From: Robert J. Bradbury (bradbury@aeiveos.com)
Date: Sat May 11 2002 - 13:21:57 MDT
On Fri, 10 May 2002, Lee Corbin wrote:
> That would be like pointing out to an oil tycoon that the Indians
> on some reservation still had 1% of the U.S.'s known oil reserves.
> All he'd think about is how he could get to it, and why it lasted
> so long. I wonder how many tons of material the last 1% of Earth
> is ;-)
Ooooh, Ooooh, a question I can answer....
The Earth weighs 5.9E24 kg, 1% of that would be 5.9E22 kg or
6.48E21 tons or 6.48E11 tons/person assuming we stabilize
at around a population of 10 billion.
Of course if you tear the guts out of the planet you lose
your gravity, and therefore your atmosphere. If you turn
the guts into computronium, you still have a *very* big
heat dissipation problem (I think we are talking cooling
towers to 100 km all over the planet here...). Then there
is also the small problem of the loss of the magnetic field.
But if you think carefully about this you will realize
that the Earth, particularly its contents, are pretty far
down on the computronium conversion schedule. Titan and
Triton probably are near the top of the list followed by
Uranus and everything further out in the solar system.
It seems likely that the stuff with the highest carbon
content will be the early targets. Iron rich planetary
cores are one of the last things you tackle.
And AI's, no matter how smart they are, "canna trump the
laws o physics", to quote Scotty. So they are still
constrained by the time it takes to get the material
out of gravity wells or the energy requirements to
move it around the solar system.
So unless you subscribe to the idea that AI's will be able to
invent magic physics (don't make me have to get my ruler to
slap your fingers now boys....) I think we get at least
a thousand years or so of gradual adaptation of humanity
to the "singularity". That makes the whole concept of
a singularity, particularly the hard takeoff scenario
(at least in the physical realm) rather questionable IMO.
Robert
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:13:59 MST