W3C

– DRAFT –
ARIA and Assistive Technologies Community Group

11 March 2021

Attendees

Present
Matt_King
Regrets
-
Chair
James Schoels
Scribe
Matt_King

Meeting minutes

Test planning - issue 385

https://github.com/w3c/aria-at/issues/385

james: we are rapidly through the patterns we decided on last time and some have changes coming.

we'd like a progress update b/c we may need to change the plan

just to keep momentum going.

Spin button toolbar, and sliders

which ones need updating in APG.

jg? we are working on sliders now

mk: we merged horizontal slider last week

James: what about other sliders

mk: the thermostat slider coming next week

jg: updates are done and ready for reivew

James: that leaves tree

mk: nav tree is stable, the other 2 trees we are still planning what to do

james: ok, we can check back on that in a couple weeks and then decide if we need to update plan

Tracking status of tests -- issue 300

https://github.com/w3c/aria-at/issues/300

James: we started this discussion last week, matt riased issue that this may not be as simple as keeping track of a binary state

mk: what is the life cycle of a test plan. When we first merge, we can mark draft

Discussion of AT release timing and versions and how that can impact the process.

mk: process is something like cg makes draft test, then agrees read for AT dev review, then test is reviewed and agreed as ready to be mark final

mk: stages could be: draft, ready for wide review, final

james: thsi raises question of what happens when AT vendor disagrees with a test during the wide review window

mk: the wide review window proposal is 120 days

james: let's say that near end of wide review window, sr developtells about several problems, what do we do?

Sina: maybe first 60 days of wide review is for raising issues, then 60 days emain for fixing issues

james: so we always have at least 60 days to fix issues

hadi: How do we have regular conversation with AT vendors to include them in the process and give us real time feedback instead of batching it all up. Are they onboard?

mk: We have been keeping them up to date, but have not had a lot of content for them to chew on.

James: we can have it in real time, but having deadlines is important to keep the project moving.

James: are we going to target during test runs for every public release of an AT

next step: work out what needs to change in working mode and james will post in comment in issue 300.

Revisiting command sequences - issue 363

https://github.com/w3c/aria-at/issues/300

James: we d3ecided we do not want open-ended sequences. we will always have exact number of key presses.

Worked out syntax in the issue.

we did not talk about commands that need to be done in a certain time frame, like typeahead in a combobox

e.g., type :"ap" to get to apple in the combobox listbox

These two letters would need to be typed quickly.

That could be like testing the example, not the assistive tech.

Sina: that could depend on where the ARIA plays a role in the experience

It doesn't seem like this is an AT function, it is an function of the example being tested

mk: maybe punt

sina: +1, we can do this later

mk: Propose that this would be a single "type string" command, e.g., "type("ap");\

Minutes manually created (not a transcript), formatted by scribe.perl version 127 (Wed Dec 30 17:39:58 2020 UTC).

Diagnostics

No scribenick or scribe found. Guessed: Matt_King

Maybe present: hadi, james, jg, mk, Sina