From: Robert J. Bradbury (bradbury@aeiveos.com)
Date: Mon Nov 25 2002 - 18:52:32 MST
On Mon, 25 Nov 2002, Ramez Naam wrote:
Mez has provided some nice counterpoints -- so I going to try
to condense:
(a) designing and manufacturing a functional nano-assembler
(b) designing an atomically perfect "car" (or pick another product)
(c) developing a number of (a's) that can produce (b's).
Points where I may differ with Mez include:
> To be clear, (b) doesn't worry me very much. We already design cars
> with CAD tools.
True, but these are not at atomic precision and though Hal and/or Lee
may argue that isn't particularly important I would argue otherwise.
I think you have a stronger case arguing that atomic precision design
will not be available to macroscale objects in the near future than
their simulation will not be available (at least assuming some reasonable
level of redundancy).
> It's designing the systems that grow the car, and then programming
> and controlling them, that I'm worried about..
Complex issues and ones that you and I may be qualified to explore
in a knock down drag out mode. A better strategy I think would be
to acknowledge this is a very very dicey area -- one we need to master
but also one we need to be very careful of.
> I AM surprised that I have yet to see a grounded analysis of the
> coordination and programming problem.
It *is* a very hard problem. Eric raises hints of it in Nanosystems
but I don't think it has ben seriously addressed.
> It seems to me that these control and design issues will
> give us far more trouble than the issues of fabricating things at the
> atomic scale.
I believe, to a large degree, this is correct. I think Eric
recognized this even if he failed to express it in sufficient
detail. (I, as I believe Mez is, are basically "software" people
and so this is more significant to us.)
> My hunch is that the reason no one has produced a good analysis of
> problem (c) is that it's deeply tied up in complexity theory, and
> complexity theory has yet to produce good tools (so far as I know) for
> analytically determining things like the order of complexity of a
> system, or the amount of information required to control a system.
I would tend to agree with this -- complex systems and the analysis
thereof are a very key point for discovery. I think this point has
been supported in the recent Foresight newsletter which suggested
that Robin Hanson's ideas with respect to Idea Futures may have been
picked up by DARPA as a concept worthy of furher development.
Robert
This archive was generated by hypermail 2.1.5 : Wed Jan 15 2003 - 17:58:23 MST