On Mon, Jul 21, 2014 at 12:07 PM, Jonathan S. Shapiro <[email protected]> wrote: > On Sun, Jul 20, 2014 at 11:43 PM, Matt Oliveri <[email protected]> wrote: >> At any rate, Curry-style >> types are basically just extra tacked-on information about an existing >> thing. > > Until you get to qualified types, I think I agree.
You said something like that on another thread too. I looked up qualified types and it seems like a generic way of restricting the domain of type quantification. Semantically, it seems to provide predicates on types. I don't understand why this would change the nature of Curry-style typing. > Though of course the > ground functions of the runtime need to be declared in order for the rest of > the inference process to be feasible. I don't know anything about inference for Curry-style types. I hope you're not assuming I'm restricting my attention to type systems with complete inference, or anything. 'Cause I don't know how to do that. _______________________________________________ bitc-dev mailing list [email protected] http://www.coyotos.org/mailman/listinfo/bitc-dev
