scribe+

ACT Standup

Kathy: Trevor asked everyone to review subjective applicability, plus I had a PR that was approved
... will send out a call for review after this meeting.

helen: Been busy with other things.
... I'm catching up with Karen tomorrow to help her start on a manual rule

Todd: Reviewing pull requests.

<ChrisLoiselle> https://github.com/w3c/wcag-act/pull/539

<trevor> https://github.com/w3c/wcag-act/pull/539

Chris: I put some comments this morning in on subjective applicability
... Still need to sync up with Kathy and Suji

Trevor: Reviewing comments on the PR, plus spent time going over PRs

Suji: PR reviews

<thbrunet> Sorry - can't attend this morning. Work call running longer than expected.

Cancel August 17th

Kathy: Several of us won't be available on August 17th, so we're cancelling the meeting two weeks from now

Subjective applicability

Trevor: This PR is on subjective applicability; how we want to allow subjectivity in the rulers format
... The first is you should use objectivity when you can.
... It seems people like having both
... The next question was on whether the applicability section should be marked as objective or not
... Kathy and Chris suggested a sentence somewhere.
... Wilco asked whether this is relevant to the reader.

Wilco: I think as metadata separate from the rule, maybe also for the expectation

Trevor: I would be fine with any of these. I just want to make sure implementors know what they're looking at

Kathy: As long as its somewhere on the page it would be fine.
... You have on line 417 the last part you have role and common design pattern. On 419 you have subjective attribute. I think it should prefer the first two.
... I was then thinking of the heading example. In some of the examples, the heading isn't styled as a heading.
... I didn't know if there are other things that would fit under the styled as type.

Trevor: I'm trying to reduce how much subjectivity can be used.
... When possible its nice that both parts are objective.
... I would argue though that the design pattern / style attribute are not objective.

Kathy: I was trying to see if there was a way to make the last part be as objective as possible

Trevor: I think it's possible. I'd need to give it some more thought. Is the real difference between styled as and functioning. Style was appearance, and function was more meant as the way you interact with it?
... I need to think more about if there is a clear distinction between these two.

<ChrisLoiselle> hypothetical - What if something is styled as a list but functions as a nav, with a heading being present then three bulleted items underneath the heading?

Trevor: What would the requirements be? That it's wrapped in a nav region or something?

Chris: For ACT primarily its web based review. If you're going off the markup, it helps determine what tests to do. If it's a p tag but it looks like a heading, you go off of that.

Trevor: In a lot of ways that's why we're doing this. A lot of rules are that we have an element tag, and we then say what is expected of that.
... Now we're doing the opposite. When we look at it, it appears to be some thing, and we need to make sure it has the appropriate tag.

<kathy> wilco: WCAG 3 is exploring more granular ways to determine subjective things

<kathy> ... rubik is a breakdown of things. Instead of saying "sufficient", list out 5 things. If it has 3 of 5, it's sufficient.

<kathy> ... if 3 of 5 criteria are met, then it could be ex: considered styled as a heading

<kathy> trevor: any written examples?

<kathy> wilco: no. maybe plain language has some.

Trevor: I like that idea.
... We could definitely put those kinds of conditions in for meeting the subjectivity.
... That's something I can rewrite to put in.

Kathy: So that list of criteria would be required if the applicability is subjective

Trevor: So if we have HTML element is styled as a heading. So inside the "styled as a heading" we would have 4 or 5 indicators that something is a heading. In the rule you'd have something like you need to meet 3 or 4 of them.
... The number feels arbitrary to me, but it makes it more concrete to me.
... I'm not sure we'll find a 100% solution, but it gets us to be more consistent
... I'm going to make some changes to that effect. We can come back and look at it again in a few weeks.

Wilco: Might be a useful exercise for TPAC to go over existing subjective definitions

<ChrisLoiselle> great work!

Trevor: I'll work on the rewrite, and we'll get back to it later

Secondary requirements and accessibility support

<kathy> wilco: 2 rules, autoplay audio and empty headings, where this came up

<kathy> ... autoplay not as much of a problem. many browser block autoplay but all don't do it the same

<kathy> ... sometimes it blocks depending on how it was opened so not consistent

<kathy> ... can autoplay SC be listed as secondary bc sometimes the rule fails depending on the browser

<kathy> ... results vary

<kathy> trevor: causes implementation inconsistencies

<kathy> ... some examples are obvious failures that should fail the SC, but don't know how to not mess up implementers

Kathy: Secondary requirements already have an example for when its not always failing the criterion
... The links rule includes 1.1.1 because it only sometimes applies. I don't think we have a situation where we can list a criterion as secondary because it fails inconsistently
... I wonder if that opens up for more SCs to be listed as secondary where it shouldn't ber.

Helen: One example when it doesn't meet 4.5:1, and a secondary would be AAA. You let AAA in that situation.
... AAA can be something the tool doesn't want to cover, but its there in case they do.

Kathy: We have that in secondary requirements as an example. The rule is less strict than the SC

Helen: I wouldn't have it the other way around. That's when it becomes more situational.
... Technically, most rules fail multiple secondary requirements.
... Secondary requirements is one that's a reason for not running the rule.

Kathy: I'm not sure this is a matter of an SC being stricter or less strict than the rule
... With the iframes it depended on the browser. Firefox puts focus on iframes, other browsers don't

Helen: Is that focusing on browser / tool bugs?

<kathy> wilco: do we want to require of implementers that they fail even if not a failure in all browsers

<kathy> ... another ex is meta viewport that only fails in mobile devices. if only testing in desktop, then it wouldn't fail

<kathy> ... we've tried to avoid test cases where one browser passes and another fails

<kathy> ... but now we have examples that are all inconsistent

<kathy> helen: do you fix screen reader issues that already meet wcag

<kathy> trevor: would be ok if tested in only 1 browser and pass some of the failed examples, and someone else tests with 5 and fails all failed ex

Kathy: I'm not sure we have consensus that it would make sense to make it secondary

<kathy> wilco: have run into this as manual tester passing failed ex bc they don't autoplay

Kathy: We we do trusted tester, there's a field to indicate which browser you tested in
... the results you get from testing in that browser. If we list it as secondary, what do we do with implementations?

Wilco: It would let tools fail such things as a best practice instead of a WCAG failure

This is scribe.perl Revision VERSION of 2020-12-31
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Default Present: kathy, Wilco, ToddL, ChrisLoiselle, Suji, trevor, Helen
Present: kathy, Wilco, ToddL, ChrisLoiselle, Suji, trevor, Helen
No ScribeNick specified.  Guessing ScribeNick: Wilco
Inferring Scribes: Wilco

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth


WARNING: No date found!  Assuming today.  (Hint: Specify
the W3C IRC log URL, and the date will be determined from that.)
Or specify the date like this:
<dbooth> Date: 12 Sep 2002

People with action items: 

WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)