[Haskell] Re: Global Variables and IO initializers

Adrian Hey ahey at iee.org
Thu Nov 4 14:55:09 EST 2004

On Thursday 04 Nov 2004 5:54 pm, George Russell wrote:
> However virtually everyone seems to have their own patent solution,
> and we are no closer to agreement.

FWIW, what Koen describes is exactly what I want (I think :-).
So there's at least 2 votes for the "non-IO" monad approach,
provided there's an escape to IO monad (typically needed for FFI
related initialisations).

I also think we need something like..
 CIOtoIO :: CIO a -> IO a
so that CIO constructors can be used normal IO code too.

Adrian Hey

More information about the Haskell mailing list