Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Eagleon]

So, this emerged from my ass not twenty minutes ago after reading a bit of I Am A Strange Loop by Hofstadter (of G.E.B fame), and I haven't really had much time to think about it much myself. Nonetheless, it's compelling enough in its annoying existentialism that I feel the urge to inflict it on you all.

Say there's a scientist (or a pilot, or a dog, it doesn't really matter much) that has a procedure. That procedure is the procedure that runs our universe. He has a word processor with unlimited page space, and maybe a calculator for aid, but otherwise no other tools. It takes him an incredibly long time to run through the procedure of course, so let's say that the scientist is immortal, but let's not call him God, because that brings up all sorts of irrelevant things like whether or not he might be distracted by salisbury steak or not. Let's just call him Steve.

Steve begins the procedure, but of course it has as much of a foreseeable end as our own does. He doesn't understand its final result, or anything in between save the infinitesimal fraction devoted to his own existence. He continues the procedure for as long as we like, and it ends at some point. It doesn't matter where.

Where does the universe exist in this process? Is it Steve's understanding and following of the process? Is it the stored manipulations inside the document? Some combination of the two? Neither? Both? How many universes are there - you could argue that there's more than one coming into "being" very easily. What happens to the universe when either gets lost - when Steve dies, or when the document is erased?

The fact that there's a procedure at all to run our universe says that the universe is procedural. I'd rather not get hung up on the perception of randomness in our own reality - to me, elements of quantum mechanics have been proven to appear in some ways unpredictable, but that doesn't mean something predictable and mechanical isn't creating patterns that are stochastic and meaningless to us. The (assumed) fact that the universe can be stored in writing as the result of a procedure means that editing any part of this procedure but the next step is essentially meaningless for its final outcome, but the final outcome must also be editable by Steve. Can those edits be considered part of the procedure?

Is it weird that thinking about this crap makes me not want to think about it too much? If everything's one long incredibly mechanical story, how am I thinking about that? What does that mean? Where on the page does it have me thinking about the page, and what notation is Steve using to scratch that part in? Worse, if older parts of the simulation are lost as the procedure is run, the final result of the procedure may not be that interesting by itself, and for some odd reason that makes me irritable.

Yeah, I know a lot of this stuff has been hashed and rehashed by various authors (including, especially, the one I'm reading now), but I was bored so here's this long-winded post about nothing.

[ChairmanPoo]

Spoiler (click to show/hide)

[Eagleon]

Spoiler (click to show/hide)

Spoiler (click to show/hide)

Yeah, I know a lot of this stuff has been hashed and rehashed by various authors (including, especially, the one I'm reading now), but I was bored so here's this long-winded post about nothing.

[Christes]

Spoiler (click to show/hide)

Beat me to it!

[Ampersand]

Spoiler (click to show/hide)

Beat me to it!

Thirded.

Supposing that the universe is infinite, it would require converting a second universe of the same size entirely into a data storage unit just so be able to hold the relevant data about every individual point location in the universe, never mind the states of objects that may actually be in those spots. It is on this grounds that I reject the idea of the universe being not 'real' somehow. Be it simulated in a computer or the mind of god. It requires the conception of something larger than the infinite.

[Reelyanoob]

Larger than infinite is not a problem

If every particle in our universe took exactly 1 planet worth of area to calculate, that's still do-able for an "infinite" outer universe. Infinity multiplied by a factor "X" is still infinity, not larger than infinity.

Say that for a infinite set (1,2,3,...infinity). We say that "each number in this set must be repesented by 1 trillion values in another set). The new set is still of the same order, infinity.

e.g: the set of all integers has the same size as the set of all positive integers. Adding negative values does not increase the size of the set, even though we infer it's "double". This can be shown through a simple mapping equation:

Code: [Select]

+0 = 0
+1 = 1
-1 = 2
+2 = 3
-2 = 4
+3 = 5
-3 = 6
etc,etc

As you can see there's a 1:1 mapping between the two sets. When sets can be mapped like this, mathematics say they have the same cardinality.

