Meeting minutes
<PaulG> oresent+
Agenda Review & Announcements
matatk: David and Helix has registered and lauched the only training programme for digital a11y certified by the US government. Kudos!
<Fazio> I will be at Zero Project
Gottfried: IAAP will host a discussion on the Zero Project conference in FEburary in Vienna regarding overlay. With Lionel and others.
Gottfried: Gottfied will be on another panel about IAAP chapters.
Fazio: ICCHP will be at the JKU in Linz. David will be on a panel on digital a11y and mental health, chairing. Submit stuff for the ICCHP everyone!
<matatk> https://
matatk: I'll put the link in to the ICCHP website.
matatk: Please post to the list when you get links everyone.
<Fazio> ICCHP link for submissions on accessibility on mental health https://
New Charters Review https://github.com/w3c/strategy/issues?q=is%3Aissue+is%3Aopen+label%3A%22Horizontal+review+requested%22
Roy: No new charters!
matatk: Any review comment tracker, explicit comment requests or review on TR?
Roy: There is one we should finish on the comment tracker.
A11y Review Comment Tracker https://w3c.github.io/horizontal-issue-tracker/?repo=w3c/a11y-review
Roy: There are two review requests.
ARIA attribute reflection uses DOMString? but they are not enumerated attributes
<Roy> - issue: whatwg/
<gb> Issue 10037 ARIA attribute reflection uses DOMString? but they are not enumerated attributes (by domenic) [a11y-tracker] [topic: reflect]
Roy: One is related to the ARIA-WG.
matatk: I don't suppose anyone on the call has seen or made an opinion about this one...?
janina: Is ARIA aware of it? What are they responding?
<Fazio> +1
matatk: Let's flag this for ARIA.
janina: And CC us on it.
PaulG: You have ID refs, and you have booleans, and you have empty case... I'm just trying to follow along...
matatk: There is some recognition in this issue thread, but it is far too technical to dive into this right now.
matatk: How does ARIA enumerated types map to DOM string and so on...? They are struggling with some facet of DOM string which requires what those in the DOM String ? to always be defined values.
matatk: They are talking about empty attributes and such things.
<Fazio> +1 to both
matatk: I'll take an action to open an issue on ARIA and point them to this and ask them how they want us to do this in the future.
HTMLSelectElement showPicker()
<Roy> - issue: w3ctag/
<gb> Issue 900 HTMLSelectElement showPicker() (by lukewarlow) [Venue: WHATWG] [a11y-tracker]
Roy: You already replied as usual, Matthew.
matatk: This is a TAG design review, so earlier when we would perhaps normally get involved.
matatk: WE have not yet heard back from the group involved and we wouldn't expect to have done so just yet, so I think we should postpone deep discussion of this until next week.
matatk: It relates to an API to getting an input control to show its picker, specifically a SELECT object.
matatk: As the TAG we are concerned about consistency and default behaviour. From an a11y perspective, I'm concerned about the use cases.
matatk: We#ve asked for the use cases and we hope to receive information about that.
Explicit Review Requests https://github.com/w3c/a11y-request/issues
BBS Cryptosuite v2023 2023-12-15 2024-01-15
matatk: Welcome Neha!
<Roy> - issue: w3c/
<Roy> - spec: https://
<Roy> - self-review: w3c/
<gb> Issue 69 BBS Cryptosuite v2023 2023-12-15 2024-01-15 (by Wind4Greg) [LC] [REVIEW REQUESTED] [pending]
<gb> Issue 102 Accessibility Self-Review for VC DI BBS (by Wind4Greg) [a11y-tracker]
Fredrik: (That's a ne/hi, I guess.)
matatk: I did hear back about t the date format. They did answer.
<matatk> Reply on date format: w3c/
<PaulG> all they need is "2018-01-10T06:14:00Z" (JSON timestamp). Not sure why they're making it so complicated
Lionel_Wolberger: The usual low level crypto stuff, basically.
matatk: It is a very heavy technical and low level document, it seems.
matatk: Unless someone can volunteer to see if there are any areas of concern, one of us would have a look and ask Lionel if there is something we don't understand that he thinks is risky.
janina: That#s lots a loops!
new on TR http://www.w3.org/TR/tr-status-drafts.html
Roy: Nope. othing new except the year.
CSS Update (Paul) https://github.com/w3c/css-a11y/issues
<PaulG> CSS HDR https://
<gb> Issue 9306 [css-color-hdr] Add mechanism to query HDR headroom (by ccameron-chromium) [cssom-view-1] [mediaqueries-5] [css-color-hdr]
PaulG: I forgot one thing from a previous week.
PaulG: HDR, the higher colour space, and this is something that is very welcome on the web. High Def colour is something we have the technology for now with screens and all, but the web wasn't ready for it.
PaulG: It will afect contrast algorithsm and advice for that.
