Silver Conformance Options Subgroup

19 May 2022


Azlan, janina, jeanne, jeanne2, kirkwood, maryjom, PeterKorn, shadi, SusanaPallero, ToddL, Wilco
jeanne2, Wilco

Meeting minutes

<janina> Date 19 May 2022

<janina> http://lists.w3.org/Archives/Public/public-silver/

<janina> Thanks for the heads up on our IRC, Shadi!

Agenda Review & Administrative Items

JS: We are continuing work on the use cases

JS: ON a new topic, there is a new spec in W3C outside of WAI that could have a big impact on Silver testing.
… APA is working with the group working on the spec
… it should give a report of data sent to the user in the last mile
… it is part of the Performance group

Continued Discussion https://raw.githack.com/w3c/silver/use-cases-apr22-js/use-cases/index.html

email from Gregg with edits

<PeterKorn> https://www.aboutamazon.com/news/devices/how-accessibility-powers-inclusion-and-innovation-at-amazon

PK: I will probably be stepping back more and more in participation in this group.

email from Gregg with edits

Situation 1 Comments

[JSp reads)

Gregg wishes to remove the three bullets under technical standards because he thinks it is a policy issue

<janina> ack \

SAZ: We already have an example where technical standard address this to some degree. It is a gray area
… a determination of conformance can be made under certain conditions
… I would like to keep at least a placeholder here. It is not clear-cut that it is totally out of the question.

JS: There is a key principle: Since we are Conformance Options, we want to have a conformance option that expands what we have in WCAG2
… we don't currently have a way of assessing an entire site. There will be bugs, it will be less than perfect.
… how many days to fix this and other timelines, that belongs in policy
… sites with 100s to millions of pages would like to compete

<Zakim> jeanne, you wanted to say priority and severity is a technical issue

<SusanaPallero> I think we should address it, at least as a recommendation as a lot of companies don't have a policy defined for this

JSp: Severity and criticality are technical in nature. It would be inappropriate for non-technical people to be making decisions about it. It could lead to fragmentation and reduce harmonizations.

Jsp: +1 that timelines are policy

SAZ: I think the timelines are policy, but the severity and criticality are technical.

<SusanaPallero> Even though severity is a technical matter, several times teams that are not accessibility trained don't know how much can an issue be. Maybe we don't establish severity but how blocking this issue is and then the companies address that and assign severity.

JK: Having developed policies and procedures for the government, I think that people are getting things mixed around. We need to stick to the technical standard and not try to develop policy.

<SusanaPallero> how much can an issue be a blocker*

JS: We won't write policy, we are just trying to give smarter guidance on regulation.

JS: If a blind researcher needs access to hand-written documents, then they should have a way to request it.

SP: What happens in the case where there are no policy? Maybe we should be giving policy recommendations?
… sometimes teams aren't sufficiently trained in what makes an issue a blocker, so they can assign a severity

JS: Not every country in the world is going to do enforce. They may not do anything for accessibility.
… there are still plenty of countries that have no laws or enforcement. We have no way to create an enforcement mechanism.
… If we don't create requirements, then it isn't a W3C spec. It is voted by the Advisory Committee. This may change somewhat in the future, but the core will remain.

<Azlan> I have to leave for another meeting now. Sorry. Great conversation here

SP: In Argentina, we have accessibility professionals who would offer advice for creating policy, and it would be helpful for them to have a guide from W3C.

JS: Even when there aren't regulations, local advocates would find it useful to have some advice around policy.

<Zakim> jeanne, you wanted to talk about work that we have done to define severity or criticality

Jeanne: We have done work on how we could define severity and give guidance to people who are trying to make those decisions.
… It's a proposal, may not go into the draft. It's an issue we've been working on.

<SusanaPallero> Sure

Shadi: On how policies might contribute, provisions for reporting bugs. Not sure what he means.
… Which bug, report to whom?

Shadi: Maybe he means reporting by users?

