The Proceedings of the 1989 NanoCon Northwest regional nanotechnology conference, with K. Eric Drexler as Guest of Honor.

Return to NanoCon Top
Return to NanoCon Proceedings Table of Contents
Return to Jim's Molecular Nanotechology Home Page

NANOCON PROCEEDINGS page 10

VI. CRITICAL PATH PANEL

A. DELPHI Survey


E. DREXLER: In the last session, we discussed the perception of nanotechnology in the technical community and the importance of credibility for these concepts as the foundation of any sensible policy discussions. There is in the audience someone who has not been prepped for this, who is participating in a policy research project on nanotechnology at the LBJ School of Public Affairs in Austin, Texas. I would like her to say a couple of words on that project, and on the results of attempting to get a group of technical people to participate in a Delphi survey on the prospect of nanotechnology, and the reactions of the half who did not want to participate.

L. COBB: We were hired by FutureTrends to do a research project on the political, social, and economic implications of nanotechnology if it does come about. We decided to do a Delphi study, in which you send out a first round of questions, tabulate all the results, and then send to everyone who participated the answers from the first round of questions, delineating the positions and reasons of the people at each end of the survey, and get people in the second round to either defend or change their positions. The purpose of the survey was to estimate when nanotechnology would arrive, and by what path, what government regulations would be pertinent.

We questioned 200 scientists chosen from the literature prominent in the paths that Mr. Drexler says will be important for the development of nanotechnology. About half said that they would not participate because of the reputation that nanotechnology has associated with it. They called people involved with it the lunatic fringe. I find it very interesting that the existence of a science-fiction oriented group such as this is making it very hard for nanotechnology to be taken seriously in the technological arena. The paradox seems to be that if you want it to come about, you can't talk about it.

E. DREXLER: I would emphasize that I have been invited to give talks at places like the physical sciences colloquium series at IBM's main research center, at Xerox PARC, etc., so these ideas are being taken seriously by serious technical people, but it is a mixed reaction. You want that reaction to be as positive as possible, so I plead with everyone to please keep the level of cult-ishness and bullshit down, and even to be rather restrained in talking about wild consequences, which are in fact true and technically defensible, because they don't sound that way. People need to have their thinking grow into longer term consequences gradually; you don't begin there.

LINDA COBB: We have just sent out the first round for the Delphi. We have several scenario-generating groups at work. This conference has been very interesting to me because the science fiction people have been more inventive in their scenarios then have the social science people. We will later have several focus groups with social scientists to explain nanotechnology to them and have them think about its consequences.

The report will be available about August, and will be available for $8-10. The address will be announced by The Foresight Institute [P.O. Box 61058, Palo Alto, CA 94306].


B. John Cramer

This is supposed to be a panel on critical paths to nanotechnology-- how to we get from where we are to the first generation of nanotechnology capabilities. I'd like to give us a frame of reference for the idea that there is more than one way to get there by using a list of ideas of paths there that I made for a talk that I gave to the Student Nanotechnology Study Group of the University of Washington. Secondly, I'd like to talk about some of the roadblocks on those paths. I'd like then to let each panel member introduce himself and make some introductory remarks, and then move into the workshop mode with as much interaction with the audience as possible.

The paths that I see are:

The roadblocks are:

None of the possible ways is wonderful. They all have problems, but maybe by some combination we will get where we want to be.


C. Vonda McIntyre

I'm one of those weird "Sci-Fi" writers who understands that "Sci-Fi" is in quotes.

I find myself in an interesting reverse position here. I've been at similar conferences and found myself listening to panelists talking as if there would be no progress. My reaction was that "You don't understand. Things will change, become different, easier, etc."

Now I feel a little bit of how they felt. I think that we must respect people's fear. There has been a lot of talk here of lumpenproletariat and luddites. The luddites had in fact good reason to be frightened. They were losing their livelihoods. Mr. Drexler has said repeatedly here that his vision is for a world of plenty, but others of us have been talking about proprietary rights and making money. People don't hear about plenty; they hear that their jobs are going to disappear. We can not just ignore this question because we can see it happening now.


