Meeting minutes
Agenda Review & Announcements
<matatk> Irfan: I sent out an email about CSS Speech - there was recent discussion of it; there is still not widespread support.
matatk: This is more of an announcmemt for us.
matatk: We have been learning abou thte Github bot (GB). When we go through the horizaontal review section, it gets a little chatty when it thinks we are talking about our issues when we are actually talking of issues of others.
matatk: I will shut GB up temporarily when we do horizontal review for this reason.
New Charters Review
<matatk> gb, off
Roy: Nothing new here.
Explicit Review Requests
DeviceOrientation Event Specification
<Roy> - issue: w3c/
<Roy> - spec: https://
<Roy> - self-review: w3c/
Roy: Yes. There is one.
Roy: They have created an a11y request and have done a self review as well, both through comment review and a11y/request.
Roy: It provides new physical orientation DOM events.
Roy: They have marked some a11y issues in their self review.
matatk: The Devices and Sensors WG, weve worked witht them before. They maintain the Compute Pressure API spec and others such as the Device Posture spec. They are receptive to a11y considerations.
matatk: I'm not hugely familiar with this spec. It's porbably worth thinking about it for a while to see if there may be a11y considerations.
janina: We wanted the ability to lock a particular orientation. For a screen reader user, in particular, you may want to keep things in a certain mode even though you turn your device over.
<Fazio> cognitive considerations as well
matatk: It does talk about rotation. I've kind of skipped over that. WE have an expecation fo what we will find there.
janina: That is an old comment form us.
matatk: There is device orientation and then there is device motion: accelerometer stuff.
matatk: There's permissions for gyroscope and magnetometer and so on.
Fredrik: I'll look at it, happily.
niklasegger: I just wanted to add: I had a look at the documetn before the meeting. There is a section for Use Cases and Requirements. They give axamples.
niklasegger: There are issues such as shaking to clear a web form where I thought of persons with uncontrolled movement. I'll also happily look at it, perhaps in tandem?
Fredrik: Gladly.
niklasegger: Done!
PaulG: Fixed devices would also be a problem here.
PaulG: What they are uilding up to is a bunch of permissions. Tehre will be opting ins and by default these things should be off. I hope that's where they are going with this.
PaulG: How do we fall on fingerprintable permissions in this? Is that somethign we are worried about because the persons we are talking about can be fringerprinted for disabilities and so on?
matatk: Excellent point and question.
matatk: First off, this group is really hot on things like fingerprinting and stuff like that. If we discuss it with them, I'm sure the discussion will be productive.
<matatk> Design Principles: https://
<PaulG> https://
<PaulG> this?
matatk: There is guidance on issues around privacy and fingerprinting in the Design Principle doc which the TAG maintains.janSecurity and privacy are horizontal review things that every spec should get from W3C itself, so iot isn't just up to us or on us.
matatk: Short discussion on what fingerprinting is.
matatk: As to Paul#s question: I don't think that htere is any specific fingerprinting guidance around a11y. Maybe that is something we want to add as a note in the Design Principles.
<Lionel_Wolberger> s/Pincipe/Principle
<matatk> APA tracking issue for this spec: w3c/
matatk: When you, Niklas and Fredrik, are looking at this, we need to go thorugh the review process. For reporting your results there is a trackign issue that has been created. You're welcome to put thoughts on teh mailing list as you go as well as putting them in teh Github issue. Whichever works.
matatk: My plan is to comment on this issue and link back to the minutes so we don't miss them.
A11y Review Comment Tracker
Roy: Nothing new on the comment tracker.
<matatk> https://
matatk: This is he repo where we receive tracking issues from someone asking for our input. It can be anywhere on Github: if someone, whoever, asks for our input on any a11y issue, it will be created as an issue here.
New on TR - http://www.w3.org/TR/tr-status-drafts.html
CSS Viewport Module Level 1
<Roy> - tracker: https://
<Roy> - spec: https://
Roy: There is one.
matatk: Github issues come form a specific publication whereas Wiki tracking gives us more context.
PaulG: I think I'm fine with this one.
matatk: I think this is just viewport bounds and zoom.
PaulG: When zoomed, do you keep the same coordinates or do you change them and report that? Something along those lines.
matatk: This is an FPWD so we haven't een officially requested to review this.
matatk: We are brought on usually when a documetn is to transition between stages.
<PaulG> https://
matatk: I don't think this needs following as of now. It will come back to us. Plus, we're familiar with it thorugh Paul's input.
PaulG: Often a virtual keyboard is covering things and I need to tap away to see what it is covered.
PaulG: It would be interesitng to know if anyone working in AT could tell us how their calculations worked when the viewport is changed, for instance through zoom.
PaulG: Can any AT developer weigh in: hwo do you calculate your virtual cursor placement or do you hand that over to the UA? For instance.
matatk: That might be an ARIA/WG thing.
matatk: The virtual cursor isn't rendered as such. It is more like a place in the DOM.
PaulG: Is it in the DOM, in teh bitmap, on top in another layer?
matatk: We should defiinitely ask about this, and I can think of no one better to ask than the ARIA WG.
matatk: If there is an appropriate issue to link to and ask them ato chime in, maybe that's the thing to do.
Fazio: We could make a new issue...?
<PaulG> w3c/
PaulG: I had ot things in my notes concerningzoom.
PaulG: This one is the one I wanted to talk about.
nehaj: I have a little doubt here. Are we talking about the virtual keybards on mobile devices or do we also include the ones on bank websites where they don't want to know keystrokes really?
PaulG: I would assme that if someones implementing HTML virtual keybaords, you know you are using CSS zoom The problem I see comes with the author using zoom on a web page or an application thhat then the AT doesn#t know is being zoomed. When it gets viewport information without zoom appliedd, then does that still work or does the irtual ursor appear in the wrong place? Or is it delegeated to teh borwser? I dont know.
<PaulG> I think this is it w3c/
matatk: Let#s move on. We need to figure out where to report this. If I could think of how to succinclty create that as an action, I would do that now.
PaulG: Put a stamp on the link I posted, perhaps?
CSS Update (Paul) - https://github.com/w3c/css-a11y/issues
<PaulG> w3c/
PaulG: Scrollbars!
PaulG: What about making scrollbars completely stylable and resizable? Thats not standard compliant. There should be considerations for this. Or at least warnings.
matatk: Brilliant summary. I'e put a11y tracker on it as well.
Actions Checkin (Specs) - https://github.com/w3c/apa/labels/action
Dangling Spec Review Cleanup
Task Force & Deliverables Updates
Other Business
matatk: There is a lot left for next week.
Fazio: I'll be at CSUN.