The Simulation hypothesis

[DaveSumm]

31 minutes ago, Jo498 said:

"Accuracy" is a distraction and basically a moot point. Recall, there is no way for those "inside" the simulation to compare it with an outside universe (if the latter exists at all).  Simulation is actually the wrong term, one should rather say virtuality or VR.  Nobody could know if it is a simulation of something similar or just a completely fantastic VR that bears little or no relation to a universe outside (if the latter exists at all).

Nobody in this VR has ever done a physics experiment. Certain "minds" have been fed certain data that created the appearance that they were doing a physics experiment or more likely that they read about such an experiment in a pop science magazin. Recall again, there is no need for 7 billion minds. It could be only you ( e.g. an apparent accountant who never did any physics beyong middle school (in her VR "life history") being fed the impression that you live in a world with 7 billion people etc.

If the virtuality hypothesis is correct, almost everything you believed is (utterly and completely) wrong. (Think of the matrix, only far more radical as you might not be a human in a vat but something else.) It's a much more radical reversion or leap of faith than starting to believe in e.g. traditional christianity. The latter leaves almost all of your former beliefs in place (there is world with humans and animals one interacts with, one's sense impressions and current science are basically correct about its domain etc.), it only requires to expand this to a bunch of miracles and an additional larger domain of being that is immaterial and includes angels etc.

As was mentioned before, surely your logic here is that if we live in a simulation, then the Bostrom reasoning is faulty ... if we don’t, then it isn’t. So we’re left with Bostrom’s probabilities, or 100%. I don’t follow your reasoning with regard to Christianity either, Christianity makes very specific claims about the world. The simulation hypothesis is the mother of all non-specific claims; the ‘parent universe’ it postulates could have any one of a practically infinite number of physical laws, as you’ve stated.

(As an aside, I agree that aspect isn’t focused on enough, I don’t see why this reality has any shot at being the one that replicates the real universe, far more likely to be an experimental tweak ... if we had the capability, wouldn’t that be the more interesting experiment? There’s an interesting discussion to be had about how feasible that is, how many possible laws of physics there are - I forget the exact Einstein quote, something like “what I’m really interested in is whether the need for logical simplicity leaves any room at all for the laws of physics, or whether God simply set the universe running in the only way it could.”)

I’m surprised so many people call this moot. It’s a hard science view, for sure, if it has no testable hypothesis, and ultimately changes nothing. But aren’t you curious? Wouldn’t you want to know if you were starring in the Truman show? 

Another aspect I’ve often thought of; there are two components to the possibility of this being true, a) that our reality isn’t ‘real’, and b) that someone created this reality. Now b) is certainly significant, and something I’d be interested in knowing. But I think a) doesn’t really mean anything. We can’t help but think of our reality as being spread out on a grand stage, that the physical distances involved could only occur on such a stage. But all these qualities are just facts, the same as information stored on a computer. The fact that our brains process these things as being a ‘reality’ doesn’t point to the existence of one, at least not one that is taking up all this space. I’m finding it hard to explain, but in as much as ‘simulation’ means ‘not really big like we thought, actually running on a smaller thing’, I don’t think that has any meaning. I don’t think there’s such a thing as an unsimulated reality, by that definition.

[Jo498]

What I called "moot" is the focus on "accuracy". The word doesn't have any impact if one does not already agree that a) our perceived/scientifically modelled world is roughly as it appears to be and b) it is possible to simulate this perception to a certain accuracy that it becomes indistinguishable from the reality.

But if we are in VR from "births" and all the time we have no clue about a) and therefore nothing to compare reality or simulation with and therefore in b) accuracy or indistinguishability becomes a moot point. The VR could be wildly "inaccurate" or fantastic on purpose.

The "simulation hypothesis" is both specific (with the ass-pulled probabilities) and unspecific in the way you point out. But regardless of this, I struggle to understand how one could not agree that it would be momentous news to realize that I am not a human in a world that is roughly like it appears to  common sense and science but rather some pure spirit or some computational structure falsely believing to be a material human in a world that is roughly as it appears etc.

(Again, I don't know enough about Vedanta, Buddhism and similar eastern religions, but neither Judaism nor Islam nor Christianity expect me to believe that I was wrong about all this. They agree that most of my worldly knowledge is largely correct. It merely has to be supported/expanded by some spiritual or revealed knowledge about nonmaterial beings.)

I don't see how alternatives in laws of nature or the setup of the universe are relevant here. There was never a scientific theory that was not underdetermined by data. There was never a theoretical framework for these data forced by logic and maths alone. There are always choices in theory-building. (That seems part of the plight of string theory: too many options, not enough restrictions.) But a simulation would have far fewer restrictions, so nothing like this would apply.

