Accessible Platform Architectures Working Group Teleconference

05 January 2022


becky, Fazio, Fredrik, JF, Joshue108, Matthew_Atkinson, mike_beganyi, PaulG

Meeting minutes

Matthew_Atkinson: Welcoming the members back and thanking Becky profusely.

Gottfried: Is on the hunt for a student who is doing work that can be of interest to us concernign audio description.

janina: This could very well be interessting. COGA would be interested as well.

janina: The more the AD for COGA's target group corresponds to VI, the easier.

Agenda Review & Announcements

<Gottfried> Link to thesis on multi-layered AD: https://nbn-resolving.org/urn:nbn:de:bsz:900-opus4-66690

Task Force Updates

janina: RQTF: As the group will recall there are a lot of AURs unde rdevelopment. All of them are moving forward at different paces, none of them stalled.

janina: SAUR will be the first to be finalized, I believe. Perhaps as early as Feb.

janina: We may want to get political about it. It will pertain to anotehr document as well. The politics would be to ask TimeText whether they would like a joint publication with us. And regardless how they answer to this, do they believe that this is significant enough information that it should rise to the level of a W3C Statement.

janina: This may be the case. We do have amainstream argument.

janina: We have evidence for this in a paper, a curb cut situation if you like.

janina: Necessary for some, useful to all.

janina: The biggest news however is on captcha. We had general agreement today. No one's opposing the conclusion. But wow, we have published the inaccessibility of CAPTCHA twice, and no one has even mentioned that CA stands for Completely Automated.

<Fazio> That's a really interesting point

janina: Ahead of expanding the abbreviation in the docuemnt, we talk about interactiity five times. It doesn't seem to have struck anyone. What is lurking behind this is the idea that we may be asking, insisting that the industry shifting from telling computers and humans apart to user agents, using some known tools that inf act are automated or almost automa

janina: The first that is automated is Google ReCaptcha 3. Part of what pushed us to working on Inaccessibility of Captcha was work of CloudFlare which pertains to using a trusted device. Then there is PrivacyPass' approach which has a numberr of tokens which can be renewed and so forth.

janina: Once we have taken care of the technological question, we might wahnt to have some advice form security and privacy expertise in W3C... We may be able to knock off Captcha.

<Joshue108> +1 to Lionel

Lionel_Wolberger: I love the passion this ellicits. I have read the note itself. I want to say that the focus on "completely automated", you can't pin too much onj that. It is a contrived acronym to begin with. IF we really want to make an engineering argument around automated or not, this may not be the way. It's not a smoking gun in that sense. I wouldn't h

janina: We'll have to hit the right balance.

JF: Not ot drag us too far into the weeds, it's "competely automated" to the owner, not to the consumer. We want to make the completely automated situation include the user as well. The token solution moves in this direction.

JF: In terms of the CA piece, I don't think it was just a merketing term, but i treally was an attempt to aautomate verification for the owner, not the end user.

<Lionel_Wolberger> lionel says to JF: well put

Lionel_Wolberger: Personalization: we are beating the weeds to get implementations. We have one on the way and many people to approach.

Lionel_Wolberger: We are inteacting with COGA.

Lionel_Wolberger: We're stating to look at registries, an interesting new direction.

Lionel_Wolberger: W3C is operating some registries, but Janina knows some more about this.

janina: Not only can we elevate notes to W3C statements, but the W3C offers registries as well. At the core of the Content Module which the Personalization group wants to send to CR is a symbol set *like Bliss or something). The symbols which are currently used are not always interoperable. There is a system maintained by Bliss whcih we have the rights to.

janina: It's a way to interoperate across a differnert AAC symbol sets. There is a mapping between numbers and what they stand for. Those numbers canthen be associated to whatever symbol catalog you are using as a user.

Lionel_Wolberger: W3C's proven diverse processes leading to global adoption which may lead to a normative registry is very exciting.

janina: One of the things blocking us... ~The rendering tool Matthew and Lisa have put together can only render Bliss. Other systems keep changing somewhat irregularly.

Lionel_Wolberger: We saw similar tendencies in Schema. None of these were normative.

Matthew_Atkinson: It's really Lisa's too., I was just the guinea pig.

JF: Actually, I jsut wanted to say, that Schema is niormative. The problem we had is that it is really har dto update and that we don't have control over that, and that is prolematic for us as a standareds organization.

JF: The idea of tregistries was birnging things under the W3C roof so that more control could be possible.

<Lionel_Wolberger> lionel says to John: well put, thanks for the correction

Matthew_Atkinson: Anyone from C'oga here?

Fazio: No updates as of yet.

Fazio: Returnign next week with updates.

PaulG: Pronunciation: We have received good feedback. We will craft responses to this and go from there. We iwll probably update our publishing timeline soon, in a week or two.

janina: That's ARIA we're negotiatng with..

janina: Unclear how long the negotiations will take.

