W3C

Accessibility Conformance Testing Teleconference

10 Feb 2022

Attendees

Present
Helen, trevor, Wilco, kathy, Will_C, KarenHerr, Daniel, JennC
Regrets
Chair
Wilco
Scribe
JennC

Contents


<Wilco> scribe: JennC

zakim take up next

Zakim: take up next

ACT rules sheet and Surveys

ACT rules survey review completed.

Open ACT pull requests

Pull request for fixing line height examples 9 and 10. one approval, awaiting others.

Form field name, Fix markdown issue #1792 to be reviewed by Trevor.

Update for Video description rules will be reviewed by Wilco.

Update aria hidden focus #1444 will be reviewed by Will, Helen

Pull requests review complete.

Heading has non-empty accessible name (ffd0e9)

Introduction from Tom Brunet - Accessibility at IBM. Accessibility Checker Tool open sourced a year and a half ago which now uses ACT. First time attending the calls.

Introductions from each team members to Tom.

Wilco: We have spent many meetings and taken breaks from reviewing Heading has non-empty accessible name. Will: found that headings are ignored by other screen readers except one. Assumption that screen readers would read empty headings, but the majority of them do not. Kathy is leaning towards this rule not being for 1.3.1 but for 2.4.6.
... is the only reason you object to the rule is it that it's mapped to the wrong SC?

<Will_C> the only one was a very old version of VO.

Kathy: no. If you have an empty heading, that's not a 1.3.1 violation, but that's a failed example of 2.4.6. It's a failure of Headings & Labels (2.4.6).
... You have a visible heading, but an empty accessible name - that I agree would be a 1.3.1 failure.

Note: The screen reader that read the empty heading was Voiceover on Safari, older version

Wilco: If we all agree with kathy's interpretation, we have to break the rule into two.

Helen: I agree with Kathy that there are rules in here that don't technically fail 1.3.1.

Wilco: Failed example 5 and 6 are what we're talking about.

Will: If you're claiming that the image is decorative and the accessible name is empty, that's ok. I feel that putting the rule against 2.4.6 is ok.

Wilco: Labels and headings - by a thing, a heading is a label by definition. For 2.4.6 that's the definition, it labels a control. So even though there's an H1 element, in failed example in 5 and 6, they don't function as headings.

Will: Silly question, but what about 3.3.2 related to labels? Ok, that's more about inputs.

Kathy: Looking at it, when headings and labels are marked up correctly, this ensures that headings are meaningful - in the current documentation, there is a tie to 1.3.1 and 2.4.6.

Helen: I agree with Kathy because it's about headings being meaningful and descriptive. Thus, it fails 2.4.6. Just because technology ignores it, doesn't mean it's not still there. With the visual image being used as a heading, I can presume it's descriptive, but they've failed to add an alt, which is 2.4.5. So there is a crossover with other success criteria.

Will: If you fix the heading error, all of it goes away.

Helen: Trying to think of it technically. If they have not included an accessible name, and it's ignored by the assistive tech, it still doesn't mean it's not wrong

Wilco: We have a significant divide in the group. This should go to AG.

Helen, Kathy: Yes! Have another group discuss this.

Wilco: Have AG look at it and have our guidance to which direction to take it. To split the rule, it sounds like a lot of work if AG just disagrees. So the next step is we take two failed examples - maybe 6 and 1, and propose they review it.

Helen: +1~

Wilco: To clarify, it asks these two questions, and provide the screen reader data that was provided and let them review. Daniel, I appreciate you taking it.

Daniel will pull together the information as liaison to AG to submit for review.

Line height in style attributes is not !important

Wilco: Wow - six reviews conducted quickly. Let's review the responses. Kathy noticed the question for Examples 9 and 10 don't wrap in Edge.
... We do not have any implementations so it cannot go to AG until we have one.

Daniel: I'm good with us surveying this and not AG publishing this until there is an implementation.

Wilco: Question in the review - What is the reason the applicability explicitly states the rule is related to soft wrap break.

Trevor: review with Jean-Yves - who thought that the distinction for soft wrap was technically sound and needing to be specified.

Wilco: What do you call a pre-element that has a line-break? The scope of the rule is slightly narrower than considered and Trevor agrees this is the gist of it.
... Will had asked what happens when CSS is turned off. Wilco confirmed it assumes that CSS isn't changed in any way.

Helen: My thoughts in GitHub may be outdated so not sure.
... Bugs looks unresolved and now I've found a way in GitHub to review them further.

Wilco: We have one fix that needs to go in - #1793 are all approved. This will be merged. Are we all good with this rule?
... As soon as we have the implementation for this rule, should we survey again or send to AG?

Helen: What is the definition of implementation?

Wilco: We expect the rules to have at least one successful implementation when it's applied rather than just an example.

Helen: I recommend a quick review based on the implementation.

<Will_C> Will admits he knows little about it too

Wilco: You are welcome to request another survey, and we can survey it again. AG will also have a call for consensus.
... Ok - we will have another survey once the implementation is done.
... Jean-Yves may be working on an implementation so we may have one next week.

What to do with deprecated rules?

Thank you!

Wilco: I have added a new status for rules - "Deprecated". We now have Deprecated as an option on the status list. One for now - "Image filename is accessible name for image"
... Do we want to apply this status?

Helen: Yes. I am currently deprecating a few.
... I feel they should not be deleted completely, but archived and separated from the main rules so they don't get referred to accidentally.

Karen: I agree - stops this from being misunderstood as missing.

Daniel: I agree. We may need to elaborate on the rationale behind why we are deprecating the rule as well.

Wilco: Yes. The reason for the deprecation for the rule is included. I propose "Deprecated Test Rules" as a heading on the main page, under this title. That includes the reasons as well.
... On the page that lists all the rules, I propose we have a section under the heading "Deprecated rules" with the list/links of those rules.
... All right, I hear agreement. Why don't I post this as a pull request.
... I will work on that and wrap that up next week. Have a great day.

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: 2022/02/11 12:20:42 $