|
BR, -Topi
3:51 AM, March 22, 2017
George Rush said...
Sabine, it's quite plausible the sim could predict everyone's actions. Indeed, the entire future (not just human actions) may be pre-determined. Isn't Block Universe the most popular model among physicists today? If spacetime is static, sim would just be running a pre-recorded script. IOW humans would have no free will. Why are you so sure that, on the contrary, they do? Unless you demonstrate that, it's just bla-bla but nothing any scientist can take seriously. I won't address other points now, since this one is rather important. BTW Of course a sim must follow the laws of nature as we understand them. They simply describe - to pretty good accuracy - what we observe, and that's exactly what the sim must reproduce. Thanks, George
6:58 AM, March 22, 2017
Sabine Hossenfelder said...
George,
Please read more carefully. That something is determined doesn't mean it can be predicted.
That you accuse me of being sure that humans have free will is another demonstration how quickly people jump to mistaken conclusions about my presumed opinions.
Actually the simulation doesn't have to follow the laws of nature that we know - another sloppy misreading of yours. It has to, erm, simulate them. And therein lies the problem. If you still haven't understood why that's nontrivial, please re-read my blogpost.
7:22 AM, March 22, 2017
Steven H. Roemerman Sr said...
Sabine,
Thank you for this. I've been saying this for a long time. But it sounds more convincing coming from you.
11:42 AM, March 22, 2017
Srk9 said...
I think your argument is flawed. This explains why:
http://www.smbc-comics.com/comic/2010-11-09
Computing devices outside could be far more powerful than those inside. The simulation also need not run in real time or model the same universe. The simulated physics in a computer simulation containing sentient beings are the laws of physics as far as they are concerned. Also, WRT running in realtime, keep in mind that Turing completeness allows any language to simulate any other language, but makes no time guarantees.
Your argument is weak because you are trying to discredit something that is unfalisible.
12:01 PM, March 22, 2017
_Shorty said...
You're basing your opinion on what you know, and/or what you think you know. The fact that the computer you wrote this blog/rant on wouldn't be able to simulate the universe doesn't mean it is impossible for a computer that could to ever exist. The fact that something seems complicated to you doesn't mean it is complicated, period. Everything's relative. ;) The computer you wrote this blog/rant on is absolutely insane compared to computers 50 years ago. And in five years it'll be a piece of crap. Again, basing your opinion on the current capability of the computers we have now makes no sense. The fact that we think the universe is complicated and that we don't know how it all works does not mean that it couldn't be simulated by a computer more advanced than computers we have now, programmed by "people" or AI more advanced than us.
"If you try to build the universe from classical bits, you won’t get quantum effects, so forget about this – it doesn’t work." Unless you program it to do so. You don't seem to be too familiar with how computers work, frankly. Basically, your argument is similar to saying we can't have fractional numbers in computers because it is only 1s and 0s. Just because you don't know how to program something doesn't mean nobody else could be able to figure out how to program a solution to that problem either. There is absolutely nothing about the nature of quantum mechanics that would be restricted by a binary computer. And there's absolutely nothing saying a simulation has to be running on a binary computer as we know them, either. Sorry, but basically your whole rant is summed up by "I don't know computers very well, so it can't be computers." Makes no sense.
12:22 PM, March 22, 2017
Sabine Hossenfelder said...
Srk9,
So, erm, what is it then? My argument is flawed or it is weak? As to your first "criticism", you seem not to have read anything I wrote, you just produce further "it might be so or so". Write it down, demonstrate that you can reproduce the standard model and general relativity, then come back.
As to your second comment, I am not trying to discredit it, I just did.
12:24 PM, March 22, 2017
Sabine Hossenfelder said...
Shorty,
So what do you recommend? That I base my opinion on what I don't know? Like the people who believe that the universe can be simulated on a computer but who don't bother to find out whether the laws of nature actually can be simulated?
Having said that, you are assigning opinions to me that I don't hold. I'm not saying it's not possible that some day some being somewhere might be able to simulate a reality like the one we experience. I am saying that this claim isn't supported by what we presently know. It's speculation. It's a mildly interesting fiction. But not science.
12:29 PM, March 22, 2017
_Shorty said...
