Newsgroups: comp.ai.games
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!oitnews.harvard.edu!purdue!ames!usenet.kornet.nm.kr!news.kreonet.re.kr!news.dacom.co.kr!nntp.coast.net!news.kei.com!wang!news
From: bruck@actcom.co.il (Uri Bruck)
Subject: Re: WFAQ: Thinking about TownSim
Reply-To: bruck@actcom.co.il
Organization: ACTCOM - Internet Services in Israel
Date: Mon, 18 Dec 1995 22:25:42 GMT
Message-ID: <DJszsJ.JD7@actcom.co.il>
References: <49rbjh$nka@mercury.mcs.com> <49vdno$ooq@mars.mcs.com> <4a03oo$s19@mars.mcs.com> <4a0gtg$jbo@life.ai.mit.edu> <4akoe3$q2b@Venus.mcs.com>
Sender: news@wang.com
Lines: 49

jorn@MCS.COM (Jorn Barger) wrote:



>>altruistic sacrifice
>>		2 (one to commit the sacrifice, one to be the beneficiary
>>		   of said sacrifice.  Possibly more to set up the con-
>>		   ditions for said sacrifice.)

>I think the simplest version of this is *food giving*, especially
>towards one's children (eg, nursing).

>You might assign some degree of linked-ness between persons, where
>you suffer when those you're linked to suffer.  But I think the
>programming here can get tricky, like what if there's only food
>enough for one?

In Shadows of Forgotten Ancestors (Carl Sagan and Ann Druyan) there is
an interesting analysis of a concept they kin-selection (I'm not
suggesting evolving it, I'm suggesting taking it into account).
The basic survival instinct is an individual survival instinct. But
occasional sacrifices to allow relatives to survive are also
beneficial, from an evolutionary [reductionist] point of view, since
they share some of the same genes, and it is the genes we want to pass
on (on a biological level).
A balance is needed. A clan of people who jump at the first chance to
commit an ultimate sacrifice will not survive for long. Neither will a
clan that that practices extreme individualism.

One must first establish the need to make such a sacrifice -
basically, an action that will give another individual a greater
chance of survival/happiness/success etc. and the motivation to make
such a sacrifice, which would depend on kinship, bonding etc.
Uri
>This is another conflicting-motives problem...

>>clan-group feuds
>>		4 (at least two groups, and the word "group" implies (de-
>>		   mands? more than one individual per group)

>I'd like to think this one is marginal, except that Goodall observed
>it in chimps.  It implies negative values for 'bonding' with a whole
>group of same-species individuals, likely due to crowding and limited
>food.  Cannibalism is also a possible plot variant...


>j


