Self-Replicating Space Habitat graduate school purpose and plans from 1988

Contents
* Plan of Graduate Studies
* Statement of Purpose [Submitted with original application]
* Ultimately ironic letter to my advisor considering how fast things fell apart and I was probably his worst student
* Projects to make happen at Princeton over the next five years
* Comments on the use of technology for social change

I'm posting this stuff here for archival purposes and in case they give others some ideas or encouragement for their own efforts. It's part of my scanning my own old paper archives. This was my proposal for graduate studies at Princeton University twenty years ago (and in some ways includes a proposal for creating a mini-Google and a mini-World-Wide-Web. :-). I fixed up some typos from the originals.

My graduate studies derailed in the first months over a conflict with the Director of Graduate studies, who was a different one than the one who admitted me as his last act (as they rotate), and the new one wanted me to follow a more conventional plan of Civil Engineering and Operations Research studies. Had the book "Disciplined Minds" been around then, I might have understood the graduate school cultural dynamics better and either avoided graduate school or plotted a better course through it. See:
  "Disciplined Minds: A Critical Look at Salaried Professionals and the Soul-Battering System That Shapes Their Lives" by Jeff Schmidt
    http://www.amazon.com/Disciplined-Minds-Critical-Professionals-Soul-Battering/dp/0742516857
    http://www.disciplined-minds.com/

And as with any story, there were lots more factors as well (including personal failings on my part). I'm still thankful to my graduate advisor at PU for giving me a chance, and I remain sorry for any difficulties I caused him. I think today I would have approached the situation with more compassion for everyone involved, especially those locked into a "disciplined mind".

Anyway, after these plans for graduate studies at Princeton fell apart, then I worked on the other plans in the late 1980s for an open source research institute and then a open source company which I have previously posted to the Virgle mailing list. And then I narrowed my scope down to just making a garden simulator, and then spent years working for others to recover financially from just that one garden simulator project.

The good news is that now, twenty years later, all or most of the hurdles have fallen that otherwise needed leaping before being able to comprehensively design self-replicating space habitats, and all the computer and informational resources I thought I needed then are now available for cheap or free. For example, for only a few thousand dollars, I have the equivalent of an early 1990s supercomputer in my office with terabytes of storage and a high speed color scanner and a network connection and access to Google and Wikipedia and so on. So, what I outlined in the 20th century is more and more doable in the 21st century for less and less cost. So, item 13 (the major goal) is now approachable without needing to do much on the other prerequisite items listed.

Many things change in twenty years. For young people now, it is also probably hard to understand in the 1980s how much the spectre of nuclear war as depicted in the movie "The Day After" haunted the times and discussions about society, and how much that influenced my approach.

Since I wrote all that, I broadened my approach to focus on systems that had more a balance of self-sufficiency and participation in a network. And the recent online book I wrote (inspired in part by the Virgle joke) also shows a broader perspective:
    http://www.pdfernhout.net/post-scarcity-princeton.html
And maybe I have more faith that people being nice to each other is a big part of any solution than displayed in what I wrote at the time. :-)

--Paul Fernhout

Plan of Graduate Studies

Paul D. Fernhout Plan of Graduate Studies 1/10/88

This document outlines my plan of studies for my graduate program at Princeton. I will describe my goal as outlined in my statement of purpose, my progress towards that goal so far, and my future plans to get closer to my goal. I am grateful for the opportunity Princeton University has given me to pursue these studies, as indicated by its acceptance of myself and my statement of purpose and its financial support.

As outlined in my statement of purpose, my lifetime goal is to design and construct self-replicating habitats. These habitats can be best envisioned as huge walled gardens inhabited by thousands of people. Each garden would have a library which would contain the information needed to construct a new garden from tools and materials found within the garden's walls. The garden walls and construction methods would be of several different types, allowing such gardens to be built on land, underground, in space, or under the ocean. Such gardens would have the capacity to seal themselves to become environmentally and economically self-sufficient in the event of economic collapse or global warfare and the attendant environmental destruction.

During the past semester, I have written one paper on this concept, entitled "The Self-Replicating Garden". Its thesis is that this concept provides a new metaphor for thinking about the relation between humans and the machinery that constitutes our political and technical support systems. Writing this paper has helped me organize my thinking and has given me a chance to explore the extensive literature relevant to the design of social and technological systems.

