Cellular Automata vs. Agent-Based Models

[jalex]

Dawn Trembath writes:
> 
> Maybe some direct questions would clear this up for me the fastest.  Why
> was Swarm implemented using an Agent-Based approach as opposed to a very
> unrestricted CA approach?  

I do not want to speak for the designers of Swarm, but let me
contribute my $.02 anyway.

Although I am not sure what you mean by an "unrestricted CA approach",
note that CA models assume the agents to be situated in a spatial
world of n-dimensions.  Agent-based models, construing the term
"agent" broadly, need not occur in any explicit space.

Suppose you were modeling interactions between certain
political and economic institutions.  In such a model, though the
agents are well-defined, as well as the relations among them, it would
likely be artificial to impose a spatial framework.  The
agent-based approach adopted by Swarm allows the Swarm libraries to be
used as a foundation for such a model, whereas this might not be as
easy were Swarm built using a CA approach.

My point is that unrestricted CA's are a strict subset of
agent-based models.  Given that, why not adopt the more general
approach, thereby making Swarm available to a larger body of users?

Cheers,

Jason Alexander
Logic & Philosophy of Science
University of California, Irvine


                  ==================================
   Swarm-Modelling is for discussion of Simulation and Modelling techniques
   esp. using Swarm.  For list administration needs (esp. [un]subscribing),
   please send a message to <address@hidden> with "help" in the
   body of the message.
                  ==================================