D. Mike Thomas

I program computers using a technology called artificial intelligence. Several people have talked about the software problem-- how will you program and control these devices. The technology to do so is growing very quickly but is nowhere near the complexity necessary to control nanodevices. But there are new technologies in artificial intelligence, neural networks etc, that are leading toward self-programing and learning systems.

In the near term, how we will communicate with the nano devices is a problem...


E. DREXLER: One small interjection. Not all nano-mechanisms are more complex than devices that we are familiar with. Some are simple; some are complex. It is important to distinguish those categories.


OK. At this point we can do something about the simple instructions. Long term, the technology needs to evolve a lot. We will need to see things like parallel processing, more work with interfaces -- getting a neural network running is one thing, but how do you talk with it and understand its reasoning?

We're not there, but the path is fairly straight-forward. The time frame probably matches that for development of the nanodevices.


E. Eric Drexler

I would like to comment on [John Cramer's] paths to the development of nanotechnology.

I tend to think of the first two paths mentioned as different approaches to a folding polymer path to nanotechnology, in one case using proteins, in the other case using chemically synthesized polymers that are like proteins or DNA or something else.

The miniaturization approach is a silicon technology. There my inclination is to say that miniaturization is not what nanotechnology is about. Nanotechnology is about control of the structure of matter rather than about scale. All that small systems can do in terms of manipulating atoms is get you smaller hands to grab things with and to position things more accurately. But that is an illusion because the STM and the atomic force microscope (AFM) can already position things to the required accuracy without themselves being small. They show that in the micro-manipulation path there is no requirement for the manipulators themselves to be small, although the actual tools at the end of the manipulator need to be.

In the STM area, people do keep coming up with serendipitous developments. A likely development there is some hybrid of a folded polymer approach to making sophisticated tools, and then using one of these positioning devices to use those tools.

The category of trapping is appealing because people are working with single atoms, and you can show pictures of single atoms. One thing I can say for shock value is that there is no virtue in handling atoms one at a time. You would like to handle vast numbers of atoms in parallel and have all of them do what you want. Furthermore the ion traps cannot position the atoms that they are working with to atomic precision...


J. CRAMER: That's not true. The laser traps are trapping ensembles of atoms that have a regular periodic structure to them...

E. DREXLER: So they're electrostatically confining each other...

J. CRAMER: Exactly. You could have a group of carbon atoms that assembles into a diamond rod.


That's something that I hadn't looked into-- an interesting point. The other problems that you cited are also quite substantial. So that's an interesting area to keep an eye on.

I tend to focus on learning techniques for the folding polymer approach because chemists handle vast numbers of molecules at a time. A trillion molecules is a very small amount for a chemist, but a very large amount if you're building them one at a time. In most cases you would prefer to have many rather than a few.


F. Jim Lewis

I'm a molecular biologist, working with tumor viruses and oncogenes. To put into perspective the progress and the possible near term future of the folding polymer approach, I'd like to make a few observations about where the biotechnology industry, as I see it, is at the moment, and how that differs from where we need to be for the folding polymer approach. Nevertheless I see reasons to be optimistic that major advances are not far off.

The first generation of biotechnology has been involved in finding ways to identify and isolate the genes for various valuable proteins that are made in nature in very small amounts, and to find ways to make them in very large amounts.

The second generation, which is already well under way, is using rational design to make more effective biological molecules. An example is the biggest selling product in biotechnology at the moment, tPA (tissue plasminogen activator) which is an enzyme that dissolves blood clots.


J. CRAMER: It essentially turns off heart attacks!


Yes. There is a lot of work now to design a better tPA by making minor, but important, modifications.

To move toward nanotechnology, we need the approach that Eric mentioned three years ago in Engines of Creation [Editor's note: And five years before that in his paper on molecular engineering]. Instead of worrying about the very difficult questions of how enzymes work, which would be the straight forward extension of what biotechnology is doing now, we need to design proteins that will have simple folding rules and use them as building blocks to make more complicated tools. There has been substantial progress in this area, as Eric alluded to on Friday night. Last summer in the journal Science, a group at DuPont reported the success of designing from scratch a protein sequence that had no counterpart in nature, but which would fold into a particular structure that is a subunit in many enzymes. [Editor's note: "Characterization of a helical protein designed from first principles" by L. Regan & W.F. DeGrado. 1988. Science 241: 976-978.]

