← Back to context

Comment by Locke1689

15 years ago

It's not good, it's very bad. Haskell researchers spent years, in fact, attempting to make sure that it was both decidable and ran in less than exponential time.

The fact that some people actually think an undecidable type system is an advantage is simply an astounding disconnect from reality.

Did they succeed? I vaguely remember a lecture last year where the lecturer told us that type checking in SML was exponential in the worst case ..

  • Type annotations have reduced those cases to very rare instances, but yes there are still cases where the type checker can produce exponential runtime.