Rudy Rucker's "white light" novel discusses some of these ideas. it's related to "cantor's theorem" (metioned in White Light). There's a little more to it, especially with definitions of exactly what infinity is, but this is a basic primer.

[RedKing]

Taoism essentially views the universe this way -- as a grand self-regulating, procedural system (the Tao) which produces all natural phenomena in the universe. Without a consciousness and without an "overseer". So in terms of the analogy, the program is running on autopilot. And for most interpretations, wasn't programmed. It's either just always been there, or arose spontaneously out of random noise.

[Another]

Re:Simulating  the Universe with a finite array of discreet math objects.

I see a principal problem here:

While the volume of theoretically observable part of the Universe is finite, total number of particles in it is finite and the amount of thermodynamical entropy is also finite - there is absolutely no way to describe and store somewhere the state of the Universe with arbitrary high precision. You can't even do it for a single particle due to Heisenberg's uncertainty in describing its coordinates along with its impulse. That is not just that you can't simultaneously measure them - they as far as we know can't simultaneously exist in precisely defined states. The Universe can not be knowable with arbitrary high precision.

If you want to do just "good enough" simulation of the Universe and not arbitrary close to perfect - you would need to fill bits of information you do not have (and in some cases can not have in principle) with some kind of very good random number generator. At quantum level said random generator would need to account for lots of paradoxically sounding correlations so it would need to either itself be a quantum mechanical device the size of a Universe or be simulated like one. Once we are talking about infinities in computational power and storage - infinitely nested finite simulations may be not be such a big deal but that still sounds complicated.


Now, the concept of being a Boltzmann's Brain is scary if you think too much about the probabilities involved.

[Eagleon]

If you want to do just "good enough" simulation of the Universe and not arbitrary close to perfect - you would need to fill bits of information you do not have (and in some cases can not have in principle) with some kind of very good random number generator. At quantum level said random generator would need to account for lots of paradoxically sounding correlations so it would need to either itself be a quantum mechanical device the size of a Universe or be simulated like one. Once we are talking about infinities in computational power and storage - infinitely nested finite simulations may be not be such a big deal but that still sounds complicated.

I disagree that your generator would need to be particularly large, just carefully suited to the task. Actually, I've always thought uncertainty looks a hell of a lot like a compression algorithm - get rid of data you don't need, fill it in with data that makes sense when you do need it, and correct for errors on various levels as they arise. There could even be error correction we never see, such as resets to previous states. If time/processing capability is less of an issue than memory and if you could find small parts of the simulation to do this with, it would help tremendously. Since we have no idea what's simulating us or what rules they have to follow we can't really say, but it's what I'd do

[Another]

If you want to do just "good enough" simulation of the Universe and not arbitrary close to perfect - you would need to fill bits of information you do not have (and in some cases can not have in principle) with some kind of very good random number generator. At quantum level said random generator would need to account for lots of paradoxically sounding correlations so it would need to either itself be a quantum mechanical device the size of a Universe or be simulated like one. Once we are talking about infinities in computational power and storage - infinitely nested finite simulations may be not be such a big deal but that still sounds complicated.

I disagree that your generator would need to be particularly large, just carefully suited to the task. Actually, I've always thought uncertainty looks a hell of a lot like a compression algorithm - get rid of data you don't need, fill it in with data that makes sense when you do need it, and correct for errors on various levels as they arise. There could even be error correction we never see, such as resets to previous states. If time/processing capability is less of an issue than memory and if you could find small parts of the simulation to do this with, it would help tremendously. Since we have no idea what's simulating us or what rules they have to follow we can't really say, but it's what I'd do

You are likely right about possibility to pull this off with a relatively simple generator and some additional techniques like error correction.

I was mostly worried that quantum entangled states are not exactly random or semi-random or reducible to normal random numbers without completely breaking some basic properties of QM. In a certain sense everything in the observable Universe is quantum entangled with everything else though the level of entanglement is usually really negligible small. Things like [natural] superconductivity and present day bank-to-bank quantum cryptography channels could probably be accounted for by additional specialized modules.

