W3C

Accessibility Conformance Testing Teleconference

15 Nov 2018

Attendees

Present
Wilco, MaryJo, Shadi, Charu, Kasper, Jey, Romain, SteinErik, Anne, Kathy, Moe
Regrets

Chair
MaryJo
Scribe
Wilco

Contents


Issue 293: Mapping "Background" and "Assumptions" sections from Auto-WCAG to ACT Rules Format

<maryjom> https://github.com/w3c/wcag-act/issues/293

MJM: start with something simple
... There is a section called Background in the Auto-WCAG rules
... question, should this be part of the ACT RF
... Do all rules have a background?

AT: Yes

MJM: Should it be required, or a "may"

WF: I'd suggest may, and because of it, it shouldn't have to be in there

AT: There are other optional sections. Auto-WCAG should be in line with the ACT rules format

MJM: How is this section used?

<shadi> example rule: https://auto-wcag.github.io/auto-wcag/rules/SC4-1-1+SC4-1-2-aria-allowed-attribute.html

<anne_thyme> Example of Auto-WCAG rule that has "Background" section: https://auto-wcag.github.io/auto-wcag/rules/SC4-1-1+SC4-1-2-aria-allowed-attribute.html

WF: The idea is borrowed from techniques, it gives more information

MJM: Gives linkes to references that describe it further

AT: I raised this because I'd like to see consistency between the two. It's a weird misalignment
... there was a discussion at TPAC about putting things in the background section, while it isn't part of the spec

MJM: It could be put in as a MAY
... I worry about consistency, want things to be named the same and things like that
... You also mention difference between "Assumptions", which should probably align as well

CP: We have quite some stuff under Rule Structure. I don't know if it's just a matter of making sure we have the same under both
... could we have it under Assumptions... It is titled differently

MJM: I don't think it could go under Assumptions
... It's like references. Places with more information about
... In the example, it has references about the test. It's related information, not necessarily what is tested
... The background is not explaining assumptions, which are about the actual test.
... It could potentially go into the description, but I can see background keeping it cleaner

CP: Maybe a subsection under accessibility requirements

-1

SAZ: One question is what is called background is more like general references. I don't think they are related to the accessibility requirements.
... For example references to ARIA. I think it would be good to have a section for such references.
... The spec talks of limitations, assumptions or exceptions. In auto-wcag we've only used assumptions. Maybe we need to tweak the description of this section in the spec
... just call it Assumptions? And say that it sometimes includes limitations and exceptions. Or change them in Auto-WCAG
... Third is the question of if there should be a background section. If we have references described, do we need this section?

Also, do we want an acknowledgement section?

SAZ: Such as about who authored the rule and who reviewed the rule.

WF: I question the value of putting things in that aren't required. What is the purpose of putting them in, since they can't change if we put them in now.

SAZ: I'm not sure all of these should be optional. For example, if we had a reference section, if it has to be a "should" requirement.
... We should encourage that, not a must but a should. Similar for acknowledgement, that you should show who developed the rule, who was involved in its creation.

AT: When ACT RF says what it must consist of, does that mean it can't contain anything else?
... Is this the minimum requirements, or is this the only thing that can be in an ACT rule

MJM: I'd think it would be an "at least"

SAZ: Does anyone think it isn't "at least". I think most people would think it's at least. Does anybody think differently?

<Skotkjerra> +1 to Shadi

<cpandhi> +1

+1

<maryjom> +1 to "at least"

WF: Seems we can do some editorial work to make this clearer, but it doesn't change the requirements.

AT: I'll create a new issue for this

KE: If I wanted to relate an ACT Rule to a sufficient technique, is there a place to indicate that?

MJM: I think this additional background section would be a good place to put that

SAZ: I see the background as it's being used in Auto-WCAG as separate from how it related to techniques
... I agree that we want to be able to reference a technique, but this related to the accessibility requirements section

MJM: Background is kind of broad

SAZ: Kathy, did you mean the rule is related to a technique, or as a background?

