this post was submitted on 27 Aug 2023
63 points (84.6% liked)
Asklemmy
43159 readers
2066 users here now
A loosely moderated place to ask open-ended questions
Search asklemmy ๐
If your post meets the following criteria, it's welcome here!
- Open-ended question
- Not offensive: at this point, we do not have the bandwidth to moderate overtly political discussions. Assume best intent and be excellent to each other.
- Not regarding using or support for Lemmy: context, see the list of support communities and tools for finding communities below
- Not ad nauseam inducing: please make sure it is a question that would be new to most members
- An actual topic of discussion
Looking for support?
Looking for a community?
- Lemmyverse: community search
- sub.rehab: maps old subreddits to fediverse options, marks official as such
- [email protected]: a community for finding communities
~Icon~ ~by~ ~@Double_[email protected]~
founded 5 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Oh, I've got something unexpected for you. I got bored listening to people going on and on about the Universe being a simulation, so I built a machine that can check if you're in some (but not all) classes of a simulated Universe. Take that, Plato!
It works using Bell's Theorem and the limitations of Turing machines. The former shows that for some types of measurements, the outcomes are purely probabilistic, and do not contain hidden deterministic variables (seems like a dumb way to design a Universe for simulation!). Then Turing machines are capable of computing anything that can be defined as an algorithm (but only things that can be defined in an algorithm!).
Since the former provides a non-deterministic measurement that the latter cannot model, you make a machine that produces those measurements, and ties them to something on a bigger scale (stock market purchases, social media posts, whatever). Then you keep the measurements, and attempt to determine whether they were actually generated with an algorithm (this is very hard but possible in theory -- sort of like reversing a hash). Meanwhile in the Universe upstairs, for the results to appear random enough so that analysis is nontrivial, it must consume extra computation in whatever device is running the simulation. In short, I'm doing a thing that consumes extra processing power.
So logically, if this Universe is being simulated, and someone is observing the machine closely that is doing the simulating, I can pass messages encoded in CPU activity. Just like in computer security where you can extract encryption keys by looking at a video of your power LED!
So if I'm not actually typing this on my laptop right now -- but really am sleeping/dreaming with some equivalent to an EEG monitor or MRI or whatever in some other Universe -- someone may be able to conclude whether this Universe is being generated by my mind or not. In the former case, they may have their answer, but have not woken me up yet. At that time, regrettably, your continued existence is not guaranteed. Sorry about that, but there was science to do!
If you would like to use the device I have built, you can do that. Because of course it's sitting on my desk, connected to a server via MQTT, and subsequently connected to a Lemmy bot. If you message kong_ming on my instance (with a non-blank message), it will use the device as an entropy source to generate an I Ching reading. You know, to help you answer personal questions with only a vanishingly small likelyhood of causing the Universe to cease existing. Sometimes it is down for maintenance, although I just checked in on it now and it seems OK.
I also have a second such device, at an undisclosed location, that is designed to work with a coffee machine. You know, for those times you want coffee that's simultaneously caffeinated and decaffeinated until you drink it.
Anyway, there you go. I've added a feature to the fediverse, I guess.
If our universe is being simulated on somebody's desktop computer then yeah maybe, but if you were going to simulate a whole universe wouldn't you at least have a decent random number generator to prevent easily-detectable patterns?
If we assume a malicious intelligence simulating a universe with goal of not being detected, all bets are off -- this technique only works in a few cases, and that's probably not one of them. Also if it's true, I think we have bigger problems :D
There are a whole bunch of other assumptions too -- like the universe running the simulation has entropy and time that work the same way as ours. It's no magic simulation-detecting bullet -- but it's the only technique I could think of to make any progress whatsoever on the underlying philosophical question! A mote in the eye of a fictional God, so to speak.
In the hard sense, there are no such thing as random number generators on computers. With sufficient starting entropy and computing power, you can generate a mostly reasonable approximation. However, this must use more computing power than not doing it, which is the "signal" we're sending out to be detected by a fictional observer in the scenario the OP presented.
Interestingly, this technique is used to exfiltrate data from secure computers -- e.g. by making the CPU do slightly more work sometimes and modulating that to send data e.g. by radio emission, hard drive noise, power LED brightness changes, and so on. Here's a generic one for you, if you're curious: https://thesai.org/Publications/ViewPaper?Volume=9&Issue=1&Code=IJACSA&SerialNo=25
Also, there are sometimes interesting and strange artifacts even with our everyday "random" number generators. I read a really neat paper about that ten years ago, comparing the artifacts of random number generators across operating systems, which sadly I can't seem to find for you presently. There's an OK example for you here though: https://www.random.org/analysis/ under "simple visual analysis".
That kind of weird pattern is pretty typical of most 'random number' functions used in software that aren't security-facing (and sadly sometimes even ones that are). For cryptographically secure random numbers (more like the image to the left than the image to the right on that site), they are more computationally expensive to produce.