Anything can be simulated. You need to stop thinking that something can't be simulated because your Dell is a binary computer. There is a reason I quoted that specific sentence of yours. Because it is wrong. Absolutely nothing about a binary computer restricts it from simulating the probabalistic location of an electron. Why you would think it does is beyond me. Like I said, you don't seem to understand how computers and/or programming works. Your opinion is most certainly based on that fact. But even learning a tiny little bit about computer programming would show you how that opinion is not a very good one.
12:34 PM, March 22, 2017
George Rush said...
You say "if something is determined doesn't mean it can be predicted". Well, no, but it sure makes it a lot easier! Look, forget that superfluous word "determined". The question is: is it reasonably possible that sim could predict human actions? I re-read your blogpost, as requested. Here's the only relevant statement: "If the programmer could predict in advance what the brain will investigate next, it would be pointless to run the simulation to begin with." Fine, I can't think of a better argument. But surely we don't want to try to guess the hypothetical motives of a hypothetical programmer?
Re. "accuse" and "sloppy" - fortunately I deal with young people all the time, and know how easily you guys take offense. (I was much, much worse at your age.) You were saying human actions can't be predicted. Incorrectly, but reasonably, I equated that with free will. Apparently you don't believe in that. It would be best to merely say so - don't you agree? Ok, forget free will. No problem. But then, you can't use it to deny the prediction hypothesis.
Conclusion: as far as we know, it is indeed possible that sim can predict human actions. If not, please give some logical justification for your view. At the moment I have no idea what it would be.
You say "Actually the simulation doesn't have to follow the laws of nature that we know - another sloppy misreading of yours." No, it's not sloppy misreading, but sloppy language. I meant that its results must "follow" those laws, not that its unknown physical mechanism must obey the laws of our universe (which is obviously wrong).
You say "If you still haven't understood why that's nontrivial [simulating the laws of the universe], please re-read my blogpost." Ok, I did. Of course it's nontrivial for us, with our primitive computers, but sim could be (let's say) 10^100 times more powerful. It can merely solve those exact laws - not for the whole universe, just for the observations made by humans. You seem to think that implies space and/or time must be discrete. No, it doesn't - as Aristotle pointed out long ago. Happy to explain his (correct) reasoning, if you wish.
Certainly, you have some knowledge about this issue I don't. Let me admit that right up front. OTOH there may be things I know that you don't. That's conceivable, isn't it? At my age, the goal of conversation is not to prove who's got the bigger {whatever}, but to learn - and, to teach. To share information, achieving a consensus opinion if possible. Doesn't that make sense?
12:59 PM, March 22, 2017
George Rush said...
Sabine,
I googled you and found some really awful things from Lubos Motl. I see also you're arguing with Scott Aaronson. Finally, some on this blog are hassling you. Obviously you're very busy and have more important irons in the fire than me. It's understandable that your patience wears thin. Please don't feel obligated to address my points, it's not important. Although I don't happen to agree, I admire your taking a stance and fighting it out. Someday when you have the time I'd like to continue our conversation. Good luck, and best wishes ...
George Rush
4:11 PM, March 22, 2017
Joscha Bach said...
Dear Sabine,
Your post raises a multitude of very different and controversial points. This makes it very exciting (and the comments heated). Thank you for the brainfood!
1. Computational vs. computable:
A lot of readers point out that any physical theory is a computational theory, i.e. that the standard model and general relativity describe the universe as computational, even though it might not be (Turing) computable. A quantum computational universe would arguably be in BQP, i.e. it is effectively computable but not efficiently. The universe could also be hypercomputational (for instance because it is continuous or even acausal), but while that would be bad news for current computer science, it means that it is possible to physically implement hypercomputers or acausal computers (and universe-simulations), no?
If the universe is not computational, what else would it be? Is it possible to express a noncomputational theory of the universe in a formal language, or to even think about it?
2. Must the universe be computable?
Digital physics seems to be a minority position, and fraught with difficulties. I think it is great that you take a strong stance and argue that it is impossible in principle, for instance because Lorentz symmetry can in principle not result from discrete operations. I am not a physicist, but it seems to me that this is a highly surprising result and not trivial to prove, regardless of the difficulties to get discrete models to work. Until such a proof exists, it seems to me that rejecting digital physics may be justified only pragmatically?
3. Methodology of philosophy:
In the face of the absence of a generally accepted proof that the universe is uncomputable, and a considerable community that thinks it is: is it acceptable to consider the implications of a computable model without having solved the problem of deriving the standard model in that framework? This seems to be unfair to ask of a philosopher.
