Bootstrapping the Future: Smart Tools and Self-Expanding Systems
The Seed Factory Project,
6485 Rivertown Rd, Fairburn, GA, 30213
7 Apr 2018
- History is marked by people combining their labor with increasingly advanced tools to better meet their needs. Better tools have yielded benefits like easier work, a higher quality of life, and improved safety and health. The last two centuries have seen gradual replacement of manual labor with powered equipment. More recently, technologies like automation, robotics, software, and artificial intelligence have enabled "Smart Tools", which can replace much of the remaining labor. Our modern economic system is based on trading the fruits of specialized labor for the other things people need and want. If most labor can be replaced by smarter tools, we question whether the current system can continue to function. A production system that meets people's needs directly, using self-expansion, replication, and smart tools, may offer a solution.
- We first look at the history of these processes, leading up to the current economic system. We then identify economic problems with this system, including the potential for large-scale labor displacement by smart tools. Past attempts and recent proposals have not fully solved these problems. We propose a new approach, based on the same processes which have always been used, but shifted to networks of owner-operators and their tools. They bootstrap from smaller and simpler starter sets called "Seed Factories". They use their labor and skills to increase the diversity and scale of their tools, and upgrade to more capable ones. This includes making copies of existing tools as needed, and eventually adding smart ones. The growing networks are used to meet some needs directly, and generate a surplus that can be traded for the rest. Replacement of their labor by smart tools is not a problem for the owners of such systems, since they still benefit from the products being made. In this way the tools can solve the very problem they create. Systems that can bootstrap from a starter set, and grow to whatever size is needed, are not limited to solving economic problems. We finish by looking at other ways this approach can build a better future.
1.0 - Historical Processes
Self-expansion, replication, and tool use are processes which predate civilization, and even modern humans as a species. People have always exploited them in the past, and continue to do so today. A short review of their history is in order for a couple of reasons. The first is to understand how modern civilization arrived at its current state. The other is to identify what processes and methods we have available. Then we can combine them in new ways, and adapt them to meet present and future needs.
1.1 - Biology
Living Things on Earth have grown and reproduced for 3-4 billion years. They ingest materials and use available energy to convert them into new parts for themselves. Some of the new parts are for self-maintenance, and some are for growth. At some point in their life cycle they have grown enough to make copies at the cellular level by division. More complex organisms reproduce by means of simpler and smaller seeds or embryos, which then grow to maturity. The instructions for how to grow and replicate are generally stored in DNA, a dual-strand polymer with complementary pairs of units. The sequence of units direct the construction of various Proteins, which in turn perform a variety of functions necessary for life. One of these functions is separating the two strands of DNA molecules, and assembling complementary units to make a matching strand for each (Figure 1). When a cell divides, each copy then contains the full set of instructions so it can continue to live.
Self-reproducing organisms have DNA sequences 0.1-670,000 million units in length. Changes to the sequences, followed by differences in how the changed organisms live and die, have resulted in all the various species that exist now, and have existed in the long history of our planet. There are several scales at which these changes can happen: at the individual unit level, to longer segments of DNA, and by wholesale reshuffling in Sexual Reproduction. Until very recently, the changes were unplanned and undirected, so the evolution of species was random and slow.
All living things directly modify their surroundings by ingesting food, rejecting wastes, and using local energy sources to sustain themselves. Many species also indirectly modify their surroundings by building non-living objects outside their bodies, such as Nests, Burrows, and Dams. Once built, these structures can passively perform functions like shelter from the environment or protection from predators. The non-living objects don't need constant biological activity to sustain them. Communities of living things, together with the non-living portion of their surroundings that they interact with, form linked Ecosystems. These are dynamic entities through which energy and materials flow, often in cycles with feedback loops. As living things ourselves, people also interact with and modify our surroundings, build non-living objects, and participate in ecosystems with other living things. We depend on the self-expansion and replication abilities of other living things to supply our food, and other products we need and use.