Having made this strong start, I must maintain the momentum of this project. After talking with professors at Princeton and exploring the alternatives, opportunities, and facilities available here, I remain convinced that Princeton is one of the best places for me to pursue this research. As outlined in my statement of purpose, and detailed further in other correspondence with my advisor, it is my intention to learn everything I can about the management of the large amount of complex information necessary to make possible the design, construction, and operation of these self-replicating systems. In particular, this will involve further library research, construction of a prototype information management system, and the construction of a prototype self-replicating biologically sealed system.

Towards this goal I have discussed projects with three Princeton Faculty members, who have all agreed to participate. The faculty and the projects are as follows:

Professor Alain Kornhauser (Civil Engineering): "The use of computer graphics and simulation in interdisciplinary research". My work with Professor Kornhauser will be the glue that holds all the other ideas together. This project will entail the evaluation of computers and the role they can play in multidisciplinary research, as well as what particular information recent studies have uncovered in appropriate man-machine interactions between the researcher and the computer.

Professor George Miller (Psychology): "The strengths and weaknesses of humans in solving problems". It is critical for me to understand the limits of human problem solving ability. Without a good understanding of this information, it would be near impossible to design a technical system that people can use effectively.

Professor Steve Slaby (Civil Engineering): "The role of innovation in structuring sustainable political and technological systems". Since the self-replicating habitat requires extensive innovations, it is important for me to understand the social and political consequences of these innovations and how they relate to the historical interactions between society and technology.

In addition to these three faculty members, I have also discussed my project with Professor Freeman Dyson of the Institute for Advanced Study. While Professor Dyson does not want to be formally responsible for supervising or supporting my work, he has said he finds my work interesting and is willing to discuss and critique it.

There are several additional faculty members who I would like to include in the project from the Civil Engineering, Politics, and Biology departments. I have not yet discussed this project in depth with them or asked for their participation so it would be premature to mention their names at this point.

My plan of studies is to arrange study contracts with individual faculty members, to count as reading or research courses under the relevant graduate departments. These paper titles will be roughly as follows:
1. "The use of computer graphics and simulation in interdisciplinary research" - Kornhauser
2. "The strengths and weaknesses of humans in solving problems" - Miller
3. "The role of innovation in sustainable political and technological systems" - Slaby
4. "The construction and maintenance of biologically sealed structures: biological analysis"
5. "The construction and maintenance of biologically sealed structures: structural analysis"
6. "The development of a prototype information management system"
7. "The economics of self-sufficient communities"
8. "The issues involved in developing a prototype self-replicating system"
9. "Construction of a prototype self-replicating habitat: Final report of success and failure"
These will be in addition to the two papers I have produced this semester:
10. "The self-replicating garden" - Professor Frank von Hippel
11. "Simulating Intelligent Interacting Objects in C" - Kornhauser, (published in AI Expert)
(It is my goal to publish at least six of these papers. I have already published one so far.)

I will construct a generals examination committee from the professors I work with on these projects, who will examine me on the material I have researched. Also, if I can arrange the required funding of $3,500 tuition, I would like to spend a semester at the New Alchemy Institute. The Institute offers a four course accredited semester on sustainable design, with courses on sustainable agriculture, energy efficient building techniques, community development, and a semester project on-site in small-scale technology.

This is a non-typical plan of graduate studies. However, it is totally in accord with my stated purpose in coming to graduate school at Princeton and makes the best use of my time and the resources available to me here. In addition, it appears to be within the written requirements of graduate studies as outlined in the Graduate School announcement. If I succeed at it, I will have two working prototype systems, and at least six publications. This would be a very significant set of accomplishments. All I am asking is a chance to let me try to achieve these goals. I acknowledge that as with any non-typical endeavor there is a high risk of failure. I am willing to accept that risk, and the possibility that I may not be able to accomplish what I have set out to do. In that case, if the university did not want to grant me a degree, I would accept that decision.

Statement of Purpose [Submitted with original application]

lifework n : the entire or principal work of one's lifetime; also : a work extending over a lifetime (source: Webster's New Collegiate Dictionary)

Lifework is a term that seems to be out of fashion these days. Not many people will admit to having one. I will. I have committed my life to the work of designing and constructing self-replicating habitats.

