If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(io9)   Physicists say there may be a way to prove that there is no spoon   (io9.com) divider line 170
    More: Interesting, simulations, physicists, strong forces, physical changes, lattices, spoons, cosmic microwave background, quantum chromodynamics  
•       •       •

6038 clicks; posted to Geek » on 10 Oct 2012 at 9:46 PM (1 year ago)   |  Favorite    |   share:  Share on Twitter share via Email Share on Facebook   more»



170 Comments   (+0 »)
   
View Voting Results: Smartest and Funniest

Archived thread

First | « | 1 | 2 | 3 | 4 | » | Last | Show all
 
2012-10-11 02:09:46 AM

Uchiha_Cycliste: what do the voices say I should do?


farm4.static.flickr.com
 
2012-10-11 02:14:28 AM
Yup, I trust the voices.
nite
 
2012-10-11 04:09:05 AM
If it is all a simulation then it's a damn boring one. The way the laws of physics are set out there is no room for magic or the fantastical. Who would run a simulation of us unless they were trying to decide bothering to brew us up is worth it. If we were just for entertainment you would wonder why all possibilities of the imagination are not explored.

Unless those behind the simulations are so driven by cold rational logic they invented us to do all the dreaming for them....
 
2012-10-11 04:33:42 AM

NetOwl: What it really means is that the universe can be modeled by a manifold isomorphic to R^4 x X, where X is a Calabi-Yau manifold (say, of complex dimension 3). (R^4 is the Minkowski space.) A manifold is, briefly, a second countable hausdorff space that is locally homeomorphic to a subset of Euclidean space (i.e., there is a continuous map with a continuous inverse from each open set in M to an open set in R^n). A smooth manifold has a smooth structure (so we can do calculus on the manifold), meaning we have a an open covering {U} and maps from each set in the covering to R^n such that on the intersection of U and V (both elements of the cover), the homeomorphism to V can be composed with the inverse of the homeomorphism to U,and the result is a smooth (infinitely differentiable) map from R^n to R^n. Replace R with C and "smooth" with "holomorphic" and you have a complex manifold, which is what we use in physics. (Actually, if you're paying attention, Minkowski space is R^4, and X is complex. Thus, we're really uing the product of a real manifold with a complex manifold.)

Given some simply connected manifold of dimension n and a Riemannian metric g with g irreducible and non-symmetric, we have one of seven cases holding, depending on the holonomy group of g. In the case where this is SU(n), we have a Calabi-Yau manifold (thus, in the real world, this is SU(3)).

The holonomy group is sort of intuitive. Given a rank k vector bundle E over M and a connection g on E, then given some piecewise smooth loop L based at a point x, the connection defines a parallel transport map T from E_x to E_x, which is linear and invertible and thus an element of GL(E_x). The holonomy group is just all such T_L, where L is any piecewise smooth loop at x.

In other words, the manifolds we use to model reality depend on the way loops behave!

That's a summary of the very basic elements of the math you need to understand this stuff.


I didn't know they had mad libs for advanced mathematics.
 
2012-10-11 04:45:09 AM

NetOwl: What it really means is that the universe can be modeled by a manifold isomorphic to R^4 x X, where X is a Calabi-Yau manifold (say, of complex dimension 3). (R^4 is the Minkowski space.) A manifold is, briefly, a second countable hausdorff space that is locally homeomorphic to a subset of Euclidean space (i.e., there is a continuous map with a continuous inverse from each open set in M to an open set in R^n). A smooth manifold has a smooth structure (so we can do calculus on the manifold), meaning we have a an open covering {U} and maps from each set in the covering to R^n such that on the intersection of U and V (both elements of the cover), the homeomorphism to V can be composed with the inverse of the homeomorphism to U,and the result is a smooth (infinitely differentiable) map from R^n to R^n. Replace R with C and "smooth" with "holomorphic" and you have a complex manifold, which is what we use in physics. (Actually, if you're paying attention, Minkowski space is R^4, and X is complex. Thus, we're really uing the product of a real manifold with a complex manifold.)

Given some simply connected manifold of dimension n and a Riemannian metric g with g irreducible and non-symmetric, we have one of seven cases holding, depending on the holonomy group of g. In the case where this is SU(n), we have a Calabi-Yau manifold (thus, in the real world, this is SU(3)).

The holonomy group is sort of intuitive. Given a rank k vector bundle E over M and a connection g on E, then given some piecewise smooth loop L based at a point x, the connection defines a parallel transport map T from E_x to E_x, which is linear and invertible and thus an element of GL(E_x). The holonomy group is just all such T_L, where L is any piecewise smooth loop at x.