Although the protein folding problem is by no means close to being solved, it has suddenly moved from a peripheral issue to a mainstream issue of molecular biology. Molecular biology started in the early 50's, with Watson & Crick, at the level of 3-dimensional structure, and then went into something very different - genetics and linear sequences of information, etc. Now it's back to focusing on 3-D structure. At the American Association for the Advancement of Science meeting last month in San Francisco one of the small number of major symposiums was devoted to progress on the protein folding problem.

I have a sense of the general atmospherics that is very similar to how I felt when I entered molecular biology as a postdoc in the early 70's. At that time, nobody had any understanding of how to deal with the complex genetic information of a higher organism. There was a feeling that if you started by studying a few very small tumor viruses, you would somehow figure out in a very mysterious way how a complex organism like a mammalian cell would work. Somewhere in the hinterland of science there were people working on restriction systems in bacteria - how bacteria defend themselves from foreign DNA molecules. Three years later the entire methodology of recombinant DNA had evolved. I think there is reason to think that there might be substantial progress in the protein folding problem as well.

One other category to mention is the renewed emphasis on trying to understand how catalysis works. Two developments that have changed conventional biochemical attitudes are abzymes and catalytic RNA.

Abzymes are a hybrid between antibodies and enzymes. Nature has only designed a few thousand enzymes to do a relatively small number of chemical reactions, but the immune system can make about ten million different antibodies. People have found techniques to make antibodies to compounds that mimic the transition state of the reaction that they want to drive. They found that these antibodies have catalytic activity.

RNA has existed in the molecular biology paradigm as the middle man in the flow of information between DNA and protein. Several years ago, it was found that RNA can have enzymatic activity, as do proteins. This has become an intensely studied area and people are talking about RNA as the primordial molecule in biological evolution.


G. Bruce Robinson

I'm in the chemistry department at the University of Washington. The old Chinese proverb is that the journey of a thousand miles begins with a single step. What scientists have to worry about is what do I do when I go into the laboratory today. Most good scientists should look to what they could accomplish in 20-30 years, but then try to build a path to that on a daily basis. We have to get there based on the things that we know how to do already. It is thus an enormous task for all of us to consider what we can do now that will lead in the direction that Eric Drexler has indicated.

I find it interesting that many scientists did not want to respond [in the Delphi survey, see above] to questions about nanotechnology policy because I know from my own experience that many scientists are looking towards these kinds of technologies, but perhaps only in terms of a specific contribution that they might want to make toward that goal.

We have to drop a few steps and start with self-assembling systems as intermediates toward self-replicating systems. What is really required is more control in the technologies that we have now over the ways in which molecules assemble. What molecule do you want to assemble? What are the problems in assembling it currently? What new techniques are necessary? These questions will lead to an enormous multitude of paths, most of which will be blind alleys. There will be a few individuals that find the "NorthWest Passage."

Another problem that scientists have is, once they've made something, to prove that they've made it. This usually requires techniques where 1013 molecules is a small number, although sometimes it can be done with ~1010 molecules.

One last comment: the more you talk about more distant technology, the less that it is correlated with what we can do in the laboratory. We have to build backward from a vision to what we can do now. Building the very first step is quite difficult.


H. Discussion

J. CRAMER: I'd like now to strongly encourage participation from the audience.

AUDIENCE: I was reading in The Tomorrow Makers [Grant Fjermedal, 1986, MacMillan Publishing Company, 866 Third Avenue, New York NY 10022] that pioneers in AI often had hobby interests as children that gave them a real head start on their scientific careers. What could you suggest for children today that would give them an edge in molecular technology 15-20 years from now?

J. CRAMER: You're suggesting there should be a kit left under the Christmas tree that would be about nanotechnology?

AUDIENCE: Over in Bagley Hall [UW, Dept. of Chemistry] there is a scanning tunneling microscope that almost looks as though an experimenter could put one together in his basement.