1.2 - Technology
Some animals actively use objects which are not permanent parts of their bodies as Tools. Tools let them perform tasks more effectively than their unaided bodies can. Technology includes both the knowledge needed to make everything, and all the tools and products derived from that knowledge. Most animal tool-use is passed genetically or arises spontaneously. They lack a knowledge component, so we don't call them technology. Knowledge is learned, saved, and passed on to others. We don't know exactly when the first technologies arose that were passed on this way. Before that, they arose spontaneously each time, and were then lost when that individual died. Many of our relatives among the primates have been observed using simple tools such as sticks and leaves to gather food. Some groups of wild chimpanzees use tools to chop large fruit into smaller pieces so they can be eaten. Neighboring groups do not do this, suggesting knowledge is being learned or passed on within the group that does. It therefore would count as a very early technology.
We presume our early human ancestors also made this kind of transition from spontaneous to learned tool-use at some point millions of years ago. This was likely among the Hominid family of primates to which we belong, but an earlier species now extinct and distinctly different from us. It is hard to pin down exactly when and how the History of Technology began. Most tool materials are perishable over such long periods. Unmodified stones are durable enough to survive. But the ones which were used as tools, like the chimpanzees are observed doing today, are also indistinguishable from all the other stones which were not. Modified or characteristically worn stones are the earliest technology identifiable as such. We know this technology was being passed on, because of the persistence of particular techniques and their geographic spread over time. If they were spontaneous inventions, which were then lost when the inventor died, the timing and distribution would be more random.
Modern Humans appear to be the last surviving species or subspecies of the genus Homo, although DNA analysis indicates some interbreeding with close relatives. We emerged in anatomically modern form about 0.2-0.3 million years ago. This is long after the appearance of early technologies like stone tools, controlled fire, and using them for preparing and cooking food. Compared to other tool-users, modern humans seem to be unique in several ways. These include (1) the number of different objects we make, and their complexity, (2) significant transfer of knowledge from person to person, and indirectly by way of external records, (3) extensive use of tools to make other tools, and (4) creating tools which can carry out tasks to some degree on their own. These differences have combined to produce accelerating growth and an increasingly complex society.
Technology Accumulation and Acceleration - The key advantage technology has given us over genetic evolution is the speed at which it can accumulate and spread. It helped our ancestors evolve to modern form, out-compete our relatives, and build a civilization which dominates the planet. We are even at the point of artificially modifying genetics, doing so much faster than natural changes happen.
Tool marks on fossilized bones and shaped stone tools date back about 3-4 million years, which starts a period we call the Stone Age. These artifacts indicate making and using tools was intentional by that time. The spread of more advanced stone tools over Africa and Eurasia, and their similarity and persistence, indicates knowledge was being passed on. This makes them the first identifiable class of technology, if not the first to develop. Stone tools are harder and sharper than our body parts, and replaceable when broken, but limited by the muscle power we can supply. The next great advance was harnessing Fire. This increased usable energy, and allowed transforming materials in ways not possible using only stone tools. Fires occur naturally, so it is uncertain when we first learned to use and pass on fire-based technologies. Evidence from caves containing human fossils, hearths, and the remains of fires indicate the technology had been mastered no later half a million years ago, but possibly starting a million years earlier. Aside from the immediate uses for warmth and discouraging predators, fire led to cooking. Various cooking methods increase available calories, reduce toxicity and pathogens, and preserve some foods for later use. Stone tools, fire, and cooking gave us a survival advantage. It also allowed us to evolve to modern form, with large energy-demanding brains.
