W3C

Accessibility Conformance Testing Teleconference

14 July 2022

Attendees

Present
ChrisLoiselle, Daniel, Helen, JennC, kathy, Rachael, ToddL, Wilco, Will_C
Regrets
Daniel
Chair
-
Scribe
Will_C

Meeting minutes

ACT rules sheet

Wilco: meta element 1831 may be ready

<Wilco> https://github.com/act-rules/act-rules.github.io/pull/1831

1831 - needing one approval

Wilco - 1831 can be merged when we have approval

Wilco: I will ping JYM

Wilco - after merge it can go into CFC (1831)

Wilco: Helen needs help with Object and iFrame.

Helen: trying to determine when ALT is ok.

Helen: part of the feedback is about defining data type defining accessible name

Helen: I moved the bit about objects and Alts into 'accessibility supports' section

Helen: JYM said to remove it, but I moved it and added a note. I have a bit more work to do.

Helen: Adding examples is a bit out of my wheelhouse

Wilco: We need volunteers

wilco: I'll take it.

Helen: akn7bn - iframe - asked for a new PR with feedback

Helen: JYM has put two comments on it we need other reviews of

Jenn: Scrollable elements will be reviewed tomorrow

Open ACT pull requests

Wilco: mumbling...

Wilco: 1890 - assigned to wilco, helen and daniel

Wilco: 1886 - svg, need a third

Wilco: will respond to question in it

Wilco: 1884 - JYM already approved. many people will review

Wilco: 1869 - JYM already approved. many people will review

Wilco: Todd is getting a PR

Wilco: 1804 to Todd

Wilco: Should be all the pull requests.

zakim take up next

Priority issues

Wilco: This is a new agenda item

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

Wilco: may need to recur

Wilco: Last week, talked about an implementer skipping certain test cases

Wilco: don't want that for the future

Wilco: Some of these issues have been open for 2 years

Wilco: in addition to other work. It's important that we don't have known issues

Will: what's the level of expertise?

Wilco: most are fairly technical

Helen: I would like to take one - volunteering for 1810

Wilco: This one is more of a discussion

Wilco: lets have it now

Wilco: issue with treeitems. ARIA isn't the rule, it's 1.3.1 and should be a failure. Semantically it's wrong. and leads to unpredictable results

Wilco: when you have different roles under a menu, the styling will look very weird

Will: wouldn't this fail 'certain children'?

Wilco: Yes and no, and you can argue that. But you could change treeitem to button

Helen: this is an inferred rule. So this is assumed knowledge. B/c it's not within the spec, it's ok. That's what he's saying.

Wilco: double problem of a bad thing to put in a menu and also missing parent

Wilco: Do we think this fails 1.3.1?

WIll: fails

Kathy: not familiar enough, but in the behavior between menu item and tree item.

<Helen> My search for the definition got this: https://developer.mozilla.org/en-US/docs/Web/Accessibility/ARIA/Roles/treeitem_role

Wilco: Think there is a major potential for 1.3.1 issue

Helen: some properties may not trickle down when there is a missing parent. Lacking tree role may mean that expand/collapse is not inherited

Kathy: Would putting a tree inside a menu affect the expand/collapse?

Wilco: Assign this to Kathy for a little bit of testing.

Helen: picking up my marbles and leaving

Wilco: #1172 - can someone do a little testing?

Wilco: if you have a lang attribute and an xm lang is empty, what scenarios might cause a problem?

Wilco: need to make sure that it defaults back to HTML lang if there is xml that is missing a lang

Wilco: just need some testing. Anyone?

WIll: I will heroically take it on

"Optional" accessibility requirements

WIlco: This is kathy's

Kathy: We talked about last week how we can allow implementers to map to SCs in certain situations when the rule might not be a primary rule. Like a AAA rule written for a AA criterion

Kathy: Our rules format requires that we must list the SCs that fail in our rules.

Kathy: Atomic rules might not fail because there are separate conditions that need to be checked and it's only checking one condition in this SC

Kathy: Here is an example where I've tried to make some edits

Kathy: Header section, no changes, but under there, I have changed it to 'mapping conformance requirement' which is a mapping of SCs that may apply. adding bits about when the outcomes have passed

Kathy: Making the listing of an SC in a rule narrower in scope, to satisfy the conditions when it has passed. THis removes the AAA SCs, and moves them to a new section called 'Requirements (optional)'

Kathy: 4.4.2 is for supplemental, related, etc...

Kathy: 4.4.2 allows for further testing when the outcome is failed. But also allows AA SCs to pass, and creating a section for further testing

Wilco: Has opinions

Helen: One Q - the more that can be picked up programmatically vs, manual, should the further testing be for manual? Since machines don't get nuance

Helen: Is this more for the black and white rules?

Kathy: 'Further testing is needed' is for times when there's going to be another rule that really determines whether the SC should be considered not satisfied

Wilco: THe reason we put that in there was the thought that rules are tech specific. We can't prove we tested everything.

Wilco: There are always ways rules can fail. Pretty much never happens that rules can tell you it passed

Kathy: I see that. I originally had left 'or further testing is needed' in. but it gets more wordy.

Wilco: Britney joined early.

Wilco: let's wrap this up? Should we put this into a PR

Kathy: I can do that

WIlco: next week, Surveys are due today. Please look at the rule sheet and fill out where your name isn't

WIlco: I'm out next weekend but you know I'll join anyway

Helen: I am bad

Wilco: wrapping.

Minutes manually created (not a transcript), formatted by scribe.perl version 192 (Tue Jun 28 16:55:30 2022 UTC).