David
On Wed, 12 Nov 1997, Jerry Balzano wrote:
>
> Like you, I am _very_ interested in the activities of building and playing
> with models, not just as a part of science curricula but "across the
> curriculum". Model building, to me, can be thought of as a kind of
> formalized version of story-telling. This is just one of many reasons why
> I am so intrigued with the variety of "agent-based" modeling software,
> because with them you are modeling a system at the level of the behaviors
> of the agents that constitute the system. Compare modeling an epidemic by
> working with differential equations vs. with a StarLogo model. In the
> former case you have to abstract out state variables like number-infected
> and number-susceptible, develop some notion of how they change over time,
> and translate this into mathematical notation. In the latter case you
> start with a story like, "You come into contact with someone who has the
> disease. You catch the disease. As you wander through the world, you
> infect other people. When you get better you might not catch the disease
> so easily again." One can translate this story reasonably directly into a
> set of StarLogo commands (an extension of "playing turtle"?), certainly far
> more easily than most people can reason about the aggregated state
> variables whose changing values emerge from these low-level interactions.
>