[ExI] Computronium planet.

[Anders Sandberg]

On 12/10/2012 17:52, Giulio Prisco wrote:
> Recently a friend observed that if our universe is the fastest machine
> able to compute itself (this assumption seems necessary to avoid
> causality violation paradoxes), then our matter is _already_
> computronium, and we just cannot squeeze more computing power out of
> it.

It is not obvious that this is true. Sure, you could argue that if a 
subset A of the universe can calculate itself faster than A, then you 
get a paradox. But it could be that due to limits on measurement 
precision (due to no cloning theorems and similar stuff) subset A can 
not reliably be programmed to calculate subset A, just some random 
version A' - and it is likely that computronium has a  particularly 
chaotic dynamics, so the evolution of A' is not going to be like the 
evolution of A.


-- 
Anders Sandberg,
Future of Humanity Institute
Philosophy Faculty of Oxford University