>I think that the report should definitly rule out the possiblity that every
>evaluation of randomIO uses the *same* initial seed.
I agree. In fact, I would be most happy with a function getRandom :: IO
Double that returns a random number between 0 and 1. I have not been able to
hack such a function using Hugs' randomIO function. What I have done instead
is encapsulating the JavaScript random function in a scriptlet and then
using that via H/Direct's Automation binding. Not exactly a very portable
solution :-) Anyway, I don't see why anyone would want randomIO to return a
list of random numbers, given that you are in the IO monad so that randomIO
could return the next element of this list at every subsequent call.
Erik