4. Simulation vs. computation:
If the universe is computable, it does not follow that it is a simulation. Even the inverse is not true, i.e. if the universe implements hypercomputational capabilities, then it is not clear why no hypercomputer running a universe simulation can be built.
But if we grant that the universe is computational or even computable, it is still extremely unlikely to be a simulation. I think here you have the strongest point to make against Bostrom: if the universe is a simulation, it is probably not one that is run by a posthuman civilization to learn about its own evolutionary history, because generating a googillion galaxies for the sake of simulating part of a planetary surface for a few thousand years is implausible.
5. Artificial intelligence:
You write "It’s not that I believe it’s impossible to simulate a conscious mind with human-built ‘artificial’ networks – I don’t see why this should not be possible. I think, however, it is much harder than many future-optimists would like us to believe. Whatever the artificial brains will be made of, they won’t be any easier to copy and reproduce than human brains. They’ll be one-of-a-kind. They’ll be individuals."
As a computer scientist, I find this baffling. If you build a neural network, even a very large one, what will make it hard to copy and reproduce it?
4:20 PM, March 22, 2017
Delta said...
A lovely rant!
P.S.: intelligencia -> intelligentsia
4:25 PM, March 22, 2017
Steve Baker said...
The simulation hypothesis seems feasible to me. BUT it's probably unfalsifiable...which means we probably shouldn't worry about it because it's up there with religion, the tooth fairy and other unfalsifiable things.
I think I see things in the laws of physics that a computer programmer would have thought to be a good thing to do. Having a finite speed of light, for example, makes parallelism in the calculations much easier and provides for a natural "edge" to the observable universe without needing a "wall" or some other unlikely-seeming boundary. The Big Bang is another handy thing - it provides a finite limit to history and a simple way to populate the simulated universe with fun "stuff" without hard-coding it all.
The fuzzy nature of things that happen at quantum scales just screams "round off error" to me - and although the things we know about what happens at those tiny scales don't really match the nature of the digital computers we've made - that doesn't make much difference. The laws of physics in the "parent" universe would have to be a lot different from ours in order for them to have the computing power they'd need. So all bets are off.
But all of this is essentially unfalsifiable. So we should make a note of an interesting hypothesis - and get on with figuring out what we can about the universe as we see it.
If evidence ever does come to light to prove that we're in a simulation - then we can start to be concerned about it.
5:02 PM, March 22, 2017
Stephen Pirie said...
Sabine
No need to be annoyed, the idea (our universe is a simulation) is easily dismissed.
The idea that we exist in a simulation rests on the ability to objectify (apply an algorithm to) the infinite. To objectify requires one to be apart from that which is being objectified.
But to be apart from the infinite means being outside of it, and "it" extends everywhere. Max Planck explained this simply with his "we can't get behind consciousness".
In other words, the infinite is consciousness and it is "immathematical", it's the realm in which poets, writers, artists, creatives, inventors dwell (but not too long, lest they be locked up in a mental health care facility, as more often happens to artists and the like). It is a realm that cannot be thought (reasoned) it can only be felt.
It is not a rational (computational) dimension. Both "reason" and "rational" have the Latin root "ratio" - in other words, the rational and the reasonable are subsets of a greater immathematical ground.
I think Bohm came to this realization, when he wrote that "The actual operation of intelligence is thus beyond the possibility of being determined or conditioned by factors that can be included in any knowable law ... Intelligence is thus not deducible or explainable on the basis of any branch of knowledge (e. g. physics or biology). Its origin is deeper and more inward than any knowable order that could describe it."
Artists get it. Bohm got it, as did Planck. Maybe those who conjecture the universe is a simulation will eventually "get it" as well, when they discover the poetic, the artistic within themselves.
The writers of "Contact" also "got it" when they had Jodie Foster, in a state of reverie, lament "they should have sent a poet".
7:02 PM, March 22, 2017
mark said...
I cant believe I missed the news that epistemology had finally been sewn up.
Let the kids have some fun.
9:58 PM, March 22, 2017
Cairo Silver said...
Sabine, if it comes down to 'Why don't you read what I explained' you're either tired from the day and don't have enough energy for charity (understandable, happens to us all), or you just decided to read uncharitably.
