3 ms·

The issue is he makes a bunch of apparently empty claims about his hypergraph stuff being related to physics. The pictures are pretty and the hypergraph stuff

The issue is he makes a bunch of apparently empty claims about his hypergraph stuff being related to physics. The pictures are pretty and the hypergraph stuff is perhaps interesting in its own right, but with all the hand waving about QM and GR, I looked pretty hard a couple of years ago and couldn't find any explanation of the most familiar QM phenomena like the double slit experiment. It's almost like free association. I'd like to see this new theory solve one problem, and by that I don't mean an open research problem. I mean pick any homework problem from an introductory QM course and apply this stuff to it. As Feynman used to say, you have to get the numbers out. As far as I can tell, that hasn't happened.

dTal 5y ago
Even that would prove nothing. Wolfram himself makes a pretty big deal about how cellular automata, being Turing-complete, can model anything. I can give you any number of cellular automata that model our current understanding of quantum physics right now, no problem. It only becomes interesting if it can make different, better predictions than our existing theories.
- throwaway81523 5y ago
  > I can give you any number of cellular automata that model our current understanding of quantum physics right now, no problem Ok, I'm interested in seeing something like that. If you have a link, I'll take a look--thanks. (If you're thinking of 't Hooft's, then I know about it, but its predictions are in serious conflict with mainstream physics, from what I understand).
  - dTal 5y ago
    It's not really that interesting. Can you write it in a computer program? Then you can write it in a cellular automata, because many of them are Turing complete: https://en.wikipedia.org/wiki/Rule_110#The_proof_of_universality https://en.wikipedia.org/wiki/Rule_110#The_proof_of_universa... https://conwaylife.com/wiki/Turing_machine https://conwaylife.com/wiki/Turing_machine And yes, this is a cheap trick. That's the point. "Porting" our current theories to some new, whacky substrate, peeking at the textbooks all the while, is Not Interesting and will not lead to new results.
    - SiempreViernes 5y ago
      As using CA like a substrate is clearly "theory free" in the sense it doesn't impose any restrictions on how calculations should be done, do you concede that CA doesn't actually provide any theory as the OP lamented?
      - dTal 5y ago
        ...concede it? That was my point... Wolfram's new physics is based on "hypergraphs", not quite cellular automata, but the point stand that they are an entirely general framework, and therefore useless for the stated purpose.
    - throwaway81523 5y ago
      No I don't see how to implement QM as a computer program while preserving basic complexity invariants of Turing machines. That is the point of quantum computing: that the complexity classes for quantum computers are different than those for Turing machines. If you take the view that that the whole Universe is one gigantic entangled quantum state, the cellular automaton simulation would get pretty bogged down.
      - tsimionescu 5y ago
        Quantum Computers are no more powerful than Turing Machines, complexity has little to do with this. Quantum Mechanics, being a linear theory (if we exclude measurement) is actually relatively easy to describe computationally. The computations are incredibly slow, but otherwise they are perfectly in line with any other formulation of the theory.
        throwaway81523 5y ago
        > Quantum Computers are no more powerful than Turing Machines They are more powerful in the sense that P-time for quantum computers (aka BQP) is hypothesized to be larger than P-time for Turing machines. That is different from every classical model of computation (RAM machines, etc.) that we usually talk about. It's hard to call something a cellular automaton if it has to expand exponentially in size as it evolves. Thus, 't Hooft concretely predicts that if his CA model is right, then quantum computers can't work. Idk whether Wolfram addresses this issue.
  - lanternfish 5y ago
    With the correct initial state, any Turing complete cellular automata (like CGoL) can model any computable system. We can computationally model prevailing quantum models (albeit incredibly slowly), so an unbounded CGoL field would fit the criteria.
m12k 5y ago
> I looked pretty hard a couple of years ago and couldn't find any explanation of the most familiar QM phenomena like the double slit experiment I was following links from the OP, and while reading his Wolfram Physics one year update[1] I came across this: https://www.wolframphysics.org/bulletins/2020/08/a-short-note-on-the-double-slit-experiment-and-other-quantum-interference-effects-in-the-wolfram-model/ https://www.wolframphysics.org/bulletins/2020/08/a-short-not... [1] https://writings.stephenwolfram.com/2021/04/the-wolfram-physics-project-a-one-year-update/ https://writings.stephenwolfram.com/2021/04/the-wolfram-phys...
- throwaway81523 5y ago
  Thanks, that helps, I haven't made sense of it yet but at least it shows that the issue isn't being ignored.