[Luzifer's right hand]

If we assume that we exist in a simulation it seems hubris to assume that said simulation is especially complex. 

Maybe we exist in a relatively simple simulation compared to the complexity of the "real" universe.

For all we know our universe runs on hardware that is the equivalent of a Tamagotchi and are closer to ants than to the makers of said simulation. 

[DaveSumm]

38 minutes ago, Jo498 said:

What I called "moot" is the focus on "accuracy". The word doesn't have any impact if one does not already agree that a) our perceived/scientifically modelled world is roughly as it appears to be and b) it is possible to simulate this perception to a certain accuracy that it becomes indistinguishable from the reality.

But if we are in VR from "births" and all the time we have no clue about a) and therefore nothing to compare reality or simulation with and therefore in b) accuracy or indistinguishability becomes a moot point. The VR could be wildly "inaccurate" or fantastic on purpose.

Sorry, I wasn’t really talking to you specifically after the first paragraph. I wondered how people could think of the whole hypothesis as moot.

38 minutes ago, Jo498 said:

The "simulation hypothesis" is both specific (with the ass-pulled probabilities) and unspecific in the way you point out. But regardless of this, I struggle to understand how one could not agree that it would be momentous news to realize that I am not a human in a world that is roughly like it appears to  common sense and science but rather some pure spirit or some computational structure falsely believing to be a material human in a world that is roughly as it appears etc.

Yep, totally agree.

38 minutes ago, Jo498 said:

I don't see how alternatives in laws of nature or the setup of the universe are relevant here. There was never a scientific theory that was not underdetermined by data. There was never a theoretical framework for these data forced by logic and maths alone. There are always choices in theory-building. (That seems part of the plight of string theory: too many options, not enough restrictions.) But a simulation would have far fewer restrictions, so nothing like this would apply.

Well, if Einstein’s hypothesis was correct and the need for logical simplicity leaves no room for choices in creating a universe, then there is only one type of universe to be created. There are no experiments to be run, you either create the one and only universe or you don’t (with the caveat that the people within it discover science and thus confirm it to be logically consistent). I’m not saying I think that’s true (and I don’t think Einstein was either) but it would certainly inform the discussion of what type of realities one could simulate, and therefore what probabilities we can attach to the prospect of our simulated universe matching the ‘real’ one.

[rotting sea cow]

23 hours ago, Mentat said:

The easiest way to simulate a human experience would be to interface with the brain and make the brain believe it was seeing, hearing, feeling and smelling something that it wasn't.

Creating an entirely fictional experience in the way someone creates a Hollywood movie may seem complicated, but it doesn't necessarily have to work that way. Imagine you can record an hour of my life with a device and that someone else can then relive that hour with a similar device experimenting everything I felt just exactly the same way I did. Why, you could even record a persons entire life and then have that be the simulation. You could have millions of people living that one same life in a simulation.

It may seem that this wouldn't work because it would be a completely passive experience (the person in the simulation just gets to live a life vicariously, but doesn't get to make any decisions), but I think there's a good chance it would (the person in the simulation would simply get 'overwritten' by the original person who lived the life and come and regard their personality and decisions as their own).

What you are describing is not a simulation!

 

[Rippounet]

2 hours ago, Jo498 said:

(with the ass-pulled probabilities)

Yeah I haven't checked it yet but I smell a fallacy in the math there somewhere, not unlike the fallacies in the math behind intelligent design.

2 hours ago, Wolfgang I said:

If we assume that we exist in a simulation it seems hubris to assume that said simulation is especially complex.
Maybe we exist in a relatively simple simulation compared to the complexity of the "real" universe.
For all we know our universe runs on hardware that is the equivalent of a Tamagotchi and are closer to ants than to the makers of said simulation. 

And funnily enough it could be the other way around and we (or a creator) could have created this simulation in order to experience something that cannot be had in "reality".

[Luzifer's right hand]

3 minutes ago, Rippounet said:

Yeah I haven't checked it yet but I smell a fallacy in the math there somewhere, not unlike the fallacies in the math behind intelligent design.

And funnily enough it could be the other way around and we (or a creator) could have created this simulation in order to experience something that cannot be had in "reality".

I simulation that can run something simpler than "reality" should be easier to make though than the opposite though. 

[Rippounet]

1 minute ago, Wolfgang I said:

I simulation that can run something simpler than "reality" should be easier to make though than the opposite though. 

Indeed. But that doesn't mean it's not improved in some ways.
For instance our video games are simplified versions of reality but they offer experiences -thrills- that cannot be had in reality.
Assuming that in "our" world the universe is not detailed (everything beyond the Solar System is an illusion of sorts) it's easier to imagine that despite the overall simulation being "simpler" it is "improved" in some respects.