[Ampersand]

Larger than infinite is not a problem......

I suppose my point was really that it would be far easier to just to have an actual universe that it could every be to simulate one.

[alway]

You can never simulate any energy larger than the energy used to simulate it, no matter what sort of 'quantum magicks' you invoke, and here's why:

In order to actually simulate it, you can not throw out any information. After all, if you were to throw out the information, you wouldn't actually be simulating it. Reducing the complexity and calling it 'close enough' won't cut it; it's merely trying to disguise the fact that you have reduced the information present in the simulation. It's still a simulation, but no longer is it a simulation of your universe. For example, if there was an ideal* computer within the simulation which was twice as powerful as your current ideal* computer, even your 'close enough' protocols wouldn't work simply due to the fact that to simulate something more powerful than that which the simulation is run on would require an introduction of information for which the computer on which it runs would be unable to provide in real time.

Thus the best we can do is create simulated universes which contain less information than ours. And within those simulations, even lower order universes could be created**. Note that when I speak of information in this case I mean information over the entire time and space of the universe; thus a larger universe than our own could be accurately simulated, but even using our entire universe as a computer it would run more slowly due to the limits of computation.

On the other side of things, our universe could be a simulated universe nested within 1 or even a near infinite number of universes. They may all be finite like ours, and merely increasing in total information content, or at some point could switch over to universes with entirely foreign laws of physics in which infinite computation is possible. However, unless such a thing could be detected (and it probably can't), such speculation is of little use.


*In this case, ideal computer means a computer which is somehow able to make use of every possible bit of computational ability inherent in its structure; this is impossible, due to the way computers work, but even in this overly optimistic estimate, my point still stands.

**An example of this would be Dwarf Fortress, which is itself a simulated universe of a sort, in which the construction of computers is possible (as demonstrated by the programmable 8-bit computer someone made). These computers, if constructed large enough, could theoretically run simulations themselves, although at a much slower rate than the computer in our universe would run them.

[Another]

Just to get one line of this discussion to its logical limit - we are in a simulation. The so-called "computer" performing the simulation of the Universe in real-time is the Universe itself. It is by the way the most efficient and accurate "computer" possible. To take it farther - it itself is a simulation performed on a giant computation machine highly specialized to the task and utilizing every trick possible which is again - our Universe. We are definitely inside a never ending chain of simulations.

[Eagleon]

You can never simulate any energy larger than the energy used to simulate it, no matter what sort of 'quantum magicks' you invoke, and here's why:

This is assuming the universe that's simulating ours behaves the same way and follows the same rules regarding conservation. It's all assumption.

I guess my initial reason for putting this up is that there are four contradictory layers here as I see it - Steve's universe, Steve's passing understanding of the process in its execution, the process itself, and the result of the process recorded in the word processor. I guess my question is, where and what would you be, as an element in the simulation, if you were in an experiment like this?

The last makes the universe a story, however grand - a record to be glanced through, even changed. You could be blotted out, censored, and the final page wouldn't be any different. If the process is what's important, where "you" reside, it means that everything, you included, is part of the same function. The Tao, maybe. I think that's closest to what I believe, but it's still a little bit unsatisfying in that it completely breaks the assumption that we 'exist'. And if it's Steve, well, what happens when he goes out to lunch and stops thinking about it all? Is it the same universe when he comes back? It's a fairly meaningless distinction, since you could interpret yourself to be in whatever position you'd like, but that in itself is interesting. How often do you have the choice?

[Ampersand]

I guess my question is, where and what would you be, as an element in the simulation, if you were in an experiment like this?

Because these things are impossible for us to know with any degree of certainty, we are forced to behave on the assumption that our experiences are real. I do not know for example that I am not a brain in a vat, but the answer is irrelevant; I am forced to confront the reality that I experience on the assumption that it is real and regardless of whether my external reality is real, I AM certain that I am real, and my personal wellbeing depends upon my interaction with the external reality that I experience.