Re: poly: Malign Probes

 hanson@econ.berkeley.edu (Robin Hanson) writes:
>> I've been assuming that detection is more extensive than just probes
>>defending their own systems. They are watching several neighboring systems,
>>and can at very least detect most objects whose velocity exceeds the
>>escape velocity of those systems.
>
>First, why should probes have to come from neighboring systems? If probes
>can come from 1000 light years away, you'd have billions of star systems to
>have to be keeping track of.

 A probe that tries to come from 1000 light years away gets destroyed by
another system near to its launch point.

>I thought the whole intuition behind needing to hide was that you just
>couldn't tell where attacks might come from. A small antimatter probe
>could come out of no where and blow you away. If you can see where attacks
>are coming from, you can more plausibly defend and retaliate, making hiding
>less needed.

 I'm assuming that defense is not effective, even when it is possible to
identify the source of the attack (as in mutual assured destruction with
nuclear missiles), and that retaliation happens (the value systems of
most probes place destruction of expansionist life above self-preservation,
and self-preservation above other kinds of destruction).

>> As for interception, I'm assuming that when a system sends out probes,
>>one or more neighboring system sends out probes which follow and destroy
>>the first set of probes plus the system that sent out those probes.
>> I'm not sure whether to assume that the retaliatory probes also attack
>>each other or whether they have some way of recognizing each other's
>>retaliatory intent.
>
>If systems are willing to not attack a system that sends a retaliatory probe,
>why would such a system need to hide? You could take apart planets & star,
>as long as you didn't send out non-retaliatory probes. If you're hiding from

 I think the "retaliate on launch" strategy may be the only criterion
needed for malign probes to be stable. To also explain the absence of
visible Dyson spheres, there must be an additional criterion sufficient
to cause retaliation, something like "industry in that system is producing
more than X joules of heat; if I don't launch a first strike soon, they
may acquire the ability to send out more probes than I can destroy".

>neighbors, you should fear they would detect you when you send retaliatory
>probes.

 Either detecting the existence of life is not something that triggers
retaliation, or retaliation triggers a widespread burst of mutual
destruction.

>> I think you're implying that such detection systems and retaliatory
>>ability need to look conspicuous. I'm assuming that as long as they are
>>passively waiting for signs of expansion, they are easy to camoflage.
>>Setting up such systems undetected is hard, and only succeeds in the
>>rare case where life is able to colonize a new system by correctly guessing
>>that there are an abnormally low number of malign probes in neighboring
>>systems, causing expansion to be safer than normal.
>
>If stars which had such systems were rare, colonized systems would be rare,
>so there would be lots of systems out there for the taking. So a fast
>colonize, reproduce, and get out of there before the bombs come strategy
>could be feasible.

 forrestb@ix.netcom.com (Forrest Bishop) writes:
>> As for interception, I'm assuming that when a system sends out probes,
>> one or more neighboring system sends out probes which follow and destroy
>> the first set of probes plus the system that sent out those probes.
>> I'm not sure whether to assume that the retaliatory probes also attack
>> each other or whether they have some way of recognizing each other's
>> retaliatory intent.
>
>In this scenario, the probes seem to be identifiable by their
>trajectories.
>Do recognize that is is very difficult to alter a trajectory in free
>space.

 When I talk about retaliatory probes attacking each other, I mean their
home base sending out additional probes rather than a single probe trying
to destroy additional targets (I'm inclined to assume that they are destroyed
in the process of their retaliation).

-- 
------------------------------------------------------------------------
Peter McCluskey  |  pcm@rahul.net  | Has anyone used http://crit.org
http://www.rahul.net/pcm           | to comment on your web pages?