The Universe Is Not a Computer | Hacker News Reader

The Universe Is Not a Computer

technologyreview.com

17 pointsca98am7926 comments

There are more things in heaven and earth, Horatio...

The following statements are not equivalent: 1) The universe is computational or digital in structure. 2) The universe is a giant computer. 3) The universe is a simulation.

#2 implies #1. #3 implies both #2 and #1. #1 does not imply #2 and #3. The universe can be digital in nature, without being either a simulation or a "giant computer".

Also I think it ironic that the article uses quantum mechanics to show how the universe cannot map to computational processes because of the problem of determinism, at the very time when quantum mechanics is being used to create computational devices that leverage the property of non-determinism.

Also "dangerously wrong"? Where's the danger, exactly? This whole article boils down to the question, "But what if they are wrong?" Ok, what if? What if Supersymmetry is wrong? What if String Theory is wrong? That's the way science works. You propose ideas. You test them. You throw out the bad/unprovable/wrong ones.

Exactly. A lot of what happens in the universe is a computation by formal definition of computation (which is rather broad and all encompassing). However, universe itself need not be a computer at all.

I felt a sense of urgency when reading this article, as though it were trying to be the Catcher in the Rye, preventing us from trying to wake up Inception-style.

I guess it is "dangerous" because that is what we are teaching children at school nowadays, and young people at college, narrowing their views to this generally accepted concept.

I think you misunderstand the paper. And, in fact, I think the main point of the paper should be completely uncontroversial to anyone with even a lay understanding of GR.

The point of the paper is that apparently the yout's of today are being taught that the universe started at the beginning of time with an initial state of S0, and then the state of the universe evolved (and continues to evolve) from that point point in time towards the future. There is one special point in time, which is now, and the past states of the universe no longer exist and the future state of the universe do not yet exist.

In GR, however, this is not the way things work. Instead, all states of the universe always exist, the past states are still there and the future states are already there, and the present is not a special point in time; it just happens to be where we are. The fact that we don't currently experience the past or the future is no different from the fact that we are also not in Idaho.

The universe, instead of evolving from the past to the future is the ever-exising solution to some equations. The distinction becomes more apparent when you think about the existence of closed timeline loops in GR and how those could exist without paradox. It's easy! Pardoxical situations are not a solution the equations.

Is this view right? Who can say? What's dangerous about only presenting the "computational" view is that it limits the imagination of what is possible. Both models should be presented so that young minds evolve in a more flexible manner.

Ah. This makes more sense to me. I believe the issue here is with the terminology, and (as usual) poor reporting.

So what you are saying is that the author is objecting to that view which limits one's understanding of the universe to a fixed-state system.

IIRC, the author didn't even really object to the computational model (aka "Newtonian Schema"). He just argued that it should not be the only one taught, as the "Lagrangian Schema" has some distinct advantages.

I think the author means to say, "Since physicists input initial states to get to a meaningful output" or "Since the physicists assume cause and effect are sequential", it leads to the grander assumption that Universe is a giant computer. One way of seeing the Universe is as a stacked geometry of inclined planes in infinite number of permutations and combinations. Physical phenomena like light/gravity/time are just rolling balls on these inclined planes taking any number of paths. Thus pairs of effect-effect or cause-cause are also possible.

If you say "the universe is/isn't computable", then you might get somewhere exploring the idea.

If you say "the universe is/isn't a computer", you're in the terrain of metaphysics again.

Typo in the first sentence: "now one physicists" -- which also undercuts the title since this is the opinion of one professor, and that his point is that assuming the universe works like a computer may limit your perspective when trying to model/understand it.

Terrible title for an interesting article.

At one point in the distant future, entropy will drive our universe to be nothing but a nearly endless array of evenly-spaced neutrinos. At that point, the computer will be ready for computation.

For now, it's still booting. That the boot process takes so long is only due to our relative perspective.

After reading the article, who assumes the universe is a computer? The only times I heard it is people suggesting the possibility, not believing in it. I'm sure there are some people who actually think the universe is a computer, but this is not a widespread problem, or anything really.

The article is conflating "the universe is a computer" with "the universe can be described as a computational process, generating later states from earlier ones." I could certainly see this assumption creeping in unnoticed in various places, though I have no clue as to whether that's the case or what the ramifications are. If it's a correct assumption, of course, then it doesn't matter that it's been assumed all over.

Math, computation, and physics are not things we crafted from nothing; they were explicitly made as tools to describe and predict the universe around us and we continue to use them because they work so well.

But relativity killed the simple, clean model of Newtonian physics by kicking out the underlying assumptions, and I like the idea that at least somebody's considering the possibility that maybe our deepest, most fundamental premises are completely wrong.

I have no idea how anyone who's gone all the way through to a physics Ph.D. is going to be able to unlearn those assumptions though, that sounds incredibly difficult.

I think what the article's really referencing is the Church-Turing thesis: http://en.wikipedia.org/wiki/Church%E2%80%93Turing_thesis

A lot of people conjecture that, if true, it implies that there exists no deterministic process which cannot be modeled by an algorithm.

The article also references the probabilistic nature of quantum phenomena. However, quantum mechanics is deterministic under the Many-Worlds interpretation. Observations don't collapse waveforms, our quantum states are deterministically entangled with that of the system under measurement.

The jury is still out on the proper interpretation of quantum mechanics, but quantum phenomena doesn't rule out the deterministic state-change metaphor a priori.

It's been my impression that, MWI aside, quantum mechanics being probabilistic does not imply that the processes it models are probabilistic. The map is not the territory, and a map that is known to be incomplete is especially not the territory.

Stephen Wolfram's ideas on New Kind of Science (NKS) suggest that the universe is based on simple computational rules. (If I might paraphrase a thousand plus page book I've only partially read into a sentence.) Strangely, the paper in the OP doesn't mention that.

We humans learned how to compute from the natural universe.

Seth Lloyd, in "Programming the Universe", seems to argue the universe is a computer or is like a computer. The universe computes "its own dynamical evolution". "As the computation proceeds, reality unfolds."

The statement that translating an outcome in a mathematical model into the physical world is, in quantum mechanics, a probabilistic operation simply depends on one's viewpoint. The Schrodinger Equation IS deterministic. As always, this is simply a matter of how one views quantum mechanics.

Reads more like a thinly veiled attempt to backdoor talk about god into science again.

It does not. It does not say or imply anything of the sort. In fact the point of the paper should be "obviously true" to anyone who has even a lay understanding of GR.

I'm sorry but WTF.

Maybe I have my computer science theory all wrong but I distinctly remember this statement: the universe is Turing complete.

That means that it is at least a computer - it may be something more - but it must be at the very least a computer.

Secondly WTF is with non-computable arguments. Given arbitrary memory and ability to do computation - anything can be computed.

Thirdly what's this mapping crap - the universe's probabilistic state does not need to be mapped - it's just information moving around.

The Lagrangian method he uses can be computed. Looks just like an expensive optimisation problem.

I'm not sure that "the universe as Turing machine" is anything more than a theory. The universe can contain Turing machines -- we're both using one right now -- but I don't think we can say for certain that the universe is one.

Also, not everything is computable. Alan Turing proved this in his paper "On Computable Numbers, with an Application to the Entscheidungsproblem." The Halting Problem is a related example of the limits of computation.

http://en.wikipedia.org/wiki/Halting_problem

Computers are an existence proof - the universe is indeed at least a Turing machine.

The halting problem can be effectively solved for deterministic finite machines - it would just take a long time time.

Reply on news.ycombinator.com