From clockwork to computation in Wolfram Physics with Jonathan Gorard

This is the second of a series of excerpts from my recent conversation with Jonathan Gorard, who was instrumental in the founding of The Wolfram Physics Project.

I asked Jonathan why he found the computational approach to physics so compelling.

In his answer, he broached a wide range of fascinating topics in the philosophy of science:

• how we moved from a clockwork paradigm in the age of clockwork to a computational paradigm in the age of computation;
• how saying that the universe is computational is different from saying that the universe is a computer;
• how our adoption of mathematics as the basis for physics has biased us to think of space-time as continuous;
• how the history of science might have been different had Turing been born before Newton;
• how the Wolfram Model can be thought of as a way of building a constructivist foundation for physics.

This led us to discuss a couple of the deeper questions of Wolfram Physics:

• is it possible to know whether the universe is continuous or discrete?
• does the hypergraph really exist?

—

Jonathan Gorard

People and Concepts mentioned by Jonathan

—

Kootenay Village Ventures Inc.

[Self-Duality]

I’m loving the historically-tinged progression of this conversation! Can’t wait for more segments 😌💭 Jonathan also displays an uncommon hybridization of groundedness and open-mindedness!

[louisgiokas]

Another good example of models not being a statement about ontology is Maxwell's first conception of electromagnetism. His first model was mechanical. He did not really think that electromagnetism was somehow mechanical, but that was the physics everyone knew at the time.

[Zayden.]

follow-up question: Wasn't there a necessity of the growth of machinery, of the industrial revolution take place, for Turing to be able to eventually conceptualize a Turing machine and develop his ideas? The industrial revolution of course took place with extensive use of Newton's ideas. Basically I think there's a certain determinism, a necessity, in the way scientific ideas of continuity and discreteness in reality have developed. It was necessary for Newton ideas to form and the industrial revolution to happen, in order for Turing to come up with his ideas.

[TheWooTubes]

I found this one reassured me about some concerns I expressed yesterday. Is the universe discreet or continuous? Or s it the always elusive Option 3, 'something else', like particles and waves? We could take calculus more literally, that they meet at infinity :-)

[parker]

Essentially the distinction between epistemology and ontology is necessary for correctly guided scientific paradigms.

Chris Langan and Bernardo Kastrup are two philosophers that take the concept of epistemology and ontology further and connect them with mentality center stage.

[YarUnderoaker]

I wonder if it is possible in a hypergraphic system with invariant causality to have such a branch of causality in the form of a premordial neural network that learns from other self developing branches?

[Youtubelaschool]

Ok the fact that the univers is computational doesn't mean we are living in a computer but at least is ask the question. Dont you agree ?

[pallharaldsson]

9:28 ".. does that let us do useful stuff that would be harder to do [than] if we assumed a different data structure, a continuous space time". Einstein's spacetime can't be a data structure [in the computer scientific sense]?! You would need infinity precise coordinates, each taking infinite amount of memory. We can only approximate reals in a computer, so if the universe is a computation, or at least it a good analogy, then it seems it must be discrete in space and time, e.g. with the hypergraph data structure. It seems plausible. The hypergraph can't of course neither be infinity large, i., e. each index seemingly stored in 128 or some X amount of bits... Turing machines allow infinite tape (though only an abstraction, still not each location/index on the tape unbounded), It's not plausible the whole world would run in discrete step in lockstep across the universe. [You could imagine the world parallel like on a multi-core computer, but then you have a different problem, sync across cores, or different tapes in a many Turing computers, how would they send messages to different cores...]. But time could run in discrete steps at each location, then giving a pretty good approximation of all of the universe running at the same speed, or even different speeds if required by the physics, if GR is actually correct.

[nb]

The map is not the territory, even if it perfectly models it.

[williamschacht]

Now days why couldn't set-valued analysis and non-smooth analysis be applied to the models that Jonathan referred to when stating that historically models are formulated in terms of continuous functions etc., if the actual phenomena is discontinuous? Is this type of analysis "harder" than a discrete analysis?

Intuitionism (or any formalism) suffers the "curse of Godel." Why would this be a good (analogous) idea for the Wolfram model, as Jonathan states?

Yes I would also agree that it doesn't matter if the hyper-graph really exists as long the comparison with our universe "matched." It's just a model! Jonathan mentions an isomorphism between the discrete and continuous models, so the claim is that a hyper-graph structure will make derivations/formulations etc. easier than using a continuous model?

[kyaume]

The ideas in these videos on what you call Wolfram Physics are interesting, but the idea of discrete spacetime is by no means new. There are other areas of mathematics and physics where such concepts have been and are being elaborated. Not so much maybe in the direction of hypergraphs, but more along the lines of quantum geometry, discrete differential geometry and discrete integrable systems. I don't want to sound glibe or irrespectful, but this Wolfram Physics seems to lack Physics (at least so far). It lacks laws of physics, and concrete models. OK, it gives a geometric/combinatorial scheme to create possible discrete structures which we can think of as possible universes, which is interesting and imho has potential, but it lacks a model system (at least as far as I have seen, please correct me if I am wrong and have missed out on a development). To put it bluntly: it lacks equations. Maybe they exist, and I would be interested seeing them. But if this is so, I think it would be useful to elevate this theory by linking it to modern developments in the areas I mentioned above.

[GEMSofGOD_com]

It's discrete based on discrete moments of meaningful change-to-change complexity growth and observation. Computational time being initial ordinals. Not precisely a model, but a limit to what can even be called real. It's all dunsamentally discrete. Period.