In other words, the manifolds we use to model reality depend on the way loops behave!

That's a summary of the very basic elements of the math you need to understand this stuff.


unrealitymag.com
 
2012-10-11 05:32:31 AM

NowhereMon: If it's not a simulation why are there arbitrary limits on things like maximum speed and minimum temperature?


Minimum temperature = zero heat. This is the "zero" in "absolute zero".

The speed limit thing is a bit harder to explain, but if you look at the curve of how much energy it takes to move a mass at speed x, it hits infinity at light speed, and impracticality shortly before that.

There *may* be ways around the speed limit via string theory and the like. There most definitely is not a way around the temperature limit, because you can't have less than zero heat - that quite simply doesn't make logical sense.
 
2012-10-11 06:27:01 AM
FTA "According to Silas Beane and his team at the University of Bonn in Germany, a simulation of the universe should still have constraints, no matter how powerful. These limitations, they argue, would be observed by the people within the simulation as a kind of constraint on physical processes."

Like a limit to speed. Nobody is really happy that the universe has a speed limit. Very convenient.
 
2012-10-11 06:33:24 AM
So, the rapid expansion if the universe was just a server upgrade?
 
2012-10-11 06:49:19 AM
an endless loop of simulations creating simulations. Good luck with that.
 
2012-10-11 07:27:12 AM

The All-Powerful Atheismo: The upshot of string theory is that it directly predicts, as a result of the theory, all of the masses, charges, and other force constants of all the particles known in the universe, and also includes gravity as an essential part (something the Standard Model doesn't do at all). These values are not constants that are shoehorned into the theory like in QED.


No it doesn't, and no it doesn't.

First, the original hope for String Theory was that there would be precisely one mathematically consistent formulation and thus the constants would fall out. The latest theory proposes around 10^500 possible universes, one of which happens to have the constants we see. In some versions of the theory, all of those universes actually exist, and we happen to be in ours. String Theory has gone from being a "theory of everything" to a "theory of anything". In brief, String Theory could hardly be further from "predicting" the constants of of the universe.

Second, the claim that String Theory contains (or even requires) gravity has been bandied about since its earliest days, but is also completely false. One of the things that got pioneers excited was that one of the permitted modes for the strings had the same quantum numbers (spin, charge, etc.) that a graviton would have if gravitons exist. That's it. It's not even proof that this string mode *is* a graviton, let alone that the theory describes gravity. Somehow, in the popular retelling, "gravitons are permitted" got inflated into "String Theory predicts gravity!". In fact, almost everything currently known about String Theory is from approximations built in a fixed spacetime background that is not curved by the gravity of the particles it contains, i.e. it is not "background independent". The absolute essence of any theory of gravity is its description of how spacetime evolves, and we have the barest glimmer of what a background-independent String Theory would look like. It would actually be closer to the truth to say that absence of gravity is an essential flaw of String Theory as currently formulated.

In fairness, anybody could be forgiven for not knowing this, since both claims are frequently made in the media and in popular accounts of String Theory, and often given a free pass by science journalists. But they simply are not true.
 
2012-10-11 07:29:22 AM

Baron Harkonnen: What kind of computer could simulate an entire universe down to the subatomic level? It's ridiculous to think about.


Programing silly play a video games only make things visible/render them when they are being looked at it saves resources
 
2012-10-11 07:48:05 AM

Baron Harkonnen: What kind of computer could simulate an entire universe down to the subatomic level? It's ridiculous to think about.


See here

http://en.wikipedia.org/wiki/Scheduling_%28computing%29
 
2012-10-11 07:54:07 AM

dready zim: FTA "According to Silas Beane and his team at the University of Bonn in Germany, a simulation of the universe should still have constraints, no matter how powerful. These limitations, they argue, would be observed by the people within the simulation as a kind of constraint on physical processes."

Like a limit to speed. Nobody is really happy that the universe has a speed limit. Very convenient.


No, not like that at all. "Constraints" is really a poor choice of word here -- "artifacts" might be better, like what happens sometimes when you digitally process an image.

Here's an overly-simplistic analogy that might help: You're playing a video game that simulates a world and you notice that sometimes, straight lines don't look perfectly straight. On closer analysis you discover that lines are straight when they go vertically or horizontally, but when they are diagonal you can see a little jaggedness in them. As a result if you, if you draw a right-angled triangle and traverse the hypotenuse very carefully, following along each jagged step, you find it is slightly longer than you would expect from Pythagoras' Theorem. Conversely, if you rotate the triangle so that the hypotenuse is vertical, now you find it is slightly shorter than expected. The larger you draw the triangle, the easier it is to measure the discrepancy. The video game universe appears to have a preferred direction. You hypothesize that this is because the universe is actually quantized (pixelated) on a grid (lattice), and distances depend very slightly on orientation to the grid.