J. CRAMER: As a matter of fact, in the student Nanotechnology Study Group there is a student who is doing just that, but hasn't quite succeeded yet.

E. DREXLER: Tinker Toys are a bad model for molecules because they don't have tetrahedral coordination. Buy a kid a molecular model kit early on and let them get used to the geometry of molecules.

J. CRAMER: Also there are some pretty nifty computer programs coming along that will do molecular modeling on your own computer screen. Right now these are the property of professionals that spend many thousands of dollars, but soon they should get much cheaper.

AUDIENCE: Perhaps some of the advantages expected from nanotechnology could be used as an argument for putting more money into the educational system in this country, which is in such disrepair right now.

G. FJERMEDAL: The guy in our group who has the nanoscope all but built is here: Rick Burton! With a mail order catalogue, he got the piezo-ceramics, he worked in the physics department with the milling machine, he built a board to connect it to his PC clone. What have your costs been so far?

R. BURTON: I think one can be built for less than $500.

G. FJERMEDAL: Rick showed me a catalogue last night where they had a rock-bottom one for $69,000.

J. CRAMER: Maybe we should point out that initially, when the STM was developed, it was believed that it would need expensive high vacuum equipment. It turns out to work quite well in air or under water or other media.

AUDIENCE: I'd like to point out that the technologies we're talking about - STM and biotechnology - are less than 30 years old so that when we're building general assemblers decades in the future we will probably have very different technologies.

J. CRAMER: That's a good point. Looking in the future is for steering, not charting a path. At every step, the details have to be revised. Someone pointed out that, if after the Civil War, the federal government had decided to build a machine to put music in the home of every American, what you would have gotten was some sort of one-man-band, automatic pipe organ rather than a modern stereo because no one at that time would have been able to chart an accurate course to get from that point to where we are now.

E. DREXLER: The reason for making a 5 year plan is not to tell you what you will be doing in 5 years; it is to tell you what to do today in order to get somewhere interesting. Next year you make another 5 year plan. The reason for a 20 year plan is to help you make a more sensible 5 year plan.

V. McINTYRE: Is there a drawback to being in the forefront of this? The USA was in the forefront of television and now we're stuck with resolution considerably less than the Europeans have.

J. CRAMER: In cases where it is necessary to establish standards so that everyone has to march in step, the first wave of technology is likely to get locked into a form less desirable than what comes later. It isn't clear that applies in this case.

B. ROBINSON: It seems that the leading edge of a technology moves around the world; i.e, from the US to the Japanese, who are now scared of the Koreans. You may be first in the beginning and then again several rounds later.

AUDIENCE: Does anyone know where the Japanese are headed with their research program?

E. DREXLER: I don't know in detail. The grand visions that they have sketched for their Human Frontiers program (which they subsequently funded at a miniscule level) sounded a lot like steps in the direction of nanotechnology: molecular mechanisms, building complex molecular machines, etc. Their Human Frontiers program is intended to be an international cooperative effort, and I hope that this country and other countries will be cooperative.

J. LEWIS: If you look at the technical literature, the Japanese are present but not over-represented. If you're worried about Japanese competition, the real challenge is not at the level of basic research.

J. COVINGTON: The Japanese had a very ambitious fifth generation computer program for artificial intelligence, but it didn't work because the standards that they established in the beginning didn't fit in the end. We could get in the same mess by saying now, for example, that biotechnology is the way to go when instead we need to be flexible all the way along.

J. CRAMER: A very good point. Freeman Dyson gave a talk here last year and described the fallacy of premature choice. In areas where a lot of money is being spent, e.g., NASA, and there are several alternatives, funding agencies show an irresistible urge to kill all the alternatives except the one favored (usually for political reasons). You end up marching down one path that might turn out to be a blind alley. One wants to proceed along a broad front when you don't quite know where you're going. I don't think that should be much of a problem for nanotechnology because the up-front investment is small.

