| String Theory Discussion Forum |
| [ String Theory Home ] [ Forum Index ] |
Re: M Physics 21: Complete TheoryPosted by DickT on December 21, 2003 at 13:41:14: In Reply to: M Physics 21: Complete Theory posted by kx21 on December 20, 2003 at 12:08:53: Yep. And you might want to look up BSS (Blum, Shub, and Smale) and their idea, the BSS machine. Briefly the BSS machine is modeled on the Turing machine except that everything that makes the Turing machine digital and sequential is replaced by something continuous valued and direct-access. So the BSS machine can do real (complete) arithmetic and computation. It's not a real machine any more than Turing is, but an intellectual device for modeling real computation as Turing modeled digital. Definition: Let a BSS-halting set be defined as any set of inputs to a BSS machine for which it gives a definite output. (i,e, it doesn't run forever) If a set of inputs is a BSS-halting set, and also its complement is a BSS-halting set, then it is DECIDABLE. Theorem: Any countable union of subsets of the real numbers, each of which can be defined without using universal quantifiers, is decidable. Which means if you perform any finite computations with real numbers, you don't have to worry about Goedel. I am sure you of all people will see the point when I say that I can't imagine any final physical theory that can't be defined with finitely many computations. You don't have to evaluate constants like pi, that's all internal to the BSS machine. So, I believe is taking limits - that's a single operation on the BSS machine. Bottom line, I think physics can be defined in a complete way. But the current way it is defined could leave gaps for Goedel to sneak in (so I think - I don't know of any specific gaps).
Follow Ups: (Reload page to see most recent)
|