Meeting minutes
Review agenda and next meeting date
<Matt_King> View the agenda at https://
<Matt_King> Next meeting date is June 13.
Status of Coming Site Updates
Matt_King: I have three things ready to go right now
Matt_King: A typo in "grid pattern"
Matt_King: jugglinmike's removal of the reference to Internet Explorer
Matt_King: And most significantly, a fix to the Date Picker Modal Dialog
Matt_King: For efficiency's sake, I'd like to get a few more ready
Matt_King: I have three in mind; they're listed in the agenda
Matt_King: Among those, the one I would really like to land is the new Landmark pattern page
Matt_King: Mark and Kurt are currently assigned as reviewers; I'd like to get at least one of those two to take a look. That would bring the total number of people who have reviewed to three
Matt_King: If we can get at least two of these three, then that would be enough to put something together for Shawn to publish
Matt_King: I believe she publishes on Tuesdays, so we would have to give it to her on Monday
jugglinmike: I will work with Carmen at Bocoup to identify someone to create a new publication branch
Matt_King: Okay, but if we don't get at least two more pull requests merged, then publication can wait another week
Pull request reviews
Remove duplicate SkipTo library
github: w3c/
Matt_King: We have at least three other scripts in "content/shared/js". They all have script tags in the head of each HTML file. I'm not clear on why we would treat skipTo.js differently than the other shared JS files
Alex_Flenniken: I was operating under the assumption that we were going to remove it from Practices. That was wrong, and so I'm wondering if this PR is even needed anymore
Matt_King: There's something downstream that was causing two skipTo links to appear on the production site
Alex_Flenniken: Why don't we close this PR, and then I'll submit a fix for that problem, separately
Matt_King: It seems that it is only duplicated on some pages
jongund: I noticed that, too, and I added some information about that to the issue
updated skipto.js script to 5.1.6
github: w3c/
Matt_King: I know there's a version 5.2, so I'm wondering if we should merge this as-is. Maybe at a later time, we could upgrade again
jongund: This pull request should work according to the needs of the WAI site
Matt_King: Okay, I'm going to go forward with that, and we can add this to the list of changes for publication
Matt_King: When you're ready jongund, we can upgrade to the parameterized version of skipTo.js. That will simplify maintenance going forward
Listbox Examples: Update scrolling of listbox item with focus into view when page is magnified
github: w3c/
Matt_King: jugglinmike is now assigned for reviewing code and tests
Matt_King: can you do that review in time for the publication next week?
jugglinmike: Yes
New Issue Action Planning
Select-Only Combobox Example - listbox closes when clicking on scrollbar
github: w3c/
Matt_King: Can anyone reproduce the reported behavior? And do they agree that it's a bug?
Andrea: I'm happy to look into this one
Andrea: I can confirm the reported behavior, and I agree that it's a bug. I'll be taking a closer look at the proposed fix
Select-Only Combobox Example Tabbing as selection
github: w3c/
Matt_King: The feedback that we got is that tabbing doesn't provide any information about new selections
Matt_King: This has always been the behavior I expect from selects. When you hear "selected" on the one you want, and you tab out of it.
Matt_King: It seems equivalent to the native behavior to me
jongund: Do we need to document that, then?
Matt_King: The reporter says that tabbing to select is not native behavior. That's right; selection follows focus, so you've already done the selection just by pressing the arrow key
Matt_King: I don't quite understand what JAWS-test is saying. Tab moves focus out; that's always an expected behavior. I wouldn't ever expect that tab would not move focus
Matt_King: The input itself is not a separately focusable element
jongund: On my system, you have to press Enter and then Tab to exit a select drop-down. Kind of like a dialog
Matt_King: I don't know if I've encountered that behavior before. I wonder if that's recent
jongund: Then again, in Firefox, when you hit tab, it moves focus and changes selection
jongund: It looks like Safari models what Chrome does, at least in the latest version of macOS
Matt_King: I'll be honest; I prefer what the APG does
CurtBellew: Same here. It's also what we do at Oracle
Matt_King: I'm trying to think of any other time where tabbing doesn't move you to the next thing. Nothing comes to mind...
CurtBellew: I can't think of anything, either
jongund: How about a dialog box with only one element inside?
Matt_King: Yeah, I suppose so. That feels like an abberation, though; I don't know if it's necessarily a good behavior to model elsewhere
Matt_King: The bottom line here is: should we make any change to what we're doing?
Matt_King: One suggested change is to issue some kind of notification. The other suggestion is to change the tab behavior.
Matt_King: I think we agree that we want to leave the tab behavior the way it is
CurtBellew: Yup
jongund: Yup
Matt_King: Okay. What about adding a notification that would occur when tab is pressed?
Matt_King: Is there anyone who would argue in favor of that?
[silence]
Matt_King: To me, it feels like a little bit of an anti-pattern for screen readers to, upon receiving a tab key press, tell users about the thing they just left
Matt_King: Is that something anyone here has experienced elsewhere?
Matt_King: You folks who implement design systems: is that something you've built before?
[no participants voiced experience]
Scope of carousel example
github: w3c/
jongund: I think we had another discussion about this. The slider seems to be more like a scroll feature. On the WebAIM discussion list, they were talking about course selection
jongund: They were calling that a slider and wondering where the focus should go
jongund: It seems like this is similar; when you have more than one thing showing at a time
Matt_King: We call those "h-scrolls" as short for "horizontal scrolls", and they are a lot like a carousel, except that they don't automatically rotate
jongund: Maybe we need a horizontal scroll example
Matt_King: When they say the entire container, maybe they mean the visible set
Matt_King: It doesn't seem like this specific example fits our carousel pattern very well
Matt_King: I don't know how common this pattern is and if it's something that we should be focusing on, but if someone wants to take it on...
jongund: I think it's pretty common. It also came up on the webaim list
Matt_King: I would love to have a soft of incubator practice for new things like this. Something to promote a wider variety of input
Matt_King: I think it's pretty clear, though, that the carousel pattern isn't going to solve this problem as it is