Meeting minutes
Setup and Review Agenda
https://
Matt_King: Next meeting: December 10
Matt_King: Any requests for changes to the agenda?
Matt_King: Hearing none, we'll stick to the agenda as planned
Publication planning
Matt_King: Here's the milestone for the next publication: https://
Matt_King: We have four things planned. Two are done and ready to ship, and two are in the works
Matt_King: In the works we have the high-contrast work and the changes related to the slider
Matt_King: The pull request for the slider is looking good, so I expect to merge that soon
Matt_King: That just leaves the high contrast practice, which we'll talk about in a moment
Matt_King: That seems like everything, unless someone here knows about something that I'm not aware of
PR 2991 - Practice Page for Supporting High Contrast
github: w3c/
Matt_King: I expected to be further along on editorial work for this
Matt_King: There is still some editing that I think needs to be done, but I want to get reviewers assigned so that people can start looking and providing comments
Matt_King: There's also the tests to consider. We have six tests which are failing, and we need to resolve that within the next few days (if we want to publish next week)
Matt_King: Slipping on this would also change the review schedule
Matt_King: It's really more like three tests (since it's running for both push and pull request). There are linting failures for CSS and HTML. Also the link checker is failing. Why is that...?
Matt_King: howard-e do these failures prevent the preview from including the latest changes?
howard-e: No, the shouldn't
howard-e: I'm also going through and leaving a review on some changes
howard-e: I just took a look at the failure from the link checker. That might be spurious, so I'll try re-running. I wonder if the way the link is written is confusing the link checker
howard-e: The others seems to be related to formatting
Matt_King: Formatting issues should be easy to fix
jongund: Got it. I'll fix them
Matt_King: I would like to get at least two (but ideally three) people to read this text in detail and identify anything that ought to be changed prior to publication
Matt_King: Because this is a practice page and not an example, your review is essentially editorial in nature. We want to be sure that the text is clear and that it serves the goals of the APG effectively. The main thing is we want this to be useful to our audience
CurtBellew: I'd love to review this. After briefly skimming it, it looks fantastic
Matt_King: Great! I'm adding you as a reviewer
<siri> can you give link to bug?
Adam_Page: I can take a look
Issue 3154 - Developing a Keyboard Interface for charts and interactive graphics
github: w3c/
Matt_King: This issue was originally reported to the mailing list
Matt_King: The first question, "what kind of patterns exist that are related to visualizations?"
Matt_King: I think the answer to that is, essentially, "there are none."
Matt_King: There's also a question about which things should be focusable
Matt_King: The only thing that I've seen out there that's reusable and that I have experience with is Highcharts
Matt_King: I won't comment on my personal opinions regarding that library
Matt_King: I don't know if we want to dive into this space with the APG
Matt_King: I don't know if we can make effective guidance here given our current capabilities. It would not be straightforward--it would take a lot of effort.
Matt_King: Can we answer these questions directly, or do we just feel it's out of scope for APG, or does anyone in the Task Force feel inspired to take this on?
jongund: I think we would need a group of experts (including, for example, someone from Highcharts) to gather perspectives...
jongund: Does anyone know of other experts in the field of accessible visualizations
CurtBellew: We have a ton of experts like that at Oracle
CurtBellew: Their accessibility techniques have varying levels of success
CurtBellew: Moving from one node to another (for example) involves a lot of keyboard knowledge
siri: Is their work publicly available?
CurtBellew: Sure. If you search for "Jet" in Oracles public library, you'll find a ton of visualization components that demonstrate some of these strategies
Matt_King: The reporter talks about pan, zoom, rotate, adding data points, removing data points, etc. Do your patterns address those sorts of things?
CurtBellew: I don't know about some of the distinctions there (e.g. pan versus zoom), but I think the patterns might address some
CurtBellew: Similarly, I don't really know about "drilling down." I can't think of any examples where we do something like that. Though to be fair, there are a ton of components and only one of me.
Matt_King: Do you use roles in the components? I mean, beyond the "application" role at the highest level?
CurtBellew: Yes, we do. It all depends on the application--how it's built and what's included
CurtBellew: Typically, anywhere you want to include some kind of a label, we'll have an ARIA role
CurtBellew: I'm not familiar with Highcharts, so I don't know what to compare it to
Matt_King: In Highcharts, to that are labels and descriptions. I don't always agree with their design decisions, though I generally recognize why they made them
Matt_King: I imagine they consider landmark regions the best-supported way of annotation. I just don't like how many regions that creates
jongund: I would be willing to organize a kind of task force meeting in early 2025. I know a contributor to Highcharts, and it sounds like CurtBellew might be interested in joining, too
Matt_King: We touched on this at TPAC this year. James N was trying to corral interested parties. He might appreciate your assistance in that effort, jongund
Matt_King: It's not clear what we're trying to deliver, yet. Is it something for ARIA? Something for APG? Something entirely separate?
Matt_King: Here at APG, it might end up looking like a practice page similar to what we're working on for "high contrast."
Matt_King: Right now, I think it's about getting enough people together with the interest and finding enough time to figure out some kind of deliverable.
Matt_King: I don't know if there's consensus around whether ARIA itself needs new features to address this
Matt_King: For the purpose of this specific issue, it seems like we don't have answers for the content. The APG just doesn't address these questions right now, and I think that this is an area where we would like to be able to provide some guidance, but we don't have the guidance to provide right now.
Matt_King: I don't think there's anything more we can say at the present moment
Matt_King: It is useful to provide a link to the Oracle components, perhaps CurtBellew can provide that in the issue
<jongund> https://
<CurtBellew> https://
Matt_King: And maybe also a link to Highcharts
Matt_King: Between the two of those, that's potentially some help for the reporter
arigilmore: Carbon charts also does some work in this direction. I'll share a link there, as well
<arigilmore> https://
CurtBellew: Please keep me in the loop, jongund!
jongund: Sure. My role will just be bringing experts together
jongund: We'll want to collect what people have done at this point and understand the commonalities and differences. That might inform a future extension to the APG