Like point free notation, I worry about what somebody somewhere is doing to
it.... :)
The existence of a well understood community standard (add a type signature
to your functions and only use monad operators with the laws) helps a lot -
but both pieces are optional. I suppose the shorter and more declarative
nature of Haskell functions makes it a less urgent point though.
On 4/16/07, Donald Bruce Stewart
clifford.beshers:
Donald Bruce Stewart wrote:
david:
Ah... so the secret is in the hidden variables. On some level I am beginning to fear that Monads resurrect some of the scariest aspects of method overriding from my OO programming days. Do you (all) ever find that the ever changing nature of >>= makes code hard to read?
You always know which monad you're in though, since its in the type. And the scary monads aren't terribly common anyway.
Also, the monad laws impose a level of sanity that most OO frameworks do not, right?
Ah yes, and we have the 3 laws of monads. If you break these , the monad police will come and lock you up.
-- Don