dTal 5y ago

Even that would prove nothing. Wolfram himself makes a pretty big deal about how cellular automata, being Turing-complete, can model anything. I can give you any number of cellular automata that model our current understanding of quantum physics right now, no problem. It only becomes interesting if it can make different, better predictions than our existing theories.

throwaway81523 5y ago
> I can give you any number of cellular automata that model our current understanding of quantum physics right now, no problem Ok, I'm interested in seeing something like that. If you have a link, I'll take a look--thanks. (If you're thinking of 't Hooft's, then I know about it, but its predictions are in serious conflict with mainstream physics, from what I understand).
- dTal 5y ago
  It's not really that interesting. Can you write it in a computer program? Then you can write it in a cellular automata, because many of them are Turing complete: https://en.wikipedia.org/wiki/Rule_110#The_proof_of_universality https://en.wikipedia.org/wiki/Rule_110#The_proof_of_universa... https://conwaylife.com/wiki/Turing_machine https://conwaylife.com/wiki/Turing_machine And yes, this is a cheap trick. That's the point. "Porting" our current theories to some new, whacky substrate, peeking at the textbooks all the while, is Not Interesting and will not lead to new results.
  - SiempreViernes 5y ago
    As using CA like a substrate is clearly "theory free" in the sense it doesn't impose any restrictions on how calculations should be done, do you concede that CA doesn't actually provide any theory as the OP lamented?
    - dTal 5y ago
      ...concede it? That was my point... Wolfram's new physics is based on "hypergraphs", not quite cellular automata, but the point stand that they are an entirely general framework, and therefore useless for the stated purpose.
  - throwaway81523 5y ago
    No I don't see how to implement QM as a computer program while preserving basic complexity invariants of Turing machines. That is the point of quantum computing: that the complexity classes for quantum computers are different than those for Turing machines. If you take the view that that the whole Universe is one gigantic entangled quantum state, the cellular automaton simulation would get pretty bogged down.
    - tsimionescu 5y ago
      Quantum Computers are no more powerful than Turing Machines, complexity has little to do with this. Quantum Mechanics, being a linear theory (if we exclude measurement) is actually relatively easy to describe computationally. The computations are incredibly slow, but otherwise they are perfectly in line with any other formulation of the theory.
      - throwaway81523 5y ago
        > Quantum Computers are no more powerful than Turing Machines They are more powerful in the sense that P-time for quantum computers (aka BQP) is hypothesized to be larger than P-time for Turing machines. That is different from every classical model of computation (RAM machines, etc.) that we usually talk about. It's hard to call something a cellular automaton if it has to expand exponentially in size as it evolves. Thus, 't Hooft concretely predicts that if his CA model is right, then quantum computers can't work. Idk whether Wolfram addresses this issue.
- lanternfish 5y ago
  With the correct initial state, any Turing complete cellular automata (like CGoL) can model any computable system. We can computationally model prevailing quantum models (albeit incredibly slowly), so an unbounded CGoL field would fit the criteria.

m12k 5y ago

> I looked pretty hard a couple of years ago and couldn't find any explanation of the most familiar QM phenomena like the double slit experiment I was following links from the OP, and while reading his Wolfram Physics one year update[1] I came across this: https://www.wolframphysics.org/bulletins/2020/08/a-short-note-on-the-double-slit-experiment-and-other-quantum-interference-effects-in-the-wolfram-model/ https://www.wolframphysics.org/bulletins/2020/08/a-short-not... [1] https://writings.stephenwolfram.com/2021/04/the-wolfram-physics-project-a-one-year-update/ https://writings.stephenwolfram.com/2021/04/the-wolfram-phys...

throwaway81523 5y ago
Thanks, that helps, I haven't made sense of it yet but at least it shows that the issue isn't being ignored.