The Future of Haskell discussion at the Haskell Workshop

[Adrian Hey]

On Wednesday 10 September 2003 10:51, Ketil Z. Malde wrote:
> And now, let's just screw any backwards compatibility, and re-engineer
> the records system=B9.
>
> I don't need any of this, and it makes my life harder.  Are you guys
> going to keep at it, until I regret ever using Haskell?

I can't speak for any Haskell implementors or whether or not they intend
to keep going at it. But maybe there are people who will regret ever
using Haskell if nothing is done about the current records/modules
situation (me for one). I don't mind waiting a while, if it's still
unclear what should be done or can reasonably done given current
state of the art re. the necessary type theory. But my understanding
of the original summary was that I might as well give up hope of
ever seeing anything like this in Haskell, for fear of upsetting
the status quo. I don't like that idea much.

I think if you want to use a language which is close to state of the art
you have to accept some change. Better this than see it permanently
crippled by backwards compatibility constraints. That said, I don't
see why any backwards compatibility problems can't be managed with
suitable compiler switches or whatever, which seems to be what
Simon.M. is proposing for ghc.

Regards
=2D-
Adrian Hey