W3C

– DRAFT –
Cascading Style Sheets (CSS) Working Group Teleconference

12 September 2025

Attendees

Present
ChrisL, dbaron, emeyer, SebastianZ
Regrets
-
Chair
-
Scribe
florian

Meeting minutes

<astearns> zoom details: https://lists.w3.org/Archives/Member/w3c-css-wg/2025JulSep/0229.html

<astearns> please don’t use the zoom link in the calendar invite, use https://lists.w3.org/Archives/Member/w3c-css-wg/2025JulSep/0229.html

<astearns> github-bot, take up w3c/csswg-drafts#12715

[css-2025] Publish Snapshot as Group Note

<github-bot> OK, I'll post this discussion to https://github.com/w3c/csswg-drafts/issues/12715.

astearns: proposed resolution is to publish a new snapshot as soon as the edits are in

RESOLUTION: republish snapshot with the edits decided on today

<astearns> github-bot, take up w3c/csswg-drafts#12699

[css-2025][css-contain-2] Add CSS Containment 2 to Rough Interop

<github-bot> OK, I'll post this discussion to https://github.com/w3c/csswg-drafts/issues/12699.

SebastianZ: we're still in working draft, but there's relatively good interop, with 88% coverage

SebastianZ: not in any snapshot

SebastianZ: still has lots of open issues

<astearns> florian: I agree. rough interop is the right bucket

<astearns> florian: draft needs work and republishing

RESOLUTION: css-contain-2 goes into rough interop

<astearns> github-bot, take up w3c/csswg-drafts#12698

[css-2025][geometry-1] Add Geometry 1 to Rough Interop

<github-bot> OK, I'll post this discussion to https://github.com/w3c/csswg-drafts/issues/12698.

SebastianZ: geometry-1 has very good interop, with 98%

SebastianZ: not in any snapshot, and has a bunch of open issues, and needs republishing. Last TR is from 2018

SebastianZ: The question is whether to add to rough interop, or if we should do another publication first

<Zakim> ChrisL, you wanted to wonder about publishing

ChrisL: not clear to me whether we have active editors

ChrisL: I can put the changes list up to date, but not much more

astearns: would like a publication before we put this in the snapshot

astearns: we have a lot of tests, but whether they provide good coverage is unknown

ChrisL: sounds like we should defer

SebastianZ: I too could volunteer to update the changes list

RESOLUTION: update changes list, find active editor, republish

<astearns> github-bot, take up w3c/csswg-drafts#12711

[css-2025][cssom-view-1] Add CSSOM View 1 to Rough Interop

<github-bot> OK, I'll post this discussion to https://github.com/w3c/csswg-drafts/issues/12711.

SebastianZ: cssom-view has reasonable interop, but has many many github issues

SebastianZ: also needs to be republished

SebastianZ: I have update the changes section

SebastianZ: and I have included the inline wpt

SebastianZ: issue 8559 is about republishing

SebastianZ: I think we should add to rough interop

ChrisL: I think we should resolve to republish

<astearns> florian: impression this was maintained as a living-standard-ish thing

<astearns> florian: seems to be in a coherent state

<astearns> github-bot, take up w3c/csswg-drafts#8559

[cssom-view-1] Update WD from 2016 with current spec.

<github-bot> OK, I'll post this discussion to https://github.com/w3c/csswg-drafts/issues/8559.

<ChrisL> +1

RESOLUTION: publish updated WD of cssom-view-1

RESOLUTION: add cssom-view-1 to rough interop in the snapshot

<astearns> github-bot, take up w3c/csswg-drafts#12754

[css-2025][mediaqueries-5] Add features from Media Queries 5 to Safe to Release Exceptions

<github-bot> OK, I'll post this discussion to https://github.com/w3c/csswg-drafts/issues/12754.

