The Q-DOS of AI, i.e., Mentifex

Mentifex mentifex at scn.org
Mon Jun 8 14:28:24 EST 1998

One day in IIRC 1978 my moment of tension and anxiety at a meeting
of the NW Computer Society was relieved by the author of Q-DOS, a
genial and easygoing fellow named Tim Patterson.  I was working on
<a href="http://www.scn.org/~mentifex/"> my AI project </a> and so
I held back when the chair asked for volunteers for the newsletter.

Tim Patterson, whom I recall in the flesh only that one single time
but whom I have since read about countless times in computer history,
answered the call and gave of himself, as the good guys always do:
Bill Gates wants to rake in fifty billion dollars?  No problem --
Tim Patterson of Seattle Computer Products sold him Q-DOS for chump

Q-DOS was the "Quick and Dirty Operating System" that became
MS-DOS after the done deal.  Strangely, now Q-DOS repeats itself:
<a href="http://www.scn.org/~mentifex/aisource.html"> Mind.forth AI </a>
for robots is about to implement some rather drastic shortcuts on
The Road Ahead to artificial intelligence, namely, a quick-and-dirty
method of creating a Forth array to represent the cortical sheet.

Instead of using CREATE...DOES to generate an array of arrays --
one for each concept in the artificial robot brain, we will rush
the product out the door (as we are wont to do here in Seattle)
by letting one big array "lexicon{}" be indexed on time "t" but
contain many concepts numbered "n" as nodes of nouns and verbs:

  /^^^^^^^^^^^\ Overview of an AI Mind for Robots /^^^^^^^^^^^\
 /visual memory\           ________   logical    /  auditory   \
|      /--------|-------\ / syntax \  memory    |episodic memory|
|      |  recog-|nition | \________/------------|-------------\ |
|   ___|___     |       |     |flush-vector     |    _______  | |
|  /image  \    |     __|__  / \  _______       |   /stored \ | |
| / percept \   |    /     \/   \/ Verbs \------|--/ phonemes\| |
| \ engrams /---|---/ Nouns \    \_______/      |  \ of words/  |
|  \_______/    |   \_______/-------------------|---\_______/   |

More information about the Neur-sci mailing list

Send comments to us at biosci-help [At] net.bio.net