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

[Anders Sandberg]

Dennis May wrote:
> Spike wrote:
>  
> > The MBrain is a target, in a sense, but certainly not a helpless 
> target. 
> > This one has a built-in speed of light self-defense system.
>  
> Been through all that kind of discussion before.  Whatever defenses
> you build the resources outside to overcome them can be as large
> as it takes.  Beams of diffuse anti-matter dust are my favorite.

It is trivial to overcome any defenses by simply postulating even bigger 
attack resources. But cost-effectiveness will get you.

In order to damage a 1 AU shell composed of 1 m^2 diamond hard objects 
each object need to be hit with (say) the explosive force of 0.1 kg TNT, 
or around 2e-9 kg of antimatter. A circular beam with a diameter 2 AU 
with one such pellet per square meter (which would of course only 
seriously damage part of the sphere) would need 1.4e14  kg of 
antimatter. That is a biggish mountain of antimatter you need to make. 
And for an entire surface attack you need 4 times as much.

I can see a Dyson shell brew up that much antimatter in a few weeks, but 
anything lesser would find it tough. And if you already have a Dyson, 
the opportunity costs of spending your energy on antimatter to mess with 
the neighbors (aeons in the future, from the subjective timescale of the 
fast processes online) will be very high. You better have a very good 
reason, especially since it is pretty obvious who is responsible.


Mbrains will no doubt look quaint as we get ever better ideas of how to 
arrange matter to do useful things, but so far they represent one of the 
more well analysed development directions. If you find them silly, why 
don't you propose a better approach and show us your engineering?

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