Carver Tate wrote: > Do they really think this is possible? How accurate do you all think > this could possibly be? > > http://blog.wired.com/defense/2008/02/pentagon-wants.html One way to look at it is that such models provide rigor in encoding intelligence -- situational awareness.
Even if all you get is a geographical database of where resources are, what major classes of relevant actors are, and there interconnections, that can be useful by itself. It's just that an agent model also gets you the possibility of testing longer-term and indirect consequences of possible actions in the virtual world. They may turn out be poor predictions, but if that happens you can see if it is feasible to improve the model, or just decide not to try to make certain sorts of prediction. Marcus ============================================================ FRIAM Applied Complexity Group listserv Meets Fridays 9a-11:30 at cafe at St. John's College lectures, archives, unsubscribe, maps at http://www.friam.org
