On Friday, 24 April 2026 at 12:01:27 UTC-4 [email protected] wrote:

It is unfortunate that this detailed review didn't happen at the time of 
the landing of the specs, or the shipping of this feature in Chrome, but 
it's good that it eventually did happen.


Yeah, +1 to this. It's a complicated feature building on a complicated 
feature, so there being rough edges in the initial spec aren't surprising, 
so the 'failure' here is entirely in review by others.

I noticed these issues when I tried to document the differences between 
popover=auto and popover=hint, and discovered that I couldn't make it make 
sense. Behaviours would flip unexpectedly, and I couldn't justify why those 
behaviours existed. I was surprised that it felt like I was the first 
person to review this feature in this detail (i.e., just below the surface).

This level of review should have happened within Chrome, but it also should 
have happened by other orgs.

My guess is that, given the spec is complicated, it was difficult to figure 
out the resulting behaviours from the spec, so it was missed in the HTML 
PR. When I came to it, I had the advantage of two implementations to play 
with. But that could have been done much sooner.

That aside, I'm really happy how this has turned out. I know it's a pain in 
the arse to revisit this stuff after it feels 'done', so I'm really 
grateful to Mason for working on it.

Jake.

-- 
You received this message because you are subscribed to the Google Groups 
"blink-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion visit 
https://groups.google.com/a/chromium.org/d/msgid/blink-dev/fed50fba-77dc-4100-b8ce-ea3ce357a19bn%40chromium.org.

Reply via email to