3 ms·
I don't agree with Bostrom's argument and wrote an undergrad paper arguing against it. I argued it from a physical point of view and I don't think we are in a B
by cgriswald 7y ago
I don't agree with Bostrom's argument and wrote an undergrad paper arguing against it. I argued it from a physical point of view and I don't think we are in a Bostrom-style simulation.
That said, if you accept that a Bostrom-style simulation is physically possible, I think Bostrom is still right, and we would most likely be the center of the simulation.
His basic argument is that if there is an original universe with humans, they might like to simulate their ancestors. And they are likely to run many such simulations, in at least some of which, humans survive. Those survivors, in turn, are also likely to run many such simulations, etc. on down the line. This means that such a high percentage of humans are simulated, that you'd have nearly infinitely better odds of winning the lottery than being a non-simulated human. To me, this means we should conclude we are the center of the simulation, since the fact that we exist implies we are probably in an ancestor-simulation.
> Why simulate the whole universe just to simulate us?
> Maybe this is about simulate the universe expansion/contraction.
Why simulate the whole universe just to simulate the expansion/contraction? There seems to be little need for conscious beings (or life at all), unless life and/or conscious beings is required in some way to explain expansion/contraction. (But if so, wouldn't that mean we ARE in fact the 'center of the simulation'?
- dmichulke 7y ago> His basic argument is that if there is an original universe with humans, they might like to simulate their ancestors. And they are likely to run many such simulations, in at least some of which, humans survive. Those survivors, in turn, are also likely to run many such simulations, etc. on down the line. I read somewhere that the smallest computer to simulate the universe necessarily has the size of the universe. This would that either the real universe contains much more information (likely more than a few orders of magnitude more so that a trivial part of that Ur-universe can be allocated to simulate our universe), or recursion is not possible.
- cgriswald 7y agoEven if humans were to colonize the entire Milky Way, the interactions between two neutrons 13 billion light years away would still be completely outside the scope of the simulation, because those interactions would have no hope of being detected by a human. So you'll always have massive savings versus simulating the entire universe. It's really human computing where it all falls apart.
- RHSeeger 7y agoDoesn't that imply that it needs to be physically possible to simulate a universe that is as complex as your own universe. I would think that it should be possible to prove that isn't true.
- cgriswald 7y agoSee my response to the sibling comment. You don't need to simulate the entire universe. You only need to simulate those parts that can possibly affect a human being. There's no need, for instance, to simulate photometry for stars in the Andromeda galaxy because we lack the resolution to measure the movement of those stars.
- inflatableDodo 7y agoI think that if your metric is not total number of universes, but available quantity of causal interactions, then Bostrom's argument falls apart on probability as well. More stuff can happen in the top level universe, unless you are dedicating more than half the top level universe to simulation, so you are more likely to exist there.
- koonsolo 7y agoSince history goes back for a very long time, maybe the future of the simulation too. So why would you think we are the evolutionary "end state", and nothing comes after us? Maybe cats are running a simulation of their ancestors?