> Continuing Simon Peyton Jones' points on the State of Haskell:
>
> > | Standardization
> > | ---------------
> > | As painful as it may be, I think that we need to formally standardize
> > | Haskell via one or more of the standard standardization organizations.
> >
> > I'm more dubious about this. I have not met a single person who's problem
> > with Haskell was that it isn't an ISO std.
>
> Speaking as an employee of the UK national standards laboratory, I would
> strongly discourage standardisation unless there is a very good reason
> for it. Good reasons are customer procurement requirements or that multiple
> dialects are holding back widespread use - neither of these applies to
> Haskell.
>
> Standardisation of a programming language takes about a decade and is a
> soul-destroying process (destroying the souls of both the participants and,
> often, the language itself).
I'm against it, but I should note that Scheme is an IEEE standard,
and *that* took far less than a decade and didn't destroy anything.
So I'd say: if you're going to do a std at all, start w/ an IEEE
one. Also, it may be easier to make an ISO std from an existing
std (there's a "fast track") than to start w/ ISO.
[I'm not en expert on any of this, but I was involved in the
development of the ANSI Common Lisp standard and in the ISO
work for a Lisp standard (as well as following what happened
to Scheme).]
> The Haskell committee is impressive in having developed a definition of
> about the same level of clarity as an ISO standard in a remarkably short
> time.
Note that the IEEE Schere standard was taken almost directly from
the "Revised Report" -- the Scheme equiv of the current Haskell
definition. I don't see why the analogous thing couldn't be done
for Haskell.
BTW, one reason for making a standard is that, if you don't, someone
else might.
-- jeff