Self-replicating habitats could be built in space, underwater, underground, in the desert, or elsewhere. A habitat would consist of many square miles of enclosed land used for living, farming, and recreation. Examples of such enclosures are a cluster of kilometer high geodesic domes in the desert, a collection of hundred meter diameter spheres underwater, and a ten kilometer diameter dough nut-shaped O'Neill habitat rotating in space. The land area inside the habitat would have tall trees, and grassy fields. It would be designed explicitly to seem extremely open and natural to the thousands of people living inside. Each habitat would have its own power plant, running from solar, nuclear, or ocean-thermal energy. Each habitat would have attached manufacturing facilities for processing raw materials into finished goods. The raw materials would come from the area immediately surrounding the habitat, like the sea floor, the desert sands, or asteroids in space. Each manufacturing facility would have all the equipment needed to replicate itself and the attached habitat.

Why do I want to build these habitats? Most people would agree there is at least a one percent chance the human race will wipe itself out within the next century through a nuclear or biological war. The issue isn't even necessarily about our politicians making mistakes. The fallibility of the Soviet missile command computer technicians is what worries me most. Like anyone else familiar with computers, I know how easy it is to make a mistake with one. Beyond accidental warfare, expanding populations and industrial pollution threaten our lives just as much. I feel that even if there is only a one percent chance of ecological disaster over the next century, I want to do my best to ensure human survival in that case.

Most people do not think about these issues, or if they do, rapidly dismiss the problems as too large and impossible to do anything significant about. I feel I have an alternative to apathy or despair. Some habitats in space or underwater would probably survive a nuclear war. Unlike bomb shelters, they would provide an intact technological and cultural base from which to regrow our civilization. If there is not a war, they would still serve the useful function of providing more living space for expanding populations. Being a closed environmental system, they would also make people focus on recycling industrial pollution back into raw materials, leading to safer industries and a cleaner environment.

This may sound like an exotic idea filled with problems and promise, but why should I personally bother to do anything about it? Why should I risk my personal future to see it through to its conclusion? Perhaps the answer lies in my undergraduate experience. The university I went to placed a great emphasis on its students committing themselves to a life in the nation's and humanity's service. The best contribution I can make in the nation's service is to put all my personal resources behind this self-replicating habitat project, to get other people involved in it as well, and to make it happen as soon as possible. I can think of few other higher services than making sure humanity survives so that it can continue to debate the meaning of life, politics, and religion. I must admit though, this is also the neatest project I have ever worked on or thought of, and I can't stop working on it. Utopia is another word that is out of fashion in America today, and some people will say what I propose is a Utopian dream. Perhaps it shows how far utopianism has fallen, because the only major benefit I see of my Utopia is that humanity avoids extinction for a while longer. Nonetheless, while not concerning politics or religion, I guess this is a Utopian dream. But isn't it time for Americans to dream pleasant dreams again, rather than the nightmares of 1984 and "the day after"? Isn't it time to start looking forward to the future with hope?

Rather than trying to combat nuclear bombs with more weapons (as in the currently proposed Star Wars defense system) I would propose we combat those forces of destruction with more powerful forces of construction through millions of self-replicating habitats providing exponential productivity, massive redundancy, and far flung dispersal. Self-replicating habitats are a surer, safer road to survival than building more missiles or high powered lasers. Only at a place like Princeton can such an idea can get a chance to grow from the tiny seed it is now into an oak tree that may someday shelter our children. I admit that the above has been mostly an emotional appeal. I don't yet have the hard numbers to back it up. Part of what I will get from my graduate education is the knowledge of how to put numbers and hard facts behind this rhetoric, assuming this idea withstands academic scrutiny, which I have faith it will.

My graduate training at Princeton will help me achieve this lifework. As my undergraduate advisor, George A. Miller put it, a Ph.D. will make the things I want to do in life easier, for many reasons. To begin with, I do not have all the skills I need to do more than begin this project. My graduate eduction will help provide me with those skills. Beyond skills, with a lifework as strange sounding as self-replicating habitats, I need all the credibility I can get, and a Ph.D. will help give me that. Finally, being at Princeton will allow me to discuss my ideas and my project with other intelligent people. Perhaps I will even find some other people who want to work on this project with me. At the very least, I will have the ghost of John von Neumann (the father of self-replicating machinery and computing) looking over my shoulder as I work. If for this reason alone, there could be no more fitting place for this project to start than at Princeton.

