> A Matter compiler is a device which takes a program...then uses the stored
> atoms to build whatever the program specifies.
>
> It would seem to me that this would sidestep the "Grey Goo" scenario. (Whew!)
No, it just makes it somewhat less likely. See below.
> To me...this seems more easily obtained...since self-replication appears
> HARD...whereas we already have the Matter Compiler Technology....only on a
> Macro scale...(assembly lines...etc) so all we gotta do is shrink
> it.....severely.
The problem is that to build a matter compiler you will likely need
self replication. I'm not 100% sure it is mentioned in Stephenson's
book (likely at the scene where Hackworth makes the Primer), but to
build a macro-object you need a large field of nanoassemblers (which
likely put it together using recursive assembly). How do you build
such a field? The most likely explanation is that it is built by
nanodevices; by assumption the assemblers are able to build most
molecular structures, nanodevices included, and would hence be a
cost-effective method of making assembler-fields; making each
assembler-field using bulk technology would make MCs prohibitively
expensive.
So, we have self-replicating nanodevices (assembler-fields) anyway.
In the book is it very likely that the designs for them are not in
the public domain and only used by the manufacturers of MCs, but it
is not a perfect protection against runaway replicators. Stephenson
simply ignores this problem, but it is clear that in order to get a
widely usable, cheap nanotechnology you need replicating assemblers
if only in the factory.
-----------------------------------------------------------------------
Anders Sandberg Towards Ascension!
nv91-asa@nada.kth.se http://www.nada.kth.se/~nv91-asa/main.html
GCS/M/S/O d++ -p+ c++++ !l u+ e++ m++ s+/+ n--- h+/* f+ g+ w++ t+ r+ !y