Comment by cynicalkane

When I write in Haskell, I find myself mentally glossing the returned monadic state, along the lines of, "Oh, an M x is just an x that does monady stuff to get the x". This becomes natural once you get the hang of do-notation and sometimes monad combinators. So I'm not really thinking about the monadic state in the return value a lot.

It's not really any less natural than thinking stateful programming, except now the state is a reified thing, which I think is strictly advantageous once you get used to it.