– DRAFT –
Silver Conformance Options Subgroup

<janina> Date 03 March 2022

Agenda Review & Administrative Items

Janina: AGWG decided to take week off for CSUN. Not sure if the Friday call happens, but we'll not be presenting on the 15th.
… AGWG has agreed to presenting on the 29th.
… We'll have to go to Silver call at least once before.
… The main purpose is to continue to perfect the document.

Shadi: We didn't finish situation 2 last week. I want to get back to it today.
… On Friday we brought this document to Silver. There was no opposition, so I interpret that as positive.
… There were some clarifications, so I took another stab at the introduction part.
… Explain a little more, thinking about what falls on the spectrum between technical and policy.

<PeterKorn> no objection

<shadi> https://www.w3.org/WAI/GL/task-forces/silver/wiki/Substantial_Conformance/Example_Scenarios

<PeterKorn> finished

<janina> finito

<maryjom> done

<KimD> done

<DarrylLehmann> done

<GreggVan> done

<ToddL> done

Kim: I can probably not help with plain language

Gregg: I suggest we take this to the committee, and then ask how to do it in plain language.

Janina: We're only required to have plain language summaries.

Shadi: We want to be clear to a broad audience.
… I like the suggestion of taking it to the plain language group.

<SusiPallero> I can help as I helped some banks in Spain remediate their contracts in plain language, and if I can understand it in my poor english then it must be easy. I could test the translation motors too.

Shadi: Please do get me suggestions, or flag it if a sentence is too complex.

Gregg: Not sure about the word authoritative. Saying way has an authoritative role in policy gets arguments.
… Under problems, have 1 example to get heads in the right direction of what it is.

<PeterKorn> @Gregg - 2nd bulet says "technical guidance"; what kind of guidance is the third bullet guidance (vs. "authoritative")?

Gregg: Next "can" sounds like it's decided. Maybe "might" to invite conversation.
… Saying fully accessible is meeting the guidance contradicts what we've always said about guidance.
… I think we should call that fully conformant.
… Reasonable effort, I'm not sure where we're using that.

<PeterKorn> +1 to "reasonable efforts standard". I would also drop "standard" from that; "reasonable efforts concept" or "framing" maybe?

Gregg: I don't think that should be cited if we're going to include more.

Janina: We've kind of avoided getting into the "why" of this document. It might be useful to have a few sentences.
… Looking for patterns, looking for ranges, how to define what is critical and what is less so.
… Helping figure out how to produce a real-world situation conformance model.
… I expect the Tuesday call will challenge us on "why"

Peter: Maybe we should say we avoid the word accessible. It's a short hand that can mean different things.
… We could say conforms means you adhere to some number of technical standards. Fully conforms means to conform to all of them.
… Noting it's not fully accessible. It makes the document more dry, but likely to get less attacked.

<janina> q/

Gregg: +1

Peter: I think it might be helpful.
… Regarding relevance to PwD, I wonder if it's still a useful concept in a prioritisation context.
… If I'm fed gov, working through a bunch of sections. I might prioritise accessibility of social security over door handle procurement.

Gregg: Having accessible door handles may mean they can or can't get in.
… I have a concern, but I don't have a recommendation. Don't like to do that but don't know how to solve.

Peter: Maybe the answer is to highlight that concept is not ready.

Shadi: Other thoughts?

<PeterKorn> (I need to briefly step away from my computer; still listening)

Shadi: I want to prioritise comments that impact the presentation of the document.
… Looking at the list of situations. Example 1.1 would read making content fully conformant requires specialists.

<janina> 1.1 requires specialists and presumes unicode for ancient gliphs

Wilco: not liking the language approach read out

Shadi: I was surprised we changed to fully accessible

Janina: I like the shift to talking about conformance. It's what we'd be producing
… We define standards / guidance. Things do or don't conform. In 1.1 it struck me it's not just calligraphy, you need to know a unicode character.

