I just got back from the National Academy of Sciences meeting that was mandated in the 2003 Nano Act to study "molecular self-assembly." As far as I can tell, it went extremely well. I spoke in the second session, on definitions, but I was there for all six sessions. More details on that later; for tonight I'll just give initial impressions.
The committee focused directly on molecular manufacturing--which is good, because the wording in the near-final version of the law was "molecular manufacturing" before a nano business organization lobbied to change it to scuttle the study.
(In what follows, MM is molecular manufacturing; NNI is National Nanotechnology Initiative.)
There were some skeptics on the committee, and some who seemed more willing to accept that MM could work. But everyone there was clearly very smart, and everyone seemed intellectually honest. I'm pretty confident that MM will get a fair hearing--that they won't write a report which merely endorses the unscientific and off-topic objections that have been floating around for the last decade or two. Whatever they produce will be the result of open-minded and very careful study.
At the end of the second day, the committee went around the table with each person giving their personal reactions to what they'd seen so far--two sessions of history, one of definitions, and two of technology. A few members had to leave early, and a few said they were still absorbing what they'd heard. I took notes, but not quotes, and I was scribbling in a hurry, so I won't give people's names lest I mis-attribute an idea to them. But this is what I heard:
A thousand miles starts with a single step. We're busy proving A->B. We have no proof for E->F. (Where F is molecular manufacturing.)
There was impressive research in the last session. It somewhat alleviated my fears [about MM being impossible?]. We need a wide net for nanoscale tools--we don't know where is the tipping point for a breakthrough. Maybe we need a DARPA-like high risk/high reward pot to see what's possible.
I did biopolymers and thought it was nanotechnology; now I see we need definitions.
I've been through tech revolutions--lasers and computer chips. Big dreams are good; hats off to Drexler. However--it drives me crazy when people do computations without taking physical issues into account. If you say "manufacturing" you inherit physical issues--tolerances, yield, etc.
I work with nanotech companies--I see and appreciate proofs of concept--there are lots of [research?] directions. We need to survey the landscape widely to avoid being surprised by competitive innovations.
We're hearing that MM is a done deal--but we're being asked to consider if it's possible. I can't forget big picture/system. The devil is in the details. Why don't we have advanced e.g. medical products already?
The books Discoverers and Creators talk about history of technology. We owe a debt of discovery / another chapter in those books to Drexler; he's why we're sitting here. We're at a critical point: what will that chapter look like?
Thanks to the professors [in the tech sessions] for showing that the first steps are done. I advise patience to the advance thinkers. I did a heretical research. In 1972 I was told: show the first step. I did it. Then the engineers didn't care. So I showed them samples. *Today* my material is first being used in jet engines. (I had given up in '96.) We need demonstrations and patience.
I'm a scientist and engineer. Want to thank the panelists and visionaries. And Roco, for vision. Nanomanufacturing is in the NNI vision. [Drexler interrupts: nanomanufacturing is different from MM--the NNI is trying to muddy the waters.] I hope we can find a way to get materials into jet engines more rapidly. [Previous speaker interrupts--No! Be conservative with jet engines!] Let's se a grand challenge: an artificial ribosome. How stripped-down can we get a cell with a reasonable error rate for manufacturing? Keep a comparable goal in mind.
We're still in the process of identifying tasks. I did structure/function relationships of proteins and enzymes. Cells do have error correction--I'm not sure we're near achieving that. We need much more terminology work.
The NNI has come a long way. There's a lot going on.
Thanks everyone. I appreciated the civil tone of the discourse, and the enthusiasm. The passion for ideas was good to see. As someone else said: we haven't talked about: is MM desirable? At some point we need to look at societal needs we could address with this tech [something about public funds] - given fundamental research, keep in mind "to what end?" To the extent researchers can do that, it'll be helpful - resonant - public issues: health, energy.
I see progress toward MM: biology, chemistry, mechanical. We need to know more about fundamental limitations; I was impressed by a presentation on thermal noise.
I'm sure different people will see different things in the meeting, and these brief personal summaries don't even begin to give a full picture of the structure and content of the meeting. But I think they show the spirit of questioning that we need to begin a full and careful discussion of the technical merits of molecular manufacturing. I don't expect the committee to say either "Yes it's workable" or "no it's not." But "We looked at it for months and couldn't rule it out" would be a significant step forward.