Hacker News Reader

Comment by cwmma | Hacker News Reader

A lot of these experiments start from a premise that the simulating universe has similar physics to the one being simulated and thus any short cuts might be noticed, but the outer universe could be be more have more complex (or just different) physics so the fact that ours doesn't match up with theirs doesn't mean ours isn't self consistent.

Sort of like how in Permutation City they create a simulated universe via cellular automata and then when they tell the residents of that universe they are simulated they don't believe them.

View on HN

Yea. I don't see how it isn't a reductio ad absurdum either. If we're in a simulation then we know nothing about the physics or structure of 'the real world', including whether things like galaxies, planets, and evolution of advanced civilizations even exist or can exist.

Everyone is not making that assumption. Asking if photona behaves more predictably while not one is watching doesn't assume there is light in the parent universe. Similarly the article opens up the question that the root universe might have infinite computing power! That would be a very different universe indeed.

The simulation has probably a purpose like solving (or understanding) issues in the outer universe. It would be reasonnable to build the simulated universe as similar as possible to the outer universe.

That's a big assumption.

What purpose does Goat Simulator http://www.goat-simulator.com/ serve to solve/understand issues in our universe?

How do you know our universe is not a child's toy for a race of deity-like beings? Or an art project?

Maybe the outer universe is populated by immortal godlike entities and has the equivalent of the guys who made Dwarf Fortress saying, "hmm, I wonder what would happen if there were a race of pseudo-intelligent creatures that just arbitrarily died after just a few million seconds?" Or, "I wonder what would happen if the fundamental building blocks of the universe did not behave predictably, but instead made a random choice to determine the outcome of every interaction?"

And maybe that programmer is just about to come back from his coffee break, look at what happened, and say, "hmm, that was boring", kill the process, and move on to other things.

When you're talking about something that could be powerful enough to simulate our entire universe, assuming that our sense of scale matters at all to it is pretty much the definition of hubris.

Also, time may not be relative to "outside". For all we know we might be in worldgen (historygen?) phase for a "Humans vs Aliens in the Year 50k" game.

There's a question as to whether ethics/morals/compassion scale with power. I think there's a solid argument to be made that human power has scaled near exponentially over the last 200 years while human ethics seem to be coupled to it (slavery and genocide are almost universally reviled where they were common place before, per capita violence is way way down, international and intercultural cooperation via trade, communication, etc have never been higher).

Are the two things tightly coupled? Seems difficult to prove, but if they are, they folks running our simulation are probably going to treat us a lot better than we treat each other.

Maybe, maybe not. The problem is (as parent posits) that a hyper intelligence with the means of simulating an entire universe could just be so utterly alien to us that we can't make any predictions about it's reasoning. Perhaps they would turn off our world/simulation with no more of a thought than what a human would give about autoclaving a petri dish with some bacterial growth.

It's only human to try to ascribe human feelings and morals to such beings - but the difference in scale is just too big. Comprehending why such a being does or doesn't do something would be almost certainly beyond us.

Seems trivial to disprove.

Dictators have lots of power and don't follow ethical rules you assert.

People destroy colonies of termites.

Nobody worries about millions of yeast destroyed when baking a batch of bread. Sorry, hundreds of billions. My bad.

> It would be reasonnable to build the simulated universe as similar as possible to the outer universe.

... said one Tetris block to the other.

I would say the world is full of simulations, and while many will try to simulate aspects of the real world, there are just as many simulating completely or at least hugely different things from our world.

At some point a simulation may have the goal to understand the existing world. But once that field has been harvested enough and simulations on that scale have become a commodity people would probably move on from the things "that are", to the things that "could be", or are even impossible.

I work in ATC simulations. After simulating usual aircraft (like Airbus and Boeing), we try to add models for helicopters or seaplanes. Nobody is asking for simulation of flying carpets or UFO.

UFOs would make a hell of an Easter egg, though.

It would be reasonnable to build the simulated universe as similar as possible to the outer universe.

Even if that was the case we still could not tell the difference between a real universe and a simulated universe. We look at our universe and observe X, then what? Are we in a simulated universe and in the real universe something slightly different X' is true? Or is X simply true in the real universe in which we live? [And in a galaxy far away some aliens are running simulated universes in which something slightly different X' is true and observed by the simulated aliens?]

Reply on news.ycombinator.com