Why is monomorphism necessary?  The sections in the Report on this topic
are very unclear (to me).  The monomorphism restriction is particularly
disturbing.  I _really_ do not like the limitations on lambda.....  I am
guessing that part of the problem may lie in trying to support
polymorphism both in values and arguments.  There may not be a good way
to do this.  Perhaps Haskell should eliminate one?

I think that for Haskell 2 (if not Haskell 98), the whole type system
should be re-thought from scratch, based on what people think a type
system should provide.

BTW, I have trouble understanding what monomorphism means.  Could
someone explain?  Also, I am still looking for info on transcendental
types......
--

  ______________________________
/  David Feuer                  \
|  [EMAIL PROTECTED]         |
|  [EMAIL PROTECTED]                |
\  [EMAIL PROTECTED]          /
  -----------------------------



Reply via email to