On May 17, 7:57 am, Malcolm McMahon <
malc...@theriomorph.me.uk> wrote:
> On Sat, 11 May 2013 00:06:33 -0700, Graham Cooper wrote:
> > 1 year of debugging later...
>
> >
http://blockprolog.com/1st-AI-code.png
>
> > ..
> > ..
>
> > PROLOG is a Natural at generating code...
>
> > here the answer should be PNT ( 1 , ANY )
>
> > Getting Close!
>
> > ------------------
>
> > Herc
>
> I came across a group of AI people who were using a Prolog interpreter
> which was written in LISP, which in turn was running on a Pop3
> interpreter.
>
> I suspect, sometimes, that AI people come up with the least efficient use
> of computer power they can devise so that they can blame their failures
> on lack of computer power.
>
You mean the fundamental 'complexity explosion' when expanding AI
applications to work on larger domains is really just an excuse for
"It doesn't work on much!"
Block Prolog will be completed on schedule!
It's now up to proving
DOG+1 is not a NUMBER!
http://blockprolog.com/nat-s-dog.png
Herc