W3C

Accessibility Conformance Testing Teleconference

27 Feb 2020

Attendees

Present
Wilco, MaryJo, Kasper, KathyEng, Shadi
Regrets

Chair
Wilco, MaryJo
Scribe
Kasper

Contents


Result from CFC to ask AG WG to review 5 rules

maryjom: The CFC passed, so a survey has been created. The review will conclude this coming Tuesday.

<Wilco> https://github.com/act-rules/act-rules.github.io/issues/1185

wilco: We had only one comment on the CFC which noted that a note could be improved.

Rule review - Link has accessible name: https://www.w3.org/2002/09/wbs/93339/ACTLinkHasName/results

wilco: This is still open for another 2 weeks, with 4 responses so far.
... We have 1 "no" on an assumption, which I think is a first.

kasper: I think having to copy too much stuff from the HTML AAM could become a problem.

wilco: I think this one is fine though and fairly common.

maryjom: Seems reasonable.

wilco: I do agree that we have to keep this manageable and not add too many notes.

<Wilco> https://www.w3.org/WAI/WCAG21/Techniques/html/H33.html

wilco: Per the accessible name algorithm, the title attribute can be used as a description if not used for the name.
... I therefore think the rule is right and does allow for this.
... Do we need to change this?

kathy_eng: My understanding of the algorithm is the same as yours, Wilco.

wilco: There's some issues around presentational roles conflict resolution and I know that JAWS might revisit how it does it.

https://github.com/act-rules/act-rules.github.io/wiki/Tests-on-Presentational-role-Conflict-resolution

wilco: Is this something we need to address before publication? I prefer this one to go ahead. When we have a fix for all the rules affected by this, we can address it all at once.

kasper: As long as there's a resolution on the horizon, I'm happy.

wilco: I think the examples can be improved as they're a little messy.
... That also made me realise that I need to ask the CG to clean up the rules before we push them to the TF.
... Should we restrict this rule to only apply to links that actually perform navigation? This will be harder to automate, but might remove some false positives.

joecronin_: I don't have a strong opinion on this.

maryjom: So you'd want to leave out things that don't behave like links?

wilco: Yes, which would make the mapping slightly more accurate.

kathy_eng: Should 2.4.4 be mapped under this one?

wilco: I think that's a fair question. Is 2.4.4 satisfied even if is there is no link name?

maryjom: 2.4.4 seems to assumes that there is a name.

*discussion on requirements of names and whether they are required just to exist or actually be descriptive*

wilco: This survey is still open so let's use that for any further feedback.

Rule tracking table - Revisit list and assign work: https://www.w3.org/WAI/GL/task-forces/conformance-testing/wiki/Rule_Publication_Tracking

wilco: We need to update this list.

maryjom: Any survey that concluded with comments will have to have an issue opened.

wilco: Kasper will take the "link has accessible name" rule.

Silver test and conformance: discussion on their direction

wilco: So, one of the changes proposed is to take away the success criteria layer and going directly from guidelines to what is called methods.

maryjom: That is potentially a way that we can have rules directly referenced.

wilco: If there's no success criteria, what do we base rules on? We currently assume that there's some testable statement that the rule is written for.

shadi: I think there's a requirement for silver to have clear, testable requirements.
... There's a discussion on how this will pan out.
... There's also a related discussion on the conformance model.
... The benefit of ACT is that we can break down the decision space into smaller pieces.

kathy_eng: Will there be something normative under the methods?

shadi: I think there would be more broken down normative requirements, whatever they would be called.
... Underneath these requirements, the ACT rules could then sit.

wilco: We've found that creating atomic tests is a nice way of grouping things and helps define things more explicitly.

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version 1.154 (CVS log)
$Date: 2020/02/27 15:13:46 $