Re: [CSSWG] Minutes Telecon 2018-11-07 [geometry] [css-sizing] [css-shadow-parts] [css-images] [css-grid] [css-break] [selectors]

Hi Dirk, Chris

Sorry to have missed the call.

On 07-Nov-18 21:14, Dael Jackson wrote:
> Geometry API
> ------------
>    Rossen: Speaking of CR are any Geometry API authors on the call?
>            Chris and Dirk are currently editing
>    Rossen: Is there a rule that suggests we can't do CR update without
>            editors on?
>    florian: I don't think so. If they're in disagreement it's bad form
>    Rossen: They're both asking for updated CR
>    <TabAtkins> Heh, yeah, don't publish above their objections, but
>                fine to approve without them.
>    Rossen: Objections to republish CR of Geometry API spec?
>    <TabAtkins> No objection.
>    RESOLVED: Republish CR of Geometry API spec

Good to see that Geometry is to be republished: the previous /TR was 25 
Nov 2014

Looking at the procedure for updating a CR with substantive changes

there are some questions that have to be answered before filing an issue 
on the W3C transitions repo
(go to the Readme, follow the link "Updated CR transition 

- *substantive changes since the 2014 CR*
  ( looks good here, although 
the date is for the September WD not the November CR so should be split 
into two)
looks like there was only one change between September and November 2014

- *Wide Review of substantive changes
*Has there been any security or privacy review?
Did the TAG look at it?
Are implementers happy with the changes made since 2014?

- *Any changes in normative references?*

- *Issues status*
18 open, 27 closed.
are the ones closed since 2014 ok (edits made, commentors happy)?
are the open ones left to be solved after this CR refresh?

(This is the main question where I need some input before sending the CR 
transition request)

- *Any changes in implementation information?*

There are only 7 tests (but each has like 30 subtests so that may be 
okay). Who is implementing? What changed since 2014?
I just ran them in Firefox Nightly, Chrome Canary, Safari TP and Edge 
Results are much improved since 2014 and indeed the CR exit criteria are 
almost met
So I think we have good things to say for this question

Do we know that Webkit is working on the two failures?
Firefox has a lot of "passes most but not all subtests" results, are 
they working actively on this?

Chris Lilley
Technical Director @ W3C
W3C Strategy Team, Core Web Design
W3C Architecture & Technology Team, Core Web & Media

Received on Friday, 9 November 2018 01:29:07 UTC