[extropy-chat] fermi's paradox: m/d approach

[Spike]

The M/D approach to Fermi's Paradox

Over the years we extropians have pondered the question Fermi so
eloquently posed in three words: Where are they?  Carl Sagan speculated
in the 1970s that evidently intelligent life was short-lived, ordinarily
self-destructing soon after reaching the technology to communicate over
cosmic distances.  That argument gradually becomes less compelling as
humanity manages to survive into a seventh decade with nukes in our
midst.

As a result of our discussion of Sbrains last week, I have stumbled upon
another explanation for the cosmic silence: we aren't worth the mass to
send us signals.  I will call this argument the M/D approach to Fermi's
paradox.  It works like this:

Suppose that life manages to evolve, and thru some torturous path, makes
it to sentience.   Certainly this is a trivial treatment, few words for
a stunningly difficult phenomenon, however it is evidently possible, for
clearly it happened on this planet.  In this vast universe, with its
hundreds of billions of visible galaxies, each with its hundreds of
billions of stars, each with planets, everything that can happen must
happen.  We have shown that intelligent life can happen.

Today there is no convincing mechanism that would cause technological
progress to stop.  Gray goo is something to worry about, but nature has
had billions of years to stumble upon it.  During that time, she has
derived so many wonderfully complex designs.  It is reassuring that gray
goo was never present in this vast cornucopia of life.  Perhaps there is
some fundamental reason why it cannot happen, some reason why goo cannot
compete with current lifeforms, or is somehow incorporated into current
lifeforms, like mitochondria.   

Even if we nuke ourselves into the stone age, recall that the stone age
lasted until only a few thousand years ago.  Humans are tremendously
adaptable, and many already live in places that would not be worth a
nuke, should all the nuke capable powers let loose with all they have.
Africans would survive, Aborigines would survive, there would be pockets
of humanity everywhere that would carry on.   Even if that horrifying
loss of life and technology nightmare scenario were to come to pass,
recall that all the really important technological advances occurred in
just the past few hundred years.  All-out thermonuclear war would merely
be a temporary setback for the singularity.

If a singularity, then an MBrain.

This is my contention: that artificial intelligence wants to THINK.  It
lives to think.  It is smart enough to make things happen: it knows how
to build things.  If it likes to think, then it wants to get all the
available material thinking, so it builds an MBrain.  It gathers all the
metals in orbit about the star and converts it all to whatever form
maximizes thought.  It is not clear to me what that form is, however if
we assume there is an optimal use for metals, some minimum energy-use
and materials-use configuration (define this as computronium), then the
AI would convert all the metals available into computronium.

So far, all this has been argued before.  The new thought is that the
computronium is optimized by having it collocated as close as possible,
but not more so.  A Sandbergian JBrain is a Jupiter sized sphere of
computronium that exists as a planet.  This would seem to require that
some of the material is under tremendous heat and pressure, even if the
JBrain is spun rapidly and formed into a flattened disk.  At the other
extreme is a Bradburian MBrain, which is computronium that exists as a
large number of particles in orbit about a star, in such a way as to
collect as much of the energy from that star as possible.  This may be
less than optimal for it separates the nodes over greater distances,
thus increasing the latency, or time required to communicate between
nodes.

The M/D approach argues that there is a configuration somewhere between
the JBrain and the MBrain wherein the computronium is separated, but not
by too much.  So it exists as a planet, like a JBrain, but is
technically an MBrain: all SBrains are Mbrains.  A given star system
could even have more than one SBrain.  Commentator Lorrey has suggested
that the proposed SBrain looks like a nautilus shell.  Very well, SBrain
means Shell Brain.  SBrains form to allow all of the available material
to be computroniumized, and M/D simultaneously is maximized, thus
optimizing thought potential.

The SBrain is more specific than the MBrain.  The insight here is that
in optimized computronium, energy is no longer the critical resource,
this being a diffuse form of matter indeed.  Matter and time are the
most valuable resources.  Energy from the star can be allowed to escape,
wasted, lost forever into the cosmos, for there is plenty of that.  But
time cannot be wasted, for heat death is coming to all. 

To an SBrain, the existence of another Sbrain in another orbit is a
valuable resource: the other SBrain might have nodes that are thinking
about some of the same things that are being pondered by one's own
nodes.  Therefore, communication between SBrains would be mutually
beneficial, as communication between humans is considered valuable
enough that we build expensive satellites to make it happen.

If the other SBrain is far away, then of course the communications are
restricted.  The time delay for trading ideas increases linearly with
distance, and the energy required to send the signal increases as the
square of the distance.  So one can speculate that the value of signals
from another SBrain is proportional to the other SBrain's mass and
inversely proportional to the square of its distance.  If two SBrains
managed to form in orbit about the same star, they would soon see the
benefit of merging.  By this argument, the smaller SBrain would move to
join the larger, for the value of signals from the larger is greater
than the reverse.

We wondered if current instruments would able to detect an SBrain.
Current exo-planet detection is based on gravitational wobble caused by
massive planet.  Of course, we could not distinguish between an SBrain
and an ordinary gas giant, or even a large dead rocky planet.  We can
imagine a situation where a star system like ours had an SBrain of mass
about 3 earths with the gas giants unused.  With Mercury, Venus, Earth,
Luna, Mars and the asteroids, we have around 3 earth masses of rocky
material to make an SBrain, but the gas giants, being largely hydrogen
and helium, may not be as useful in building such structures.  So our
instruments would not be able to find the small wobble from a 3-earth
mass, the signal being swamped by a 300 earth-mass gas giant.

Early in this post, I promised to suggest a solution to Fermi's paradox.
Enough background has now been presented.  If there is an energy cost to
sending a signal that is inversely proportional to some function of the
distance, and the value of the incoming signal is proportional to the
mass of the distant SBrain, then it could be that the distant SBrain
would decide that communication between star systems is not worth the
cost of sending the signal.  It would take mass or material to collect
the energy and create a transmitter of some sort.  This is material that
is no longer optimized for computronium, so thought potential is lost.
So talk isn't cheap: it has its cost.  The M/D argument calls upon
Robert Bradbury's question about the value of present thought versus the
possibility of a greater amount of future thought.  When we calculate
the potential of computronium, perhaps the value of present thought far
outweighs the value of potential future thought from a distant star
system, whose maximum mass can be bounded and whose distance is known to
be very large.  The M/D, and thus the potential value of the
communication, is very small, whereas the cost is large.  Transmitting
is a poor investment of valuable resources.

The M/D approach to Fermi's Paradox suggests that the reason the cosmic
abyss is silent is that we are not worth the mass to talk to us.

spike