Niko Matsakis
Niko Matsakis
2020-08-05: Discussion seems to be centered about the question of how much to separate the "retrospective" (what has happened over the last N years?) from the "edition" (what migrations or...
Update: still waiting on folks to review the RFC.
Update: I e-mailed Mozilla legal to get their opinion.
Update: as @skade wasn't present today, no update.
One specific use case that I would like to avoid breaking: - Sometimes we have a bug fix or other change and we want to evaluate the impact - If...
+1 to adding a `-Z` flag, seems useful. I see that @pietroalbini already implemented it (though it seems to be pending review). I don't really have a lot of insight...
Interesting points. I had never considered "interpretation B" -- it still seems strikingly less useful to me, but I agree that `Iterator` makes a good case for where you *do*...
On a related note, I have mixed feelings about the pattern where all the combinator methods are defined in the trait itself, to be honest. I think the main reason...
If you ignore trait objects, then there isn't much of a reason to adopt the first interpretation -- where `impl Trait` desugars into an associated type on the trait --...
Some examples that came up in discussion today: To represent `static S: Foo = f-body`, instead of ``` (StaticDecl (term (S (static () () (rigid-ty Foo ()) fn-body)))) ``` we...