Re: Bob's 'Self-aggrandizing ignorant' Count: Was: What databases have taught me

From: Marshall <marshall.spight_at_gmail.com>
Date: 30 Jun 2006 18:45:59 -0700
Message-ID: <1151718359.763424.33210_at_75g2000cwc.googlegroups.com>


Tony D wrote:
> Bruno Desthuilliers wrote:
>
> > FWIW, I think this is just impossible. Computers are imperative. You
> > need to tell them *how* to do anything.
>
> Not necessarily: consider SKIM (a hardware implementation of the
> SK-combinator scheme), Norma (a graph reduction machine built by
> Burroughs in the mid-80s) and of course the Symbolics LISP machine.

Not to mention the fact that it is perfectly possible to use one computational model to implement another one. We can write prolog in C, for example, or python in haskell. The fact that there exists a lower layer (in this case a hardware layer) that is imperative and untyped doesn't mean we can't use that layer to build a higher layer that is functional and typed. You could do the reverse as well, alternating computational model for as long as you cared to.

This is one place where an "it's turtles all the way down" argument doesn't work.

Marshall Received on Sat Jul 01 2006 - 03:45:59 CEST

Original text of this message