LAST EDITED ON 03-16-13 AT 06:29 AM (EST)This one's for you, AyaK, in our mutual interest in the capability to make predictions based on fact and history. It's a quantitative science, so it lends itself to statistically-based models and data mining across multiple disciplines (baseball, cards, elections, economics, earthquakes, epidemics, population growth, global warming, etc.).
You might find this conversation from Edge.org interesting.
I'm in the midst of reading Silver's latest book, The Signal and the Noise: Why Most Predictions Fail – But Some Don't and I highly recommend it.