Meeting minutes
Review agenda and next meeting dates
Matt_King: No CG meeting on Wednesday, April 10
Matt_King: Next community group meeting: Thursday April 18
Matt_King: Next automation subgroup meeting: Monday, April 8
Candidate review decisions
Matt_King: We have a goal of six plans in the "Recommended" phase by June 30th
Matt_King: We already have two plans in "Candidate review" and two that I think are ready to move to "Candidate review"
Matt_King: For "Command button", the testing is complete, so I'm assuming that there would be no objections to advancing that to "Candidate review"
Matt_King: And the same goes for "Toggle Button"
Matt_King: Any objections?
Matt_King: Hearing none, we'll advance those, now
IsaDC: I'm on it
Matt_King: I think we need to talk with Apple about whether conveying the "alert" role is a "SHOULD" requirement or a "MAY" requirement
Matt_King: We're already positioned to move that discussion forward with Vispero and Apple
Matt_King: we have three candidates for the final test plan to target for this milestone: dialog, action menu button, and slider
Radio test plan
Matt_King: we have a target to complete testing for this by the 10th of this month
Matt_King: We have Hadi, Joe, IsaDC, and Alyssa all assigned to this in the test queue
Alyssa_Gourley: I noticed that when I went to test this that there was no output pasted in from the bot
Matt_King: that seems to be stuck with responses for 2 tests missing
jugglinmike: I believe that if you assign the run to yourself and then assign it back to the bot, it will pick up where it left off
Matt_King: I started to do this here on the call, but the page became unresponsive, and now, the page won't load at all. Now, I'm getting an "502" error
Matt_King: Can we get that automated collection job to finish and then assign it to Alyssa_Gourley
jugglinmike: Sure
Matt_King: Hadi is assigned on JAWS, and Joe_Humbert is assigned on macOS
Hadi: I don't have the exact text in front of me, but when you "tab" to a radio button, JAWS reads the instruction--how to handle the radio--followed by the essential information about the roles, name, and state...
Hadi: ...when I am considering "excessive verbosity", I do not know if lowering the verbosity will eliminate that instruction. Does anyone know about that?
Hadi: What do folks think about getting the instructions every time you enter a radio group?
Matt_King: In the past, we have said that that kind of tutorial material is a design decision by the screen reader and thus completely acceptable
Matt_King: normally, when we say "excessive verbosity", we're talking about double (or even triple!) speech. It's less about making a judgement call
Matt_King: That's how we've used it historically (I think in order to avoid editorializing too much), but I'm certainly open to other perspectives from folks here on the call today
Hadi: It might not always been so cut-and-dry, though. For instance, reading the landmark of a containing element. Do we have any agreed-upon threshold for "excess"?
Hadi: I think that some guidance on these more subjective aspects is important to reach consistency
Matt_King: I don't know if we've formally defined any such guidance, but I think that we generally ask, "is this justifiable" rather than "is this justified"?
Joe_Humbert: I am not assigned to this test
Joe_Humbert: murray_moss is
Matt_King: Oh! I'm sorry; I misspoke
Matt_King: IsaDC is assigned to two others in addition to "radio", so maybe it would be good to have Joe_Humbert take on this one instead
IsaDC: Sounds good
Joe_Humbert: Okay, I have assigned myself
murray_moss: I'm still on track for completing this next week
Matt_King: Are there any more questions on "radio button" planning?
Hadi: As I was running the "radio" test, at some stage (it may have been "tabbing" in virtual mode), I wondered: did JAWS switch from "virtual mode" to "PC mode"?
Matt_King: Yes! I'm glad you saw that. We recently changed the way we format tests so that we can express expectations that are specific for individual commands.
Matt_King: ...the behavior your observed is a candidate for that new way of authoring tests
Hadi: I also wanted to mention that JAWS was behaving differently when I was on that page, so I restarted my computer to make sure I was in a good state
Hadi: Somehow, though, that behavior--switching from virtual mode to PC mode--it was doing that even when I was arrowing down from the field.
Hadi: Could it have been related to the test environment? Have you seen it before?
Matt_King: I've seen it before. Not in our tests, but out in the wild. Mostly related to "edit" fields
Matt_King: there's nothing in our test environment that should cause that to happen when it's not supposed to
Hadi: Thank you!
Matt_King: Any other questions?
Matt_King: Hearing none, we can adjourn.
Matt_King: Thank you everybody. Happy testing. I love our new tests and how beautiful they are.
Joe_Humbert: What tests are next?
Matt_King: "Dialog", "slider", and "action menu button". I'm pretty sure "dialog" is going to be the first of those, but we shall see. They'll probably all be ready at close to the same time, so we may be able to choose