How do you prove the thing you're sitting in front of now is a computer? Then I'll know what you accept as a proof.
11:17 PM, March 22, 2017
Sabine Hossenfelder said...
_Shorty,
It is evident from your comment that you don't have the faintest idea what I'm talking about to begin with. The logic of your argument goes like this: Because I can calculate the outcome of a Bell-type experiment on my laptop this means quantum mechanics is a classical, local theory. Hey, I have broken physics! Please line me up for the Nobel!
Look, I don't have to know how a computer works to tell you what you can't and can't do with classical, local operations. For me a computer is just another model for reality, and models for reality is what I deal with for a living. I am sincerely sorry to hurt your computer scientist's pride, but in the end it's all physics. Now go and look up Bell's theorem. If you manage to disprove it, please publish your results and then come back.
I have no clue what you think my age has to do with any of that.
4:19 AM, March 23, 2017
_Shorty said...
I said nothing about your age.
I'll say this again. This:
"If you try to build the universe from classical bits, you won’t get quantum effects, so forget about this – it doesn’t work."
shows that *you* don't know what you're talking about. The fact that you don't know why doesn't mean it isn't so. Sorry, but that's the truth. You don't know what you're talking about, or that sentence wouldn't have come from you.
5:01 AM, March 23, 2017
Ashutosh Rai said...
Sabine,
As far as I get it, I think your main argument is the complexity of simulating the physical laws as we observe them. But is it not possible that the actual physical laws are much more complex, and we have no way to probe them (because we are in a computer) and what is our universe is actually a much smaller part of the world of the future beings? That is, they don't have to simulate the whole universe or all of its laws for us to be in the simulation.
6:23 AM, March 23, 2017
Jim Cross said...
I think of the simulation hypothesis as philosophical speculation and don't try to apply science to it.But if you want to do that I wouldn't disagree with your main points.
You wandered off the main point into some other areas about artificial brains and consciousness.
We will likely create (and to a big extent already have) hardware/software that do very specialized complex tasks - for example, drive a car. However, even when I drive a car, most of what I do is unconscious. Recently my consciousness is much more focused on listening to Warren Zevon while I commute to work and the actions of driving are almost automatic. Much of what the human brain does is not conscious.
So what would it mean to simulate consciousness? Do we mean that the artificial brain would possess qualia? Or are we simply saying the artificial brain has acquired some critical threshold of abilities to perform complex tasks that it now seems to be conscious?
6:44 AM, March 23, 2017
Sabine Hossenfelder said...
_Shorty,
You're right to the extent that my sentence failed to spell out explicitly that I was assuming the underlying interaction is local. You can do it classically if non-local. In that case, however, nobody knows how to get quantum field theory - see my note added. If you believe that you can prove it can be done, please enlighten us how, because we poor and stupid physicists haven't figured out how to do it. It's pretty clear though from your previous comments that you don't even understand the problem.
7:23 AM, March 23, 2017
_Shorty said...
Anything you understand can be modeled on a computer. Whether a computer/network fast enough and large enough exists today isn't relevant. And just because you don't understand it doesn't mean it is impossible for anyone to ever understand it. I know full well that you know more about physics than I ever will. This does not change the fact that anything that is understood can be translated to a computer algorithm and thus simulated on a computer. And, no, I'm not saying *we* understand everything that would be required to build a simulation of our universe. What I am saying is that if it were understood well enough it could be simulated. The only things we couldn't simulate would be the things we do not know enough about. "Classical bits", as you want to call them, are just fine and dandy. Modelling the behaviour you want is not limited by "classical bits." It doesn't matter if "classical bits" are not "quantum bits." You can still descrive quantum behaviour via "classical bits." You just need to know how to write a suitable algorithm. Just because you don't understand how to write it doesn't mean it is impossible to write. Anything can be modeled. Anything. As long as you understand what it is you're trying to model.
7:53 AM, March 23, 2017
Sabine Hossenfelder said...
_Shorty,
I give up. I merely post your last comment to demonstrate the communication failure.
7:56 AM, March 23, 2017
_Shorty said...
Keep it simple then. The two-slit experiment's results are due to quantum effects, correct? As I understand what you're saying, you believe you cannot model that experiment on a computer with "classical bits." Is that not what you're saying? Because that sure seems to be what you're saying.
8:02 AM, March 23, 2017
Sabine Hossenfelder said...
