2 ms·
> The bootstrapping process is not laughably bad, the developers have merely prioritised their own ergonomics Ironically, I would disagree here; the language d
by steveklabnik 6y ago
> The bootstrapping process is not laughably bad, the developers have merely prioritised their own ergonomics
Ironically, I would disagree here; the language developers have not prioritized their own ergonomics, which is why building rustc is a big pain.
I do agree that making it easier to bootstrap in that way would not necessarily help them... there is a sense in which you're right, I just found the framing kind of funny, since I'd argue the same point but for completely opposite reasons.
- chrismorgan 6y agoBuilding rustc may be a bit of a pain (big pain? It’s more than five years since I last built rustc from source, but I don’t recall having any particular difficulty with it), but I was speaking specifically of the bootstrapping part, which I don’t feel is at all terrible. Specifically targeting version n − 1 means that you can adopt new language features, which you couldn’t if you were targeting version 1.0 forever, or writing it in another language. That’s what I was meaning by prioritising their own ergonomics—allowing them to use new features. With that clarification, I’m interested in whether you agree or disagree. You’ve been a lot closer to the action than I have.
- steveklabnik 6y agoYes, I do think you're right about this generally, which is what I was getting at at the end. This doesn't have to be all-or-nothing, though, if you check out matklad's post I linked, he points to a version where the standard library still gets to use unstable features, but the compiler does not, which is an interesting hybrid that may retain the best of both.