This is from the wikipedia entry on chaos theory, regarding its discovery:

"Edward Lorenz was using a simple digital computer to run his weather simulation... To his surprise the weather that the machine began to predict was completely different from the weather calculated before. Lorenz tracked this down to the computer printout. The computer worked with 6-digit precision, but the printout rounded variables off to a 3-digit number, so a value like 0.506127 was printed as 0.506. This difference is tiny and the consensus at the time would have been that it should have had practically no effect. However Lorenz had discovered that small changes in initial conditions produced large changes in the long-term outcome."

I wanted to share this not because I think it's interesting such a small mistake can have such a dramatic effect on results. That seems to me so obvious it need hardly be stated.

What is interesting is that it was so easy to ignore this concept that the entire meteorological community did so until sometime in the 60s. Lorenz is actually credited with the "discovery that meteorology could not reasonably predict weather beyond a weekly period (at most)."

So who's to say that tiny mistakes aren't being made in other areas of science all the time? Wouldn't all the tiny mistakes being made in gathering and reporting of evidence ultimately add up in the same way as Lorenz's weather simulation?

Reply