/implementations/

Trevor: CFC pretty much only +1s
... that's going forward

Wilco: All the implementors +1'ed the CFC as well

ACT rules sheet https://docs.google.com/spreadsheets/d/1OSkPFocXk4K3zYLnwS78WLsWO4PvE5yRcsauyefuIUI/edit#gid=0

Wilco: On sequential focus, there's an open PR. I need to resolve the feedback.

Daniel: I'll see what's changed in PR 1925 to avoid conflict

Helen: On iframes, Jean-Yves is working on it.

Daniel: On page title, I merged 1912 this morning.
... We changed the definition, moved something out of the background

Wilco: I think we can put it out for CFC

Trevor: We have a couple from Will, I'll meet with him today to work on that
... I have an open PR on automatic changes. I'm waiting for feedback from Carlos
... I'll probably open a separate PR on changing the applicability. I think there are ways to simplify

Tom: Still working on Text enhanced contrast

Chris: On CSS transforms. Trevor helped me with a PR. It has been approved, PR 1935

Wilco: I'll merge 1935 then

Open ACT pull requests https://github.com/act-rules/act-rules.github.io/pulls/

Trevor: PR 1950, needs reviewers, updating rules to ARIA 1.2
... lots of link updated, couple entries in the glossary.
... PR 1945, consistency to support and assumptions wording.
... I went through the assumptions, and I tried to give more consistent wording.

Daniel: I'll put the italics back.

Helen: For some cognitive impairments, italics are difficult to read.

Trevor: Is there other styling we could use?

Helen: For readability it's probably better to have it plain. Italic suggests it's important to know there are no assumptions.

Daniel: No underscores then

Trevor: 1941, remove outdated support notes
... 1935, I think this is editorial, it can be merged without call for review
... 1933: looks like it's approved
... 1931 is approved
... heading is descriptive, there are a couple change requests.

Helen: His last update was he hadn't had a chance

Trevor: 1925, remove exception from focus visible. That's the item we took to AG.
... 1917 looks ready, but we'll wait for Kathy to get back
... 1916, definition for essential text change. I think this PR may be ready to go. I need to get some approvals.
... Assigned to Wilco, Daniel, Helen
... 1855 from Helen, Jean-Yves is working on that

Wilco: 1742 is getting picked on again by Mark, this was something Chris had assigned.

Priority issues https://github.com/act-rules/act-rules.github.io/issues?q=is%3Aopen+is%3Aissue+label%3APriority

Trevor: Support on title attributes, sounds like that's finished

Wilco: The ARIA 1.2 priority issue is getting addressed. No progress on the other two
... The xml:lang issue can be closed as that's getting deprecated

Text has minimum contrast

<dmontalvo> https://www.w3.org/WAI/standards-guidelines/act/rules/afw4f7/proposed/

Trevor: Will commented on text with a 1:1 contrast

Will: Might be out of scope

Wilco: Should be out of scope, because it's not visible

Tom: There was a separate survey on contrast.
... There was a todo about changing highest possible contrast
... technically it's correct, but trying to figure out better wording

Trevor: Do we have a PR?

Tom: Was out for a few weeks, I'm getting caught back up

Trevor: I can put this in, two birds / one stone
... We still have that todo, but once it's fixed on the second rule, it should be fixed here too
... On Kathy's comment, adjust the contrast on the letter "K"

Wilco: I think I put a text-shadow on the text for that

Trevor: I don't think the open issues need to be addressed
... I wonder if we should put these rules out before secondary requirements?
... I wonder also if we should check color filters or things on the OS.
... At least for a while, Chrome wasn't great with windows contrast mode. It could make things with good contrast into poor contrast?
... Might not be relevant.

Helen: Not all browsers allow you high contrast. I think that's beyond this.
... You cannot control what people have on their devices.

Trevor: Agreed, I don't think it needs to be said in the rule.
... Have we decided we'll go forward with rules like that?

Wilco: I think we can move forward with these rules without secondary requirements. I'm not even sure AGWG will let us use secondary requirements until it's spec'ed out.

Tom: I can be liaison on this.

Better define how rules related to page states

<trevor> https://github.com/act-rules/act-rules.github.io/issues/1953

Trevor: The next conversation I wanted to have was on applicability.
... When we proposed writing an update to the format, we'd been debating allowing subjective applicability.
... There are two ways to go. If we want a subjective applicability, it changes how we can define states in more human intuitive ways, but less quantitative.
... If not we have to rely on specs with defined states.
... Defining states, and transitions between states can be difficult. There are more or less an infinite number of ways that can happen.
... Back in 2021 we talked about state and decided to look at CSS pseudo classes. These match an element when they're in a particular state.
... You can say in CSS, if someone focuses an element, you can change the style.
... CSS states are for the most part pretty concrete.
... CSS gives us very concrete, very objective definitions. The problem is that pseudo-classes are constraint. There's a finite state, if we need things outside this list, we may not be able to write rules for it.

<Will_c> I have to drop to lead a demo

Trevor: An example of something that's hard is form validation done by javascript.
... Another thing would be a dropdown menu. Something that expands / animates. I don't see how to tie the action of clicking a drop-down with the actual state change.
... I think there are three paths;

1. Continue with CSS states. 2. An extension of that, looking at other specs, for example ARIA.

scribe: 3. Or we open and allow subjective applicability.
... Things from specs are well defined and objective, but it can't represent all states.
... The second, we saw with CSS that creating understandable rules can be difficult. You're using CSS as a proxy of the element.
... the other option, subjective applicability gives us much more options on defining states, but makes it difficult to determine state consistently, and may not be backward compatibility?
... I want to start thinking about what we can handle with specs, vs what we can handle with subjective applicability

Wilco: We could also consider input aspects as a way to work around objectivity. We did that with language and with accessibility tree. Those aren't well defined, but we're taking them as input

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: 2022/10/18 10:11:32 $