Saturday, 4 June 2016

The Matrix of Musk

The ever ingenious Elon Musk claimed that "There's a billion to one chance we're living in base reality,"  on stage at the  Recode's Code Conference,and that it is "overwhelmingly likely we're just characters living inside a simulation". I'm glad we both enjoy the same type of science fiction - first "Foundation" and now "The Matrix"!

However, if that is the case it mean that his life's work, especially his excellent achievements with SpaceX and his Mars colonization thoughts, are essentially wasted. If we are in a simulation - an idea considered by Bishop Berkeley in the 18th century and perhaps more practically by Plato's Allegory of the Cave, then achievements in this world would essentially be hollow and arbitrary.

If you do actually act on such beliefs it dangerously leads to solipsism  - so for example, being a sociopath is a perfectly reasonable and justifiable option (after all other people are not really real and don't have feelings of lives in a solipsistic world view) and any evil inflicted is just part of a game.

Alternatively  one's own efforts could be much better put into hacking the sim - either to find out more about it or if you were resigned to existence in this universe to 'break the game rules' to achieve knowledge, immortality or faster than light travel. Just when I see a new game can't help but try to work out what the underlying algorithms are and then reach for a hex editor. To be consistent, now, Elon needs to start thinking this way about the universe.

So, if we were in a simulation, how would we know? Well, simulations fall into two broad categories, complete and incomplete. In a complete simulation every action, every item, in simulated in as high or higher fidelity than the laws of physics being simulated can exist. There are no approximations. An example is a chess simulator which fully simulates the laws of chess. In this sort of simulation all observed features emerge following rigorously applied laws from starting conditions.  Complex features of the universe are not explicitly coded. There is no stating condition of complexity (like a flower or a 747) at the start or 'creation'.

In an incomplete simulation compromises are made. Typically but sometimes whole sections of the simulated system are approximated. Almost every simulator which we are familiar with falls into this category. Games and flight simulators are good examples. If we look closely enough at the screen, or indeed any object we can see pixels, building blocks of the simulation. In a complete simulation these should be subatomic particles. If we follow a timeline back in "Call of Duty" or indeed in a flight simulator, we get back to the start of the game or exercise. Everything before that is assumed and programmed into the simulator. In a complete simulation we would go back to see the protagonists breakfast, his or her birth, the emergence of their species, the origins of life etc. . .  A simulation of this type would be like the one in the film "The Matrix" or like the belief that God buried the fossil bones of dinosaurs in the rocks in order to test the faith of sinners. Observed features of the universe would be explicitly coded rather than emerging following rigorously applied laws.

So what is Musk's hypothesis - that we are in a complete simulation, or an incomplete one?
Observed features of the universe

  1. Observed features of the universe emerging following rigorously applied laws. (and no granularity of simulation coarser than the in-universe features (like quanta) that it simulates). In that case we simply can't tell, we can never tell, and it doesn't actually explain anything (as we have to ask - does the simulator have to follow 'laws' of it's own - and if so is the simulator being simulated?)
  2. Observed features of the universe explicitly coded rather  emerging. This is testable and it would be interesting to formulate the series of test that would allow us to recognise it.



1 comment:

  1. Very fascinating topic.

    As a sci-fi loving geek designer of a zero-g space simulator on the one hand and spiritual bunny-hugging skydiver on the other I have developed a faith-based position that working for "God" (or simply the lighter side of the darkness) rather than my "self" produces better results .. well, at least it seems to at least attract help.

    Although it probably pays to act as if both base reality and simulation^level^n are true (finding non-zero winning conditions of being that work in both/all .. such as "be kind", "be good", "do no harm") my own gut-feel is that I should act as if this is base reality and take some level of responsibility for at least keeping it stable for later generations to also "play happily" in.

    I like to imagine we're a distributed network of emergent solipsistic "infant"[1] goddesses and gods, at least in relation to all other life and thus need to learn to be nicer god to ourselves and all other life. The only ethical reason to hack the system that I can think of is to try reduce suffering and increase happiness.

    [1] https://en.wikipedia.org/wiki/Solipsism#Solipsism_syndrome

    Musk also really likes The Culture series. My own favourite is "The Player of Games".

    Dr. Mark: "Just when I see a new game can't help but try to work out what the underlying algorithms are and then reach for a hex editor."

    Okay, challenge! Here's a new one for you to crack ;)
    https://www.facebook.com/SpaceBallUnity3D/

    Latest builds here: https://tinyurl.com/y7st2bt5

    ReplyDelete