I’d be surprised if there were a product structure that was delicate enough to need counterfactual measurement (but could last long enough to be considered a product). But you make a good point that it’s not just final structures, but processes that have to be observed. And more generally, there are probably other applications that I don’t know enough to think up. That’s why I more-or-less retracted my comment.

Brian Wang says that Dwave is coming out with an analog quantum computer system… this year… with 16 or maybe 64 qbits. Go to http://advancednano.blogspot.com/search/label/quantum%20computer and search for dwave to find relevant stories.

Diamondoid MM has a lot of (expected) advantages beyond being relatively easy to simulate. High strength and stiffness, and superlubricity.

There are areas where it’s fair to accuse MNT of avoiding potential functionality to keep analysis simple. Electronics is a big one. But the choice of diamondoid isn’t. In fact, if you read Nanosystems closely, Drexler didn’t define “diamondoid” as pure-carbon diamond lattice. He was talking about a broad range of covalent solids, certainly beyond detailed analysis at the time. As far as I know, the more recent emphasis on diamond-only came from Freitas and/or Merkle, and its purpose was to make bootstrapping easier.

As to the benefits of easily characterized systems (which I think is more about digital/multistable than about few degrees of freedom, though the two tend to go together): Regardless of the manufacturing system, predictable device and subsystem operation will be extremely useful for product design. The most complex/intricate things humans have built BY FAR are computer hardware and software, and they would be impossible without levels of abstraction, which in turn depends on fully characterized (Boolean) operations.

There will be lots of useful nano-products, even programmable-materials products, that won’t need that kind of intricate predictability. But for products that do need it, well… if you can’t characterize, you can’t design. You can punt and use evolution, but it remains to be seen whether evolution can be a general-purpose R&D tool.

Covalent bonding is highly non-linear and in many cases produces metastable systems that might as well be called stable. In effect, it should provide a family of digital operations with ignorable error rates, making structures that are completely identical (modulo isotope-sorting, if necessary, and below the few-micron scale where background radiation causes too-high damage rates per unit). I think that fully-known nanostructures (engineered to have fully-known functionality) will be centrally important for a number of directions of nanotech advancement. And additive covalent synthesis of covalent solids is the best way I know of to achieve that. If the Ideas Factory comes up with another way to achieve fully characterized construction, that’ll be great! (Though I’ll still want to ask about material properties and exponential manufacturing.)

Of course, the approach I’m promoting here denies the necessity of biological complexity for making useful products. I could write several essays on why biological complexity is overrated. It’s too easy to point to biology’s successes and attribute them to its complexity, but those successes come with subtle limitations (e.g. ATP synthase needs a volume of water on each side to be efficient), the complexity doesn’t always contribute to the success (maybe it was necessary to let the success evolve), and often it’s better overall to over-engineer in order to simplify (products compete in the design cycle, not only in usage).


Ps. In large systems, there will be faults. But I think fault tolerant design to deal with relatively rare errors in otherwise perfect constructions can preserve fully characterized operation, a lot more easily than statistical aggregate constructions can be treated as fully characterized. So what about thermal noise, which certainly provides an inescapable statistical variance? I have answers for mechanosynthesis, for noisy motion trajectories, and for mechanisms “jumping the tracks”, to explain how in each case the important features can be fully characterized and thermal noise can be functionally ignored above the atom scale, but this comment is already far too long.


0 Responses to “Chris Phoenix 4”

  1. Leave a Comment

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: