Hacker News new | ask | show | jobs
by luc4sdreyer 3019 days ago
I think it's very likely: https://en.wikipedia.org/wiki/Simulation_hypothesis#Ancestor...

It's not a testable claim right now, but it might be in the future: https://en.wikipedia.org/wiki/Simulation_hypothesis#Testing_...

1 comments

Your link only provides a way to falsify a theory, not provide evidence for simulation.... which is impossible by definition.

Not to mention the idea of a simulated universe is, you know, philosophically boring and implies mostly false things in most peoples' minds.... like an anthropomorphic scientist god. In reality, it would change virtually nothing about how we view our universe.

> not provide evidence for simulation.... which is impossible by definition.

I disagree.

Science cannot prove that a theory is 100% true, only that it is not wrong by repeatedly testing the predictions made by the theory. The best we can do is say: this theory (e.g. general theory of relativity) is the best explanation for data we have so far and every experimental prediction it has made has come true.

The paper (https://arxiv.org/pdf/1703.00058.pdf) linked on the wiki page proposes four experiments that seeks to test the simulation theory. None of these experiments will single-handedly prove the simulation theory, but if they all pass we can only keep testing and seeking alternate explanations.

If the experiments keep passing and no alternative theories can be found, then we're either in a simulation, or the universe just happens to behave exactly in the way that a simulated universe would, but it isn't. I agree that we can never truly know which is true, but at that point the difference is reduced to semantics. It's like saying electrons don't really exist, instead they're just wave/particles that are exactly like electrons in every way except for some immeasurable quality.

This assumes the entity running the sim will never directly influence it, of course if that ever happens (requiring a _very_ high burden of proof) it'll be proof of the sim.

I’m saying the semantics of a simulation are boring. I think the actual tests are pretty cool. :)
Dunno about you but if I were pentesting a machine and found out it was a VM, then maybe I'd try to get a foothold into the hypervisor, host, or sibling VMs. Not boring.
That metaphor is ill fitting unless there is an anthopomorphic entity designing flawed software. It smells a hell of a lot lite Pascal’s wager.