The only way I and others can manage a task of this complexity will be to use computers to help us collect data, analyze it, reorganize it, design from it, and control the resulting system. One of the most productive ways to interact with computers is to use graphics, so I want to learn everything I can about interactive computer graphics and its use in exploring data, simulating systems, making design decisions, and managing a complex system. Princeton's program in Operations Research and Statistics in the Civil Engineering and Operations Research Department, in conjunction with the facilities of the Computer Graphics Laboratory, provides one of the best opportunities for me to learn what I need to know about managing a complex amount of information and using it productively.

The person I specifically want to work with, Professor Alain Kornhauser, is one of the most knowledgeable people from whom to learn the art and science of using interactive computer graphics to accomplish real world ends. Additional faculty from the Human Information Processing Group and the Cognitive Studies Committee, in which Dr. Kornhauser participates, also have much to teach that I should learn.

Now that I have mentioned why Princeton is one of the best places for me to do my work, I shall mention why I am one of the most qualified people to work on this project. I have extensive informal training in robots, automation, and manufacturing. My father is a certified manufacturing engineer and I have learned much from working with him on projects. I have extensive skills in programming in C and Lisp, and so I will not stumble over these hurdles. I have spent a year doing computer graphics related research with Professor Kornhauser, and I am familiar already with the equipment in the graphics laboratory. My formal background in cognitive psychology has prepared me for advanced work in statistics, information management, and human-computer interaction. I have published a conference paper on the subject of self-replicating factory control systems, which was a derivative of my undergraduate senior thesis. I will be presenting another paper in May on a graphical simulation of self-replicating robots I developed. I have another paper on self-replicating robot simulation currently being considered for a workshop on AI and simulation. I have a more general paper on graphical expert systems I wrote with Professor Kornhauser last year. I have even gotten my current company to tentatively agree to provide a small research grant of a few thousand dollars a year over the next three years towards equipment and computer time for studying self-replicating machinery design. My greatest disadvantage is my lack of formal engineering discipline. All I can say is that I really want to design and build self-replicating habitats, and if learning formal engineering is what it takes, I am willing and motivated to learn.

Paul D. Fernhout
Raleigh, NC

Ultimately ironic letter to my advisor considering how fast things fell apart and I was probably his worst student

[To:] Alain Kornhauser 6/18/88 Professor, Civil Engineering

Dear Alain,

Once again let me thank you for going out on a limb to get my application accepted to Princeton for the PhD program - especially given such a offbeat statement of purpose about self-replicating habitats. I had said to myself,
"The admissions committee probably will throw this in the trash, but if they don't, I will really want to go back." For whatever reasons, I got in, and now I am very motivated to come back to Princeton and achieve my goals. Of the three times I have gone to Princeton (as undergraduate, as staff, and now as graduate student) my attitude is the best and most appropriate to the work at hand this time.

I am planning to be the best graduate student you have ever had. I have set several goals: to get an A in every course I take; to have ten journal publications accepted before I graduate; to produce a useful information management system on the IRIS; to get an NSF graduate fellowship; and to bring in on average at least $100,000 a year in outside funding into the department. Since research grants have about a one in ten chance of being accepted, this last goal means I will be writing about a million dollars in grant proposals a year related to information management tools. I would like to do enough as a graduate student so that if I continue on in academia, I will be in a good position to get tenure somewhere within a few years after graduating. This is an ambitious set of goals, but with your help, I should have some chance of succeeding at least partially with most of them.

Accompanying this letter is a list of specific projects that I would like to make happen during my next four years at Princeton. I would like you to help me prioritize these and schedule them in such a way that the ones that are the most important, interesting, and beneficial to the both of us and Princeton get done the soonest and with the most probability of success. I realize that to do all of these is beyond the ability of just myself, for there aren't enough hours in the day.

One of my approaches will be to get projects started while simultaneously finding interested students or faculty members to help work on them. Others of these projects will entail me carrying them through from start to finish. Even if only a a few of these projects succeed, we will have accomplished much. I look forward to September.

Sincerely,

Paul D. Fernhout

Projects to make happen at Princeton over the next five years

By Paul D. Fernhout 6/18/88

Projects to make happen at Princeton during the 1988 to 1993 period:

(Included are guesses at development time and suggested figures for grant proposals to cover the research. Many of these projects don't require any funds, but as long as we are going to do the work anyway, we might as well get someone to pay something that we might put towards other things or towards general computer support. Almost all of this funding would be obtained from sources outside of Princeton. My aim here is to get all the necessary resources needed to design and construct self replicating habitats in place at Princeton so that I may use them. As a side effect, the entire research and teaching community at Princeton will be enriched.)

