Rossen: it has to do with timing and update of autofocusable elements. The important discussion is happening in issue 4992 in whatwg... Has to do with how the browsing context gets updated... if you look a the diff in terms of what this change is doing - it changes when style and layout gets flushed. Previous to this change it's defined as flush after update to style - at least style has happened. The problem is that some autofocus elements could depend on styling... The argument is you shouldn't have to update layout or style to make this observable. so... with that in mind..

... I looked at the patch that was landed in blink - it's a straightforward change that moves the update style and layout...

Dan: any controversy?

Rossen: my take on this is that ... if you want to allow focusability of generated content ... e.g. focusing into list items... desirable feature... not sure how you can compute... does seem like a good change.

Peter: nothing in the style that affects whether it's focusable?

Rossen: yes.

Peter: if something is display: none it's not going to be focusable

Rossen: another reason why you need style computed...

... previously it was doing it before style and now it's doing it after style.

Dan: so shall we say it looks good to us? Anything for us to do here?

Rossen: I don't think so. it's a relatively important change for a focusability and accessibility PoV but only improving things.

Rossen: the other items in this issue addressed by Tess....

Peter: design review on User Activation V2.

Tess: i think that would be reasonable - to reference the previous design review. I'm comfortable with not double checking.

[Rossen to summarize discussion and close the issue]