[Luzifer's right hand]

1 minute ago, Rippounet said:

Indeed. But that doesn't mean it's not improved in some ways.
For instance our video games are simplified versions of reality but they offer experiences -thrills- that cannot be had in reality.
Assuming that in "our" world the universe is not detailed (everything beyond the Solar System is an illusion of sorts) it's easier to imagine that despite the overall simulation being "simpler" it is "improved" in some respects.

Or maybe not even our Solar System is that complicated compared to "reality". Think Tamagotchi instead of Matrix.   

How would the Tamagotchi know that it is incredible simple?

[Mentat]

3 hours ago, rotting sea cow said:

What you are describing is not a simulation!

 

Why do you think that?

[Kalbear]

2 hours ago, Mentat said:

Why do you think that?

Because the brain in your example exists. We are talking about a situation where there is no brain in a jar. Your brain isn't being tricked into seeing things that aren't there - you are being tricked into believing that there is such a thing as a brain, and a you. 

[Mentat]

1 minute ago, Kalbear said:

Because the brain in your example exists. We are talking about a situation where there is no brain in a jar. Your brain isn't being tricked into seeing things that aren't there - you are being tricked into believing that there is such a thing as a brain, and a you.

But there has to be a 'you'. Otherwise who is being tricked? It might be that I'm not who I think I am, but there must be a ghost within this simulation or the whole proposition stops making sense. The ghost might be an AI, a brain in a vat, a human in a pod a la Matrix or whatever else you fancy, but that doesn't really matter.

If reality is a simulation then whatever is outside the simulation (really real) is unknowable, as is the nature of the ghost in the simulation. There must be something out there to support the simulation, though, and something in there to be fooled by it. Otherwise you have, indeed, lost me. 

[Kalbear]

23 minutes ago, Mentat said:

But there has to be a 'you'. Otherwise who is being tricked?

Two possibilities. The first is that it is someone else simulating you in order to determine how the real you would behave. The second is that there is no real you, and there is no one being tricked and someone is simply running an experiment to see how something behaves. You're thinking Matrix, when you should be thinking the Sims. 

 

23 minutes ago, Mentat said:

If reality is a simulation then whatever is outside the simulation (really real) is unknowable, as is the nature of the ghost in the simulation. There must be something out there to support the simulation, though, and something in there to be fooled by it. Otherwise you have, indeed, lost me. 

Sure, there's something running the simulation, but it isn't being tricked any more than I trick my Xbox into playing Gears of War 5. 

[sologdin]

fallacy in the math there somewhere

the purported probabilities, such as--

Quote

the probability that we are in the one and only base reality, rather than in one of the millions or billions of simulations to follow is, well, millions or billions to one

--beg the question.

[Free Northman Reborn]

4 hours ago, Mentat said:

But there has to be a 'you'. Otherwise who is being tricked? It might be that I'm not who I think I am, but there must be a ghost within this simulation or the whole proposition stops making sense. The ghost might be an AI, a brain in a vat, a human in a pod a la Matrix or whatever else you fancy, but that doesn't really matter.

If reality is a simulation then whatever is outside the simulation (really real) is unknowable, as is the nature of the ghost in the simulation. There must be something out there to support the simulation, though, and something in there to be fooled by it. Otherwise you have, indeed, lost me. 

The idea is that the entirety of “you” is a line of code inside a highly complex computer program (simulation). And it is so advanced that you are self aware. You are an AI basically.

[Kalbear]

37 minutes ago, Triskele said:

tl;dr - the extremely dangerous is that the users discovering that they're in a simulation could ruin the purpose of the thing from the creator's POV so they'll shut it down that is to say end our reality.  I have not pondered this stuff as much as some of y'all, but something jumps out to me there which is:  are we not already at that point?  

Nah, we're not there. We don't have actual proof of living in a simulation. We just have people guessing as such, and most of them are not particularly taken seriously. In particular, there is very little evidence of anyone attempting to actually break or game the simulation. 

[The Mother of The Others]

On 9/17/2019 at 6:49 AM, Liffguard said:

 maybe my reasoning is off somewhere, but I can’t see how there would be endlessly recurring simulations. And that puts a limit on how likely it is that our universe is a simulation.

If reality is a fractal, every sim-within-a-sim could possess the same level of detail, no matter how many layers of the russian doll you peeled away, every level of magnification would show you a 'complete' universe.  Whatever complete means.

[Kalbear]

24 minutes ago, Triskele said:

But while this makes sense on a certain level, doesn't it also presume to what extent the Creators (caps on purpose) are monitoring our existence?  Put another way, are the Creators so unsophisticated that they wouldn't see that we the things we are are now, finally, semi-self-aware and publishing articles and writing books and podding podcasts in our little created reality? 