M. THOMAS: I'd like to respond to the comment about the fifth generation project. Indeed they locked themselves into some technologies and languages that appeared to be the most favorable but turned out not to be adaptable. But they learned from that, the project is on-going, and they sent many people over here to learn our approaches, and they are even with or ahead of us in some aspects of AI. They just started a new initiative, the sixth generation, to develop biological computing devices.

G. FJERMEDAL: Carnegie-Mellon University recently instituted a nanotechnology studies institute. A few years ago, I talked to its director. At the time he could not get American funding, but he had a large group of Japanese businessmen willing to underwrite his entire project. One project concerned rhodopsin, the protein in the rod cells of the eye that reacts to light. They were going to paint some onto a disk to get optical storage. He finally got US funding, but the Japanese are very alert to developments in these fields and anxious to fund American scientists.

B. WEBB: It seems to me that money gets allocated to R&D in four ways. (1) Tradition. That's obviously not going to be effective for nanotechnology...

J. CRAMER: Since I have a lab that's been funded by DOE for the last 20 years, let me comment. One of the investment strategies for government funding is to invest in track records. To give money to people who have done good things in the past so that they can do more good things, and that is not such a stupid thing to do.

B. WEBB: ...(2) By fad. (3) Things that please the generals. I don't think that we want to see nanotechnology developed as a weapons system. (4) By bottom line-- somebody thinks he can make a profit from the R&D. I've heard a lot about social implications and technologies, but not much about products. I'd like hear from the panel some thoughts about near term products that would excite the market-place.

J. LEWIS: Certainly one large area is the pharmaceuticals industry. Nothing that they are doing is nanotechnology, but there is a large interest in protein engineering. That will not lead to building assemblers, but it will lead to a much better understanding of how proteins fold, which is a step in that direction.

J. CRAMER: I would say that the pharmaceutical industry has a ball and chain hanging on them, that their products have to be approved by the FDA. I think you will see spin-off's from that technology in areas where you don't need FDA approval. Non-drug biologicals might have a lot more growth potential.

G. BEAR: I think that devices that clean, paint, and refurbish surfaces will be an area that will require fairly simple nanomachines and will make a lot of money.

M. THOMAS: Based on Eric's studies, there are obvious applications to computer science. If we can shrink computers and put a Cray on everybody's desk, computing will gain enormous momentum.

E. DREXLER: Regarding pharmaceuticals, I was invited to give a talk at Upjohn. I told them that one possible application of protein engineering and learning how to catalyze new reactions is to develop a catalyst to put inexpensive reactive molecules together in a novel way to make much more cheaply a pharmaceutical that has already passed FDA approval.

AUDIENCE: If the approach to advancing research toward nanotechnology produced spin-offs at incremental stages during the natural development of the technology, I think government would be more inclined to sponsor non-defense research.

L. COBB: I think that a lot of paths that you talked about are already being funded by traditional means. Almost everyone of the participants in our survey is working on a path [toward nanotechnology] even though they might refuse to label it nanotechnology. Nevertheless, if Mr. Drexler and others are correct, these people will end up developing nanotechnology. As Dr. Robinson said, there are many different paths, and trying to correlate where you are now with where the technology might be in 50 years is nearly impossible. I think this is another reason many of the scientists refuse to participate in the survey: they refuse to say what will happen in the distant future when they are too cautious to say something they are nearly sure will happen soon.

J. CRAMER: I'd like to comment on the response of scientists to your survey. As one who is both a scientist, and thus not supposed to indulge in speculation, and a science fiction writer, and thus somewhat in the other direction, I can understand both sides. Scientists do not get rewarded for making outrageous speculations; they get punished for it by and large. If you say something that is surprising to most people in your field, you damn well better be able to back it up with a lot of evidence, logical arguments, and experiments to test whether your ideas are correct. Thus the sociology of scientists is such that they are not likely to joyfully participate in the sort of survey that you are describing.

L. COBB: We had to promise not to reveal their names.

J. CRAMER: People like Greg [Benford] and I can work both sides of the street and perhaps get away with it because our colleagues don't watch too closely what we're doing on the other side.

AUDIENCE: Does anyone subsidize weird science?

