> I'm following this list since several months now and i'm rather
> surprised that only assemblers using nanotechnologies in their
> implementation are discussed here.
>
> Probably we could build a more "macroscopic" one. It will sure be less
> efficient than a nanotechnological one, but it's really worth to study
> too.
Yes, a macroscopic assembler (even a limited one) would likely be very
useful. I'm quite interested in the work done by Klaus Lackner and
Christopher Wendt (see http://www.discover.com/archive/index.html) in
how a hive of robots could build a growing solar cell plant.
Still, the comments about nanoassemblers made by John K Clark are
reasonable; they avoid a lot of the problems inherent in macroscopic
assemblers.
> I would like to see some reviews of the interesting ideas from Joe
> Michael at http://www.stellar.demon.co.uk/index.htm about Fractal Shape
> Changing Robots. The basic idea is so simple that i'm wondering how it
> can be possible that nobody come with that before.
Actually, Forrest Bishop had a similar idea independently
(http://www.speakeasy.org/~forrestb/home.html).
I think Joe Michael has had a good idea, but he is making it a great
injustice. He makes bold claims for it, but provides absolutely no
proof, calculations or even estimates despite claiming that his work
is "academic". I have repeatedly asked him about certain sticky
details (like how to deal with dirt in the connectors, how to program
the system to do useful transformations etc) but not got any
answers. This makes me think he has very little ideas about how to
really implement the robots, but is very eager to proclaim their
greatness.
It is a far cry from the nanotechnologists, who at least admit their
own shortcomings and do their best to calculate what is possible or
not, and how to get there.
-- ----------------------------------------------------------------------- Anders Sandberg Towards Ascension! asa@nada.kth.se http://www.nada.kth.se/~asa/ GCS/M/S/O d++ -p+ c++++ !l u+ e++ m++ s+/+ n--- h+/* f+ g+ w++ t+ r+ !y