you don't need to do either. just run performUnsafeIO if you really want to spit out log statements randomly. I like having to do logging the Haskell way, actu

you don't need to do either. just run performUnsafeIO if you really want to spit out log statements randomly. I like having to do logging the Haskell way, actually. it allows for more coherent and easier reuse. What if want to reuse some code, but do logging in a specific way? By allowing the caller to decide where logging goes you can readily accomplish this.