From: Robert J. Bradbury (bradbury@aeiveos.com)
Date: Mon Nov 25 2002 - 18:03:36 MST
On Mon, 25 Nov 2002, Ramez Naam wrote:
> From: hal@finney.org [mailto:hal@finney.org]
> > Drexler's proposal was to try to restrict the set of designs
> > to those which could be modelled relatively cheaply. That's
> > one reason they use so much diamondoid, it's stiff so the
> > atoms don't move around much.
But that wasn't the reason Eric (or Ralph) focused on it.
Covalent bonds are much stronger that hydrogen bonds.
(That is what makes diamondoid stiffer and also gives it
such interesting material properties.)
I do not believe that Eric cared very much (at least early
on in the '80's and early '90's) about whether or not you
could do the simulations and have some "faith" in the results.
> > I think the moving parts do need some careful simulation, I'm
> > pretty suspicious of some of Drexler's exotic bearings and
> > such.
I think the work by Goddard's group at Caltech and the Ames group
have nailed the "exotic" concepts down as being fairly sound.
It may also be true that some of the bearings may require
credit to Ralph Merkle (I'd have to go check the detailed
references in Nanosystems and Nanomedicine VI).
> > But for the most part these can be embedded within a
> > diamondoid matrix so you can use the part in a modular
> > design without having to simulate the whole system at once.
Boeing gets a new airplane off the ground (usually) without
simulating the "whole system at once". (At least I suspect
that to be accurate -- it certainly was before we had computers.)
> At this point I'd like to make clear that I'm not doubting the
> feasibility of building an assembler, someday.
Gee, I'm glad we got that out of the way.... :-)
> I am, though, surprised that many of the problems that make the design
> and control of such a device an extremely challenging problem have not been
> widely raised. Illustrating some of those issues is my goal in this thread.
I think they *have* been raised, though perhaps not widely because most
of the scientists & engineers who "claim" to be involved in nanotech
have not read the primary literature Eric has written.
The issues *are* out there -- they have been known since the
1960's with the design of the Apollo command module. Boeing
deals with them in the design of any modern airliner. Its
a fundamental problem of dealing with a complexity that exceeds
a single human mind.
It comes down to a fundamental question of whether you can
do divide and conquer on the problem. I think for almost all
questions with regard to molecular nanoassembly the answer
is yes. (Nature seems to manage it quite well). There will
be some problems that are too complex to be handled by that
strategy but I don't know yet how to evaluate them.
It may be that this should be an area of new computational
science investigation -- the class of problems that cannot
be solved (in a reasonable time) by a divide and conquer strategy.
> 1) Assemblers process information to follow their instructions,
> monitor their environment, etc... This involves many and very
> complicated moving parts and in effect brings all the complexities and
> difficulties of prediction of software engineering into play.
Agreed. I think this is the complexity problem that Eric *does*
outline but tends to get lost in abstractions and summaries.
> 2) Assemblers make and break chemical bonds. From a molecular
> modeling standpoint, this automatically forces you to use more
> expensive simulation methods, as the cheapest methods simply do not
> handle the breaking or formation of chemical bonds. [snip]
Yes, but I think you are overstating the point -- as Eric and Robert
are always pointing out there is somewhat of a difference between
"atomic" bonds and "molecular" bonds. The modeling problem is
much smaller if you are dealing with two "large" molecules interacting
with each other when they happen to interact compared with the modeling
of a couple of small molecules interacting through some random brownian motion.
The problem of modeling the activities V-B12 in a cell is very
different from modeling the action of methane (or some other much
smaller molecule) in your cells.
> 3) Perhaps most problematically, in at least some proposals assemblers
> reproduce.
You should divide this into two very distinct problems.
1) The biological nanoassmblers already reproduce. [That is a topic
probably best reserved for some offline discussions.]
2) Nanoassemblers don't have to reproduce. There are some quite good
papers by Josh Hall and Ralph Merkle over the last decade that provide
clear directions with respect to the "broadcast architecture" (so
self-reproducing nanoassemblers are impossible" and the limits of
specially programmed nanomanufacturing assembly lines (so self-reproduction
is clearly less profitable).
This does not make the "worst scenarios" impossible -- but they
do point out how prepared societies can deal with the worst
scenarios that nanotechnology seems to present. So they are
not much different from the scenarios we already face.
Robert
This archive was generated by hypermail 2.1.5 : Wed Jan 15 2003 - 17:58:23 MST