> M. Pour-El and I. Richards [Computability in Analysis and Physics, Springer,
> Berlin, 1989] have proven that even though solutions of the "wave equation"
> behave deterministically, there exist computable initial data (initial conditions)
> with the weird property that for a later (computable) time the value of the "field"
> is "non-computable" (non-computable evolution). Randomness in physics
> correspond to mathematical uncomputability.
I haven't seen this result, but it is well known that Newtonian physics
has similar properties. In fact you can do an infinite amount of
computation in finite time using Newtonian point sources. I wouldn't be
surprised if this QM result relied on similar unrealistic idealizations.
This archive was generated by hypermail 2b30 : Mon May 28 2001 - 09:59:43 MDT