Yes, it's hard: it's impredicative polymorphism. Dimitrios and Tim and I have a plan for a modest but reliable form of impredicative polymorphism, which we hope to have ready for the Haskell Symposium deadline. It is relatively simple, like QML, but a bit clumsy. That's the tradeoff. Simon | -----Original Message----- | From: haskell-bounces@haskell.org [mailto:haskell-bounces@haskell.org] On | Behalf Of Barney Hilken | Sent: 05 March 2012 12:22 | To: haskell@haskell.org | Subject: [Haskell] Higher types in contexts | | Is there any deep reason why I can't write a polymorphic type in a context? I think | the record update problem can be (sort of) solved if you could write: | | class Has r Rev (forall a. [a] -> [a]) => HRClass r where | setHRClass :: (forall a.[a] -> [a]) -> r -> r | | but polymorphic types are not allowed in contexts. Is this one of the problems SPJ | considers "Hard" or is it a feasible extension? | | Barney. | | | _______________________________________________ | Haskell mailing list | Haskell@haskell.org | http://www.haskell.org/mailman/listinfo/haskell