Hara Ra wrote:
> I agree that it is ambitious. However, building the components needed to
> pickup and join Lego blocks should be feasible with a limited budget.
> have no objection to placing something on the blocks to identify their
> types, such as the barcodes you suggest. The idea is a low cost approach
> which is affordable for amateur nanotechnolgists. And, of course, to
> identify the higher level problems which as yet remain undiscovered.
>
I think a critical issue with a macro assembler is positioning. Once you can position components accurately and verify that they are there, you can also position work tools and build the components. I don't think lego blocks give you a lot of inherent clues about their position. Still, I agree availability is a big plus. A lot depends on the functionality of the active components they provide.
I've been thinking about macro-assemblers for a long time. The model I have been thinking about I call the Universal Fabricator Replicator (UFR), on the principle that any engineering project needs an obscure acronym :-). It useful feature is that it can disassemble anything it builds. Building one off things out of generalized robust components seems likely to be more expensive than using bulk technology. But when you can download a new design for you living room, and simply disassemble the old one to make the new, you are into a whole new ball game. Like nanotech, you drop much of the bulk transport and central manufacturing. Even with nanotech, it may be energetically cheaper to assemble large objects from largish (millimeter?) components.
Bernard