Eric Dedieu escribió:
Still more importantly to me, I understand that anyhow if I intend to use IO or random numbers, I must design my strategy from the beginning as "encapsulated in a monad". Something like:
class (Monad m) => Strategy m a where ...
That's not true at all, you can always pass this data to your strategy entry points and let haskell get it lazily, though it is not as intuitive as other aproaches, Ie. I need to get the next pick for my IA so I can use a function called get IA wich would take as arguments a random seed (or an infinite vector of random numbers using laziness ;) ) and the previous user inputs and strategy outputs. Of course this is just retarding the monadification of the function as soon or later you'll have to embed it into a IO monad to get the desired results. Anyway, making the strategies completely deterministic can also be useful when debugging.