Re: NANO: Institutional Safety

Rik van Riel (riel@nl.linux.org)
Mon, 15 Nov 1999 15:22:55 +0100 (CET)

On Mon, 15 Nov 1999, Robert J. Bradbury wrote:
> David Blenkinsop <blenl@sk.sympatico.ca> wrote:
>
> > or for using them to build an overwhelming force of arms.
>
> In a nanotech environment, the concept of an "overwhelming"
> force of arms is very questionable. You have to guarantee
> that you have disassembled *every* last little bit of nanotech
> in an enclave that can have berserker potential.

Then there is the question _why_ you would want to build an overwhelming force...

If you have a number (say 20) of economically equal consortiums rivalling with each other, they'll all do what they can to keep their part of the pool.

This includes making sure that the others play by the rules and that no consortium grows too strong. Since every one of those consortiums will usually have 19 others opposing it when it doesn't play by the rules, there's a pretty good incentive to play it safe.

A feudal system like the one in the Dune books also seems like a good candidate for a semi-stable government system. When light speed becomes a barrier in the speed decisions are taken, the current system of decisions-by- instant-information won't work any more.

We need a system that will keep on going, even when the individual consortiums are in fierce competition and unable to quickly exchange information with each other.

If FTL travel will ever become possible, we might even end up with the situation where a weekly postage shipment will be faster than radio transmissions. Just think about what that would do to our current information age thinking!

The ideal goverment system would be strong enough to keep going independantly of the will of the players.

> > No fighting over the resources of asteroid Ceres for them,
> > their Coalition has voting Ceres a mineral preserve, or
> > something, and that is that!
>
> If you presume that we can be as "rational" as possible, then
> it makes sense to cooperate (game theory says this is most
> efficient).

War is already economically unfeasible. It's been a few hundred years since we last had a war where the winner also came out with an economical benefit.

As more and more countries are converted to market economies and the fundamentalists can no longer compete (economics will always defeat idealism) wars will cease to be commonplace.

With wars, we have more and more to lose and less and less to gain. Currently this is because the information age has caused our economies to become very much interlocked with each other.

In the far future, when/if the light speed becomes a barrier to instant communications and the exchange of information like we have now, wars will be economically unfeasible because the various parties involved will all be so distant from each other that the agressor will automatically lose its reputation with the others and will be faced with distrust from the others.

Also, such a spread out mankind will make it impossible to form superpowers. That is because the communications delay inside such an empire will be too large to have every subpower in such a coalition react as one or even have them have the same opinion on things.

Coalitions like that have failed to do that even during the cold war; the Romans and Gengiz Kahn (sp) have shown us that the maximum feasible communications delay between different parts of an empire is about one week, with the time to military intervention in those remote areas having a maximum of about 4 times that.

Nanotech could make the restrictions even stricter, with the time to react being shorter, the gains to be smaller and the risks becoming so large that nobody in his/her right mind would ever risk fighting a war.

The threat of nuclear mutual annihalation was enough to stop us from beginning a third world war. I believe that with nanotech the balance will tip even more into the risk-avoidance zone.

(damn, I should work on my english ;))

> You don't but a sovereign state can't "nano-nuke" another state
> if it potentially has hidden berserkers or strong allies. You
> only get to be "untrustable" once,

Indeed. When a cup full of nanites could be enough to destroy a major part of a civilisation, who will be taking that risk?

> I don't need the surface of a planet covered in berserker bots to
> feel safe, I only need a few hundred of them scattered in very
> different locations. Or you simply need strong allies.

Every consortium should have diplomats stationed at the others, each having a bunch of bezerker nanites with them, to be unleashed when the civilisation they are stationed has crossed the limit.

Could the threat of mutually assured destruction be any stronger than this?

> > Basically, we need the Coalition of Polises to come around and just
> > bomb the heck out of anyone who violates the Neutral Zone conventions --
>
> Well, say you are one of the powers -- what constitutes a good
> development plan for the available resources and what constitutes
> a bad development plan

A good plan doesn't waste resources (in the eyes of the others) and doesn't threaten the others. As soon as one consortium starts implementing a plan that endangers the others or is too wasteful with resources (again, in the eyes of the others), it will be curbed.

Rik

--
The Internet is not a network of computers. It is a network
of people. That is its real strength.