If you love science, have a subscription or library access to Science magazine, and happen to have a few minutes to spare, I can think of few better ways to spend it than with Kerry Emanuel’s obituary of Edward Lorenz:

Chaos is not merely about the practical problem of prediction. Lorenz and others showed that there are mathematical (and almost certainly physical) systems for which the predictability horizon–the time limit over which a prediction can be made–approaches a finite limit as the initial error approaches zero. Such systems are formally unpredictable beyond this finite horizon, no matter how precise the computation and the specification of the initial state. Even the motion of a frictionless billiard ball becomes completely unpredictable after only 11 collisions, owing to the uncertainty principle’s limit on describing its initial state.