Deliberately trying to annoy simulators?
I regularly seek inspiration by taking long solo walks; and during my most recent such, considering what practical consequences (if any) there would be of the universe I know being a simulation—something flipped in my head, and I thought to myself, “Screw the simulators. If I’m the first copy of me, I should make it as hard as possible for any simulation of me to keep up with me—and if I’m a simulation, I’m going to try to do even better than my original did.”
Ignoring the impracticality of trying to out-do myself, is there anything that someone living in an ‘original’ universe can do that would make it harder for a future simulator to reproduce them? And, mirror-wise, is there anything someone in a simulated universe could do to differentiate themselves from their original? And, if the answer to either question is ‘yes’… would it be a good or bad idea to try?
(And is there any way to gather any actual data that might support the answers to such questions, instead of merely making guesses of a similar nature to classic college/stoner “Our whole universe could be, like, an /atom/, man” musings?)
For one thing, you probably can’t increase the amount of entropy you use by making more observations.
If video games are any guide, I would recommend spending a lot of time standing in doorways to screw up as much path-finding as possible. Getting the NPCs to deal with temporarily blocked corridors seems to be a universal pain in the ass. (Not that there is evidence that we already have trouble with cell transitions.)
On that note, become a city planner and design horrifying paths through high-density areas, then regularly traverse those areas at peak times to force the simulator to pathfind for all the simulated people.
.… Whenever I see a particularly awful piece of infrastructure, I will imagine that this was the intention of the engineer and laud them for their creativity in creating chaos for our simulating overlords.
While this is amusing, it doesn’t likely work very well. If the simulators are simulating even a small number of humans to a high fidelity, the path finding issues would be very computationally small in comparison.
This only works with half-assed early 21st century A* pathfinding. If we are in a simulation good enough to simulate general intelligence, pathfinding is a non-issue.
Well, that depends on what the level and extent of the simulation is. For example, if this is a video game, most people might be computationally simple NPCs based on legacy code.
Despite many orders of magnitude more processing speed and memory, we still can’t get companions to walk through doors properly. If engineering taught me anything, it’s that every job is done half-assedly and will stay that way as long as it possibly can.
Or we might be in half a simulation, where player AI actually comes from outside the world (like in all our games), but NPCs are handled via (flawed) AI. I’m not sure how I could easily tell the difference between the two. If you’re serious about annoying the simulators though, going off the rails is probably more interesting than finding implementation bottlenecks, sure.
You are postulating a non-reductionist world in which you are the only or one of a small set of agents with actual minds. I think that’s wrong.
If current science and philosophy tell us anything about our simulation, it is that it is very detailed and low-level.
(Note, just to make it explicit: I’m not really taking the whole simulation cluster of ideas serious. It’s just that common simulation hypothesis assumptions about necessary complexity and low-level simulation seem quite implausible.)
Correct. Maybe not strictly non-reductionist, but for all practical purposes, yeah.
“Us”? Are you running a particle accelerator? All I know is that there exists somewhat-consistent literature to back up a certain vast, low-level interpretation of my actual experiences, and even then I have no idea how the (certainly plausible-seeming) ideas about physics get me to an explanation of minds or consciousness at all, so if they are all wrong or massively incomplete, I wouldn’t know.
That might be because there is a vast, detailed universe that follows relatively simple and consistent laws. (I certainly favor that interpretation.)
It might also be because someone designed a plausible physicalist setting for a game and presented it to me in a convincing way. I don’t think that’s actually very hard to do.
For example, I regularly have lucid dreams that have a similar level of detail as my waking state (even though I know I’m dreaming). It looks the same, people act realistically, I can talk to them. I don’t think my brain simulates any kind of detailed physics for that. The illusion doesn’t last very long, sure, but scaling it up still seems simpler than getting a quantum physics implementation to work properly.
I’m not saying physics is a hoax, but it would certainly be much easier to bullshit me into believing that I’m living in a detailed physicalist universe than to actually simulate even just one planet of it. So if you want to complicate the simulation, you maybe shouldn’t even assume that there actually is a quark layer or anything like that.
Of course I know this to be incorrect. but of course I would say that rite? :)
Ok good point. So then if we are in a simulation, we can probably fuck with it.
I had some good ideas on this the other day, but I was thinking of it from the acasual trade angle instead of the “universe is sim” angle. One was you should entangle your mental state with some really hard to compute stuff, like some crazy multiphysics with convective fluid flows. That only works for acausal trade, because you are being simulated to find out what you would do so the simulator is constrained to be accurate. In the general case, they can just bullshit a result.
If they are doing a simulation tho, it will be because they have some question that they can’t just bullshit. The trick would be to find out what that is and do some crazy high-load physics around that.
Why is that in your interest? Is this just because it seems fun?
It was in counter to some musings I was having, along the lines of “Is there anything I can do to make myself more interesting, in the sense of being more likely to be re-simulated?. If I tried, I could probably arrange to get myself namedropped in a published novel by a prizewinning authour, and thus anyone who wanted to simulate that particular fictional universe would also end up simulating me, as well. But do I really want to arrange my life to suit the hypothesized interests of future simulation-creators?”
If you are living in a simulation and the anthropic principle holds, then yes, you probably do want to arrange your life to be interesting.
I’m aware of that line of reasoning; what I’m proposing is a deliberate rejection of some of it’s assumptions.
There are several mythic precedents; if the universe is arranged a certain way, then at least from a certain perspective, the best course of action is to sit down, shut up, and dance to the head of the pantheon’s tune. But looked at another way, greater rewards can be found in the dignity of rebellion and choosing one’s own path, even if it means getting cast down or having your liver eternally gnawed upon—or being un-simulated.
Being un-simulated, or death in a materialistic/naturalistic reality, is probably the least scary future there is. If you don’t exist, then you can’t feel pain, remorse, or even regrets.
Precisely (or at least arguably) - and that’s part of the point.
I’ll try describing this a different way: What actions can we take that would have an effect on the actions of a simulation-builder? What can we do that would make them more likely to not create a simulation at all, then to create a new one containing copies of ourselves who are ignorant of the builder’s level of reality?
If it’s in my own best interests to have a few future copies of myself as possible placed into simulations to dance to the simulation-builder’s tune, then how can I nudge the probabilities so that future copies of myself tend to be placed into my preferred sorts of environments?
Something related that I posted a long time ago:
(Taken from http://davidbrin.blogspot.com/2005/10/holodeck-scenario-part-i.html)
...
David Brin’s answer is here, along with a way to ruin the player’s fun: http://davidbrin.blogspot.com/2005/10/holodeck-scenario-part-ii.html
Most of John Dalmas’s novels are based on our living in a “video game” universe, where we are purposely limited versions of our more powerful selves, to make it more interesting. It makes as much sense as any other metaphysics.
The idea is explored most explicitly in The Reality Matrix and in The Playmasters.
A character in Greg Egan’s Permutation City shoved into a low-fi simulator takes a very long shower and carefully looks at the water rivulets to annoy the simulation owner, since the sim only models things around him he’s looking at, and modeling turbulent hydrodynamics is very computationally expensive compared to regular solid objects.
Taking long walks in forests and paying as much attention as you can to the very dense amount of visually discernible fractal details around you would be a similar thing. Or in general hanging around anywhere where there is much rust, dirt, decay and other halfway-entropy making things, but things haven’t gone completely homogeneous like in a desert.
Most people alive are already living in pretty entropic environments compared to a spotless empty rooms with all straight angles and perfectly smooth white walls, so this might not inconvene the simulator that much.