From: Robert J. Bradbury (bradbury@aeiveos.com)
Date: Sat Jul 28 2001 - 21:30:00 MDT
Wow, in an interesting mis-application of the "intelligence" of
the net the original message, attached below, got truncated
when inserted (even as "source"!) into the Javien database.
(No fingers being pointed here -- it could be a problem on my end.)
I don't know if the list subscribers got the truncated version
or not but I'm throwing it into the thread again (as well
as personal mailboxes) to make sure I'm not being misinterpreted.
As a followup to Brian's comment regarding what Samantha was
pointing out -- I may have failed to follow the thread back
far enough to have understood the details in full. If it is
true that post-singularity intelligences can significantly
rewrite the laws of physics as we now know them, then all bets
are clearly off. However, I personally wish people wouldn't be so
willing to take the universe as we know it and think it is
something that can so easily be discarded. I'd sooner
vote that its all a simulation (which can of course be
transcended if the simulation manager decides to resurrect
us in the "real" universe). And nothing we "know" about this
universe need be any reflection of what the real universe
may be like if that is the case.
Robert
-------- Original Message has 43 lines from this point -----
> Date: Sat, 28 Jul 2001 15:16:31 -0700 (PDT)
> From: Robert J. Bradbury <bradbury@aeiveos.com>
Samantha wrote:
> I think it is a mistake to assume that all that is knowable will
> be known almost immediately after reaching Singularity.
Tsk, tsk, tsk, you folks need to read the literature...
>From Drexler, 1995 (emphasis mine):
> "The shortcomings of current fabrication processes, when considered from
> the perspective of *ultimate* limits, are dramatic. The number of
> distinct covalent three-dimensional, highly polycyclic structures (here
> termed `diamondoid') that can occupy a volume of one **cubic nanometer**
> has been estimated to be greater than 10^148 (Drexler 1992). This
> volume can contain more than 100 atoms, each chosen from one of many
> elements, and even a pure-carbon structure with exactly 100 atoms
> could exist in many stable bonding patterns.
>
> Most of these structures are irregular and asymmetrical, and hence cannot
> be made by techniques that produce crystals. Since 10^148 is **greater**
> than the number of particles in the observable universe, random
> generation processes would be ineffective. to say nothing of inefficient."
If there is anything to be "known" about all the capabilities of
all the possible nanomachinery that you can assemble in a cubic nanometer
then I don't think you are going to know it even "thirty physical
seconds" after the Singularity (as Eliezer qualified).
One thing I'm pretty clear of is that *if* we want to come even
close to exploring that phase space, then sub-atomic computing
*is* likely to be necessary. If so, you have now crossed the
boundary into the realm of "magic physics".
Samantha is right in that "optimizing" the computronium takes
a very long time because of the requirements for moving materials
around the solar system, breeding the right element mix for all
the nanomachinery, etc. Those things are going to take time even
when you have stars to burn.
Robert
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:09:15 MST