Keith Cirkel

We just rolled out the polyfill in[ `Primer::Alpha::Overlay`](https://primer.style/view-components/components/alpha/overlay) on PVC. I believe the polyfill is ready to go, but it is [not without its caveats](https://github.com/oddbird/popover-polyfill#caveats). It still suffers from the...

> The Chrome team are aiming to ship in 113, which will be main in May IIRC. Both [mozilla](https://github.com/mozilla/standards-positions/issues/698) and [safari](https://github.com/WebKit/standards-positions/issues/74) are supportive of this API. [Mozilla has started implementation](https://phabricator.services.mozilla.com/D171719)...

> > The Chrome team are aiming to ship in 113, which will be main in May IIRC. Both [mozilla](https://github.com/mozilla/standards-positions/issues/698) and [safari](https://github.com/WebKit/standards-positions/issues/74) are supportive of this API. [Mozilla has started...

I have a refactor PR that’ll be ready soon which will resolve these without regressing the other cases we’ve run into.

I did some work on trying to create a polyfill for Temporal Durations so we could move to those but the work stalled due to the complexity of the Temporal...

This is a good idea. Additionally I think we can afford to start with a `console.warn` and then throw a RangeError for a breaking release. Down with non-iso8601 dates!

I concur but an error would be a breaking change. I’d like for our breaking changes to have a lead up so IMO a console warn should happen in a...

That does seem confusing. Would you kindly add a set of failing tests around here: https://github.com/github/relative-time-element/blob/3a8f30f1a66edbe8d4e76c721ff1767c49b0ed27/test/relative-time.js#L1756-L1762 If you raise a PR where the `expected` is your desired text, and we...

A PR adding attributes+props for `timeStyle`/`dateStyle` would be welcome.

I'm not sure I share those opinions; if you're extending a class it's fairly common to have to override parts of the class to deal with things like private state...