W3C

Accessibility Conformance Testing Teleconference

20 May 2021

Attendees

Present
Aron, KathyEng, Wilco, Daniel, Trevor
Regrets
-
Chair
-
Scribe
Daniel

Meeting minutes

Joint meetings with Silver

Wilco: This is a friendly reminder that we have a joint meeting with the Silver TF and AGWG
… Focus will be to explore how we make the methods more like ACT Rules
… We sent two rules ahead of time for people to take a close look, WCAG-TF people should be familiar with those at the time
… What do people think about last week's meeting?

Trevor: We got a bit off on tangents, probably should have been more on ideas exploration

Wilco: I think that's fair, a lot of opinions out there

Wilco: That tends to happen sometimes in meetings, difficult to steer the discussion

Kathy: I appreciated all of the conversation, I was thinking that there would be more familiarity with testing, but it was explained pretty fairly that they need some help on that area

Wilco: It seems to me like that groups has done a lot of work but now they need to focus more on how it should be tested

Kathy: Including definitions, and trying to write this proposed resolution, that was getting very detailed, I was surprised that it could not be written out in a way that everybody agreed

Wilco: I think there is general agreement that we need definitions, and ACT needs to be involved on how and what should be defined

Aron: We need definitions, but the methods are not going to be normative. I think we need definitions at an outcome and at a method level, there is a need for linking the definitions in the outcomes with the definitions in the methods
… Not an easy task anyway

Wilco: I think I am going to approach the meeting like we need to provide help especially in testing experience

Aron: I quite like the format (that is focusing on the outcomes, critical errors), but links need to be clarified. Currently WCAG is kind of binary, so that transition needs to be worked out

Wilco: There is great stuff on what is being created, now we need to polish it a bit

<Daniel> Very sensitive topic, especially when it comes to requirements. Good opportunity to explore the ties between ACT and actual WCAG requirements

Kathy: WOuld ACT be involved in holistic testing?

Wilco: I don't know yet. We represent people who will need to do that testing.

Kathy: We are not fully supportive of holistic testing within the current testing outcomes defined in Silver

Wilco: I have had some comments in that regard too. I think at this moment it is too early to have conversations around that topic.

Aron: How is holistic understood in this context?

Wilco: We don't know yet. We may have a bronze level which will be like things are done today, there might be screen reader testing after that, but it is not yet decided
… A maturity model is not going to be part of WCAG 3 as it is defined now, but it could easily change also

ACT rules checks sheet

<Wilco> https://docs.google.com/spreadsheets/d/1OSkPFocXk4K3zYLnwS78WLsWO4PvE5yRcsauyefuIUI/edit#gid=0&fvid=507619053

Wilco: Orientation rule.

<Wilco> https://github.com/act-rules/act-rules.github.io/issues/1608

Wilco: We decided to wait for Susan's response on that, discussed last week

<Wilco> https://github.com/act-rules/act-rules.github.io/issues/1605

Aron: Link in context is descriptive, I raised an issue, example on the how-to page saying that headings provide context for links
… You could use a screen reader dedicated shortcuts to get back to the heading which provides the context. If when going back the heading does not provide any context, then it would fail

Wilco: My understanding is that that would be a failure
… This used to be a sufficient technique quite a while ago but they changed that

Aron: If it was a failure, the description does not point it out

Wilco: These examples look like failures to me. I was not involved in AGWG when it changed
… I think there is a problem with this technique

Aron: When you arrive at the link and it is ambiguous, but when you get back to the heading and it does provide the context, it is also a programmatic context
… It is not veery intuitive, it does fit into the programmatically determined context definition

Kathy: Within that definition, HTML says context is given by paragraph, list, table cell, or header that is associated with the table cell. Heading is not included there, maybe this is why the technique is advisory

<kathyeng> In HTML, information that is programmatically determinable from a link in English includes text that is in the same paragraph, list, or table cell as the link or in a table header cell that is associated with the table cell that contains the link.

Wilco: If headings are missing, would landmarks be missing too? I think what we are doing in this rule is consistent with what is documented. There is an argument for this, but we should make it to AGWG, if they decide to include heading in the list then we should update the rule

<Daniel> If so, that should be carefully qualified as the previous heading or something, otherwise there could be situations where you would need to go back several heading to find the context

<Wilco> https://www.w3.org/TR/WCAG21/#interpreting-normative-requirements

Wilco: Examples are not normative

Aron: It looks like heading might be programmatically determined because examples are not normative

Wilco: But there is informative context that says so, even though there is no normative text. I would suggest you open an issue with AGWG

Migrating proposed rules to the WAI website

Wilco: Working on proposal to migrate rules. It will take a while before the whole automated system sets up, but we could have an intermediate solution so that somebody manually copies the 50-ish rules
… Do we want to do this?

<Daniel> Probably, we may want to split that.

Wilco: It is a bit technically involved. I think I'd just do that. I will talk to Shadi and Hidde about it

Trevor: I am good with it

Wilco: That is not going to have any implementation data yet
… We need to wait until the process is set up

Kathy: I am good with it as well, this may help collect more implementation data

Follow-ups

Trevor: Will not attend next Week.

Aron: Me neither.

Minutes manually created (not a transcript), formatted by scribe.perl version 131 (Sat Apr 24 15:23:43 2021 UTC).