Joshue108: The functionality subgroup will meet wtice a week from now. We met yesteraday.

Joshue108: Functioanl User Needs subgroup.

Above passage from Josh was for FAST.

<Fazio> sorry

janina: Lisa is committed to be on our calls more now.

janina: COGA are expecting some joint conversations with Personalization. Personalization wants to get to CR as well. The conversation we want to have evedntually...

janina: We will have a joint chat or go join their call or vice versa or both. But I think at the momment we want to make sure we move towrad CR with the content module.

FAST Update

Joshue108: Nothing more to add than what was already said. If anyone wants to join the call, that would be great. It's at 9 o'clock EST on Tues and Thurs. The Thurs slot may change.

New Charters Review https://github.com/w3c/strategy/issues?q=is%3Aissue+is%3Aopen+label%3A%22Horizontal+review+requested%22

MichaelC: The only new charter is not new anymore - it'#s our old friend Second Screen.

janina: We have one clarification left to send to Francois. John, please send it to both the APA and Francois.

JF: I will get ti done befor ethe end of the week.

janina: How muych playtime do we have on this?

MichaelC: They can't take it to management before next Tuesday. EOW won't kill them.

JF: I'm mindful of the urgency.

janina: The hardest thing might be the right thing to point to in the HTML and the MAUR.

A11y Review Comment Tracker https://w3c.github.io/horizontal-issue-tracker/?repo=w3c/a11y-review

<MichaelC> https://github.com/w3c/imsc-hrm/issues/23

MichaelC: A new issue about IMSC hypothetical render model. We do want to track it.

<MichaelC> https://github.com/w3c/csswg-drafts/issues/5594

MichaelC: Media Queries perfers to reduce motions leads readers to believe all animations should be stopped.

MichaelC: Do we want to lookk at this?

janina: I think I'm okay with it.

MichaelC: ...in which I'll do tat.

zkaim, next item

new on TR http://www.w3.org/TR/tr-status-drafts.html

<MichaelC> https://www.w3.org/WAI/APA/wiki/CSS_Containment_Module_Level_3

MichaelC: This is a FPWD.

MichaelC: I took an ction to review the Level 2 one. We shoudl probably review Level 3.

MichaelC: It enables heavy optimizations by the user agents.

Matthew_Atkinson: It's alike media queries but you can apply them to a part of the page. You can do sophisticated things.

MichaelC: It can be dangerous but doesn't have to be?

Matthew_Atkinson: We should have a cursory look to see if this introduces anything problematic.

Matthew_Atkinson: We don't have Amy anymore...

janina: My only concern is if you could restrict navigation. Beyond that I don't necessarily have a concern at this stage.

Matthew_Atkinson: That would be covered by WCAG. Off the top of my head, I can't think of a new type of problem you could create.

MichaelC: Shall we let it sit unti lwe find a next CSS wiz?

janina: We might also revisit our request for a liaision with the CSSWG.

janina: In short, we need an Amy replacement.

<MichaelC> https://www.w3.org/WAI/APA/wiki/CSS_Fonts_Module_Level_5

<MichaelC> https://www.w3.org/WAI/APA/wiki/CSS_Fonts_Module_Level_4

MichaelC: We filed a comment on Level 4 here three years a go which is stil lopen. I own#'t ask for anything more now.

<MichaelC> https://www.w3.org/WAI/APA/wiki/EditContext_API

MichaelC: I'll remind them.

MichaelC: Next one, another FPWD, Edit Context API.

MichaelC: We'll wanbt to review this.

janina: It allows the authors to interfere with the text input process, to put i t cattily.

ACTION: Fredrik: to review CSS Edit Context

ACTION: fischer to review CSS Edit Context

<trackbot> Created ACTION-2313 - Review css edit context [on Fredrik Fischer - due 2022-01-12].

JF: Referring to the abstract, it seems like somethint we shoudl take a look at.

<MichaelC> https://www.w3.org/WAI/APA/wiki/IMSC_Hypothetical_Render_Model

MichaelC: Next is IMSC Hypothetical Render Model. Lionel was supposed to review that in December.

janina: I believe he did.

<JF> the draft states "...allows authors to more directly participate in the text input process." so all aspects of keyboard accessibility comes into play

Lionel_Wolberger: INdeed he did.

<MichaelC> https://www.w3.org/WAI/APA/wiki/UI_Events

MichaelC: Next is UIEvents. We looked at it three years ago and want to look at ti again, they have republished three months ago.

MichaelC: Should we just drop it?

Matthew_Atkinson: "It is not expected that htis meet with full backwards compatibility". That's interesting.

JF: My only concern here is that, looking at the overview here, it is talking about event systems and event listeners. Is it tracking back only to the DOM or to the Accessibility Tree as well?

JF: I don't know if we want to do a really deep dive, but a llittle clarification may be useful.

