Sorry for the delayed response...
There are several different dimensions of what can be built. One is the materials used. One is the complexity of the product. One is the size of the product. There are others, but let's stick with those three.
The first nanofactories will build special materials with special inputs, as you say. The materials may not be high performance (e.g. DNA rather than diamondoid).
By the time we get to nanofactories that can build macroscopic objects, we will probably be able to build very complex objects. If we can build a diamondoid fork using a nanofactory, we can probably expect to build motors and computers out of diamondoid, and integrate them into e.g. a mini-airplane. (This may not be as true for other materials - DNA motors are slow and weak.)
On Tue, Apr 20, 2010 at 6:41 AM, Erin Casson <solidstatefusion@...>
The first nanofactories and industrial assembler systems will likely be special purpose units that require special inputs of chemical fuel and prefahricated building blocks, and which will require special programming. Their productive output will likely be very limited, but as time goes on and as the systems are refined they will gradually become more and more general purpose in their input and output as well.
We should expect early products to be things such as computer chips, inert, simple structural materials such as diamond fiber beams and household goods such as diamondoid flatware (stronger harder tougher and lighter than steels and does not rust ever, nearly indestructable).
The products will become more complex, more intricate, and "smarter", such as shape-changing active materials and robotics and much more.
Ultimately there is no reason why nanofactories could not be designed to produce food items. The early systems will not be able to, though.
Would you all tend to agree with the above, as related to some of the limitations and products on early nanofactories?
Director of Research, Center for Responsible Nanotechnology, http://CRNano.org