[extropy-chat] Wired article on Drexler

[Hal Finney]

Emlyn wrote:
> Why isn't all of terrestrial biology usable as proof? It makes a pretty
> solid proof of concept as far as I'm concerned.

The problem is that by itself, biology is only a proof of concept
of biology, or perhaps of biotech.  But nanotech claims to go beyond
biotech (or else it would not deserve a different name).  It proposes for
example to do construction in a vacuum, which Drexler calls a eutectic
environment.  It proposes to do chemistry using mechanosynthesis.
It proposes the construction of a range of products which biology has
never produced, and which probably can't be produced by biology.

If all nanotech could do was what biology can do, then it would not
be nearly as exciting and revolutionary.  So there must be limits to
how much biology can be said to be a proof of the nanotech concept.


Similarly John Clark wrote:
> I respect Smalley, having a Nobel Prize and all, but I would be much more
> impressed with his argument that assembling macro sized objects with atomic
> precision if fundamentally imposable if I didn't know of a pretty damn good
> counterexample to his thesis, life. Granted Drexler proposes a different way
> to do it, but I find it difficult to believe that random mutation and
> natural selection just happened to stumble onto the one and only way to
> go about it; I have a hunch intelligence can do better.

Smalley does not say that "assembling macro sized objects with atomic
precision is fundamentally impossible".  I'm sure he is aware of
the existence of living organisms.  Smalley specifically criticizes
Drexler's design proposals using assembler arms to produce the kinds
of products described in Engines of Creation.  In his debate with
Drexler, http://pubs.acs.org/cen/coverstory/8148/8148counterpoint.html,
Smalley discusses the limits of ribosomes for assembly (in an apparent
misunderstanding of Drexler's point, a point which I have never seen
satisfactorily explained [1]).

Hal


[1] Drexler: "In particular, you have described molecular assemblers as
having multiple 'fingers' that manipulate individual atoms and suffer from
so-called fat finger and sticky finger problems, and you have dismissed
their feasibility on this basis. I find this puzzling because, like
enzymes and ribosomes, proposed assemblers neither have nor need these
'Smalley fingers.' The task of positioning reactive molecules simply
doesn't require them."