W3C

Accessibility Conformance Testing Teleconference

20 Jan 2022

Attendees

Present
WIll_C, JennC, kathyeng, Wilco, trevor, Helen, Daniel
Regrets
Chair
Wilco
Scribe
Will_C

Contents


zoom issues. start without me

<Wilco> scribe: Will_C

WIlco: Explains to will about scribing...

Group agrees scribing is fun

CFC: Accept "headers attribute refers to cells in the same table element"

Wilco: should we extend time on this?

Group agrees

RESOLUTION: Extend CFC until next meeting

ACT rules sheet and Surveys

78fd32 has all approvals, needs one week call for review

Trevor: will send out call today

307n52 - presentational children. No Karen today

Changes weren't done, Karen is in a conflicting meeting

Will: Having difficulty with github, will make meeting with Wilco to resolve 80f0bf

WIlco: in future, when opening pull requests related to issues, please put Pull# in google sheet

a25f45 - in CFC.

Open ACT pull requests

no new pull requests since last week

#1775 for Wilco to look at, #1773 too

PR 1760 review ended today

Helen: PR 1747 - was broken ad Helen fixed.

PR 1747 - too many references to 1.3.1. Those have been removed, except for one. Will C approved

Jennifer C, Trevor B, and Carlos to review too

PR 1740 is going to be merged.

group agrees it's ok to merge. It's an editorial change.

Question: Are we going to start having a 'related rules' section?

Wilco: Yes. and don't need to go back to retroactively add it. Just going forward, for closely related rules

Will C: review question.

WIlco answered.

Heading has non-empty accessible name (1ec09b)

Wilco: I have changed the tab names in the sheet.

In Tabs, removed rule ID, and gave two word summary instead

Daniel: It helps with screen readers.

Most of rule has been got through. Rule requires that all headings have accessible name

Some of group don't think that should be required because SR consistently ignore empty headings.

Will C: I think this should be an issue. Will try and recreate blank headings in the accessibility tree of NVDA

Wilco: My vote changes if Will is right.

Kathy: I agree. But I also have a different comment about a restriction.
... Should this rule apply to visible headings only?

Trevor: I was trying to test the heading

Helen: I do not think hidden headings apply. It's a legacy software issue, rather than a failure of checkpoint

Daniel: It's a fair point. FIrst thing that needs to happen is that there is some visual indication of a heading.

Wilco: The page structure needs to be available in a programmatic way.
... 1.3.1 is about information relationships and structure. The visual structure needs to be available in a programmatic way. But if you have not needed headings in the heading list, that's a problem. Means structure of page is no longer programmatically available

if hidden headings end up in the heading list, it seems like an issue

Trevor: I tested one browser and NVDA and blank header did NOT show up
... I wouldn't be against adding a small section to this rule about ignoring the style section of headings. Like if you were to make a heading 15px with a blue background, to make a division section, it's an empty heading that breaks up a page that is maybe structural?

Kathy: That might be a different rule. This rule is heading structure is there, but no accessible name.

WIlco: Yes, this is examples like aria hidden headings, images in a heading without alt. but there is one with an empty heading. Example 6

Kathy: Is a heading still a heading if there is nothing on the screen

Helen: Technically this is not invalid markup.
... More of a best practice than a failure.

Kathy: This isn't a visual heading so that it wouldn't need heading markup. part 2 of 1.3.1 - if its not a heading, don't give it heading markup.

Wilco: I propose - if it turns out there is a good number of AT that include empty headings in heading list, it's reasonable. if not, then deprecate it.

+1

<Helen> +1

<dmontalvo> +1

<kathyeng> +1

<JennC> +1

<trevor> +1

RESOLUTION: Will to test failed examples in AT, if it turns out these are ignored in heading list, the ACT TF recommends the rule to be deprecated.

<dmontalvo> +1

Wilco: JAWS, NVDA, Voiceover and talkback should be tested for this.

Kathy: If it's just visible headings only, then the applicability of the rule would be narrower.

Wilco: there are two examples without visible headings.

Kathy: If it comes back that AT does ignore these, I would still think that 1.3.1 is dealing with visible presentation only
... I would still fail 5 and 6 regardless of AT test results.
... These are things that have heading markup that should not have heading markup.

Daniel: Example 4 is failing because of misuse of role="presentation"

<dmontalvo> Will: That seems like a misuse of role presentation and that's why this should fail

Kathy: Good example because you visually have something in the image.

Wilco: Can you suggest a change to figure out where your opinion differs from what the rule says?

Kathy: in applicability, adding 'is visible'.

WIlco: 1, 2, 3, 4 are visible and would fail.

Kathy: 5 isn't visible and would fail 1.3.1 but not for this reason.

Wilco: we need to spend another week on this one.
... Needs to restate a negative comment.
... Should we clean up passed example 3?

Already a rule.

We have a hard stop at 10

Summary of Action Items

Summary of Resolutions

  1. Extend CFC until next meeting
  2. Will to test failed examples in AT, if it turns out these are ignored in heading list, the ACT TF recommends the rule to be deprecated.
[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version 1.200 (CVS log)
$Date: 2022/01/24 07:13:35 $