This work is sort of like that. The hypothesis is that if we were to simulate a universe, we would probably do so by creating a lattice to represent the universe and performing calculations at each point on the lattice. If we ourselves are inside such a simulation, it might be possible to detect the existence of the lattice because certain physical processes would proceed very slightly differently depending on their orientation to the lattice. For example, if you could draw a really big triangle in space, photons traveling along the diagonal would have to follow the jagged path, and would take slightly longer to reach us than expected. You would be forced to conclude that either the speed of light varies very slightly in certain directions, or that distances do. The latter conclusion would be the signature clue that our universe is computed on a lattice.

If we are in fact on a lattice it has to be extremely fine-grained, because otherwise we would have noticed already. So in order to see its effects you need really big, cosmically big distances -- and that's what this work is about. Of course, in reality detection is much harder because you have to find two processes that originate from the same spot and are affected differently by the lattice, but this is the essence of the idea.
 
2012-10-11 07:58:43 AM
Each quantum interaction in the universe can be viewed in terms of a computable operation. In fact, the difference between a quantum computer and the system its modeling is academic (the simplest way to simulate the behavior of a pair of electrons is to go out and get a pair of electrons). Seth Lloyd demonstrated that Turing's theories can be extended to quantum computers; that is to say a quantum computer is a super-set of Turing's universal computer.

This means a number of things:
1) All quantum computers are quantum Turing machines
2) All quantum computers can emulate any other quantum computer in linear time
3) Anything that can be computed can be computed by a quantum computer
4) All physical interactions are quantum computations

So here's the real problem: any simulation of a physical system in a quantum computer would be nearly identical to the actual physical system. The only difference is that the computation time would be scaled linearly. So if in the "real" universe, it takes 1s for some event to occur, in our simulation, it might take 2s of real-world time- but within the context of that simulation, all of the clocks run twice as slow as in the "real" world, so they would still measure only 1s has passed.

Last time I checked, it is impossible to write a program that can tell if the executing Turing machine is a "real" Turing machine or a simulation of a Turing machine. So that means that these physicists have to make assumptions about what kind of universe is executing the simulation, and what kinds of short-cuts the designers in that universe might have made when building our simulation.
 
2012-10-11 08:05:58 AM

czetie: The hypothesis is that if we were to simulate a universe, we would probably do so by creating a lattice to represent the universe and performing calculations at each point on the lattice


Which isn't true. First off, that's a terrible way to implement physical models. Second, since they can resource-manage the simulation, it'd be challenging-but-practical to make the simulation to locally increase the resolution based on the interactions required. For example, when you start probing at the Planck length, it could start generating sub-planck scales. In order to keep the simulation in sync, it can just reduce the clock speed for the rest of the universe while it does these more cumbersome calculations. But the rest of the time, it can just run at scales larger than Planck lengths.

The strongest argument for a simulated universe is quantum indeterminacy, if you ask me. Instead of tracking the state of particles, you derive a new state when the interaction probability crosses a certain threshold. That'd be a huuuuuge memory savings.
 
2012-10-11 08:17:59 AM
This far into the thread and nobody's posted A Bunch of Rocks? Fark, I am disappoint.

imgs.xkcd.com

/Hot like a very hot thing.
 
2012-10-11 08:21:58 AM

WyldKarde: /Hot like a very hot thing.


The only thing hot about xkcd is the way it burns yer brain if you let it touch too long.
 
2012-10-11 08:42:01 AM

Uchiha_Cycliste: The All-Powerful Atheismo - you're cool

WaitWhatWhy: - you're cool

NetOwl -fark you

Just Another OC Homeless Guy - you;re cool.

I'm out.


What a tool you are, man.
 
2012-10-11 08:46:57 AM

t3knomanser: Each quantum interaction in the universe can be viewed in terms of a computable operation. In fact, the difference between a quantum computer and the system its modeling is academic (the simplest way to simulate the behavior of a pair of electrons is to go out and get a pair of electrons). Seth Lloyd demonstrated that Turing's theories can be extended to quantum computers; that is to say a quantum computer is a super-set of Turing's universal computer.