1. Various projects at Brown, MIT, Stanford, Bellcore, Xanadu, and Apple have produced hypertext information management systems. These seem to be the future of information management. I would like to put such a system on the IRIS [an early Silicon Graphics Workstation model of which there were many in a networked classroom lab] so that people at Princeton can use it and extend it. This should take a year's work, and probably cost around $50,000. NSF or Apple should fund this. We could put a hypercard compatible system on IRIS in the case of Apple.

2. Graphics consumes as much computing power and data storage as you can throw at it. With the IRIS's we have quite a bit of computing power. We still don't have ready access to massive quantities of data storage for data bases and scanned images. Tapes on the mainframe or IRIS are too inconvenient for the immediacy needed in exploratory data analysis. We need to put in an optical disk jukebox system (WORM or Erasable) so people using the IRIS's and the Hypertext system have as much storage as they can handle. A simple system of around 10 Gigabytes (10,000 Megabytes) should cost around $20,000. We might also convert the GE robots to putting optical disks in drives. They would be good at that. Eventually we could put in a 10 Terabyte (1000 Gigabytes) Optical Disk Jukebox system on the ethernet for the entire campus to use. IBM makes such a system. To explore the possibilities and get a prototype 10 Gig system running should take six months and cost around $50,000. IBM should fund this.

3. With a lot of data storage, we need an easy way to get graphics data into the system. We need to purchase several text scanners and start our own project on a neural net based text and image recognizer so that we understand the technology. A nice mix would be a top of the line Kurzweil scanner ($75,000), two medium resolution ($20,000 each) and ten low resolution PC ones ($3,000 each). Color scanners should be investigated. The neural network processing could be done on the IRIS's. This should all take a year to get well integrated with the Princeton network, and would cost around $200,000. Kodak should fund this.

4. As it currently is, the robot lab downstairs cannot be used for research. To produce a research quality facility we need to acquire two PUMA 560s ($40,000 each) with Lord Force/Torque sensors ($15,000 each), multiple end effectors ($30,000), an AGV ($50,000), four or five miniature mills and lathes ($5,000 each), and a material storage system ($30,000). The system would need a dedicated Unix box on the network with multiple serial ports, like an IRIS ($30,000). This all would come to about $300,000 and take a year to get set up so that it is an easy to use manufacturing testbed. Siemens, GE/RCA, or NBS might fund this.

5. For research on self replicating habitats, it is essential to have some space for materials research at the Forrestal campus. The research effort would be aimed at extracting aluminum and other substances from generic soil. This project will take two years and cost $200,000. Various materials companies like ALCOA might fund this, along with possibly EPRI or DOE, since this system would use a lot of power. This project could be integrated into the new Materials Science Program research agenda.

6. For research on the closed biosphere aspect of self-replicating habitats, Princeton needs a Biosphere II like test facility. (Biosphere II is a 2 acre terrarium like project going on in Arizona funded at thirty million to explore closed life support systems. Eight scientists will live in it for two years.) A few large Plexiglas covered geodesic domes should be built at Forrestal campus nearby the extraction facility. This would cost about $300,000 and would take a year. Issues to be explore would be environment/habitat energy transfer, construction methods and costs, air and water quality, maintenance, etc. A related issue would be to make these structures earthquake and disaster resistant. This should turn into an ongoing biosphere III project at Princeton within three years, involving faculty and students from Biology, MolyBio, Civil Engineering, and Energy and Environmental Science. EPA, NSF, and FEMA should fund this.

7. We need Common Lisp on the IRIS's. LISP provides the immediacy of APL but enjoys more widespread support. The cost per station is $1750 for Common Lisp from Franz Inc. The Princeton Computer Center should be able to get a site license for around $10,000. Lisp will need someone half time to support it, meaning about $15,000 a year indefinitely. The computer center should pay for this. If not, the cost could be bundled under other projects.

