Summary: Wolfram's model doesn't handle violations of Bell's Inequality, which has been observed in nature. Violations of Bell's Inequality require true randomness in the underlying universe. Hypergraph state machines assume an underlying mechanism that has no randomness. In other words, it functions like the kind of hidden variables theory ruled out by Bell's theorem. They haven't gotten anything like Hilbert space (quantum states) out of Wolfram's model.
There's no doubt that he really, really wants to find a theory of everything. But he hasn't found one, which he admits himself.
What he claims to have found is a framework which can be used to produce such a theory. But what he's pushing seems to be little more than the idea that simple rules can have complex results - the same thing he's been pushing for years. He seems to have moved on from cellular automata, though, essentially acknowledging that his previous ideas were wrong, at least in the specifics.
But the richness that can arise from simple rules is hardly a new idea. Many programming languages demonstrate that. The lambda calculus demonstrated it in the 1930s, but no-one seriously claims that it underlies the physics of the universe.
There's actually a surprising amount in physics that can be proved mathematically from first principles, given some basic premises. The constant speed of light gives us special relativity (and Pythagoras' theorem gives us its equations.) Noether's Theorem gives us conservation laws. The inverse square law follows from the simple mathematics of an abstract 3D space.
Just producing a single new result along those lines would be a major leap forward for Wolfram's ideas. But so far, I'm not aware of anything like that.
And when he does try to relate his ideas to physics, it usually amounts to nothing more than speculation and handwaving. Here's an example from the linked post, about the nature of space:
> "Well, I think it’s very much like the picture above. A whole bunch of what are essentially abstract points, abstractly connected together. Except that in the picture there are 6704 of these points, whereas in our real universe there might be more like 10^400 of them, or even many more."
This is pretty meaningless, since it has no connection to anything observable or testable.
All sorts of mathematics can be used to model a problem. An example is string theory, which can be used to derive correct results about the universe. But that doesn't mean that the strings it postulates exist, because there are other models that can derive the same results without strings.
Wolfram seems inclined to conflate the particular mathematical models he's using with the physics of what he's modeling.
There are cases where that can make sense, like the examples I gave above - in those examples, an observable result about the universe is derived mathematically without needing to postulate new, unobserved theoretical entities (like strings). Such derivations can actually explain something, not just describe or model it.
But in the string theory case (just to use it as a convenient relevant example), what we're almost certainly dealing with is a model that happens to work because it is equivalent in certain important ways to other models that also work. (Occam's razor can come in handy in these cases.)
Wolfram's very approach, where he starts with some framework and then tries to fit it to physics, seems almost guaranteed to produce this kind of result: he may be able to model something, but there's no reason to think that the framework he's using is a uniquely meaningful reflection of physical reality. It's very much a case of "When all you have is a hammer, everything looks like a nail." Wolfram is obsessing over a particular type of hammer.
I'm not an expert at all, but I think the "one model to rule them all" is an sf problem that no serious physicist really work on...because rules are models to describe the universe. Knowing a unique model that match exactly means we are able to know everything about the universe. But how one can know that there is no more phenomena to be discovered, and that all models match exactly ?
slowmovintarget|2 years ago
[1] https://arxiv.org/pdf/quant-ph/0206089.pdf
[2] https://www.youtube.com/watch?v=wd-0COLM8oc
Summary: Wolfram's model doesn't handle violations of Bell's Inequality, which has been observed in nature. Violations of Bell's Inequality require true randomness in the underlying universe. Hypergraph state machines assume an underlying mechanism that has no randomness. In other words, it functions like the kind of hidden variables theory ruled out by Bell's theorem. They haven't gotten anything like Hilbert space (quantum states) out of Wolfram's model.
cuuupid|2 years ago
It remains to be seen if wolfram’s theory outperforms earlier attempts
antonvs|2 years ago
What he claims to have found is a framework which can be used to produce such a theory. But what he's pushing seems to be little more than the idea that simple rules can have complex results - the same thing he's been pushing for years. He seems to have moved on from cellular automata, though, essentially acknowledging that his previous ideas were wrong, at least in the specifics.
But the richness that can arise from simple rules is hardly a new idea. Many programming languages demonstrate that. The lambda calculus demonstrated it in the 1930s, but no-one seriously claims that it underlies the physics of the universe.
There's actually a surprising amount in physics that can be proved mathematically from first principles, given some basic premises. The constant speed of light gives us special relativity (and Pythagoras' theorem gives us its equations.) Noether's Theorem gives us conservation laws. The inverse square law follows from the simple mathematics of an abstract 3D space.
Just producing a single new result along those lines would be a major leap forward for Wolfram's ideas. But so far, I'm not aware of anything like that.
And when he does try to relate his ideas to physics, it usually amounts to nothing more than speculation and handwaving. Here's an example from the linked post, about the nature of space:
> "Well, I think it’s very much like the picture above. A whole bunch of what are essentially abstract points, abstractly connected together. Except that in the picture there are 6704 of these points, whereas in our real universe there might be more like 10^400 of them, or even many more."
This is pretty meaningless, since it has no connection to anything observable or testable.
All sorts of mathematics can be used to model a problem. An example is string theory, which can be used to derive correct results about the universe. But that doesn't mean that the strings it postulates exist, because there are other models that can derive the same results without strings.
Wolfram seems inclined to conflate the particular mathematical models he's using with the physics of what he's modeling.
There are cases where that can make sense, like the examples I gave above - in those examples, an observable result about the universe is derived mathematically without needing to postulate new, unobserved theoretical entities (like strings). Such derivations can actually explain something, not just describe or model it.
But in the string theory case (just to use it as a convenient relevant example), what we're almost certainly dealing with is a model that happens to work because it is equivalent in certain important ways to other models that also work. (Occam's razor can come in handy in these cases.)
Wolfram's very approach, where he starts with some framework and then tries to fit it to physics, seems almost guaranteed to produce this kind of result: he may be able to model something, but there's no reason to think that the framework he's using is a uniquely meaningful reflection of physical reality. It's very much a case of "When all you have is a hammer, everything looks like a nail." Wolfram is obsessing over a particular type of hammer.
logtempo|2 years ago
catlifeonmars|2 years ago
baq|2 years ago