_Shorty,
Seriously? After all I've explained you didn't even bother to look up Bell's theorem?
8:07 AM, March 23, 2017
_Shorty said...
No need to look up anything. You think because you can't do something, nobody can. Not even someone with more knowledge and technology than you. *shrug* I honestly do not know how you cannot get that through your head. You are saying "I personally don't know how to do "A" right now. So nobody anywhere ever will be able to, either. Because the knowledge I possess right now says to me that it is impossible, and that means it will always be impossible." Scienticians. Heh.
8:26 AM, March 23, 2017
Sabine Hossenfelder said...
_Shorty,
As I said, you're welcome to try. But unless you manage to come up with a way to circumvent Bell's theorem, I can't take your claims seriously. (Neither will any other physicist.)
8:31 AM, March 23, 2017
_Shorty said...
As I said, you just don't get it. Your argument is no different than someone from 15,000 years ago thinking it is impossible for people to fly, so nobody will ever fly. And yet, planes.
The fact that you personally can't build a universe simulator right now because you can't see how it could possibly be done doesn't mean the universe you're in can't be a simulated one. It could be. You don't know. There's tons of crap you don't know. Just like everyone else. Your argument that it is impossible because you don't understand how it could be possible makes no sense. I'm no physicist, but I get that. Da plane, boss! Da plane!
8:52 AM, March 23, 2017
JimV said...
"Because the knowledge I possess right now says to me that it is impossible, and that means it will always be impossible."
a) That is a gross mischaracterization of what Dr. Hossenfelder has said.
b) In fact, certain things *have* been proved to be impossible, and will always be impossible. For a trivial example, if N is an integer, N*N+1 (N-squared plus one) will never have three as a factor. (Therefore you should read and understand Bell's Theorem.)
c) If your counter-position is that anything that seems impossible now might be possible in the future, that is an example of an unfalsifiable hypothesis (as long as the future of the human species still exists), which science has long known are not worth considering.
9:39 AM, March 23, 2017
Plato Hagel said...
Interesting analysis.
The interesting thing I find is that in order to valid the world as a "real object," we are using data from our measures. So how you use that data reinforces the belief that what is real, conforms to the data from the measures you use. But you create measure in order to validate? You see the circle?
So data "as information" may be called bits? You may not like to call bits data, so is there some "other way" information can be represented?
Consciousness, implies a range of perception that the mind is capable of? Any physical object, is focused reality conforming toward objectification. Consciousness creating measure is a subjective perception is what the mind created?
10:00 AM, March 23, 2017
Sabine Hossenfelder said...
_Shorty,
I am not saying it's not possible that we live in a simulation. I am saying that it's a non-trivial statement that isn't easy to make compatible with all we know about the laws of nature. Or some of us know, anyway.
It is really remarkable that you continue to talk down to me after you've just publicly demonstrated your utter ignorance of even basic physics.
11:25 AM, March 23, 2017
Bill said...
Hello Ms Hossenfelder,
I read your interesting blog post. I have to say, I am interested in physics but being a mechanical engineer I can't claim to understand even a tiny percentage of what you are. I'm here to make a question on the topic. I remember this from when it got published a couple of years ago: https://arxiv.org/pdf/1210.1847.pdf
It is supposed to be an effort to devise a scientific way that will lead to experimentation that will help us understand whether we live in a simulated universe or not. Not sure about the results, etc. Can't even understand the essence of the paper. Can you tell me if it holds any worth? Also, would there be any way that we can work our way scientifically to set up an experiment that will give us indications towards one way or the other? Because from your ramblings, I understand that you think we are not working on that direction, but follow a sallow approach of belief. Is it possible that we can prove one or the other, or do you feel this will remain a philosophical question that scientists should leave aside for now?
Thank you :)
12:28 PM, March 23, 2017
Stephen Pirie said...
_Shorty
"Anything can be modeled. Anything"
Rubbish. Actually, "Complete and utter rubbish".
Tell you what, why don't you disprove not only Bell's Theorem, proving that local reality is unable to explain the world we experience, but also Gödel' Incompleteness, Heisenberg's Principle of Indeterminism, and Chaitin's Randomness Theorem, in which he states "some mathematical facts are true for no reason, they are true by accident, or at random. In other words, God not only plays dice in physics, but even in pure mathematics, in logic, in the world of pure reason. Sometimes mathematical truth is completely random and has no structure or pattern that we will ever be able to understand."
