[Haskell-cafe] Equality constraints in type families

[Manuel M T Chakravarty]

Hugo Pacheco:
> Since I was the one to start this thread, I have managed to  
> implement what I initially wanted as F a :: *->* with F a x::*, and  
> the cost of not having partially applied type synonyms was not much  
> apart from some more equality coercions that I wasn't expecting.
[..]
> Generally, I love type-indexed types.
Glad to hear that!

Manuel