GHC 6.12.1 and impredicative polymorphism

Simon Peyton-Jones simonpj at microsoft.com
Tue Feb 9 10:01:50 EST 2010


| Hello Simon and others,
| 
| unfortunately, I missed this e-mail.
| 
| Yes, removal of impredicative polymorphism hurts me, since impredicativity
| plays a crucial role in the Grapefruit FRP library at the moment. This is
| described in section 5 of my paper “Signals, Not Generators!” [5]. It’s
| probably possible to use a workaround involving a newtype wrapper, in case
| polymorphic fields in newtypes are still supported. However, this makes things
| more awkward for library users.

I plan to support impredicative polymorphism in some form; but you may need more type annotations.

I'll announce when there's something to test (it'll be a couple of months).

Simon


More information about the Glasgow-haskell-users mailing list