[bitc-dev] monads

[Jonathan S. Shapiro]

On Tue, 2008-03-25 at 16:55 +0100, Pierre THIERRY wrote:
> Scribit Jonathan S. Shapiro dies 25/03/2008 hora 10:34:
> >   2. Every valid Haskell program *would* have a direct transcription
> >   to BitC *if* BitC provided support for monads.
> 
> For the few that I know about monads, it seemed to me that they were
> only needed to have side-effects in purely functional language. As BitC
> is an impure one, like Scheme, aren't Haskell programs already trivially
> transformable to BitC?

I agree. The motivation for monad support comes from two places:

 1. A pure science motivation, allowing direct comparison of
    implementation techniques within a single language.

 2. A state reduction motivation, allowing our compiler to witness that
    certain apparently stateful procedures are actually not stateful at
    all, thereby reducing the burden of stateful software verification.

shap