jgraham

Skipped already means something specific (we didn't try to run the test). I'd be opposed to reusing it for "we ran the test, but the test decided it wasn't relevant"....

Per the discussion in https://github.com/web-platform-tests/wpt-notes/blob/47c3ea1a941c0b0f78869ff2d00a2e2f93594636/minutes/2023-11-07.md?plain=1#L24 we might want to try a different implementation approach. We likely need to try with existing tests to see which approach is most robust.

I think the next step is for @gsnedders to give some feedback on what can work from WebKit's point of view.

I will note that in the past I have been _very_ skeptical about making this work across multiple repos. The problem is that each repo needs to be using exactly...

The linter should be safe enough, I think, because it's generally possible to find some code format that satisfies multiple versions of a linter, unlike autoformatters which generally define a...

Unless we have a way to address the concerns in https://github.com/web-platform-tests/rfcs/issues/171#issuecomment-1777545307 we _can't_ adopt a formatter.

https://phabricator.services.mozilla.com/D186092 is gecko updating black a few months ago. It reformatted 290 files. I think that's sufficient to prove that we'd need a system that required every user to have...

I think usually with `tentative` tests, we expect there to be a standards document, and the test is `tentative` because the pass condition presumes one outcome of an ongoing technical...

In practice I don't recall anyone previously trying to merge an RFC without any kind of positive signals before. However there have been a lot of RFCs that went a...

Cases like #101 are basically why we said that it's OK to add testdriver endpoints as long as they have backing in specification text. Both #101 and #177 are cases...