8. Using LISP, I would like to develop a graphical general purpose network analysis tool. Networks can be used in everything from architecture to robotics to system design. Princeton should lead the way in producing one tool that has many interacting subparts that allow anyone who can conceive of a network approach to a problem to be able to readily build and use the network in seconds to minutes. This tool will take two years to develop, involve several students and faculty, and probably cost around $300,000 in computer time and stipends, etc. I am currently talking with the T.C. Howard, president of Synergetics, an architecture and engineering consulting firm originally started by Buckminster Fuller. His company has much expertise on the use of network analysis in building structures, as well as in general design issues in engineering. We have discussed a SBIR to the NSF that would involve me at Princeton developing a few parts of this network tool and then using it to encode his expert knowledge about reticulated structure design and engineering project management. I have also talked with someone who is partner in a company that uses networks for personnel management. He might be interested in funding some work on that, or in a personal interest of his in using networks to represent biological ecosphere relations. Conceivably, as per the above, this project could be funded by a variety of small sources all kicking in around $20,000 to get their module written and in use. The NSF, DOT, EPA, and specific companies could all be approached. We must also integrate George Miller's Wordnet into this system. Network modules: Structural, Management, Project Management, WordNet, Ecosystem, Design, Hypertext, Communications, Transportation, Community Interactions, Factory Work Row, Expert Systems, Chemical Processes, etc.

9. Beyond hypertext and network analysis on the IRIS, we need more powerful information management tools incorporating CAD/CAM, CAI, expert systems, and graphical simulation. We need to give serious though to the development of an integrated package on the IRIS. This project would take 3 years, involve tens of people, and cost over $500,000. The result should be a system that allows experts in many fields to be immensely productive and creative. General Motors is a likely candidate to fund this.

10. Beyond these individual tools, we need a good system for allowing multiple people to work together on a project on the IRIS's (or across campus). This entails producing workgroup software for managing shared databases, electronic mail, and other communications and coordination systems. To get into this will involve at least a year and probably $50,000. Various electronic mail companies and software houses, as well as major companies and governmental agencies may be interested in funding various parts of this. Suggestions are Dupont, Da Vinci Systems, Apple, IBM, DEC, NASA, and Ford.

11. Beyond these computer workstation based tools, we need a good portable tool that someone can use when they are in class, or at home, on the road, or in the library. This could be based on a PC Laptop and should include a hand scanner. The system should readily transfer data between itself and the larger and more powerful information management tools mentioned earlier. This project will take six months and cost $20,000. A laptop manufacturing company might fund this, since it will be mostly software which will sell more laptops.

12. I would like to establish on ongoing self replicating habitat research group at Princeton. This group would invite speakers, fund graduate students, fund $10,000 level skunk work projects, give lunch time colloquia, and try to attract faculty and students to Princeton interested in related issues. This group will need ongoing funding ramping up from $1000 to $100,000 a year over five years. I would like to get various farsighted companies to contribute towards a million dollar endowment for this group

13. Using all the above resources that will be created over the next five years, I would like to go through several iterations of designing and prototyping a self replicating habitat with constantly increasing levels of closure. Closure is the amount of processed goods that must be imported into the system for it to replicate. Examples of early bottlenecks will be computers. I hope to have 95% closure by mass in the first prototype, 98% in the next, 99% in the next, 99.5% in the next, 99.9%, and finally 100%. This will be over the next ten years. The specific cost for this development will be 10 million dollars. NASA, SSI, the UN, and various other sources may contribute towards this. An essential first step will be a feasibility analysis so I have the figures and documentation to convince others that this project can really succeed. This first analysis will take six months and cost $10,000.

14. Finally, I want to port my self replicating robot simulation to the IRIS. This should take about two weeks. This system will provide a testbed for people to learn about the effect evolution and survival have on shaping intelligence.

This is a long list, and there may be other things that should be added to it. The question now is which of these things to do first, and how to go about doing all of them. I want to get as many people at Princeton excited about this project as possible, so that the entire endeavor acquires a momentum of its own. As soon as possible, I would like to devise a better organized and more detailed plan that I can show to other people in the department, as well as the trustees, President Shapiro, and Dean Kobayashi. I would like to involve as many people as possible in the project, including students, faculty, staff, deans, alumni, and trustees. Even if they do not participate in the research or funding, I want to keep them informed and involved from the start. I feel this endeavor has the capacity to improve the world, and I want as many people as possible to be favorably disposed to it and to help it along whenever possible. Without the support of a major institution like Princeton, such a complex project would stand little chance of success. With Princeton behind this project, there is hope.

