Agenda for Dec 1 meeting

[foolip]

Here's the agenda for our meeting on Thursday, Dec 1:

• Interop 2022 investigation project scores
• Proposal review, revisiting 4 proposals that we didn't resolve in the last meeting
• Interop 2023 proposal grouping (email)
• Update grid/flex test-suites for 2023 #242. Views of the existing + proposed tests:
  • Flexbox
  • Grid
• Giving feedback on proposals that are not accepted. First idea is to divide up proposals ~evenly among volunteers, at least one person for each engine. Get text reviewed by the other volunteers. Fall back to a wordsmithed version of Logistics of proposal selection #243 (comment) if necessary.

Previous meeting: #249 (Nov 24 meeting was skipped due to Thanksgiving)

[jensimmons]

Could we add briefly discussing the current Investigation project scores.

[foolip]

@jensimmons sure thing, I'll add it.

[nairnandu]

Attendees: @boazsender, @chrishtr, @dandclark, @foolip, @jensimmons, @nairnandu, @ntlm, @robnyman, @tantek, @zcorpan

Brief notes:

Proposal review follow-up

• [@foolip, @chrishtr] Need more time to review 145 (URLs) and come back with a position
• [@foolip] Tl; dr - exclusions being explored are file:// URLs, mailto, javascript:/ URLs and ShadowRealm API
• [@jensimmons] Need to re-visit 182 (Animated AVIF) and 175 (WebCodecs API)
• [@jensimmons] WebCodecs - revisit next week
• [@jensimmons] Animated AVIF - very little confidence that tests are working properly. Continue to exclude. We are looking at tests for all of AVIF.
• Feedback on 211 Accessibility tree outputs - changed position to Neutral

Interop 2022 investigation project scores

• [@jensimmons] Are the scores up-to-date?
  • Editing, contenteditable
    Issue updated
  • Pointer and Mouse events
    Issue updated
  • Viewport measurement
    [@jensimmons] raised an issue - Update progress score  interop-2022-viewport#22
    AI: [foolip@] connect with bramus@ on updating the score

Interop 2023 proposal grouping (email)

• [@ntlm] Content visibility and contain intrinsic size different from container queries? “CSS Containment level 2” as the name maybe?
  • Spec Level 2 adds content visibility and paint containment. Not sure about contain-intrinsic-size.
  • Level 3 has container queries
• [foolip] Container queries should be its own thing.
• [nt1m] how about split by property (contain/content-visibility/contain-intrinsic-size)
• [jensimmons] for being able to explain to web developers, I think a single “containment” bucket would make sense. They’re additional parts of the same thing. For typography, it’s different, they were just a few things related to typography that we put together.
• [foolip] Web Components, does anyone know this well?
• [danclark] I think Import Assertions make sense, they’re a prerequisites for things like CSS modules and HTML modules.
• [zcorpan] Web Workers isn’t really related to web components
• [nt1m] Import Maps seems related to modules, so we could create a “modules” bucket.
• [danclark] if we have that, then we could add import assertions to that as well.
  Web Compents and Modules is an option.
• [jensimmons] Smaller features. I think we could take :modal and :nth and do a pseudo-classes area, but not including :has(). So it’s “here’s :has(), and we’re putting some attention on pseudo classes as well”.
• [nt1m] Masks feel more web compat than its own feature.
• [foolip] I suggest labeling the tests and then looking to see if it’s almost all tests, or just a handful.
• [jensimmons] masking comes up in state of css, suggesting it’s a struggle, so its own focus area could make sense. But let’s label tests first.
• [foolip] AI for self, create google doc with grouping.

Giving feedback on proposals that are not accepted

• Revisit in the next meeting

Jan launch

• [jensimmons] We should be thinking about PR coordination now. Last year we all wrote blog posts, and at the last moment there was the idea of press outreach. That needs more lead time. So we didn’t try to create a press cycle, and just published blog posts. We’d prefer to do the same thing, publish but without proactive press outreach.
• [robnyman] There’s something in the middle as well, like developer marketing.
• [jensimmons] Would have to talk to our PR folks. But obviously we’ll post about this. Different from contacting CNET.
• [foolip] call for everyone to consider what outreach they want to do. Revisit next week
• [robnyman] I can start a doc to share what we’re doing.

Scoring

• [tantek] 2-4 weeks ago there was a suggestion to have a score that only shows the tests. I think we agreed, and that we could have an investigation-only rollup score. I thought that was settled. But didn’t agree to change the main score that we have.
• [jensimmons] From memory, we landed on adding an interop score. I don’t think we’re in total agreement yet and we need some drawing.