<Zakim> jeanne, you wanted to say harmonization

Jeanne: I'm concerned about harmonization, for international organisations if they have to provide different reporting mechanisms for different countries / legal entities.
… I don't think that belongs in policy.

Janina: Agree, I hope APA can help with that, between maturity model, reporting technology already in a W3C spec.
… There will need to be something user level, to the website and to the legislator.
… That's where we don't want to see divergence.
… Expense not put into fixing bugs, but managing compliance.

Shadi: Hypothetically, a requirement that a website should have a mechanism to report issues, and that mechanism be accessible.
… The response to these bugs, when responded to, how quickly, that's more a policy thing.
… I think the technical standard needs a hook.

<kirkwood> “reasonable response time”

Shadi: I don't agree that this is all policy.

Janina: The maturity model will provide a way of capturing those kinds of reports. Databasing, reporting on them consistently, what happens in an uptake.
… Maturity model will not be normative, we'll have to convince people it's worth implementing.

Peter: Maybe to help bridge this is another rewrite of his technical standards bullet, or some of the other content.
… Say "What is or is not acceptable in particular contexts is a policy issue, the technical standard can help define things like critical errors, levels of tolerance for caption sync, etc. to help policy define what is acceptable in various scenarios".
… Start from his starting point, but have the technical standard inform the policy decisions.
… If the technical standard has levels, those are an aid to the policy decisions.

<SusanaPallero> +1

Peter: Caption sync tolerances, you might have tolerance for automated captions vs human captions.
… Surfacing to policy makers the best knowledge of what is and isn't doable today is valuable.
… That's more guidance.

Janina: I think the biggest challenge is getting organisations to bring after real-time events the captions in line.
… Maybe policy decides if it's a live event still after publishing, don't know.
… We heard from people in the media domain it's a document they'll use.

Jeanne: I think the point is policy needs a technical specification to refer to.

<kirkwood> policy is where the lawyer comes into the room ;)

Janina: If we have best practices backed up by data, it can be used.

Shadi: I think this is a crucial issue to discuss with Gregg. The more we can swing his way, looking ahead in situation 2, I have a similar disconnect. It seems to me Gregg may be more on the policy side, rather than finding the right mix.
… I think having that discussion would be good.
… I think Peter's suggestion is important.

Wilco: I wonder if we shouldn't focus the conversation more on trying, and see how far we get, rather than try to convince everyone from the start.

<PeterKorn> +1

Shadi: Not sure I agree, we should try to capture comments now and address them early, to make for a stronger proposal.

Janina: I'd like to find out from the Friday call what most people working on silver, get some sanity check.
… In the next few weeks can we clean up situations on what is technical, and what is policy.


<SusanaPallero> +1

Shadi: I think it was situation 3. I agree with it.
… We had started to go into guidance / technical requirements on an example basis. I think that will expand the effort quite a bit

<kirkwood> +1 to alignment

Peter: Before going to Friday is to make sure we have alignment with Gregg, otherwise it might not be effective.

<Zakim> Wilco, you wanted to suggest taking this to AGWG

Wilco: I would suggest we take this to AGWG, to bring them along with the work.

Shadi: I think this is critical to WCAG 3, my experience is to bring this to the group early and often.
… Keep close connection. Even if at first there is a lot of misunderstanding and push-back. It's still the most effective approach.
… Pick one situation, then to silver, and then to AG as quickly as we can. Keep bringing it back.

Shadi: We'll need to pick one to focus on.

Janina: I'd like to make a note about WAI-CC yesterday. All groups are asked not to go the full hour meeting.
… Starting next week's call I'm going to try to end 5 minutes early

<kirkwood> not much in 3 really

Minutes manually created (not a transcript), formatted by scribe.perl version 185 (Thu Dec 2 18:51:55 2021 UTC).


Succeeded: s/+1 that timelines are policy//

Maybe present: JK, JS, JSp, Peter, PK, SAZ, SP