This means a number of things:
1) All quantum computers are quantum Turing machines
2) All quantum computers can emulate any other quantum computer in linear time
3) Anything that can be computed can be computed by a quantum computer
4) All physical interactions are quantum computations

So here's the real problem: any simulation of a physical system in a quantum computer would be nearly identical to the actual physical system. The only difference is that the computation time would be scaled linearly. So if in the "real" universe, it takes 1s for some event to occur, in our simulation, it might take 2s of real-world time- but within the context of that simulation, all of the clocks run twice as slow as in the "real" world, so they would still measure only 1s has passed.

Last time I checked, it is impossible to write a program that can tell if the executing Turing machine is a "real" Turing machine or a simulation of a Turing machine. So that means that these physicists have to make assumptions about what kind of universe is executing the simulation, and what kinds of short-cuts the designers in that universe might have made when building our simulation.


I've underlined the part where the consequent is presumed. Basically, for whatever reason I don't know, people get sloppy with reality and maps of reality. The Turing Machine is really handy for representing computable series of events in reality. But mistaking the way of thinking about reality for reality, regardless of how well it seems to map, is to turn Plato's cave into Polyphemus' cave.

Basically because each quantum interaction in the universe can be viewed in terms of a computable operation does not mean that all physical interactions are quantum computations, because imagining that the universe operates according to the parameters of QED (I love that acronym...) should not imply that quantum mechanical interactions are what is actually going on in the universe.

Yes, there's something to be said for suspending disbelief for the sake of promising research, but it's that kind of sloppiness that encourages mysterians, morons, and the irrational to pervert science to their own ends.
 
2012-10-11 09:06:55 AM

Nurglitch: Basically because each quantum interaction in the universe can be viewed in terms of a computable operation does not mean that all physical interactions are quantum computations


Every quantum interaction is a computation that yields the next state of the system as its output. While quantum physics is incomplete (we don't have a model of gravity that maps to this, for example), the key point that physics is computable remains.
 
2012-10-11 09:18:27 AM

t3knomanser: Nurglitch: Basically because each quantum interaction in the universe can be viewed in terms of a computable operation does not mean that all physical interactions are quantum computations

Every quantum interaction is a computation that yields the next state of the system as its output. While quantum physics is incomplete (we don't have a model of gravity that maps to this, for example), the key point that physics is computable remains.


It would be weird if physics were not computable; unscientific, as it were. I'm talking about what physics is supposed to describe, and people confusing physics with reality. Yes, modern physics seems like a highly accurate set of theories, and having one highly accurate theory to replace them all would be peachy, but there's a serious methodological issue mistaking those theories for what they represent.
 
2012-10-11 09:21:57 AM

t3knomanser: czetie: The hypothesis is that if we were to simulate a universe, we would probably do so by creating a lattice to represent the universe and performing calculations at each point on the lattice

Which isn't true. First off, that's a terrible way to implement physical models. Second, since they can resource-manage the simulation, it'd be challenging-but-practical to make the simulation to locally increase the resolution based on the interactions required. For example, when you start probing at the Planck length, it could start generating sub-planck scales. In order to keep the simulation in sync, it can just reduce the clock speed for the rest of the universe while it does these more cumbersome calculations. But the rest of the time, it can just run at scales larger than Planck lengths.

The strongest argument for a simulated universe is quantum indeterminacy, if you ask me. Instead of tracking the state of particles, you derive a new state when the interaction probability crosses a certain threshold. That'd be a huuuuuge memory savings.


Oh, I agree with you (Again. Sigh.) I wouldn't do it that way either. It's a reasonable way to do weather forecasting, but not to model a universe, especially one where almost nothing is happening almost everywhere. I'm just saying, that's the premise of the article.

If you ask me the strongest argument for a simulated universe is the difficulty of reconciling quantum mechanics and gravity. Late at night I sometimes wonder whether the laws of physics in our universe were not actually implemented from the start but were added to the simulation as we discovered the bugs in them: for example, the simulation started quantizing radiation only once we noticed that the classical theory blows up when you apply it to a black body. And then I get to wondering whether GR and QM were added to the simulation separately, as approximations for very large scale and very small scale behaviors, without anybody realizing that they couldn't be reconciled. (Or perhaps they did realize, but assumed that they'd never have to simulate both at the same time.)

Even later at night I wonder whether the point of the simulation is to model people, or whether it's to model laws of physics to see how alternate universes might work, and we are just an unfortunate unanticipated side-effect.
 
2012-10-11 09:35:59 AM

chance4510: Let's hope the system we are running is not windows.


Don't be silly. The universe would have ended back in 1995 when it BSOD'ed.
 
2012-10-11 09:36:39 AM
Additional dimensions are easily provable...it all depends on how you define them.
Find the right defintion.

