Wow, he's turning into a crackpot: long, rambling articles with no introduction or conclusions that have no math or results to speak of, furtive mentions of how the rest of the physics community can't grasp how brilliant and far-out his remarkable work is (which could be because he doesn't publish, doesn't collaborate, doesn't go to conferences, and basically doesn't actually participate in the research community).
When the book first came out, I remember there was an air of skepticism surrounding some of its concepts. Has the view towards the book by the science community changed much in the years since? I'm not a physicist, but have always been interested in Wolfram's work out of curiosity from a comp sci perspective.
Some ideas have already become commonplace, others less so.
Simple programs as 'language primitive' for modeling was criticised by many, but is now becoming popular, particularly in biology (see for example http://fontana.med.harvard.edu/www/Documents/WF/Papers/signa...)
General ideas like computational irreduciblity and instrinsic randomness are finding increasingly direct evidence in the physical world (http://www.scienceblog.com/cms/fruit-flies-have-free-will-13...)
In terms of exploring the raw computational universe, lots of people in architecture and generative design are doing related things now.
Unforunately most people are unaware of the fundamental ideas that underly all of these threads.
Why should it be this network thing? He doesn't back up his assertion that the universe is described by one of these networks.
Juergen Schmidhuber lays out an interesting idea that our Universe is just one Turing machine, generated by an enumeration of all Turing machines (generating all possible Universes). http://www.idsia.ch/~juergen/everything/html.html
The difference is that Wolfram wants an actual, single correct model in his hands, whereas Juergen is just talking about hypothetical computations that theoretically produce our universe as well as all the rest (with no hope of being discovered by us humans)
Effectively Wolfram is pruning the search space by looking for the simplest rules that he thinks have a shot of reproducing known physical features of the universe (eg special and general relativity, with a plan for how QM may come about), and then looking within that space for the specific flavor that is our universe. If you actually want to find the model via searching programs, this is the only realistic way to do it.
"But I suspect it's really more integrated: that everything is "just space", with the particles being something like special little lumps of connectivity in the network corresponding to space."
He's not saying it's definitely a network, he's just saying that he's going to start looking there because it intuitively makes sense. This seems reasonable for a few reasons. First, networks are the most advanced form of organization we currently know of. Second, it potentially provides a very simple explanation for many things, such as gravity. Third, the idea that matter is just really concentrated nothingness is just too cool not to look at. And it sort of makes sense if you believe in zero-point energy, which supposedly both general relativity and quantum mechanics predict (according to the free energy nuts).
I read NKS. Good book! Especially for math, physics, and programming geeks.
There has been a lot of speculation that the universe is a simulation. If Wolfram manages to ascertain the rules and network that "runs" the universe, then the question becomes whether a universe is a simulation, or just acts like one. Is there a difference between something that has only computational rules and structure and a computer program?
So if our universe is essentially computational, what are the implications of Godel's incompleteness theorem? Doesn't this mean there must be an infinite number of axioms our universe is generated from?
If Wolfram is right, our universe would be a mathematical "function" not a mathematical "system". Godel showed that every system of sufficient complexity has facts about it that can't be proven. There is no reason to think though that every function within a complex goes back on itself. x = y, for example, doesn't loop around no matter how far out you go.
For my part however I don't believe it's possible for the universe to be "pure math" (that is, its reality comes from an equation and nothing else besides it) in the sense that we understand "math". Nevertheless, it might be that there is an equation that can predict everything we experience. We may as well search for it.
Wolfram's a lot smarter than I am, but I believe he also believes the function is discrete and not continuous. In fact, he makes the point in NKS that we may have went off the rails with Newton and the calculus. It may turn out that the universe is discrete and that our understanding of it as continuous is just an artifact of not having the computational power to see how discrete functions can effectively look continuous at a much larger scale (sort of a reverse proof of the integral)
It blows my mind. But if you're a genius, might as well have a hobby, right? I'm just glad he took this up instead of bowling.
Just thought I would add that clarification in there, because I know the word "function" has a lot of meanings to people.
The problem I see is 'truth' that is not generated by the system the universe is a function within. To then maintain truth is that which corresponds to reality gets hard. Kinda trippy.
From the reports I can read on that page, the citations refer to sentences of the form:
CA produce complex/unpredictable results[n]
[n] Wolfram, S. NKoS
So yes, it's referred to a lot but I don't see anything very deep.
I think the problem people have with this book is that it was hailed as the second coming before it came out and that SW had been in his basement for the past 5 years discovering CAs that ran the universe.
What it actually turned out to be was a thousand pages of examples/cataloguing CAs and the /suggestion/ that this is how the universe works rather than anything to prove it.
People citing work would indicate that work's quality, but the complaint about "A New Kind Of Science" was weak citations, an indication of poor quality. I don't know much about the details. Nor do I know how much people have sited Wolfram's work since.
I think what upsets a lot of people is the attitude. But a bad attitude can still accompany good work, so I excuse Wolfram for his quirks. That acceptance is essential when dealing with people who know they are much smarter than the average person.
While at NYU, I saw Wolfram speak while on his big book tour. There were a few big brains at the Courant Institute that leveled major criticisms at the book. Unfortunately, there was another talk about nonlinear PDEs that most of the critics elected to attend in place of watching Wolfram. Wolfram started the talk by saying, "well, the important people aren't here to see this, but I suppose I'll continue anyway".
As one of the hoi polloi, I found the remark amazingly conceited, probably correct, and a bit charming at the same time.
wow that is incredibly naive. You think people in academia don't actively try to promote and position their work? Let alone the book and journal publishers, and conference organizers that have an obvious interest in all this.
What exactly is 'unduly' about it? Going on a book tour? Putting up a website?
That his research was funded through success in the market rather than government subsidies doesn't imply any 'unduliness' to me, but attitudes may vary.
