Re: ...The evolution of Stellar Based Superminds...

Date: Wed Jul 19 2000 - 18:33:13 MDT

In a message dated 7/19/00 6:14:51 PM Central Daylight Time, writes:

> Greg makes some points -- lets partition them.
> 1) Remote colonization by "unconscious", "undivertable" sub-SIs.
> In this case, I think it comes down to the problem of "remote"
> matter (or energy) is of much lower value than "local" matter
> and energy. It isn't the cost of getting the probes over to
> the remote star that is the problem, it is the cost of getting
> the remote matter & energy *back* to the local star where
> it is of value to you. <snip>

Fuggedaboudit. I assume moving matter over interstellar distances is a
no-starter in terms of net economic benefit from getting more atoms. Any
technology that could make it pay to go get atoms from other star systems
would make all the rest of your (and my) projections in this realm

> Even if you don't send back matter & energy, presumably you do
> communicate and the communication costs will be lower if you get
> closer. How close you get will be determined by the cost to change
> your orbit relative to the volume of communication you would
> like to do.

It depends on what you want to communicate. Consider how many people devote
themselves to the study of history. They know they'll never be able to talk
to Thomas Jefferson or Cortez or Chuang Tzu, but they spend a lot of time and
energy to this "one-way communication". Getting bits from a daughter SI
would just be INTERESTING. Carrying on a two-way conversation, albeit VERY
slowly (especially by the standards of the fastest computations possible by
SIs), might also be rewarding, although in a way we can hardly imagine.

> 2) Colonization by unhappy sub-SIs looking to expand.
> In this case, you have to make a case that the sub-SIs are
> independent enough from the SI that they have sufficient
> personal freedom and autonomy to accomplish this. England
> certainly didn't let the colonies go "willingly".

No - but it STARTED the colonies willingly. And I assume we can do better
than apply the kind of mercantilist economics that held sway at the level of
"national policy" in the First Age of Expansion. After all, the motivations
of most of the COLONISTS were quite different from those of the "Mother

> Presumably
> the SI keeps sub-SIs either happy enough or relatively
> powerless so that any sub-SI probe/colonization ships
> that get sent out get zapped before they get very far.

OK. Assuming a "Singleton" to use Nick's term. I'm not at all sure that's a
certainty, but I'll accept it for the sake of discussion.
> Remember that the more information content the sub-SIs
> intend to take with them, the more matter or energy is
> going to have to be involved. Hey -- "What is that
> Galaxy Class Starship you are building there?" "You
> are only authorized to build "Delta Flyers!" As soon
> as the SI gets wind of any solar-system changing
> activities in a nearby star, you can bet it is going
> to investigate. If SIs have the ability to dominate
> local regions of space then the sub-SIs have to go further
> away (taking longer or using much more energy, making
> them much more noticable).

I don't think that "information content" is THAT big a factor in the mass of
a colony seed. Distance probably is, if for nothing other than shielding.
But again, I'm assuming an almost "dormant" seed, so the information can be
packed pretty densely.
> If you have a moderately open society, SIs might let sub-SIs
> leave, but presumably there would have to be some very interesting
> treaty arrangements for future allocation of interstellar resources.
> (When you come down to the last 4 H atoms in the universe, who gets
> them?)

Are you sure you weren't a Spanish colonial official in a former life,
Robert? Consider how difficult it would be to police such "treaties". You
have to have what amounts to a Galactic Armada to do that. Makes for good
space operae . . . but I think it's much more of an economic fantasy than
colonization in the first place.
> It seems the only cases where colonizers have let colonies go
> freely has been done is where the cost of administering them
> was lower than any benefit they could provide or when the
> colonies unexpectedly overthrew the administrative system
> and grew so fast that the cost of winning exceeded any
> long term benefit they might provide (from the perspective
> of an analysis done at that time). It would be really
> interesting to study all of the colonizer/colony cases,
> their relative economic, military, etc. strengths & weaknesses,
> abstract them to some basic economic principles and see
> how they apply vis-a-vis the SI/sub-SI relationships.
> You get into questions like, "Would the American revolution
> have succeeded *if* America had been where France is?"

I'm sure there are lots of folks who have done some modeling of colonial
relationships, but I don't think such models would be of much use (within the
solar system, perhaps - I'm not totally convinced that there will be a sudden
"flash" from essentially no intelligence off Earth to a solar-system-scale SI
- but that's another question).
> If the pilgrims has been "ants", I don't think King George
> (a human) would have had anywhere near as many problems with
> the colonies. (Insert cartoon of King George stepping on an
> ant hill).

I think that's really a more apt metaphor, because the "seeds" I'm imagining
would be to a true SI far less than an ant is to a human.
> 3) Colonization choices
> There is a final aspect of this that I have been working on
> which revolves around the question of if you have the freedom
> to colonize, "What do you colonize?" It turns out that
> brown dwarfs and M-class stars are *much* more abundant
> than the stars we can actually see. Brown dwarfs particularly
> are going to be easier to dismantle because they are smaller
> and cooler (you can burn their H in thermonuclear reactors).
> But so far my very rough calculations suggest that they
> may take a long time to dismantle (perhaps millions of
> years). If that is accurate, Robin's theories may be correct
> but the colonization wave may proceed very slowly because
> the time constraint *isn't* light speed, but time it takes to
> dismantle large amounts of matter in gravity wells. So
> large numbers of the Milky Way's Brown Dwarfs may be in the
> process of being dismantled to contribute to the MBrain
> population. That is interesting from the perspective of
> what the Microlensing Observations *did not find* --
> no rogue planets and no brown dwarfs! Our galaxy *may*
> be being consumed and the tasty bits are going first.
> The tasty bits *aren't* the closest ones but the ones
> you can dismantle the fastest.

But I think Robin's model doesn't assume that the colonization wave waits on
full development of each node. Rather, low-cost seeds get launched as soon
as a node is developed to the point that it can launch a replicating seed.
Obviously, this parameter is crucial to the question of the "burn rate" of
the wave.
> I suspect there may be a frontier beyond which it makes
> absolutely no sense to quickly return any matter and energy to
> the parent because the transportation costs are *so* high.
> It may be better to send out a probe, construct only what
> is necessary to give a brown dwarf or star a gentle nudge,
> setup a "this system is the property of SI 293203 beacon"
> (and berserker drones if necessary), then wait 10,000,000
> years for it to show up on your doorstep rather than wasting
> it all just so you can get your next meal a bit earlier.

It seems like the slow accumulation of so much matter would have
galactic-scale gravitational effects - which would be observable.
> (Keep in mind that if MBrains switch over from stars
> as their primary energy source to thermonuclear reactors
> (as R.F. argues strongly), then they can control their
> thought (clock) rate and consume energy at exactly
> that rate that guarantees they run out just when the
> next meal is expected to ship in. (Of course taking
> into account possible disruptions due to pirate
> sub-SIs.)

That would actually, it seems result in a fairly complex game-theoretic
analysis: How much resources do you devote to "home" computation and how much
do you devote to policing of your mass-gathering activities. If pirates are
common, you'd want a big navy.

This archive was generated by hypermail 2b29 : Mon Oct 02 2000 - 17:34:54 MDT