Remember, you have to do it for each...and they have to inter-relate.

Einstein already showed how to do it for 3 to 1.

Also, it has to work at all scales...for HERE and NOW.
No, shrinking away, no moving faster than light, no it was here in the past and so on...

ready, set ...go.

/oh BTW...just because you "may" find a type of lattice...doesn't mean it's a computer similation or a hologram...CDNIC
 
2012-10-11 09:42:14 AM
Good going, io9 guys. The devel team for Universe (the largest MMO in the, well, universe) have read this article and are working on a patch that will keep us from being able to test for this bug. Thanks a lot.
 
2012-10-11 09:44:08 AM

Mister Peejay: NowhereMon: If it's not a simulation why are there arbitrary limits on things like maximum speed and minimum temperature?

Maximum speed is one of those things that hurts my head to think about for too long (the maximum speed is light speed, as seen from ANY frame of reference - light always travels the same speed relative to you, no matter how fast you are going, thus time dilation happens...) but minimum temperature is easy.

Absolute zero is when atomic motion ceases, essentially. You can't get colder than zero motion. There's no such thing as negative motion, so there's nothing colder than absolute zero. That's why it's absolute.


Maximum speed is simply the speed that cause/effect propagates. If you can go faster than that you can go or at least send a message back in time, thus cause an effect before the cause.
 
2012-10-11 09:48:15 AM

czetie: Second, the claim that String Theory contains (or even requires) gravity has been bandied about since its earliest days, but is also completely false.


You're forgetting duality; compare Type IIB superstrings on AdS5 × S5 and N = 4 supersymmetric Yang-Mills theory on the four-dimensional boundary of the Anti de Sitter space (either a flat four-dimensional spacetime R3,1 or a three-sphere with time S3 × R). The claim that a theory must require (or even contain) gravity to assert the existence of gravity is dated.
 
2012-10-11 09:49:14 AM

Nurglitch: I'm talking about what physics is supposed to describe, and people confusing physics with reality.


Ah, you're misunderstanding me. I'm not talking about our models of physics. I'm talking about physics itself. When a ball rolls down a hill, you've just built a machine that performs a computation. It has an initial state, it performs a series of operations, and it has a final state. The system is a model of it self, a computer that changes that model according to processes which can be modeled mathematically.

czetie: Even later at night I wonder whether the point of the simulation is to model people, or whether it's to model laws of physics to see how alternate universes might work, and we are just an unfortunate unanticipated side-effect.


If the purpose of the simulation is to model people, it's terribly designed for that goal.
 
2012-10-11 10:22:00 AM

t3knomanser: Nurglitch: I'm talking about what physics is supposed to describe, and people confusing physics with reality.

Ah, you're misunderstanding me. I'm not talking about our models of physics. I'm talking about physics itself. When a ball rolls down a hill, you've just built a machine that performs a computation. It has an initial state, it performs a series of operations, and it has a final state. The system is a model of it self, a computer that changes that model according to processes which can be modeled mathematically.

czetie: Even later at night I wonder whether the point of the simulation is to model people, or whether it's to model laws of physics to see how alternate universes might work, and we are just an unfortunate unanticipated side-effect.

If the purpose of the simulation is to model people, it's terribly designed for that goal.


No, physics is the model. You're confusing the model-mediated reality with reality. When you roll a ball down a hill, something is happening and you frame it as a ball rolling down a hill. You can also frame it as a computing machine. Regardless of whether your model is the result of folklore or a scientific theory, you're still using the model.

I think the issue that you're missing isn't the theory-ladeness of our perception of reality, as you seem to understand how our perception of reality is built out of an equilibrium found between model and reality, but how making a system a model of itself squeezes reality out of the loop. Yes, of course a system is a model of itself; many things are often identical with themselves. The issue is making models of models, and then expecting the results to be pertinent to models of reality.
 
2012-10-11 10:27:28 AM

Nurglitch: When you roll a ball down a hill, something is happening and you frame it as a ball rolling down a hill


For my purpose here, the sense of computation I'm using is any process which terminates in finite time and could be implemented in a Turing machine. Ergo, a ball rolling down a hill is a computer which calculates the process of rolling a ball down a hill.

Let's say I build a model of this system. I build a very detailed model of the ball, so detailed in fact that I replicate the exact quantum states of every particle in the ball in perfect fidelity. I do the same with the hill. And the air. And the gravitational, electromagnetic, and whatever other fields in the region. What is the difference between this model and the reality, aside from the fact that I happen to know one is a model?
 
