I'd like to try to shift the focus of the discussion of Standard Haskell.
The fact is that another revision of Haskell is in preparation, that the
present committee intends it to be the last, and that there are no current
plans for official standardisation. If some of those decisions turn out to be
mistaken then they will no doubt be reversed in the future, but for the time
being that's how it is.
Naturally, if there is massive opposition to the design of Standard Haskell,
and if the decision taken at the Haskell workshop is proved to be
unrepresentative of the wider Haskell community, then I will shut down the new
committee and Haskell's future will take a different turn. But frankly, at the
moment I don't see that --- I see some understandable disquiet, and discussion
of the decision, but not massive protests. I feel I've been entrusted with a
job to do, and I intend to do it.
Given that, let's make this revision as successful as possible. It *is*
different from previous revisions, in that we are explicitly trying to
simplify where we can (without losing power), and we're trying to remove
`warts' --- those little traps in the language that any user may fall over,
and that can be particularly baffling for beginners. We're prepared to go back
and re-examine decisions made in the original Haskell design, if that leads to
a useful simplification. This is an opportunity to make Haskell a better
language!
You can see the ongoing discussion on
http://www.cs.chalmers.se/~rjmh/Haskell/Display.cgi?id=0
But it isn't certain that we on the committee are aware of all the `bugs' in
the present design: the programs that fail in strange ways, the features that
are harder to understand than they need be, the situations where you know
exactly what you want to do but the language won't let you...
It would be very useful if people who have encountered problems with the
language which we seem to have overlooked would let us know about them,
preferably by mailing an appropriate committee member. But please check the
web site above first to see if the topic is already under discussion.
John Hughes