If you think you can model randomness, not some computer generated randomness, but the genuine thing, go for it.
When you're done doing that, go off and model the infinite. Report back when you're done. After which you'll need to again model, and thereafter, yet again, and again and again (ad infinitum).
3:54 PM, March 23, 2017
_Shorty said...
I continue to talk down to you because you don't understand everything you think you do, as smart as you think you are. I know something you don't know. And it is eating you up inside. And you don't like that. Because you think I'm not anywhere near as smart as you. And that's why you continue to respond. Because you hate it. And can't leave it be.
I'll remind you, you don't know what knowledge I possess. You only know what I allowed you to see. You know, with my free will. hahahaha
4:02 PM, March 23, 2017
Sabine Hossenfelder said...
_Shorty,
I am posting your comments because you highlight the problem my post was alluding to. People talking about physics without knowing a thing about it to begin with. Do you realize at all that you're making an idiot out of yourself by proclaiming you can disprove Bell's theorem while at the same time demonstrating you don't even know what the theorem says?
3:08 AM, March 24, 2017
_Shorty said...
And yet, you don't understand something so simple. I thought you were supposed to be smarter than me? Confounded by an "idiot." Hilarious. With every interaction you confirm the stuff your "peers" publicly say about you. Quite entertaining. :)
6:45 AM, March 24, 2017
Qwertie said...
I winced when I heard Neil deGrasse Tyson supporting this harebrained idea.
My problem with the simulation hypothesis is that - even if there is some way to simulate a large space including quantum effects, or even if there is some way to deceive the brains in the simulation into thinking this is what is happening, it would be prohibitively expensive computationally.
In our universe, a simulator is necessarily much larger than that which is simulated. Wouldn't a computer builder in the outer universe have the same problem?
The usual way we deal with this problem is to make a simulated universe extremely simple compared to the real world, and certainly no more complex than necessary. Thus, for example, a first-person shooter doesn't simulate the inner workings of a person; the model of a person is no more complex than a single (classical) molecule in term of its physical behavior, and things like air, walls and mountains are merely drawn, not physically simulated.
Even if our universe were classical, it is hard to overstate how simple and inadequate our own computer simulations are compared to what we would need to accurately simulate, say, the inner workings of a tiny speck of dust. We can't simulate a single bacterium at the atomic level; simulating individual proteins is a big challenge. So our technology - which is perhaps far from theoretically optimum, but not *that* far - requires, what? A processor trillions of times larger than the space being simulated? Just for a classical simulation. Presumably the outer universe could have more a favorable ratio between the size of the computer and the amount of matter in the simulated space, but that ratio would surely remain above 1, wouldn't it?
There are two possibilities. Either the "programmer" is trying hard to deceive scientists, or else the physical laws and facts inferred by scientists are real. And the facts are these: scientists have inferred a universe whose complexity at the smallest scales is dramatically, even unfathomably, greater than what is actually necessary for micro-organisms (let alone humans) to exist, and whose size at the largest scales is not only beyond our ability to imagine, but indistinguishable from infinite.
It seems therefore that we can rule out the second possibility; surely the size of "the matrix" is not virtually infinite, especially since that implies it took a virtually infinite length of time to build it. The only remaining possibility is that the "programmer" (we should really call him "The Architect" - have you not seen The Matrix?) is intentionally deceiving us.
But if one of the goals is to deceive scientists into thinking we are not in a simulation, why work so hard at it? If "the architect" were doing enough simplifications to make the simulation tractable (such as running a classical simulation with an "ether", no relativity, etc.) why bother to fool us into believing we're in a dramatically more complex universe than we are actually in? If our laws of physics were much simpler, all the processes of life could still be possible, and the simplifications wouldn't make it any harder to deceive us. The only things that really need to be "covered up" are simplifications that can only be explained by positing a computer that simulates our universe.
And finally, what would be the point of deceiving us in the first place?
Some people bizarrely seem to think that quantum effects are evidence of a lossy "compression algorithm" that makes the simulation more tractable. The reality of course is that quantum effects make everything more complex and harder to simulate, not easier. Evidence of lossy compression artifacts have never been seen.
8:30 AM, March 24, 2017
Qwertie said...
