Re:Landmark computer science proof cascades through physics ...
by Strilanc on Thursday March 05, 2020 @02:57PM
(#22497481)
> you've just disguised an unsolved math problem as a computer program
But that's the point! The halting problem hides inside of it all of the complexity of math, including the dark tricky self-referential corners. Saying that you can solve the halting problem is tantamount to saying you've solved Godel incompleteness.
If you applied the halting proof diagonalization step to yourself (modeled as a machine built of out physics) you would find that it spit out something very complicated that ultimately amounted to a program equivalent of a statement of the form "The physics machine that I am cannot prove this statement". If you proved the statement then that would disprove it, which is the core of the problem.
But that's the point! The halting problem hides inside of it all of the complexity of math, including the dark tricky self-referential corners. Saying that you can solve the halting problem is tantamount to saying you've solved Godel incompleteness.
If you applied the halting proof diagonalization step to yourself (modeled as a machine built of out physics) you would find that it spit out something very complicated that ultimately amounted to a program equivalent of a statement of the form "The physics machine that I am cannot prove this statement". If you proved the statement then that would disprove it, which is the core of the problem.