On Mon, 13 Dec 2004 18:21:17 +0900 (JST), Jacques Garrigue
> You make me curious. > By the way, if you want an example of non referentially code,
> this looks easy:
> 
>    do
>      x <- readInt
>      y <- readInt
>      return (x-y)
> 
> (The syntax and functions may be wrong but you get the idea.)
> Of course according to your definition this contains nothing that is
> not referentially transparent once you've taken the syntactic sugar.
>
> But looking at the code, it looks like readInt is executed twice
> returning different results, i.e. this function does not always return
> 0.
Ah no, the key point is that this is not a function but a computation.
In order to be not referentially transparent, you would have to run
it. But you cannot do that -- the key point about the IO monad is that
you can just compose computations involving I/O, but the only time
those computations are actually executed is when "main" is invoked by
the runtime.

> So I suppose this is just an instance of what you see is _not_ what
> you get, but wasn't referencial transparency about avoiding that?
No, besides referential transparency having a definite technical
meaning, you can clearly see what you are getting by simply looking at
the type (IO Int).

Cheers,
Michael