J. CRAMER: There is the MacArthur Foundation, which does something a little along those lines.

E. DREXLER: They wait until you're already famous and well supported by a university, from what I've been able to see.
Something I alluded to in my talk on Friday-- there is a real cultural difference between science and engineering. It makes no sense for scientists to talk about future discoveries beyond the sorts of things that they think they might find out about. In engineering on the other hand, I came out of activities during the 70's where people were asking what sorts of things could be done when we have less expensive access to space. In the aerospace field, people have a long history of looking 10-20 years ahead. We have a range of parts, we know what they can do, etc. Engineers are very much in the business of saying "Here is a proposal for an ambitious system that can do something as outrageous as flying a person to the moon" and delivering on it. That's very different from the culture of science. I don't do science.

G. BENFORD: I would like to add to the remarks about the scientific community. You shouldn't expect them to take any risks. They see no pay-off in public advocacy. However, one of the nice things about being a scientist who writes science fiction is that the media pay a lot more attention to you. If you want to have a piece in the LA Times about your opinions on, e.g., nanotechnology, there are clear ways to get that done and reach millions of people over the heads of your recalcitrant colleagues.
One reason for emphasizing products developed from nanotechnology is, first, it's sort of fun. Second, you can get people who hold stock options who might be very interested in listening to you. Third, one of the problems of technology in modern society is that if you can't reach into the ordinary life of an ordinary person in a way in which they identify you as both good and new, you have got two strikes against you. One of the problems of the nuclear industry in this country is that it has never been able to deliver a product that had its label on it. Electricity through the wall-- you don't know where it came from. Meanwhile, all the PR has been in the opposite direction, nuclear weapons, etc. They have never established their constituency. Nanotechnology could establish its constituency by producing -- a good wall cleaner!

G. FJERMEDAL: Make it so practical that Middle America would just embrace it! Bathroom cleaners, wood preservatives.

G. BEAR: Might I suggest the fabric industry. There is apparently a huge grant available for anyone who can remove rust stains once they are set.

N. SEEMAN: Nanotechnology is not today something that simply requires engineering; it also requires science. Scientists usually get their laboratories funded to solve problems within the context of science -- trying to solve day-to-day scientific problems. The reason I got involved with self-assembling DNA that I spoke about yesterday is that, frankly, I can't grow crystals to explore certain things on the nanometer scale. These research questions have arisen over the last 30 years with the advent of molecular biology. The guys in the lab will be doing more work to answer these scientific questions, not to develop rust removers. The development of nanotechnology, I believe, will come primarily from people trying to solve scientific problems on the nanometer scale. I think it's important to emphasize that what is being discussed as nanotechnology is largely the chemistry of the future.



FOR FURTHER INFORMATION on nanotechnology, write to:
The Foresight Institute
Box 61058, Palo Alto, CA 94306
For a donation of $5, you can get a packet of recent papers on nanotechnology, including ones that describe rod logic in more technical detail. For a donation of $25, you can subscribe to the Foresight Institute newsletter "Foresight Update."
[Editor's Note: For more up-to-date information, see the Foresight Institute home page.]

TRADEMARKS

® HyperCard, Macintosh, and Mac are registered trademarks of Apple Computer, Inc.
® SuperCard is a registered trademark of Silicon Beach Software, Inc.
® IBM is a registered trademark of International Business Machines Corporation.
® Guide is a registered trademark of OWL International, Inc.
® VAX is a registered trademark of Digital Equipment Corporation.


Next page of the NanoCon Proceedings
Previous page of the NanoCon Proceedings
Return to NanoCon Proceedings Table of Contents
Return to Jim's Molecular Nanotechology Home Page
Return to Web site design and authoring services

Comments and correspondence? Send email to Jim Lewis:
nanojbl@halcyon.com
NanoCon Proceedings © Copyright 1989, by NANOCON
This page is part of Jim's Molecular Nanotechnology Web, copyright ©1996 James B. Lewis Enterprises. All rights reserved.
Last updated 29June96.
The URL of this document is: http://www.halcyon.com/nanojbl/NanoConProc/nanocon10.html