SebastianZ: I triaged which features have good coverage and interop on the basis of having edited the wpt inline into the spec

SebastianZ: I'd like to go through the list maybe

<astearns> florian: the ones here seem fine

<astearns> florian: one may show we have insufficient tests

<astearns> florian: scripting has 100% pass, but it does not test much at all

<astearns> florian: this might be true for the other features

florian: nevertheless, I support this set

ChrisL: I think it's fine, but I don't have strong confidence

astearns: me too

astearns: I think these are quite widely used anyway

RESOLUTION: add the subset identified by SebastianZ to the safe-to-release set

<dbaron> FWIW Chrome has a real implementation of the scripting feature: https://source.chromium.org/chromium/chromium/src/+/main:third_party/blink/renderer/core/css/media_values.cc;l=575-584;drc=48d6f7175422b2c969c14258f9f8d5b196c28d18 though it only ever uses 2 of the 3 values

astearns: I was a little concerned about the pass rate, but most do seem to have 2 implementations

<astearns> github-bot, take up w3c/csswg-drafts#12773

[css-2025][css-pseudo-4] Add typographic pseudo-elements to Safe to Release exceptions

<github-bot> OK, I'll post this discussion to https://github.com/w3c/csswg-drafts/issues/12773.

SebastianZ: fantasai mentioned in another issue that we need to resolve on a syntax question

SebastianZ: the :: syntax has been in all browsers for many years

ChrisL: css2 had the same thing, but with a single colon

SebastianZ: so we should resolve that all the typographic pseudo elements, with the :: syntax, are safe to release

<dbaron> https://www.w3.org/TR/selectors-3/#pseudo-elements

dbaron: are these not in the selectors 3 recommendation

dbaron: which pseudos?

dbaron: ::first-line, ::first-letter, ::before, ::after are in selectors-3, ::marker is not

ChrisL: nor are ::prefix and ::postfix

SebastianZ: but I think these last two were in css2

dbaron: no

astearns: of the typographic pseudos, we're only concerned about ::prefix and ::postfix

astearns: ::marker is in a different category, tree abiding pseudos

SebastianZ: the idea was to mark all that were part of css2 as safe to release with the :: colon syntax

astearns: if they're in selectors 3, we don't need to

dbaron: the only missing one are ::prefix and ::postfix

SebastianZ: I don't remember why I added those, probably tested them before

astearns: we don't seem to have test for those, and edits seem to be pending too

<astearns> github-bot, take up w3c/csswg-drafts#12774

[css-2025][css-conditional-5] Add `font-tech()` and `font-format()` to Safe to Release exceptions

<github-bot> OK, I'll post this discussion to https://github.com/w3c/csswg-drafts/issues/12774.

SebastianZ: font-tech() and font-format() from css-conditional-5 were implemented in all major browsers since 2023

SebastianZ: I think it was part of an interop project

SebastianZ: 100% interop, only 1 editorial issue on github

RESOLUTION: font-tech() and font-format() are safe to release

<dbaron> +1

<astearns> github-bot, take up w3c/csswg-drafts#12775

[css-2025][selectors-4] Add `:is()`, `:where()`, `:has()`, and selector lists in `:not()` to Safe to Release exceptions

<github-bot> OK, I'll post this discussion to https://github.com/w3c/csswg-drafts/issues/12775.

SebastianZ: the spec was last published in Nov 2022

SebastianZ: these features are implemented in all browsers since 2021, except :has(), which was part of interop 2023

SebastianZ: tests are at 95% interop

SebastianZ: there are some issues on github, but given high interop, should be safe to add

astearns: are there tests for selector lists in :not()

SebastianZ: I don't know about the tests, but I have used the feature, and it works

SebastianZ: yes, there are actually tests

SebastianZ: and they're all passing

RESOLUTION: add :is(), :where(), and :has() to the safe to release

astearns: anything else in that spec?

SebastianZ: [lists many features]

SebastianZ: focus-within should probably go there too

<dbaron> the column selector stuff is also in selectors 4

RESOLUTION: selector list in :not() are also safe to release

florian: focus-within should be good too

astearns: let's do a new issue for that

ACTION: SebastianZ to go through selectors-4 and check for additional safe-to-release features

ACTION: SebastianZ to go through selectors-4 and check for additional safe-to-release features

<astearns> github-bot, take up w3c/csswg-drafts#12776

[css-2025][css-images-4] Add `conic-gradient()` and `<color-interpolation-method>` for gradient functions to Safe to Release exceptions

<github-bot> OK, I'll post this discussion to https://github.com/w3c/csswg-drafts/issues/12776.

