On Thu, May 21, 2009 at 9:31 PM, Viktor Cerovski <[email protected]> wrote: > Then if it is really side-effect free no function will be ever evaluated > for no key on the board will ever be stroked.
(If a theoretical framework conflicts with an observation, should you reject the observation?) In my opinion, "side effect free" can be a reasonable constraint on computation, but can not be a reasonable constration on communication. And while characterizing Haskel's IO monad as computation might be an interesting exercise, pretending IO is not about communication really misses the point of doing IO. -- Raul ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
