The Foresight Institute is holding their Conference on Advanced Nanotechnology. The first day, Friday, was about technical progress toward molecular manufacturing.
There was a lot to talk about. We heard from Christian Schafmeister, who has developed a new way to build protein-like molecules. Proteins are usually floppy, and hold their shape about as well as a string of beads--that is, only when tied in knots--and even then they're more like overcooked spaghetti than like anything solid. But Schafmeister has developed a way to make two bonds between each "bead" in the "string," so he can now engineer straight lines, curves, and switchbacks in sub-nanometer structure. The "beads" are simple, and stringing them together is even more simple--he has undergrads doing it.
Robert Freitas outlined a way to test the process of depositing carbon atoms on diamond in programmable locations. This is key to perhaps the most high-performance form of nanotechnology: using mechanically guided programmable chemistry to build machines out of diamond. This is what Smalley said couldn't be done; but it looked like someone could follow Freitas's outline and demonstrate it in the lab in (I would guess) a year or two.
Eric Drexler showed an almost-finished animation that illustrates how a nanofactory would actually work. This should lay to rest any remaining misunderstandings about "fingers" that "grab" atoms. The animation shows pointy molecules being pushed together and trading atoms between them--nothing new in that, except that the molecules are mounted on molecule-sized conveyor belts (also made out of molecules, of course). This is what Drexler has been saying for a decade, but now it's in a form that will make it hard for even the most wilfully dense skeptics to find something to misinterpret.
There were many other exciting and inspiring talks, but I don't have time to write them up; I have to get my beauty sleep because I'm giving two talks tomorrow. (One on "Clean Molecular Manufacturing," the other on "Molecular Manufacturing: Top Ten Impacts.")
Given the rapid recent progress in biopolymer engineering, I think some kind of programmable robotic chemical manufacturing system might be doable in the next few years by a reasonably small team. I think this is probably good news. Working underwater, and with lower-strength materials, it would have lower performance than a full diamondoid nanofactory; it might initially be more of a demo than an economically significant system. This would give us some time to get used to the idea of programmable nanoscale exponential manufacturing. It might even be possible to make good policy reactively. And it would spur discussion of future, higher-performance exponential manufacturing technologies.
I could lend my technical skills to help develop that biopolymer molecular manufacturing. I could contribute substantially, and I'm currently thinking this would be the most responsible thing to do. But I'd like to get a sanity-check on that.
Please don't discuss that question here. Instead, go to "Wise-Nano: Early MM Pros and Cons" and discuss it there.