In fact, my reply grew into a post: “Self assembly and nanomachines: Complexity, motion, and computational control”.

The rate at which advances are being made is very encouraging.

One problem with this approach, obviously, is that such assemblies would be static, at least in this simple concept. It is much harder to imagine how you would self-assemble machinery that is capable of motion, i.e. has more than one valid spacial configuration.

It is not trivial to specify the face-combinations that would result in a unique configuration. This can be, and probably has been, explored computationally, today.

As a more general remark, I have a concern with molecular manufacturing that I have so far not seen addressed here or elsewhere. Because of the absence of people on the nano-scale, any manufacturing has to be completely automatic. Complete automation is not impossible, and today we are coming close to it in our manufacturing facilities. However, we are taking advantage of Feynman’s “room at the bottom” in our control circuitry. That control circuitry is many orders of magnitude smaller than the machine parts (approaching nanometers these days), allowing us to put, say, the controller for a CNC machine in a small box right next to the machine.

In nanosystems, there is no such room at the bottom. Is this a problem? How can it be addressed? Forgive me if this has been discussed before, I am fairly new to this subject.

This post was mentioned on Twitter by publicpatton: http://j.mp/cpatton Self-assembling nanostructures: Building the building blocks http://bit.ly/76UYgy...