2012-10-11 10:34:27 AM

t3knomanser: Nurglitch: When you roll a ball down a hill, something is happening and you frame it as a ball rolling down a hill

For my purpose here, the sense of computation I'm using is any process which terminates in finite time and could be implemented in a Turing machine. Ergo, a ball rolling down a hill is a computer which calculates the process of rolling a ball down a hill.

Let's say I build a model of this system. I build a very detailed model of the ball, so detailed in fact that I replicate the exact quantum states of every particle in the ball in perfect fidelity. I do the same with the hill. And the air. And the gravitational, electromagnetic, and whatever other fields in the region. What is the difference between this model and the reality, aside from the fact that I happen to know one is a model?


The difference is that one is a picture of reality, and the other is reality. The picture will be incomplete, as it will only contain the elements you want to (or are able to) include in the model. I mean, I could draw a cartoon of a ball rolling down a hill and doodle a dinosaur in the margin. Doesn't mean that finding dinosaurs implies that the universe is a drawing.
 
2012-10-11 10:40:00 AM

Nurglitch: The difference is that one is a picture of reality, and the other is reality.


But there is not one bit of information that can be used to tell which is which.

Nurglitch: The picture will be incomplete


The premise of my thought experiment was that is not the case- if I replicate the quantum states of every particle involved, the picture could not possibly be incomplete. My model contains exactly the same exact qubits of information as the original system, in the exact same states as the original system. It is, in fact, indistinguishable from the original system (in fact, it is the original system, since it would have to be co-located within the original system).

If I physically roll a ball down a hill, I have executed a computation that determines the final state of that particular ball-rolling system with that particular set of initial conditions. That computation includes every qubit of information in that system, since the system is the thing doing tho computing.

If I draw a dinosaur, I have a drawing of a dinosaur. If I build cells out of atoms, tissues out of cells, and a dinosaur out of those tissues- I have a dinosaur.
 
2012-10-11 11:00:58 AM
t3knomanser:

Your thought-experiement fails because there will be plenty of information available to tell the picture from whatever it depicts, it's just not accessible to the putative observer who can't tell the difference between that model and reality.

So while we may be able to draw a very accurate picture, photographic even, but the only way we're going to make the picture indistinguishable from the reality for all perspectives is by duplicating that reality. Hence there is no such thing as a perfect simulation; you're always going to be constrained by your resources, and particularly by the accuracy of your model. You're going to build the simulation to be good enough for the observer, and the subject at hand. That's why empirical testing is important, because it is necessary to find the gaps that the picture fails to represent, mis-represents, or uses some artifice to hand-wave away. Access to new perspectives tends to resolve gaps, just as access to new techniques in representation can be used to paper over them. As an example of the former, there's the difference in perspective from the Einsteinian cosmology and the Newtonian cosmology, and as examples of the latter include forced perspective in renaissance art, and Photoshop.

The whole: "It's indistinguishable!" argument is bullshiat, and your thought-experiment highlights why such an argument is bullshiat, because either you're duplicating something, or you're relying on some limitation on the part of the observer. Computer simulations rely on limitations on the part of the observer to make it seem like they map onto reality, like all pictures. You're not going to have any kind of philosophical zombie problem, except for the kind that happens when physicists get taken in by spurious metaphysical arguments about epistemic limits.
 
2012-10-11 11:08:54 AM

Nurglitch: Hence there is no such thing as a perfect simulation; you're always going to be constrained by your resources, and particularly by the accuracy of your model.


The problem is that you're making this assumption and he isn't. You're never going to agree as long as you're working off such a vastly different set of assumptions.
 
2012-10-11 11:14:18 AM

sprawl15: Nurglitch: Hence there is no such thing as a perfect simulation; you're always going to be constrained by your resources, and particularly by the accuracy of your model.

The problem is that you're making this assumption and he isn't. You're never going to agree as long as you're working off such a vastly different set of assumptions.


Like the assumption that there's supposed to be agreement? What's the point of agreeing with someone when they start off from the wrong assumptions? I'm pointing out his assumptions are wrong, and precisely the kind of half-baked bullshiat that results in stuff like philosophical zombies. And if I'm wrong, then he'd be wrong to agree with me. So I disagree with your assumptions too.
 
2012-10-11 11:16:54 AM

Nurglitch: Like the assumption that there's supposed to be agreement?


More the assumption that there's a farking point to your babbling. But by all means, if you want to slam your dick in a door for the next hour, don't let me stop you.
 
2012-10-11 11:19:13 AM

sprawl15: Nurglitch: Like the assumption that there's supposed to be agreement?

More the assumption that there's a farking point to your babbling. But by all means, if you want to slam your dick in a door for the next hour, don't let me stop you.


