[ExI] Feasibility of solid Dyson Sphere WAS mbrains again: request

[Eugen Leitl]

On Thu, Sep 29, 2011 at 03:45:09PM -0700, spike wrote:

> Well said Anders.  I can imagine an MBrain would master nanotechnology, and

Computronium is nanotechnology. You have to speak machine-phase in 
order to be able to make the real thing. Self-assembly is *at least*
an order of magnitude less efficient in terms of volumetric circuitry
concentration and thermal stability/ability to dissipate heat from
the bulk. The difference is far more than between virus crystal 
versus paulingite.

> thus would be capable of self modification as it discovered better and
> better ways to reconfigure itself. 

There is only one way to configure yourself optimally if you want
to maximize computation. However, co-evolution can and does produce
nonlinearities which will be surprising.

Apart from sexual selection there's the issue of the fitness function
being mostly other individuals. I don't see how you can abstract
that from the physical layer, so things will get weird there.

If there's one sure thing we can predict about the future it's that 
it's going to surprise us.