I came here to raise CSS reading order because I think we need to spend some time with the CSSWG folks.

I have a good handle on that one and have been providing input. We have a prototype in Chrome. Today the WG agreed to handle display:contents cases, but, some of the weirder/harder edge cases don't make a lot of sense to bend over backwards for. It comes down to this: you can have intervening display:contents ancestor between the reading container and reading items, but only DOM siblings can be reordered relative to each other. Anything else causes problems.

Are you saying we don't need to spend time with the CSSWG folks for CSS reading order?

We probably really don't. I don't think there will be anything worth using everyone's time on in a big meeting.

Maybe we can try having me a give a summary before TPAC on a call and see if anyone sees a blocker that needs wider discussion. It's a good feature for a11y and the only friction was how to deal with weird cases without breaking browser a11y impls.