This Shorty, huh? I could ask him why people are working so hard to build large and expensive quantum computers with modest, non-general-purpose computing abilities if classical computers are all one could ever need. I could point out that as a computer engineer who recently took a course in quantum computation, I might even have a clue about this. But I certainly have no desire to talk to him.
However, if Bell's theorem just says "No physical theory of local hidden variables can ever reproduce all of the predictions of quantum mechanics" (Wikipedia) I don't see how it is necessarily a roadblock for making a simulation. The "outer" universe could play by different rules, allowing instant data transmission across arbitrary distance, for instance. What I can't fathom is why, if this is The Matrix, our universe would be unimaginably large and the laws of physics would be absurdly hard to simulate.
11:16 AM, March 24, 2017
Sabine Hossenfelder said...
Qwertie,
Yes, indeed! The 'real' universe could work by other laws than ours. And these more fundamental laws could be used to give rise to the laws we observe. Which is exactly what theoretical physicists have been working on for decades...
11:48 AM, March 24, 2017
Bill Toulas said...
Hey Qwertie
Just some questions here. Why would you say that our universe is unimaginably large? For whose imagination is this claim valid? For ours? Why would that matter for the creator of the simulation? This simulation may not even have to do with us at all. Did that occur to you at all? Also, why do you suggest that the simulation should have laws of physics that are easy to simulate? Do you suggest that what we see as "hard" would also be hard for the creator of such a simulation? I wouldn't think so personally. Also, you said something about "very expensive to simulate" on a previous comment. Yeah, having a smartphone in the 80ies would be absurdly costly, but now it costs just $50. You say you studied quantum computing. What if the creator uses something relevant to this technology, or even something more advanced? Wouldn't that make a simulation much much cheaper to run?
11:52 AM, March 24, 2017
Koenraad Van Spaendonck said...
If there is such an entity as a creator who simulates 'all of it', then i would personally be more interested in the constitution of that creator, since he would be the only 'real' stuff around. Which means the simulation is the boring part of the quest, haha.
1:56 PM, March 24, 2017
Stephen Pirie said...
I agree with Sabine, I think the majority here are either ill-informed or ignorant of the deeper implication of quantum physics.
E.g. _Shorty, and all those who "think" that we or someone may one day, or is simulating or modeling reality.
As physicist Dr Bernard Haisch points out "the Leggett inequality that was recently measured ... rules out any possible interpretation other than consciousness creates reality when the measurement is made."
Max Planck said simply, "we cannot get behind consciousness".
Likewise David Bohm, "the operation of intelligence ... is deeper and more inward than any knowable order that could describe it"
Likewise physicist Freeman Dyson "Quantum mechanics makes matter even in the smallest pieces into an active agent, and I think that is something very fundamental. Every particle in the universe is an active agent making choices between random processes"
What the mechanically minded responding on this topic don't or refuse to appreciate is - as Dyson explains - all of reality is collapsing the wave-function, all the way down to the smallest particle, and as per David Deutsch's "shadow photons", those as well.
No one will ever, ever, ever simulate all of the individual particles, including the shadow particles in the multi-verse collapsing the wave-function in their own special way, according to constraints and systems within which they operate.
As Dyson also explains "consciousness is not just a passive epiphenomenon carried along by the chemical events in our brains, but is an active agent forcing the molecular complexes to make choices between one quantum state and another. In other words, mind is already inherent in every electron, and the processes of human consciousness differ only in degree but not in kind from the processes of choice between quantum states which we call 'chance' when they are made by electrons."
I share Sabine's frustration, but due to the fact that those who are mechanically minded (those who believe or even conjecture that one day we or someone can simulate the process of choice being performed by each particle in the universe, and multiverse), and who have not availed themselves of the quantum evidence, are also people who are helping shape societies. And that affects all of us, to our detriment.
For all those mechanically-minded folk, here's a question: What reaches into the infinite recesses of possibility to begin the collapse of the wave-function (from possibility to actuality). See my diagram on this https://stephenpirie.com/sites/all/files/simple-tools-fig.6b.png
What "mechanism" can do that?
What reaches into the unutterable, unspeakable, "immathematical" depths of existence, and does that, again and again, trillions of times each second?
If you want an answer, the quickest pathway is to ask an artist, poet, or anyone who is intuitively, precognitively gifted. Then go and study the real implications of quantum physics, and marry the two (left and right hemispheres in your heads), rather than being imprisoned in your left-hemispherical world of numbers, names and notions.