Counterpoint: the simulation is of the whole goddamn universe, and we're a very small part of a very small slice of that universe. Maybe some undergrad will play with the simulation and discover some bizarre factoid and glitch on one small planet and publish the cool hack thing on reddit and it'll be internet famous for a few days, but our sentience isn't the point. 

24 minutes ago, Triskele said:

I guess I'm confused by the jump from that op-ed to your post in that I don't see how we need proof we're in a simulation in order for that to prove to Them that we've caught on.   Quite possible I'm not following.  

The jump is this: there is a difference between stating that we're in a simulation, our knowledge that it might be a simulation being a Bad Thing, our proving (somehow) that we're in a simulation, our proving  that we're in a simulation being a bad thing, and finally doing an experiment that somehow alters the actual simulation. Think of it as degrees of freedom in a video game that you're playing.

• First case: You're running around RDR, and one of your friends notes that some NPC is holding a placard saying 'we are all in a simulation'. It doesn't destroy your enjoyment of the game. The NPC isn't doing anything that weird in the game, and you can largely ignore him.
• Second case: you're playing some genetic algorithm game, and the whole point is to determine at what point does the AI figure out that there is a world beyond itself. If it's guessing, well, you'd start paying attention but you wouldn't shut it down.
• Third case: the NPC in RDR is proving it's a simulation by killing itself repeatedly and having itself respawn, and somehow carrying over knowledge of its death over and over. Still doesn't ruin your enjoyment of the game.
• Fourth case: same as second case, but in this case you've now finished your simulation because the AI has proven that it's in a sim.
• Fifth case: that NPC in RDR is able to repeatedly cause the servers to crash whenever it chooses, or cause a user's machine to attempt to calculate a crazy amount of physics collisions due to a weird glitch it found in walking, which results in people with high-end graphics cards that do physics calculations to have their card overheat and die. Or it has invented a donkey lady. In either case, the simulation is now causing harm outside of the simulation and needs to be patched.

We are in case 1 or 2 right now. We need to worry when we're in case 4 or 5. 

[Mentat]

11 hours ago, Kalbear said:

Two possibilities. The first is that it is someone else simulating you in order to determine how the real you would behave. The second is that there is no real you, and there is no one being tricked and someone is simply running an experiment to see how something behaves. You're thinking Matrix, when you should be thinking the Sims.

In the first possibilities there are two 'me'. The original 'me', who is a human out in the really real world and the second 'me' who is a simulation of the original me built for nefarious or scientific reasons. The second 'me' is still a conscious entity independent of the real me (which might well continue existing after the biological 'me' is dead). It doesn't really matter if it's some lines of code, a brain in a vat or a digital scanner of my brain. The physical reality behind the ghost is irrelevant (though going back to my original post, a digital scanner of my brain seems like the easiest way to go about this).

In the second example there is also a 'me'. I might be nothing more than some lines of code in a computer, but in terms of individuality that doesn't really make me any different than, say, HAL 9000 or the Terminator. Moreover, remember that a simulation as an experiment doesn't offer any valid data outside of its frame of reference, so the really real would have to be similar to the simulation for it to have any purpose. A good example of this is the Black Mirror episode 'Hang the DJ'. Though we never get to see outside the box there is a heavy implication that what is out there is very similar to what is in there, otherwise the whole thing makes no sense.

The Sims are also a good example because, while the individual Sims aren't currently self-aware, their world is an abstract representation of the real world their creator (programmer) inhabits. If, in a thousand years, The Sims 800 includes self aware Sims for the player to coddle or abuse, each Sim will think of itself as 'me', and they won't be wrong. The personality of these future Sims might well be a digital copy of the personality of actual humans.

11 hours ago, Kalbear said:

Sure, there's something running the simulation, but it isn't being tricked any more than I trick my Xbox into playing Gears of War 5.

No, in a simulation scenario there must be a 'creator' entity from the really real who is (or was) responsible for the simulation existing and a conscious entity that exists within the simulation and is unaware of the fact that it lives in a simulation rather than in reality (hence, is being tricked or fooled). Your Xbox is not a conscious entity (and neither is Kait, though she would be more akin to what I was thinking of if you want to use a video game as an example).

 

[Rippounet]

6 hours ago, The Mother of The Others said:

If reality is a fractal, every sim-within-a-sim could possess the same level of detail, no matter how many layers of the russian doll you peeled away, every level of magnification would show you a 'complete' universe.  Whatever complete means.

A fractal multiverse is a hilarious idea, but @Liffguard is correct: unless we are ourselves witness to the spontaneous creation of a simulation within a simulation we have created there is zero reason to seriously consider the possibility of endlessly recurring simulations.