2024 TPAC planning

[rniwa]

Are there any topics we should discuss during TPAC in breakout sessions?

[justinfagnani]

A few things I'd like to discuss:

• Finishing scoped element registries MVP. I think the previous round of open questions was resolved, but there's a newer topic of how to use scopes when constructing detached trees that would be good to cover. I can help prep things to show the current status for this one.

• Theming. It's been a while since we tried to talk about theming from a high level. A lot of people are interested in "open styleable" for theming uses cases, not necessarily the backwards-compatibility with existing stylesheets use case, so I think it'd be good if we took some of the pressure off open-styleable to be all things by opening up the theming discussion again.

• Declarative CSS modules (deduping per-component CSS in SSR). Microsoft is very interested in this now and dedicating some time to work on it.

• DOM Parts, especially goals and constraints. Chrome's experimentation is ongoing, but so far a native implementation doesn't look like a slam-dunk performance win. Are there DX goals that justify the feature, or higher level features (like templating) that we should pursue?

I don't know if they all warrant their own session or should be grouped.

[rniwa]

@Westbrook : anything to add?

[Westbrook]

The couple of things I can think of based on our recent meetings:

• Are implementors outside of Chromium still aligned on https://github.com/WICG/webcomponents/blob/gh-pages/proposals/reference-target-explainer.md? It's been evolving a bit, and the initial implementation from the Microsoft team is coming close to done. It would be great to reconfirm that other implementors are onboard and what (if any) support the community can lend in furthering the absolutely most important issue we face an custom element authors.
• We're discussing a couple of CSS selector additions that it would be good to make sure implementors are aligned on:
  • The ability to match :host:has(...), on top of the great work WebKit has done in support of :host(:has(...)): https://github.com/w3c/csswg-drafts/issues/10693
  • :has-slotted(...): https://github.com/w3c/csswg-drafts/pull/10586
• And while thinking about CSS, getting insight into any blockers that implementors are seeing around bringing ... with { type: 'css' } to the JS module system would be nice as well.

Baring any bad news, they should all fit as smaller conversations in a larger window, rather than needing their own individual blocks.

[rniwa]

Proposed scoped custom element registry as session: https://github.com/w3c/tpac2024-breakouts/issues/26

[rniwa]

Proposed open styleable shadow tree & theming as a session: https://github.com/w3c/tpac2024-breakouts/issues/27

[rniwa]

Proposed DOM parts as a session: https://github.com/w3c/tpac2024-breakouts/issues/28

[rniwa]

Proposed CSS modules as a session: https://github.com/w3c/tpac2024-breakouts/issues/29

[rniwa]

Proposed Web Components & ARIA as a session: https://github.com/w3c/tpac2024-breakouts/issues/30

[justinfagnani]

Re: DOM Parts. I wrote down one of the things I was thinking of as a motivation for DOM Parts outside of the pure performance vs a library of the low-level API, declarative templates in JS: https://github.com/WICG/webcomponents/issues/1069