SebastianZ: interop is somewhat high, with 83%

SebastianZ: there are about 9 issues on github, plus 3 in the spec

SebastianZ: given high interop and few open issues, I think it's ok for safe to release

ChrisL: conic gradients have been available for a while, the spec was lagging

ChrisL: color interoplation methods is widely implemented, except for the thing we resolved on 2 weeks ago

ChrisL: except for that, it seems good and should be in

astearns: I think we can put it in safe to release, even with that consideration, because it does seem safe

<ChrisL> +1

RESOLUTION: add conic-gradients and color-interopolation-method to safe to release

<Zakim> fantasai, you wanted to comment on different topic

<astearns> github-bot, take up w3c/csswg-drafts#12777

[css-2025][css-easing-2] Add `linear()` to Safe to Release exceptions

<github-bot> OK, I'll post this discussion to https://github.com/w3c/csswg-drafts/issues/12777.

SebastianZ: there was an issue about whether it is in shippable state, and we resolved it is

SebastianZ: implemented across browsers since 2023

SebastianZ: interop is not that high (68%), but we already said it's fine

SebastianZ: just one open issue on github, none in the spec

astearns: the shipable state issue said OK for FPWD, but did not say OK to ship

SebastianZ: Ah, right

SebastianZ: does someone remember if we discussed shipability, or publication only?

astearns: seems we only discussed publishing first

florian: so the precondition is fulfilled, but is the feature good enough?

astearns: test pass rate seems low, I am not OK with putting it in safe to release

astearns: test results show lack of interop

SebastianZ: safe-to-release is supposed to be about spec status, not interop

astearns: editor said it's good

astearns: I'm leaning to not yet

fantasai: the test seems to be largely about specified value, not behavior

fantasai: doesn't seem like a big deal

astearns: I just don't feel confident

<astearns> github-bot, take up w3c/csswg-drafts#12779

[css-2025][css-text-decor-4] Add `text-decoration-thickness`, `text-underline-offset` properties and `from-font` value for `text-underline-position` to Safe to Release exceptions

<github-bot> OK, I'll post this discussion to https://github.com/w3c/csswg-drafts/issues/12779.

SebastianZ: interop is not that high

SebastianZ: spec last published in May 2022

SebastianZ: implemeted in all major browsers, but with moderate interop (66%)

SebastianZ: only 2 issues on github, 1 in the spec

fantasai: look like fairly minor issues (spec bug, rounding question)

fantasai: not concerning from a spec perspective

RESOLUTION: add `text-decoration-thickness`, `text-underline-offset` properties and `from-font` for `text-underline-position` to Safe to Release exceptions

<astearns> github-bot, take up w3c/csswg-drafts#12780

[css-2025][css-env-1] Add `env(safe-area-inset-*)` to Safe to Release exceptions

<github-bot> OK, I'll post this discussion to https://github.com/w3c/csswg-drafts/issues/12780.

SebastianZ: spec is not published yet, not even FPWD, but there's already an issue for that 6729

florian: interop is at 88%

SebastianZ: 10 issues in total

ChrisL: this has been blocked from FPWD from 2021 because of a request on Apple to submit a list of env features, which seems wasn't done

ChrisL: there seems to be contention on the list

<fantasai> w3c/csswg-drafts#11899

fantasai: we should circle back on that

ChrisL: we should decide if we add or not, but either way, we should do a FPWD

astearns: and then we can add to the snapshot

astearns: it's safer to wait

end

Summary of action items

  1. SebastianZ to go through selectors-4 and check for additional safe-to-release features
  2. SebastianZ to go through selectors-4 and check for additional safe-to-release features

Summary of resolutions

  1. republish snapshot with the edits decided on today
  2. css-contain-2 goes into rough interop
  3. update changes list, find active editor, republish
  4. publish updated WD of cssom-view-1
  5. add cssom-view-1 to rough interop in the snapshot
  6. add the subset identified by SebastianZ to the safe-to-release set
  7. font-tech() and font-format() are safe to release
  8. add :is(), :where(), and :has() to the safe to release
  9. selector list in :not() are also safe to release
  10. add conic-gradients and color-interopolation-method to safe to release
  11. add `text-decoration-thickness`, `text-underline-offset` properties and `from-font` for `text-underline-position` to Safe to Release exceptions
Minutes manually created (not a transcript), formatted by scribe.perl version 244 (Thu Feb 27 01:23:09 2025 UTC).

Diagnostics

Succeeded: s/issue for the/issue for that

Maybe present: astearns, fantasai, florian

All speakers: astearns, ChrisL, dbaron, fantasai, florian, SebastianZ

Active on IRC: astearns, ChrisL, dbaron, emeyer, fantasai, florian, SebastianZ