Mindforth

Arthur T. Murray uj797 at victoria.tc.ca
Tue Dec 3 00:11:48 EST 2002


standley at rcn.com (spurk) writes on 2 Dec 2002:
>
> "Alex McDonald" <alex_mcd at btopenworld.com> wrote in message
> news:<ase3p1$avs$1 at knossos.btinternet.com>...
>> "Bill Harvey" <bill_harvs at hotmail.com> wrote in message
>> news:asdoss$fel$01$1 at news.t-online.com...
>> > So how do I get Mind.f to run, I take it I need a different
>> > version of MindForth?
>> >
>> >
>>
>> Have you tried version 4.2 of Win32Forth yet? You were running
>> on a later version in your previous post.
>> ---
>> Regards
>> Alex McDonald
>
> Hi.  I have just obtained  winforth and the mind.f program;
> I am having the exact same problem; I have tried both 4.2 and 4.1,
> with the same error message as the originator of this thread.
mentifex at scn.org/ATM:
Hello.  I confess to being the author of the misbehaving
http://www.scn.org/~mentifex/mind4th.html Mind.Forth program.
>
> Your ideas w/ regards to AI are fascinating, I was quite
> looking forward to trying out your program as soon as I
> discovered it.  please help : )
ATM:
In my laziness, when Bill Harvey first complained on 30 November 2002
upthread, I assumed that the problem was with the version of Forth.
(By the way, several Forthers over the last few years have reported
that Mind.Forth, with minor tweaking, works on several Forths.)

The Win32Forth that I have been using, from the Tom Zimmer site at
ftp://pweb.netcom.com/users/zforth -- now demanding a login -- says:

Compiled: July 9th, 2001, 11:29 a.m.
Version: 4.2  Build 0671

However, now that two people have reported the problem, it
must be in my AI code, sorry :-)

However, it is not an AI problem; intuitively I think that
it is an erroneous starting or stopping point in the ".echo" module.

I will try to reproduce and correct the faulty behavior "real soon now."
Any Forth expert who can beat me to it is welcome to host the code
on a public website.

The Mind.Forth code is lagging a little behind the JavaScript
http://www.scn.org/~mentifex/jsaimind.html -- Mind-1.1 release.

That lag began in August of 2002 when I tried to port into
Win32Forth a cluster of mind-modules that I had initiated in
the JavaScript AI in June and July of 2002.  For some reason,
the new modules were not working as well in Forth as in JavaScript.

Then I spent September and October of 2002 slaving away at
http://www.iuniverse.com/bookstore/book_detail.asp?isbn=0595259227 "AI4U".

Because the Mind.Forth code was lagging, in the AI textbook I
only included the Sun.27.OCT.2002 JavaScript AI source code --
which anyway is object-oriented and more modern than Forth :-( / :-)

However, I still consider Forth to be the main language of the
Mentifex "Seed AI" -- for several reasons:
- JavaScript is too limited;
- Forth can control machinery (robots!);
- Forth is in various ways its own operating system;
- Forth on Forth-chips can be ultra-fast; etc.

I am very eager to get Mind.Forth onto a 64-bit system.

Anyway, this new Mentifex AI textbook covers both the
Forth and the JavaScript AI versions.  It also is heavy
on the AI theory, which can be found in detail at
http://www.scn.org/~mentifex/theory5.html and elsewhere.

In late May of 2002, while doing the ~sixth iteration of
updating the http://mind.sourceforge.net/progman.html AI Docs,
I suddenly discovered how to make an ASCII brain-mind diagram
for almost every one of the now thirty-four mind-modules --
and now all 34 of those diagrams grace an AI4U chapter.

I will try to announce in news:comp.lang.forth when I
find and correct the Mind.Forth ".echo" (apparent) bug.

Thanks for your patience.

Respectfully submitted,

Arthur T. Murray



More information about the Neur-sci mailing list