SCN: Fwd: Artificial Societies - this is fascinating stuff

[sharma at blarg.net]

------- Start of forwarded message -------
Date: 4/11/02 9:46:14 AM

Now they are simulating societies with computers and, in the process, 
finding out that societies can be simulated with a surprisingly small 
number of rules. I first saw this concept in Steven Johnson's book 
'Emergence', which talks about how complex behavior can arise from a 
small number of simple rules. This is apparently true in practice as 
well as theory:
http://www.theatlantic.com/issues/2002/04/rauch.htm

Yet more proof that the study of 'complexity' is often really the study 
of how complexity emerges from simple systems. One of the more 
interesting examples cited in the article is the study of official 
corruption using two types of agents; citizens and bureaucrats. Could 
this kind of thing be used to simulate governance and laws in a quest 
for a utopian society? (In the case of the previous example, simply 
adding in 'jail time' for bad behavior and a couple of incorruptable 
agents into the mix was enough to inculate 'honesty' throughout the system.)

If you think that people are too complex for this kind of simulation to 
be any use you are forgetting that even individual human behavior may 
often be modelled as an emergent system. Groups of people are apparently 
far easier to model. The examples of artificial racism (and genocide) in 
the article are especially chilling. However the article centers on a 
simulation of Anasazi culture that turns out somewhat different than 
reality; proof, at least, that you must model *all* the factors. Perhaps 
a win for more conventional 'Chaos Theory'. Or not.

Lots of other good stuff in the article, for example it mentions 'Zipf's 
Law'. But I didn't find the thing I most wanted; a link to download a 
program to simulate the stock market...



-------- End of forwarded message --------



* * * * * * * * * * * * * *  From the Listowner  * * * * * * * * * * * *
.	To unsubscribe from this list, send a message to:
majordomo at scn.org		In the body of the message, type:
unsubscribe scn
==== Messages posted on this list are also available on the web at: ====
* * * * * * *     http://www.scn.org/volunteers/scn-l/     * * * * * * *