Meeting minutes
Matt_King: Next week, we'll be meeting as usual. That's August 1
Status of Site Updates
Matt_King: I have nine pull requests that have all been merged to the "main" branch and therefore ready to deploy to production
Matt_King: There's an additional pull request from Jon that I'd like to include, but there would still be value in deploying what we have next Tuesday
howard-e: I can prepare a pull request for Shawn by next Monday. I'll include Jon's patch if it's merged in the mean time
Matt_King: You can use the list in this meeting's agenda in the pull request description for Shawn
Matt_King: I'll send her an e-mail just to let her know that we're going to have a pull request for her on Monday (last time, it sounded like she would have liked more notice)
Pull request reviews
Infrastructure: Update localhost references for regression tests
github: w3c/
Matt_King: According to howard-e's research, this will be fixed by Node.js version 20
Matt_King: ...but if we merge this, we can fix it in the current release of Node.js
howard-e: That's right!
howard-e: This is a low-risk patch and easy to revert. I think it should be fine to merge now that you've reviewed, Matt_King.
Matt_King: If something went wrong, then the workflows that are running the regression tests would fail, right?
howard-e: Yes
Matt_King: Okay; I'll merge it later today
Listbox Examples: Update scrolling of listbox item with focus into view when page is magnified
github: w3c/
Matt_King: This was authored by Jon. He said he's running late and may not make it to this meeting at all.
Matt_King: The regression workflows are failing. I've tried to re-run them, but they continue to fail.
Matt_King: And they're failing in places that are not related to this pull request.
howard-e: I can take a look at what's going wrong. It looks like numbers 1, 2, and 4 are giving us trouble
<Jem> https://
Matt_King: It's not a truly necessary link, and we can probably remove it. We'll definitely have to remove it if the site doesn't exist anymore!
Jem: The site is responding for me right now
Matt_King: Okay, so the link checker should probably pass, now. Maybe that was just a temporary anomaly with that website.
PR 2723: Select-only Combobox Example: Fix scroll event listener bug by ariellalgilmore
github: w3c/
arigilmore: I was working on this with Andrea, and when we tried the suggested solution, it didn't work for mouse, and our solution didn't work for keyboard
arigilmore: We were having trouble getting the "blur" effect
arigilmore: The original issue was: when someone would click on the scrollbar in the combobox, it would immediately close the combobox.
arigilmore: By fixing that, we were having trouble getting keyboard navigation to work AND mouse navigation to work when navigating in the combobox.
arigilmore: Now, you can click on the scroll bar, and it will scroll without closing. But after that, if you click outside the combobox, it doesn't close.
arigilmore: Though it does close when you tab away. It's just a problem for the case where you've clicked on the scrollbar
arigilmore: This doesn't appear to be a browser bug because the interaction works as expected on the other examples (even though those examples look quite different)
arigilmore: I don't currently understand why the other comboboxes work while this one particular combobox does not
Matt_King: Sarah wrote the original code here; they might be able to help
arigilmore: I'll reach out to them
New Issue Action Planning
Incorporate/embed Feed example into APG Pattern Page · Issue #2747 · w3c/aria-practices
github: w3c/
Matt_King: This is about the feed example which is anomalous because we have the "feed" page and a separate page called "feed display"
Matt_King: We originally thought that putting the feed in the middle of the example page would have been a problem, and not very realistic besides
Matt_King: We thought we'd be able to make a more realistic example if we put it on its own page
Matt_King: With the new site design, I'm not sure if that's still true
Matt_King: We may now be able to embed the feed directly into the example page and get rid of the "feed display" page
Matt_King: It seems like a fairly straightforward exercise to do this
arigilmore: I can give it a try
Jem: The page says that the feed does not yet have ARIA-Practices task force consensus. Is that still true?
Matt_King: We might want to change the wording there
Matt_King: The parts of this that are kind of unusual are that because this does not have a native equivalent, we don't have patterns to go off of for keyboard commands (in particular for nested feeds)
Matt_King: For instance, to exit a feed, we use "Control + End", but we could do a lot of different things
Matt_King: This example is also distinct in that it generates new HTML content dynamically, and in that it simulates timing
Jem: I can help you if there are any problems with the template, arigilmore, because I need to better understand how templates work
arigilmore: Thanks! I'll let you know
Matt_King: I've assigned the issue to arigilmore
SECURITY - NPM Dependency Confusion
github: w3c/
Matt_King: They're asking us to publish our package manifest someplace
Matt_King: howard-e you might be the best person to address this
Matt_King: It looks like they want to pin all packages in the package.json file to specific versions
jugglinmike: what the reporting is describing sounds like the purpose of a package-lock.json file to me
jugglinmike: This issue appears to be the only public activity of this particular GitHub user account
color picker best practice · Issue #2742 · w3c/aria-practices
github: w3c/
Matt_King: There have been plenty of people who have requested a two-dimensional color picker over the years. This appears to be one of those
Matt_King: The response from JAWS-test appears to address the possibilities pretty well
<Jem> https://
Matt_King: I'm unable to access inputs on the provided demo page
[the group discusses the demo's visual behavior]
Matt_King: Ah, if I turn off JAWS's virtual cursor, I get a different experience
MarkMcCarthy: the inputs have ARIA labels, but they're on DIV elements without a role. That's probably why JAWS isn't announcing them
Matt_King: Maybe the values are changing on my system, and it's just that JAWS isn't reporting the changes
MarkMcCarthy: I'm not sure what role you would give to these, though
Matt_King: It seems like you could make a slider that changes two different values, but not three different values. ARIA doesn't support that
MarkMcCarthy: JAWS-test has covered everything I can think of
Matt_King: ARIA might already have an open issue for a two-dimension slider, but if not, it seems reasonable for ARIA to support it.
Matt_King: I think it would mean a new role
Matt_King: Or maybe a new value for aria-orientation
Matt_King: Then again, you can achieve this with two sliders
jugglinmike: But with two sliders, you can't change both dimensions simultaneously--you have to tab back and forth between them
MarkMcCarthy: Yeah, supporting that "free movement" would require something new from ARIA
Matt_King: So this could re-use the "slider" role, but maybe a new value for aria-orientation like "both"
Matt_King: That's an ARIA problem, though. Here in APG, there's not a great solution right now, other than offering two sliders. They just need to make sure their inputs have the "slider" role