Re: Overloaded unification

On Jan 29, 9:21 am, Christian Maeder <Christian.Mae...@[EMAIL PROTECTED]
> wrote:
>
> P.S. A typical example for ad-hoc overloading that one may miss in some
> functional languages are selector functions over products/tupels.
> I.e. in Haskell "fst" and "snd" only work over pairs, not over tripels
> or quadruples.

SML does something along these lines (#l is applicable to all records
and tuples, but has to be resolved statically). In practice, I find it
unsatisfactory because you need type annotations too often. Row
polymorphism is a way more elegant and expressive solution to this
problem.

- Andreas