Matthew_Atkinson: Just asking the question is pretyt straightforward. I fyou're willing to ask them that question, that would certainly be suitable due dilligence.

JF: I don't think we should just turn from this without putt9ing in some work here.

MichaelC: It's a document we've been tracking for a while.

JF: I do note that the former editor was Doug Shepard. And current editors include others who are queite a11y aware.

JF: I want to read through this carefully.

ACTION: Fredrik: action: jf to check on UI Events -- does it include AAPI events due two weeks

<trackbot> Created ACTION-2314 - Action: jf to check on ui events -- does it include aapi events due two weeks [on Fredrik Fischer - due 2022-01-12].

<MichaelC> https://www.w3.org/WAI/APA/wiki/Web_Platform_Design_Principles

MichaelC: One last spec: Web Platform Design Principles. It's a document.

MichaelC: We said let's come back in November in September. We should look at it if we want to. It#'s from the Technical Architecture Group.

janina: Is there some kind of nexus between them and FAST?

MichaelC: There is a very light nexus.

Matthew_Atkinson: It's a fairly substantial piece of document.

<Fazio> OMG

MichaelC: We need to see if this breaks any of our proposed FAST principles.

MichaelC: I think it would be good for mahy poeple to review it as well.

Actions Checkin (Specs) https://www.w3.org/WAI/APA/track/actions/open

ACTION: Fredrik: to review Web Platform Design Principles due one month

<trackbot> Created ACTION-2315 - to review web platform design principles due one month [on Fredrik Fischer - due 2022-01-12].

ACTION: janina to review Web Platform Design Principles due one month

<trackbot> Created ACTION-2316 - Review web platform design principles due one month [on Janina Sajka - due 2022-01-12].

Dangling Spec Review Cleanup: https://www.w3.org/WAI/APA/wiki/Category:Spec_Review_Assigned

<MichaelC> https://www.w3.org/WAI/APA/wiki/WebXR_Layers_API_Level_1

<Joshue108> https://lists.w3.org/Archives/Public/public-apa/2021Dec/0015.html

Joshue108: What I did with Layers was I did make a reply to this.

Joshue108: I think we need to start a discussion with the group about this. If there is re-rendering, the accessibility content will continue to be privileged within any re-rendering that my happen.

Joshue108: I have drafted this a s a response to the group to see waht they may say about this.

Joshue108: Teh URI is the second response.

JsohIt pretains to maintaining fidelity of any content.

Joshue108: I have no feedback on it, nothign good or bad. Any comments appreciated.

MichaelC: Instead of flagging considerations, I would ask questions. I don't want to raise aflag as longas the response to the question doesn't.

Matthew_Atkinson: It's asking a question, starting a discussion.

Joshue108: I'll file that in their GitHub repo.

<MichaelC> https://www.w3.org/WAI/APA/wiki/Web_of_Things_(WoT)_Profile

<Matthew_Atkinson> https://lists.w3.org/Archives/Public/public-apa/2022Jan/0005.html

<Joshue108> Review and discussion of Web Of Things Profile FWPD - v3

Joshue108: I updated my response to this. I could do with a little bit of discussion on this.

Joshue108: We have talked about middleware type applications in Web of Thigsn environments.

Joshue108: I have always had the sense that middleware thigns in WOT is a bit of wild west in terms of a11y.

Joshue108: There could b ean impact where profiling mechanisms relating to middlewqre may need some clarification.

janina: I'm not sure we're using "middleware" in their way.

janina: We were conveying that there was some kind of layer controlling devices and if that layer wasn't accessible, that would then be na a11y issue. NOt only do you have to pick a good IOT device, you have to be told what environment you will be managing it in.

janina: If you buy it from one of the giants, you will probably use their middleware. Butsmaller thigns are another matter.

Matthew_Atkinson: I think Josh's draft post is good to tart a discussion, and I would be happy for him to post it.

Joshue108: Janina has touched on a couple of thigns but that is just half of the things I wanted to touch on.

JF: Make sure this is a specific agenda item next week.

janina: Point made and taken.

Joshue108: This is something we need to tease out.

Matthew_Atkinson: It's good ot be back.

Summary of action items

  1. Fredrik: to review CSS Edit Context
  2. fischer to review CSS Edit Context
  3. Fredrik: action: jf to check on UI Events -- does it include AAPI events due two weeks
  4. Fredrik: to review Web Platform Design Principles due one month
  5. janina to review Web Platform Design Principles due one month
Minutes manually created (not a transcript), formatted by scribe.perl version 185 (Thu Dec 2 18:51:55 2021 UTC).


Succeeded: s/quweries/queries

Succeeded: s/don't have an Amy/don't have Amy/

Succeeded: s/Dsign/Design

Succeeded: s/ease/tease

Maybe present: Gottfried, janina, Lionel_Wolberger, MichaelC