6:41 PM, March 24, 2017
JimV said...
"However, if Bell's theorem just says "No physical theory of local hidden variables can ever reproduce all of the predictions of quantum mechanics" (Wikipedia) I don't see how it is necessarily a roadblock for making a simulation." (Qwertie)
I think the point was that the simulation would require qubits, not classical bits.
As for computers getting better and cheaper, Moore's Law or any exponential law works in the earlier stages of growth and development, but cannot go on forever, and I think physics dictates that Moore's Law is just about done. (Circuit sizes in the range of small numbers of atoms.) Then going to qubits would require maintaining temperatures near absolute zero, which will add huge costs (outer space itself in Earth's vicinity is too hot for good efficiency). Finally, as mentioned in previous comments, the ability to simulate every parameter (mass, energy, spin, etc.) of every particle in the observable universe (so that wherever we look we find no glitches) is mathematically impossible in this universe. One must postulate a higher universe with magical capabilities, total unobservable and forever unknown to us, with incomprehensible motives (since if we want to simulate, say, the collision of two black holes for scientific purposes, there is no need to model ants as part of the simulation - or people either, which would also look like ants to such simulators). As also mentioned before, speculations involving unobservable, unknowable, and incomprehensible hypotheses are not encouraged in science - but knock yourself out.
6:49 PM, March 24, 2017
Thomas Schaefer said...
The whole discussion strikes me as pretty absurd. The best known methods for studying the time evolution of quantum many body systems without serious approximations can do maybe 100 atoms at most. This only works by treating nuclei as point particles. If you need the dynamics of neutrons and protons you can at most do a small nucleus, say carbon. And if you want the dynamics of quarks and gluons there are no known methods at all.
Of course we all hope for the advent of quantum computing. But you still need a physical system to encode a qbit. Indeed, the easiest way to simulate a quantum system is to make it, and let the quantum hamiltonian compute its own evolution. If our simulation overlords wanted to simulate a universe, they would first have to create a physical universe to run the simulation. What, exactly, would be the point of that?
10:29 AM, March 25, 2017
Stephen Pirie said...
JimV
"the ability to simulate every parameter (mass, energy, spin, etc.) of every particle in the observable universe"
Well, that assumes a certain interpretation of quantum theory that many (I believe a majority) of physicists would dismiss.
As David Deutsch explains (in The Fabric of Reality) "Single-particle interference phenomena unequivocally rule out the possibility that the tangible universe around us is all that exists" - aka, we exist in a multiverse, not simply just that which is directly observable (noting that recent experiments have established the superposition of molecules containing around 5,000 atoms, if I recall correctly).
Now, knowing each particle is in a superposition prior to observation, who's going to simulate those superpositions?
Which also means per Thomas Schaefer above, that not only would "they would first have to create a physical universe to run the simulation" they would also have to create the multiverse, containing all those superpositions, for every photon, quark, lepton, electron, etc.
As JimV adroitly suggested, "knock yourself out".
2:53 AM, March 26, 2017
Stephen Pirie said...
Furthermore, let's for the moment imagine that some higher-order entities were to simulate the multiverse, with all those endless superpositions.
At Cambridge University, mathematician Noah Linden and physicist Sandu Popescu found that
"in the typical quantum state occupied by any group of particles the links between the particles are mostly of a nonlocal character. Quantum theory isn’t just a tiny bit nonlocal. It’s overwhelmingly nonlocal. Nonlocality is the rule for our Universe"
So even those doing the simulation are part of the nonlocal interconnectedness. In other words, they'd be part of the simulation, for everything effects everything.
That "everything effects everything" I believe led physicist John Wheeler to suggest that we're also effecting the past -- in his delayed choice experiments, he envisaged (didn't get to complete it, I understand) using galactic lenses to confirm our choices now choose which past we experience all the way back to the Big-Bang. He concluded that we live in a participatory universe, wherein everything is helping/participating in the unfoldment/evolution/creation of an unimaginably vast self-organising system (hence why galaxies are known to distribute themselves fractally).
Good luck simulating yourself, all the present multiverse, and the past ones as well (necessary for the foundational delayed-choice experimental results)
3:26 AM, March 26, 2017
|
Part2:Comments "No, we probably don’t live in a.." [re: Mod vege]
(старо куче)