Comments on the use of technology for social change

6/18/88

Alain-

Here are my thoughts on social change and how self replicating habitats relate to it.

The one thing that stands out in my mind that I learned at Princeton came from a Science and Human Affairs course taught by Professor Michael Mahoney. We studied Langdon Winner's book Autonomous Technology. As I came to understand it, Winner's thesis was that the introduction and use of a new technology had more of an effect on people's lives than the way that technology was used. The effects from the introduction of a single technological innovation like the automobile, the telephone, the aircraft, the atomic bomb, the factory, or even agriculture has done more to change the way people live than all social movements put together. What I have realized is that if you want to change the world, you should introduce a new technology.

Why do I want to change the world? I agree with Einstein and others that the atomic bomb has changed the world and our thinking has not caught up with it. Our sprawling industrial base is like a single celled organism and is extremely vulnerable to disruptions like atomic or biological warfare which may cause the end of life as we know it. Beyond that, the need to control and coordinate vital relationships across the planet has led to massive governments who must exert detailed planetwide control and which often run into conflict with each other. The failure of one or more of these major governments could spell disaster for everybody. With its high degree of planetwide interdependence, the structure of our economy leaves us vulnerable to isolated failures in command and control propagating themselves to destroy the fragile world community.

There are two solutions. One is to get everyone to think globally and accept how interdependent we all are. If everyone thought this way, the arms race would stop, and people would be nicer to each other. I don't believe this will happen. A person tried to preach this two thousand years ago and everybody got so upset they nailed him to a tree.

The other solution is to return to the Jeffersonian ideal of local self sufficiency, where government plays a role in national defense but does not have to do much about local affairs. Given our current industrial base, this is not an option either. Everyone in the United States is highly interdependent on everyone else. The farmers depend on the city dwellers. The city dwellers depend on the farmers and other cities. Production of various goods is specialized at locations across the country. The way our economy is structured now we need big governments and other institutions providing detailed control to keep it running. Big governments lead to the earlier mentioned dangers of them propagating errors and running into conflict with each other.

How can we return to the Jeffersonian ideal? Or, how can people achieve local self sufficiency? Does this mean a return to the technology of the eighteenth century? It does not. By an appropriate design, a miniature technological base could be designed that would allow ten thousand people to produce all the technological goods we have now come to take for granted. They could enclose all their lands under a few big domes and thus survive and continue to grow in case of nuclear or biological war. (This is my gut feeling - I need to do a study to back this up.) They can produce power locally from coal, oil, solar, or nuclear sources, and they can extract all needed raw materials from soils around them. This extraction will take more energy than current technology, which uses enriched ores, which is why this technology wasn't developed earlier. But the extra energy used will be worth the self sufficiency it provides. With local communities self sufficient in air, water, food, computers, schooling, medical care, and so forth, national governments can then focus on creating innovative technologies to add to these systems, and to prevent any individual group from attempting to take over others by force. Additionally, if the technological base is so designed so that it can make a copy of itself, after the first such system is produced, it can then make copies of itself in its nearby vicinity.

Accordingly, this new technological base can be introduced gradually into both existing industrialized economies and into the third world. Even if current industrial powers choose not to adopt this technology, it could readily be used to industrialize the third world. Given such a technological base, a community is most likely to come into conflict only with other communities that are their geographical neighbors.

Humanity has built in methods for coping with and limiting aggression from neighbors. There are no such limits to aggression to unseen individuals across the planet. There is no psychological reality to pressing a button and killing millions of people on the other side of the planet. Given our interdependent planet, there are plenty of reasons to be in conflict with people across the planet. This makes today's world a dangerous place. However, in local conflicts, there would be a psychological reality to killing ten thousand of your neighbors - people who your children have probably intermarried with. With such limits on the aggressive impulses, such impulses could be channeled more constructively, as in extending the current community habitat, or in innovating new process technology.

I believe I have a solution to many of the problems that confront our world. This solution overcomes many inertias other proposals have had. It doesn't entail the US trusting the Russians. It doesn't require a cut back of nuclear weapons. It doesn't entail people giving up the luxuries they are accustomed to. It does entail the creation of new technology. We are good at that. Of all the things I could create to make the world a better place and to ensure my family's survival, I think self replicating habitats are the best. That is why I am working to create them.

-Paul D. Fernhout