Mark Hammond

> In swift, this generates a Person which conforms to equatable, but Animal doesn't That's because Swift is able to directly compare each field for the record. However, it is...

> (In fact with this capability, I suspect this issue should be closed?) hrm, probably not - the original issue is about interfaces *inside* records, so that's probably still true....

A bit late to the party here, but the code which generates this has no idea about what is actually in the component interface. I don't see an obvious way...

TBH, I'm not aware of Mozilla looking at that over the last couple of years. We have very simple profiles - typically `opt-level="s"`, `lto = "thin"` and `debug = true`...

> Feel free to close this issue if it's not really on your radar for the docs. Cheers! It really just means we haven't explored that for a couple of...

Yeah, the foreign namespace is currently flat for a given crate. I'm not aware of any plans, so I think we'd welcome proposals, but I suspect there's not going to...

> There was a PR at one point to do that, Are you thinking of #1063 maybe? Frustratingly, that PR was rejected because there was a fundamental objection to something...

Yeah, the attribute name isn't ideal and we probably can do better. However, I think the objection was more from the foreign side - that something declared as a "constructor"...

> You can get on owned `Self` instance by removing an element from a HashMap. To me, this kind of blows up the whole "new" vs "existing" distinction based on...

> My proposal allows to choose the runtime and to configure the runtime you want _per function_. We've a vaguely-defined use-case that would like to choose the runtime dynamically at...