When Edward Lorenz gave a talk in 1972 entitled "Does the Flap of a Butterfly's Wings in Brazil Set Off a Tornado in Texas?," he distilled the main essence of his thoughts on predictability, interdependence and "chaos theory" in one pithy question.
Lorenz, who died earlier this month, was a mathemetician and a meteorologist who, in the early 1960s, discovered that weather simulation models he was developing were exhibiting chaotic, non-predictive behavior, despite a fixed set of variables and no apparent equipment malfunction. Two identical weather simulation machines, side-by-side, given the same variables to process. Wildly different results. How?