University of Cambridge > Talks.cam > Semantics Lunch (Computer Laboratory) > Let should not be generalised

Let should not be generalised

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Sam Staton.

The Hindley-Milner type system is a thing of beauty that is so deeply embedded in the cultural history of type systems that we hardly realise how wonderful and surprising it is.

However, in this talk I’ll argue that this beauty is a tad misleading. HM is a local optimum: before we can go up we must go down. In particular, I’ll argue that we must give up the automatic generalisation of local (nested) let-bindings. As we make the type system more sophisticated it becomes harder and harder to support auto-generalisation.

Happily, it turns out that auto-generalisation is seldom use. I have numbers.

This is all part of the complete rewrite of GHC ’s type inference engine, which Dimitrios and I have just completed—it’s in the upcoming GHC release.

This talk is part of the Semantics Lunch (Computer Laboratory) series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2019 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity