Skip to main content

Glad <dialog> focus is finally being implemented/defined in a reasonable way after a torturous years long, largely pointless, discussion.

“1. Make the dialog focusing steps look at sequentially focusable elements instead of any focusable element.

2. Make the dialog element itself get focus if it has the autofocus="" attribute set.

3. Make the dialog element itself get focus as a fallback instead of focus being reset to the body element.”

#HTML #WebStandards #a11y

github.com/whatwg/html/commit/…


Implement dialog initial focus proposal · whatwg/html@a9f103c

This implements the changes proposed here: https://github.com/whatwg/html/wiki/dialog--initial-focus,-a-proposal#dialog-draft-text Specifically: 1. Make the dialog focusing steps look at seq...
GitHub
#a11y #html #webstandards
This entry was edited (1 year ago)
in reply to Steve Faulkner

Jamie Teh
Jamie Teh
• •
Frustrating and torturous? Yes. Pointless? I don't think so. There are conflicting use cases here which are hard to manage, and even this solution isn't perfect, though it's a good compromise.
in reply to Jamie Teh

Steve Faulkner
Steve Faulkner
• •
@jcsteh I said “largely pointless” there was a lot of unnecessary discussion and time wasted. It shouldn’t have taken 6+ years was my point.
@Jamie Teh
in reply to Steve Faulkner

Jamie Teh
Jamie Teh
• •
Yeah, that's fair. I worry that a lot of standards stuff is like this. It's one reason I have (until recently) avoided being officially involved in standards groups. Hard to know how to resolve it. On one hand, it's far too slow. On the other hand, the whole point is to gain consensus and to ensure that major viewpoints are considered and doing that with many parties means a lot of back and forth.
in reply to Jamie Teh

Steve Faulkner
Steve Faulkner
• •
@jcsteh I was involved in this particular discussion for years before I tired of it. It was it’s anti-consensus nature that created blocks not consensus.
@Jamie Teh
in reply to Steve Faulkner

Jamie Teh
Jamie Teh
• •
Yeah. Thinking back a bit, I suppose the only way I managed to push for progress here was arguing that fixing this should be a condition of Firefox shipping it, which isn't a card I should have needed to play and is suggestive of a broader problem.
⇧