Quantum Computing and the Scientific Method

A while back I expounded an idea that the scientific method as generally practiced couldn't be used in a quantum world. In particular, I argued that the only way to verify a theory was to make a measurement, and that the only way to check the correctness of the measurement was to couple a prediction device with a measurement device. The prediction device might be pencil and paper or it might be a computer. Classically, the result of the prediction can be written down and used for comparison at any time in the future. With quantum mechanics, this is impossible. Now, I don't feel I made a convincing case at the time (this was about a year ago, and I drove a few of my friends nuts talking about my idea), but I found a paper by David Deutsch that does make a convincing case. All Deutsch's arguments apply to the scientific method, so long as it can be shown that the scientific method is a classical algorithm that can be implemented on a Turing machine, which seems obvious to me -- though I haven't sat down and proven it. In fact, if the scientific method can't be framed as such an algorithm, I really don't know what it is! Some heuristic? Note that Deutsch believes his arguments apply even if one accepts (viz Penrose) the Platonic reality of mathematical ideas!

http://xxx.lanl.gov/pdf/math/9911150v1