W3C

– DRAFT –
Open UI Telecon April 10, 2020

10 April 2020

Attendees

Present
Greg, Greg Whitworth, Levi, Levi Thomason, Markus, Markus Mielke, Mielke, Rob Eisenberg, Thomason, Whitworth
Regrets
-
Chair
Greg Whitworth, Levi Thomason
Scribe
Rob

Meeting minutes

<gregwhitworth> TOPIC https://‌lists.w3.org/‌Archives/‌Public/‌public-open-ui/‌2020Apr/‌0002.html

<nickrice> Present

Join Community Group

Greg Whitworth: Begins by covering the agenda and giving Rob an outline of scribe info. Greg invites folks to join the community group.

Review Charter

<gregwhitworth> https://‌github.com/‌WICG/‌open-ui/‌blob/‌gwhit-charter/‌research/‌src/‌pages/‌charter.research.mdx

Greg: Recognizes Nick's editorial changes.

Greg: Two Questions for Levi primarily and also group.

Levi: Talks about goal to provide tools and resources.

Levi: There may be room for tests as part of the spec. e.g. if a spec designates specific accessibility for a component

Greg: Lots of tools out there for things like accessibility. Perhaps better to contribute there.

Greg: Wants to clarify that our goal isn't to build things like linting tools.

Levi: Agrees.

<gregwhitworth> https://‌github.com/‌WICG/‌open-ui/‌blob/‌gwhit-charter/‌research/‌src/‌pages/‌charter.research.mdx

Greg: Recommends deleting goals 4 and 4. * Eventual browser standards for web app components * Converging designer processes and developer workflows

Correction: Goals 3 and 4.

Greg: Recommends moving these two goals to "scope of work" section

Levi: Thinks goal 3 should stay but drop the word "eventual"

Levi: One of the primary focus is to improve the web platform and ultimately get rid of custom implementations in user land.

Greg: Asks Levi to rephrase goal 3 to clarify this.

Greg: Suggests associating the goal with a phase to get rid of ambiguous "eventually" wording.

Levi: To follow up with a PR.

Greg: Reviews the scope of work in the charter

Levi: Asks clarifyication on states. Are design concepts included? Is Danger a state?

Greg: No. We're coming from two different worlds. Greg is thinking in terms of pseudo-classes while Levi is thinking in css classes.

Levi: Asks for a bullet to utilize descriptivism for design terminology. While this doesn't affect platform APIs it does relate to design token and theming interop.

Greg: Greg agrees. He wants to see the platform side and also agrees that the design system side is important as well.

Greg: Do we need to call out accessibility here since the spec template covers that already?

Levi: Believes much of this falls under "utilize descriptivism" already.

Levi: However, we want to use this to fill in gaps where things don't yet exist.

Greg: Do we need a bullet for behaviors (it seems to be already implied by the state transition wording)?

Nick: Thinks a rewording is in order. Felt it was odd that "behavior" was missing.

Greg: Is default appearance out of scope?

Levi: Maybe any appearance?

Greg: Removes the appearance bullet from "out of scope".

Greg: Moves to PR Process section

Greg: Can we use the code owners file to open up the PR approval process a bit?

Greg: Can we use this to assign an "editor" for key documents.

Levi: We need something for scaling, but maybe not cross-section on component. Lots of concerns common across components.

Greg: We're commiting changes on a specific component though.

Greg: Date picker and how it relates to something like Popup is an example of cross-cutting concerns.

So, in that case, maybe it means factoring out a new popup component.

Greg: So, in that case, maybe it means factoring out a new popup component.

Greg: Folks beginning to get engages with openui have less experience engaging with w3, so he wants to find ways to make the process more friction-free.

Greg: Let editors handle changes directly and only handle more contentious or ambiguous issues in an actual call.

Levi: Has another format.

Levi: Concensus isn't as meaningful because there aren't enough samples.

Levi: Instead of having a few people agree, could any person prescribe person and then measure agreements and implementations of that proposal.

Greg: Are we splitting hairs here?

Greg: We want every web dev to weigh in but that doesn't occur.

Greg: We should make a recommendation based on all of our research. But there are less and folks who have enought time to provide valuable insight.

Greg: We can always overturn a resolution if a pseudo standard arises based on the broader community activity.

Greg: We start from descriptivism based on our research still.

Levi: Clarifies, not saying we need thousands of people to agree.

Levi: 4 or 5 big players would be enough.

Levi: Anybody can put a proposal forward, but need the 4 or 5 who are engaged.

Levi: We need to look at those 4 or 5 to make sure they both agree and later implement.

Levi: If implementation doesn't follow, then it's not as valuable as a "concensus"

Greg: Need to look at why things don't get implemented. It might be because it's not possible as spec'd. We shouldn't wait for implementation, but we should circle back around based on implementation.

Greg: Validates that the phases/stages are important.

Levi: Was interpreting concensus as stage 3.

Greg: Oh! No, concensus on issues.

Greg: Similar to what editor's draft means.

Greg: This is why I want to lift ownership about a bit.

Greg: The editor's draft get treated like a living standard has an editor gains reputation (in practice).

Levi: Asks Greg to add a comment clarifying that concensus does not mean stage 3.

Greg: Feel free to skim through the charter.

Greg: I think we can add code owners for different areas. e.g. Make Rob a code owner of the spec template.

Greg: Doesn't want more than 3 chairs so we don't slow down things (not many advantages of being a chair).

Greg: Let's try and land the charter PR before the next sync and keep things moving.

Greg: Calls for feedback from the group so we can make that happen.

Levi: Good for merging after addressing the issues described above.

Greg: Thanks everyone!

Greg: Rob is amazing.

<mmielke> hello everyone

Minutes manually created (not a transcript), formatted by scribe.perl version 114 (Tue Mar 17 13:45:45 2020 UTC).

Diagnostics

Maybe present: Correction, Nick