Eliezer S. Yudkowsky wrote:
> To blow up a planet? Okay, I'll accept that. But what about
> bootstrapping? I waft in a dust particle from space, a few nanobots
> onto the surface of the ocean; they build more nanobots and
> drop to the
> bottom; then they reproduce a few times and build a few bombs. As for
> power, nanobots float to the top, absorb solar radiation, sink back
> down. If that's necessary at all, of which I'm not convinced..
That might work, depending on how alert your neighbors are. Certainly, the initial dust would be hard to spot. Once you start building bombs you get a lot more detectable - you need macro-scale equipment to separate deutrium from sea water, and probably particle accelerators to make tritium, and the bombs themselves are pretty big. The whole operation will eat up a lot of power, and there isn't much to work with down there - again, no free oxygen. Nuclear power would work, or a really big swarm of your solar collectors.
Hiding all this activity is a bit iffy. The problem is you're burning lots of power in an environment that doesn't contain any large, random energy sources. A good sensor network will spot you through indirect environmental effects - your operation will have a measurable effect on ocean currents, surface temperatures, local wildlife populations, water chemistry, and probably all kinds of other things I haven't thought of.
> One thing is for sure; if you want an active defense against
> that, you'd better be prepared to sterilize the oceans..
No, you just need an early warning system. Populate it with small (mm of smaller) chemical sensor platforms (capable of recognizing nanobots), with a network of larger bots operating as control & communication centers. You won't catch everything, but you should spot a growing swarm before it gets big enough to be a real threat.
> (This is pretty much academic; by this point in 'tech any
> sane person is far, far away from the Earth or any other mass not under
vis
> control, and has a laser diligently vaporizing stray hydrogen atoms.)
Or the SIs are running everything.
Billy Brown, MCSE+I
bbrown@conemsco.com