Haskell-prime Digest, Vol 2, Issue 6

[Simon Peyton-Jones]

| We didn't dare include MPTCs in Haskell 98 because the number one goal
| on that occasion was to be conservative! But if we had taken that view
when
| Haskell was first designed, classes would never have seen the light of
| day at all. I think now is the time to bite this bullet.

I'm all for biting, but I'm not sure that the timescale is compatible
with finishing by mid 2006.  I think the right way to bite is to use
associated types and they are simply too untried for Haskell'.  

Now, someone can try to define a conservative version of FDs, as has
been suggested; perhaps it could even be kind-of-isomorphic to
associated types.  Perhaps something *very* conservative would be enough
for most applications.  (It would hurt to standardise FDs if we think
we're going to discard them subsequently, but if the spec was somewhat
crisp maybe it would not hurt too much.)

I'd be happy to help review such a specification, but I don't want to
write it.  And I think it must be at least somewhat formal.
(Karl-Fillip Faxen's paper on the static semantics of Haskell would be a
good starting point, and the paper that I mentioned earlier.)

Simon