Evan Laforge wrote:
Also, I wouldn't say more generic (or rather, less explicit lifting needed) is really totally "free"... it always looks more complicated when you see some polymorphic typeclass constrained type constructor parameter vs. a monomorphic concrete one. And I think even just *looking* simple is important!
Agree. If you usually use it with Maybe, then -- the documentation should clue you in on the fact -- after you realize it is, then on the occasion that it's used with a different type than Maybe, you're likely to get confused doing type-inference-in-your-head. Also, ad-hoc polymorphism in *both* function results and arguments, quickly leads to ambiguity and the type checker may yell at you. +1 for Maybe. Conor put it best (though the argument was incomplete without the practical arguments and personal opinions and experiences) -Isaac