[Haskell-cafe] NumberTheory library

[ajb at spamcop.net]

G'day all.

Quoting John Meacham <john at repetae.net>:

> Ack. I use the MTL quite extensivly in most of my projects. I hope it
> won't disapear from the standard libraries completely.

For the record, I have no information that the MTL will disappear any
time soon.  However, the proposed roadmap is to move stuff out of hslibs
and into cabal, and that may be the eventual fate of the MTL, especially
if someone proposes something better for future standardisation.

My real point in all this is that using a state monad for the purpose
proposed in the number theory library is (IMO) overkill, despite making
the code slightly prettier, and introduces dependencies which don't need
to be there.

My real real point is that the rules for libraries are slightly different
than the rules for programs.

Cheers,
Andrew Bromage