Re: Functional Programming in the Larger or Functional?Oriented Software Engineering

Thomas F. Burdick <[email protected]> wrote in article <[email protected]> in gmane.comp.lang.lightweight:
> Haskell is a great example of the same thing.  Its evaluation rules
> absolutely require mutation under the hood, much more so than in a
> strict-evaluation language, where you could at least imagine a VM that
> avoids mutation until it uses up all the machine's resources.  But
> this under-the-hood mutation buys you a language with a lot less
> mutation going on in user-space.

I don't see Haskell as different in this regard.  Standard Haskell
(in particular, without unsafePerformIO) is consistent with
applicative-order evaluation without memoization (call-by-name rather
than call-by-need).

-- 
Edit this signature at http://www.digitas.harvard.edu/cgi-bin/ken/sig
As the choice is essentially cosmetic there will likely be no end to
the debate on what the correct decision should be.
http://successor-ml.org/index.php?title=Quoting/anti-quoting