In particular, ending a lifetime *automatically* released resources owned by borrows at it. And there are two kinds of borrows (unique and shared), which can reinterpret a type in arbitrarily complex ways--but their specs must be strong enough to prove real Rust programs correct.
My sense is that academia thought substructural type systems were a dead end at the time, and it was going to be very hard to get anything about them published no matter what. It’s just the sense I had though; could be wrong.
-
-
I think academics may have a tendency to think that if ideas don’t catch on in practice that it’s because they’re inherently flawed. But in many (most?) cases failure to achieve industry success is due to poor execution or just bad luck, not because of the ideas.
-
My theory is that many ideas are first proposed in too strong a form (a safe programming language with manual memory management! a type checker that works for any scheme code! a partial evaluator that generates a compiler!) and success is in scaling back to something workable.
- 5 more replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.