Newsgroups: comp.ai
Path: cantaloupe.srv.cs.cmu.edu!rochester!cornellcs!newsstand.cit.cornell.edu!portc01.blue.aol.com!newsxfer3.itd.umich.edu!howland.erols.net!worldnet.att.net!arclight.uoregon.edu!news.bc.net!unixg.ubc.ca!aurora.cs.athabascau.ca!news.mag-net.com!freenet.unbc.edu!news.scn.org!scn.org!mentifex
From: mentifex@scn.org (SCN User)
Subject: Re: More on the Verb Problem
Message-ID: <E4ttpA.H67@scn.org>
Sender: news@scn.org
Organization: Seattle Community Network
Date: Thu, 30 Jan 1997 15:00:46 GMT
Lines: 44


Malcolm McMahon <malcolm@geog.leeds.ac.uk> saw, then commented:

  /^^^^^^^^^^^\ Syntax Strings Together a Thought /^^^^^^^^^^^\
 /visual memory\           ________   semantic   /  auditory   \
|      /--------|-------\ / syntax \  memory    |episodic memory|
|      |  recog-|nition | \________/<-----------|-------------\ |
|   ___|___     |       |     |flush-vector     |    _______  | |
|  /image  \    |     __|__  / \  _______       |   /stored \ | |
| / percept \   |    /     \/   \/ Verbs \------|->/ phonemes\| |
| \ engrams /<--|-->/ Nouns \    \_______/      |  \ of words/  |
|  \_______/    |   \_______/-------------------|-->\_______/   |

> ba672@lafn.org (Arthur T. Murray) wrote:

>> This design for artificial intelligence will lead to AI robots if
>> enough people see the design and it is not proven to be in error.

>> Note that only the CONCEPTS of words exist in the semantic memory,
>> while the speakable, phonetic words move about in auditory memory.
>> Grammar works silently, and you hear the results inside your mind.

> Oh, for heaven's sake you write as if natural language parsing with
> formal rules was the be all and end all of AI whereas, as far as I can
> see, it's an interesting sideline which has been pretty much done to
> death starting 20 years ago.

The main novelty value of the Mentifex theory of mind lies in how
it INTEGRATES linguistic structures with the rest of the brain-mind.
If you can show that this Mentifex mind-model is either unoriginal
or invalid, you may well stop these Mentifex posts and win a Nobel.

> ITS NOT THE HARD PART, and in any case formal grammars are probably
> not the way to go. The hard things are things like ideation and the
> various kinds of memory. Even a speach act is a transaction which
> depends far more on assumptions about the state of mind of the person
> being spoken to than it does about arranging words.

The Mentifex mind-diagram and its Webspread associated theory files
do indeed address and arguably SOLVE the hard part of AI:  ideation
and the associative memory channels which give rise to the ideation.
Simply that you do not understand something, does not invalidate it.
Even if the Mentifex message ceases to propagate visibly its silent
memetic activity within hosts of infected minds will lead to Omega.
