Newsgroups: comp.ai.games
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!gatech!newsfeed.pitt.edu!ctc.com!info.ucla.edu!library.ucla.edu!news.ucdavis.edu!csus.edu!netcom.com!scottec
From: scottec@netcom.com (Scott Cunningham)
Subject: George and Martha
Message-ID: <scottecD7qo6n.MEr@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
Date: Fri, 28 Apr 1995 09:39:59 GMT
Lines: 23
Sender: scottec@netcom12.netcom.com

I just finished reading 'Terminal Man' by Michael Chriton and in it there 
is an interesting AI simulation between two computer personalities named 
George and Martha.

	George is known as Saint George because he is perpetually nice 
and Martha is extremely irritating.  The premise is that the two 
personallities exchange gifts.  Each personality has his/her own likes 
and dislikes.  The conversations were somewhat 'Eliza' like but each 
character learned from the replies of the other and changed the replies 
to suit attitude/emotion.

	Has this been tried for real and if so what were/are the 
results.  Is there code available for any major platform (or source).  
The results in the book (though somewhat dramatic) were that Martha was 
getting less and less bitchy until George finally blew a fuse and got 
violent.

	This would be useful to use in negototiation AI for strategey 
games.  I have in mind something like hostage negotiations where 
terrorists could be worn down.

	Scott