<Fazio> I'm interested to see a side by side comparison
PaulG: There may be more to it than that: migraines, astimatism and so on. Lots that we are not aware of yet.
<Fazio> I can imagine the a11y implications positively
PaulG: The examples and experiments right now seem to all be done in canvas.
Fredrik: (Almost wrote Kansas there...)
PaulG: There are two issues posted and then the HDR colour spec, its a11y section is really small.
matatk: One of the challenges with recent video specs is that they have grouped a lot of features together. Wehn you say 4k, you might just be talking to a biger number of pixels, or how a 4k Blu-Ray disc is regarded: resolution, 10-bit colour, and so on.
matatk: HDR is 10-bit coluor and the expanded reach of brightness.
matatk: With the colour, I'm not concerned, but thebrightness could be concerning.
matatk: Maybe the contrast algorithsms go haywire with 10-bit colours.
PaulG: The HDR portion of it, might not work in the current calculations. There's ltts of math and colour theoy in here.
PaulG: The experiments and proof of concepts are all being done in canvas.
PaulG: I believe, at least.
matatk: Presumably this is going on in canvas because there isn't browser support for anything more than 8-bit colour. It sounds like there would be a lot of work going on in the browser to recalculate the higher definition colour. 10-bit colour is not going to be a thing on webpages for a logn time because you can't specify it with CSS.
matatk: The canvas rendering video is just a video stream bing rendered, blurted out like video bist.
matatk: I think people may be wanting to recreate user interfaces in canvas.
PaulG: If theyh do, they will leave a lot of folks behind that yuse AT.
matatk: The real anser right now is "don't do it".
matatk: If they are going to do it, there has to e a lot of work done behind ehte scenes for it to work.
lionSounds lke people want to get 10/bit colour out. Perhaps we should think of an AUR?
janina: We were really in the same spa e with canvas before HTML5 became a thing.
janina: We didn't have anything, we didn't have the shadow DOM.
janina: I'm not surprised by any of this, really. Its nice to have it as an experimental area in canvas, but to put everything into canvas doesn't help when it comes to a11y or regulation.
matatk: This is a really interesting topic.
matatk: I think of four questions, though Janina might have added a fifth.
matatk: Do we think these are valid questions_ We may want to spend some focused time on an upcoming call deiscussing this.
<Fazio> I think we should dive deep soon
matatk: 1. I'd like to understand more of the pressure that Paul and otehrs perceive that a significant umber of people want to do this.
matatk: 2. Is it possible to make canvas fallbacks accessible? My experience is "sort of" but its ugly.
matatk: 3. What spec work is needded to make canvss fallbacks accessile?
matatk: 4. What AUR or guidance is needed to make canvas fallbacks accessile?
<Fazio> I'm curious if there are we shouldn't abandon a11y policing of canvas just because its garbage
matatk: 5. What is this used for other than web games and so on?
matatk: I'm reulctant t to give RQTF more work to do and at the sime time they seem to do their work dilligently under the circumstances.
janina: Too many blind people here to work with colour.
PaulG: Canvas is just ones and zeros and can have a consistent profile pattern fo rthe device it's on. Kiosks, TVs, and phones might be places where these implementations could come into play.
PaulG: There needs to be more work done to ensure this is usable.
janina: You don't need to make the controls to an F35 screen reader compatible. If you're lookign for speed, you have to decide whose needs are important and which ones you can forego.
<Fazio> is it even needed?
matatk: Is there going to be time for people to rewrite their app as a 10-bit app before CSS gets 10-bit colour?
matatk: A11y Object Model would play a part in it, AUR and general best practice adice, too.
matatk: Are these things accessible at the moment?
Fazio: It seems this convo as evolved from HDR to canvas.
Fazio: Do high def colours have a use case or just a novelty?
matatk: We did digress to canvas, yes.
matatk: Fallbacks to a UI is presented on a canvas, thats the underlying issue here.
Fredrik: (Let's for the record state the Accessible Object Muddleas eing a eupphemism for canvas.)
matatk: I wouldn't notice 4k at all, my resolution is so low. Most people have 4k in their eyes and I've got SD.
<Fazio> spatial computing may revolutionize this
<Fazio> Apple Vision Pro and its future competitors
PaulG: Right now we can produce 21:1 luminosity contrast maximum. You will probably be able to get far higher numbers with 10-bit colours. But ifyou should and how and what the implications are - that's interesting to discuss.
matatk: We should ask AGWG about this, if they're aware of it, if they know its' coming up and so on.
matatk: We didn't have any major things in the outstanding actions and no other major news.
matatk: We talked on and on consciously.
matatk: I wil contact AGWG. I'll think on it and post to the list.
matatk: I will see if we can get someone on the call who has expertise in these areas.
matatk: I may also ask the TAG about it next week.
janina: That may be more productive.
matatk: We still need to be on this.