Monday, February 28, 2011

Why is it so hard to predict the future?

There is this time machine paradox. If you were able to go back into the past to change things to make things "better" in the present, you may end up altering the course of history that lead to the developments that caused you to exist, or that made the technology available to you to time travel. A recursive dependency, ending up in a logic trap we can't wrap our limited brains around.

This is the same reason why it is extremly difficult to reliably predict what will happen in the future. The more precise, tangible we try to pin down future events, the more likely we will affect the outcome leading to an alternate, different future. At least if we communicate our prediction to those involved in the assumed future outcome.

The stock market is a good example of that. If someone spreads a "secret" tip, how to get rich quick in the stock market, many people will follow, do the same thing, and in effect eliminate the niche opportunity by their very actions of pursuing it.

If we develop a hypothesis as to how the future will shape up, it is most stable if the factors that influence that hypothesized future are unaware of our theory. If the factors involve self-aware agents, like people, they may adapt to intermediate events, or even the expected outcome, and by these actions possibly avoid the assumed future event/outcome.

On the scientific side, popular examples in this real are the Heisenberg's Uncertainty principle, and Schroedinger's Cat thought experiment. As Quantum Computing will advance, interesting possibilities may shape up for predicting the future. Not necessarily to our satisfaction. The human mind is a paradox in itself. If you ever pondered "how big is the universe?", what if I told you it is <this> big. Then the logically next question would be "where does it end?", and I tell you it ends <there>. "And what comes after that?".

If first-graders already ask such pertinent questions, what can our highly developed adult minds come up with?