Yeah, I think you're stupid too. We should hang out some time.
 
db2
2012-10-11 11:56:20 AM
This is what happens when physicists smoke weed and watch Tron.
 
2012-10-11 12:10:30 PM

Nurglitch: Hence there is no such thing as a perfect simulation


I'm really starting to think we're not talking the same language, so I'm going to take a stab at saying this in a way I think you'll understand.

There is always exactly one perfect simulation of any system: the system itself. If I roll a ball down an incline, I have built a perfect simulation of rolling a ball down an incline. Would you agree? Would you agree that a ball rolling down an incline exactly emulates a ball rolling down an incline? Not the entire class of possible balls and possible inclines- I mean that exact ball and that exact incline at that exact moment with that exact conditions?

All I'm saying is that when I roll a ball down an incline, that system is a system that computes the behavior of a ball rolling down an incline.
 
2012-10-11 12:31:43 PM

SearchN: [2.bp.blogspot.com image 765x941]


both are fish oil
 
2012-10-11 12:35:28 PM
I'm just going to leave this here...

http://www.youtube.com/watch?v=JkxieS-6WuA
 
2012-10-11 12:47:31 PM

Wolf892: If it is all a simulation then it's a damn boring one. The way the laws of physics are set out there is no room for magic or the fantastical. Who would run a simulation of us unless they were trying to decide bothering to brew us up is worth it. If we were just for entertainment you would wonder why all possibilities of the imagination are not explored.

Unless those behind the simulations are so driven by cold rational logic they invented us to do all the dreaming for them....


We are entertainment. Play the Sims sometime... it can be fun!
 
2012-10-11 01:03:10 PM

czetie: Here's an overly-simplistic analogy that might help: You're playing a video game that simulates a world and you notice that sometimes, straight lines don't look perfectly straight. On closer an....

.....If we are in fact on a lattice it has to be extremely fine-grained, because otherwise we would have noticed already. So in order to ...


Bravo. This is exactly why i already have you fark Favorited as "favorite: (Quantum Physics Genius)"
I'd like to subscribe to your newsletter.
 
2012-10-11 01:12:54 PM

Uchiha_Cycliste: what


There is actually a great 10 minute youtube video that explains the different dimensions and how to conceptualize them without math here:
Link
 
2012-10-11 01:18:39 PM

t3knomanser: But there is not one bit of information that can be used to tell which is which.


if only we had an architect or something to give us a hint

no worries, Science will provide all the answers we need
 
2012-10-11 01:18:42 PM

t3knomanser: Nurglitch: Hence there is no such thing as a perfect simulation

I'm really starting to think we're not talking the same language, so I'm going to take a stab at saying this in a way I think you'll understand.

There is always exactly one perfect simulation of any system: the system itself. If I roll a ball down an incline, I have built a perfect simulation of rolling a ball down an incline. Would you agree? Would you agree that a ball rolling down an incline exactly emulates a ball rolling down an incline? Not the entire class of possible balls and possible inclines- I mean that exact ball and that exact incline at that exact moment with that exact conditions?

All I'm saying is that when I roll a ball down an incline, that system is a system that computes the behavior of a ball rolling down an incline.


Yeah, that's the point of contention. If you're going to suppose that a duplicate of anything is its perfect simulation, then you're going to run into issues about confusing content, depiction, and observers. If you're going to roll a ball down an incline, then you're going to frame some event in the world in the specific terms of some mode of depiction. Your system of balls and inclines, with its motions of rolling, and possible forces of gravity and friction, are all a way of framing that event.

Basically the issue is with an over-extension of the notion of simulation to cover self-identity. We can suppose that things exactly simulate themselves, and hence are the same as we imagine them to be, or we can suppose that simulation is necessarily an exercise in approximation. In these terms the notions are equal, and indeed the notion of perfect simulation seems like a logical extension of the notion that simulations should be the same as what they simulate for some set of conditions.

However, what seems to get left out of these discussions is the value of "same" that we're using here. A simulation is one thing appearing to be another, and appearances should appear the same as themselves. When you start saying that a depiction, an appearance, is the same as what it depicts, suddenly you're drawing an unwarranted equivalence between a depiction and what it depicts.

So no, I would not agree that that ball rolling down that incline exactly emulates itself. It just is. Another ball rolling down that incline, or even the same ball rolling down that incline again might be said to exactly emulate that ball, incline, and motion, should it roll down that incline in exactly the same way, or closely enough that we couldn't tell the difference.

