Some months ago I announced the formation of the `Standard Haskell
committee' at the Haskell workshop, to make one final revision of
Haskell which would become a stable, and hopefully simpler, language.
At that time I said we were aiming to complete tidying up the design
in October. Now it's December and still no Standard Haskell -- what's
happened?
What's happened is that the proposal not to change Haskell again has
led the committee to propose much more far reaching revisions than we
originally intended. In our defence, I would say that if Standard
Haskell really is to remain stable for a long time, (and if users are
to stick with Haskell rather than a successor language), then it must
support the kind of programming we already know we want to do. But
Haskell 1.4 has serious deficiencies -- in particular the restriction
to single parameter classes is painfully frustrating. If Standard
Haskell does not drop that restriction, then users will soon move to a
different language that does. There are other extensions as well that
we can already see will be important in the near future.
So the goals of the committee have shifted: we still aim to simplify
the language and remove `traps', but we also aim to `future-proof'
the design. We still aim to produce a language which will be stable
for many years to come.
Meeting these new goals will take longer. We're shooting for a Draft
Standard early in 1998, with compilers available soon afterwards.
Thereafter there will be a period of several months for evaluation
and revision, with the final definition hopefully complete by late next
year.
If you would like to follow the progress of the committee, you can
find the entire discussion via
http://www.cs.chalmers.se/~rjmh/Haskell/
John Hughes