KE: When we write our tests we do have a reference section, which is where we list a sufficient technique
... If there isn't a sufficient technique, we don't say that, but when we have one we list it.

SAZ: You list it with other references such as ARIA?

KE: We list it separately from the criteria
... We generally only list sufficient technques, but it could also include ARIA and accessible name computation

SES: The relationship is which requirement it related to.
... A requirement does not necessarily link to a criteria, but it could also be sufficient techniques that a rule is testing
... I think this should go into the accessibility requirements section.
... We also want to indicate if it enables you to pass a criteria
... If there is a way to include a technique it should be noted how it relates to the rule.
... What criteria it links to and what the relationship is.

KE: When I see accessibility requirement, that's not where I think to list sufficient techniques
... That's my understanding what that section is for.
... It's a matter of identifying that information.

SAZ: Kathy, when you list an advisory technique, do you put anything in the accessibility requirements, or do you leave it blank?

KE: We always have a success criteria listed. It is separate. We want to make it clear that the sufficient technique is not the requirement.

MJM: Current issue is about having a background section, should it be in the rules format as another section
... It seems like it makes sense to have it there

WF: I think we should not have a background section, as it is optional, and standards shouldn't have optional components to it

MJM: If people are using it consistently, it should be part of that

CP: In our rules help we do the same thing, we have the mapping to the SCs, and we have technique references where we list techniques
... We have rules that specifically tests a failure technique

AT: I think if there are things that are useful to have in rules, it would be nice to have an overview and have some common wording so it can be done consistent
... we could maybe have a section that says what optional parts could be for rules, but don't spec it out as much.
... Having common terminology is important.

MJM: It is good to have common terminology, so that as you're reading different rules you know where to find things. If you don't have consistent terminology it makes it harder to understand the rules.

WF: I'm just concerned about overengineering things, if ACT Rules CG finds a need to change things in the future they will be stuck

AT: I think the point of harmonisation is to find common language

SES: It is good to give people taxonomy. If they have something completely different than we should add that, so that we use the same words
... If we have a lot of rules, it is important to have consistent language
... we should probably identify the top X things of people would want to describe

MJM: For example acknowledgements. If it is not normative, we can give a couple of examples
... Can we come up with optional sections? Anyone oppose the idea?

-0

<Skotkjerra> +1

AT: Would this include the accessibility support section?

MJM: We could move that there

<shadi> +0.5

<maryjom> +1

<romain> +0

SAZ: I think laying it out in the spec might provide clarification. It would help consistency

<cpandhi> +1

<kathy_eng> +1

<MoeKraft> 1-

<anne_thyme> +1

RD: I don't fully understand the change but I don't see the reason against it

CP: We want to know where to put the techniques. We don't know where to put them today

SES: We can have a first try at writing something up. We'll keep it in the issue, and when we agree we can put it in a PR

MJM: Agreed, looking forward to your draft

SAZ: This is the background section, what about limitations assumptions and exceptions?
... I'd propose to just call that section Assumptions, and leave the section as it is

<cpandhi> +1

<Skotkjerra> +1

<maryjom> +1

WF: I haven't seen any example that couldn't fit under Assumptions

MJM: I think it's helpful that the rules are consistent with the spec

WF: Concerned about our timeline, we should already be in CR. Let's hurry this up than.

MJM: It is important to get things out. We don't want to get extra delay because of publication restrictions

SAZ: Agree, ideally we would have caught these things earlier, but its better than having to get out of CR

SES: Lets set a goal to tet this proposed next week?

MJM: Next week is thanksgiving, I will not be on the call.
... If someone can come up with a proposal, I can do a survey earlier in the week to get that out.

<maryjom> https://github.com/w3c/wcag-act/issues/301

SAZ: Wilco put in use cases trying to describe what rules need to do. I think we should look at those and hopefully resolve the last issues.
... maybe people can engage in that thread on Github

SES: That is also a section about how specific we should be

WF: Meeting next week

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: 2018/11/15 20:16:59 $