<GreggVan> bullets - change CAN to MIGHT so we dont sound like we have decided these -- but are open to discussion about them at this point. 4) i see problems with "fully accessible". this contradicts WCAG which says nothing is fully accessible even if fully conform to WCAG then we say more accessible is more than fully accessible.. Peter just spoke and i like his suggestion for handling this. 5) "relevance to People with

<GreggVan> disabilities -- always bothers me. but i dont know how to handle. Yes - info for PWD SHOULD ABSOLUTELY be accessible -- but then ALL the other information on the site SHOULD ALSO ABSOLUTELY BE ACCESSIBLE if it is important to anyone - it is important to PWD. so it is a concern but i dont know how to resolve 6) Reasonable Effort standard --- is just one standard. there is also Undue burden and other measures. Many are stronger.

<GreggVan> I would not adopt one standard here - but leave that to policy people "

Janina: Yes, I prefer replacing accessible with conformance

Peter: I wonder if another way to frame this is "conforms to same" vs "conforms to all".
… I don't like partial conformance, that caries over. Maybe we can say you can say conforms to some, and part of what technical guidance can do is guide us to what might be more achievable.

Wilco: don't feel strongly opposed to the change; slight preference for "accessible"
… not about "fully conforms" vs. "conforms"

<Zakim> GreggVan, you wanted to say Summarize "Suggest 1) removing "authoritative" from bullet 3 since it is about policy and W3C/WAI is not the authority for policy. I see the idea but suggest removing. 2) Put an example after the first sentence in Problem Description - to get people's heads going in the right direction (since some developers find everything about WCAG a challenge and we arent talking about that level of challenge. 3) in the three and to

<janina> We're also teasing out that some conformance criteria are more critical than others; though that can vary by situation

Gregg: I agree with Peter. I think we're better off talking about conforming / not conforming. Declaring that something that meets WCAG 2 is accessible drives people crazy.

<PeterKorn> "I'm partially pregnant"?

<janina> partially dead?

Gregg: I think saying something is partially conformant makes no sense. Getting rid of partial conformance doesn't mean anything.

<DarrylLehmann> +1

Shadi: There is also levels of conformance.

<PeterKorn> [Shades of the Princess Bridge - "he's only mostly dead"]

Shadi: I'm hearing the group lean towards trying conformance.

<ToddL> I'm in agreement with the conversation so far. Apologies, I'm multi-tasking today.

<maryjom> +1 to Peter's suggestion

Peter: I know we can roll things back, but it might be interesting to have a document that does the conformance transplant, so we can see them side by side

Shadi: I won't use partially conformant

Janina: In the 2.x series it was unspecific. I think we're heading towards some things being critical and others we'd forgo depending on the situation.

Shadi: I'll take a stab, then we can look at them side by side

Shadi: Returning to Situation #2
… have only one example: archived content
… initially, some discussion in the group about things on the same page, but not as important [prioritzation discussion?]
… question: do we have other examples here?

Janina: Peter's idea from last time - keeping data generating now that you don't know now is important, but it may become important later.
… and then in situation #1, extend text description of what is in the scroll is an extreme edge case.

PeterKorn: Images of section of stars, and later you find there's a comet in that sector

PeterKorn: Now, by looking back you can find where you missed it at first; but if not archived, you'd not be able to look back

PeterKorn: point is we don't know whether the data is important until something later indicates examining past data makes it more important

PeterKorn: it might also be situation 3, too rapid for a11y, but might be important later

maryjom: thinking in software scenario - processes or things only used by certain roles of people: e.g. installers.
… scenarios that occur only once; is that a "rarely used" scenario?

[Peter aside: "Maintenance and monitoring spaces" from 508 refresh]

Shadi: example 3.1, about generating content.
… collect parts of 1.1 & 2.1 & 3.1 - taking pictures that may be worth looking at later
… and put them together into situation 2. And change situation 3 into user-generated content specifically