Which, I think, is where this stupid "Well, maybe we're in the Matrix!" or "Maybe I'm just a brain in a vat!" bullshiat comes from, because people have treated the distinctions between perception and actuality in a sloppy way, mixing them up, and then wasting money on the latest version of phlogiston research.

The thing that really interests me about this research is that, to some degree, it appears that the researchers understand this notion that simulations are necessarily prevented from being what they simulate, and that you can test for those limits. What they get wrong is supposing that they can test the simulation while within the bounds of that simulation. It's boot-strappy to imagine that because we can create simulations and test them for insufficiently representing their subject, that we can test our universe to see if it insufficiently represents their subject.

The funny thing is that it's almost exactly the same dog's breakfast as the Problem of Other Minds that philosophers have been successfully buying academic careers with, in that if you carefully ignore the details at issue, then you can continue to make up all sorts of pseudo-problems (zombies, other minds, consciousness, intentionality, etc) to justify grant money. Which isn't to ascribe some sort of misconduct to either the physicists or the philosophers, but just that it's annoying to see philosophy fail to get a handle on a problem for 2000 years, and then a branch of science try to dress up that problem in more fashionable rags. I preferred it when science took philosophical problems that had been solved and applied them to expand human knowledge, not replicating its most embarrassing excesses.

And in addition, fark Catharge. Seriously, fark those guys.
 
2012-10-11 01:29:50 PM

OceanVortex: Uchiha_Cycliste: what

There is actually a great 10 minute youtube video that explains the different dimensions and how to conceptualize them without math here:
Link


thanks, I'll watch it when I get home and see what I can gleam off of it.
 
2012-10-11 01:58:56 PM

Uchiha_Cycliste: OceanVortex: Uchiha_Cycliste: what

There is actually a great 10 minute youtube video that explains the different dimensions and how to conceptualize them without math here:
Link

thanks, I'll watch it when I get home and see what I can gleam off of it.


Correction, I'd recommend watching this link first instead. Link . This link is sort of the Physics 101 version which is easier to digest to start with. The previous link i included is by the same guy but is sort of his 201 version and I think overly complicated to jump in to start with.

I was like you, I had trouble conceptualizing what this all meant, but after watching the video it all makes a lot more sense because it just takes what we can already understand "the 3rd dimension" and slowly keeps adding on with realistic examples until you're at the 10th dimension.
 
2012-10-11 01:59:49 PM

t3knomanser: All I'm saying is that when I roll a ball down an incline, that system is a system that computes the behavior of a ball rolling down an incline.


Trying to follow along...

So if a system exists, then it necessarily has the property of being a model of itself. And whatever it does in the course of existing can be interpreted as a computation of that model. Fair enough.

For humans to discuss possible explanations of that computation, requires a meta-model. Since you're a human, the statement "all physical interactions are quantum computations" should be in reference to our human meta-models of the concepts "physical interaction" and "quantum computation", not the objective-models, assuming they exist.

Differentiating between objective models and meta-models, the statement "O-phyisical-interactions are O-quantum-computations", while maybe true, isn't a useful statement since the only models we can interact with, intellectually, are the "M-physical-interactions" and "M-quantum-computations".

To talk about O-objects directly, we need to throw in the caveat, "...insofar as the existence of an M-model implies the existence of a corresponding O-model, and insofar as our M-model faithfully represents it".

If you were intending your statement to be about O-objects then, absent that caveat, someone might think you were claiming to have a personal hotline to transcendental knowledge.
 
2012-10-11 02:05:36 PM

Nurglitch: If you're going to suppose that a duplicate of anything is its perfect simulation


Drop the duplication, I think it's confusing you. I'm saying a thing is a perfect simulation of the thing itself.

I am a perfect simulation of myself. A ball is a perfect simulation of that ball. When you roll that ball down an incline, you've built a perfect simulation of that ball rolling down an incline. I am saying that the act of being a ball is a computation of ballness.

Heh. Ballness.

It is not a simulation, and I'm going to pull us away from that word, because I think that's also confusing you. I have never really been talking about simulations, but that seemed to be the language you preferred.

All physical systems are computers. This isn't just something I made up- this is a natural consequence of Turing's Universal Computation. In the physical world, a system is a computer that derives the next state of that system. That is not the same thing as saying that it's a simulation. When the computer on my desk changes electrical potentials to open and close circuits to power its logic gates, it is not simulating anything- but it is computing.
 
Displayed 50 of 170 comments

First | « | 1 | 2 | 3 | 4 | » | Last | Show all

View Voting Results: Smartest and Funniest


This thread is archived, and closed to new comments.

Continue Farking
Submit a Link »






Report