The arrival of modern humans brought additional technologies at an accelerating pace. These include clothing, language, ceramics, domesticated species, and many others. There is a strong positive-feedback loop that causes this acceleration: Improved tools conferred a survival advantage on early humans. Their numbers multiplied and they spread geographically. Assuming a constant rate of inventiveness at a given point in our evolution, more people means more inventions. A greater variety of living conditions in new locations encouraged new solutions to the problems they faced. For example, cold climates led to the need for clothing. New inventions, once made, could then spread elsewhere and be further modified, supplying greater survival advantages. On top of this feedback loop was the gradual increase in the size and energy use of our brains. This presumably increased our inventiveness over the last three million years, but it also allowed for larger social groups. Modern humans form groups of about 150 people, compared to about 40 for our distant ancestors. Larger groups allowed for specialization of skills and therefore becoming more expert in each, further increasing our survival and numbers. Specialized skills encouraged developing specialized tools for each task. For example, axes and the many varieties of chisels and saws all cut a material by concentrating forces at a narrow edge. But their sizes and shapes have evolved according to the specific materials and needs of the work. Specialized tools were an improvement over the previous generic ones, so we are back to increased survival in the feedback loop.
Knowledge Transfer - DNA encodes genetic information, and is passed from generation to generation. But it doesn't change at all, or very little, in the life of a given individual. It can slowly change as organisms adapt to their environment and specific traits encoded in the DNA are selected for or against. Learning is a process which allows more variable information to accumulate within the life of a given organism. It has evolved in many forms throughout the animal kingdom (Moore, 2004). For most species, the learning is lost when the individual dies and must be done anew each generation. Knowledge transfer is more efficient since the organism doesn't have to repeat all the trial and error in accumulating knowledge. Instead it receives useful information previously learned by others, and has the opportunity to improve on it and pass the improvements to the next generation.
Early knowledge transfer used methods like observation and imitation, with later additions like gestures and pantomime to focus attention. These transfer methods didn't require special tools. Language and memory techniques are technologies that allowed larger amounts of knowledge to be stored and transferred, over longer time periods. They need training, but still did not need external tools. Art and writing further increased the quantity, persistence, and accuracy of knowledge transfer, and began to store the knowledge externally to people. These techniques now required tools like pigments and chisels to put the information on durable media like stone. Our ability to store, transfer, and copy knowledge externally has only improved with time, with more and more advanced tools to do it with. Pictorial art, such as cave paintings and figurines, didn't require specialized knowledge to understand because they were similar to the originals they represented. Pictorial symbols were more abstract and required some learning to understand. Finally, abstract symbols like numbers and alphabets required the most learning to use, but could represent ideas and new things which didn't yet exist. Along with increasing abstraction, our storage media have evolved through marked stone, clay, papyrus, printed paper, and the rapidly changing electronic methods.
Sequential Toolmaking - Parts of the body are the universal first tools, but are not made in the same way as external tools. For a given species, body parts have a fixed repertoire of actions they can perform. For example, human elbows are limited to less than 180 degrees in bending. They are also limited in terms of strength, size, durability, etc. You can use your fist as a hammer, but only for certain light tasks without injury. Many animals use natural objects, like rocks and twigs, as tools to pound or poke. Fewer will modify natural objects to suit a task. Humans, and our past relatives, seem to be the only species who first make one tool, then use that to make another. Hammerstones are natural round stones used to shape other stones, or to crush, pound, or grind other materials. They are found, rather than made, so don't represent sequential toolmaking. A Hand axe which is used to cut and shape a Digging Stick, which in turn is used to dig up food is a sequence of two tools, both of which are made.
Sequential toolmaking requires enough memory and planning to accomplish the whole series. The steps may not be kept by a single person, but they all have to exist with a society. Modern humans have developed by far the most complex sequences of tools used to make other tools, including loops which lead back to earlier and simpler ones. As our social groups got larger, their members developed specialized skills. One of the skills became making tools for others to use. Specialized toolmakers persist as a significant segment of industry, since modern civilization can't function without all the tools they make. This includes making better tools than the ones we already have, a process we expect to continue in the future.
Self-Operating Tools - Domesticated species may be considered the earliest self-operating tools. They are improvements over acquiring food in the wild, and agriculture is an important technology, so they fit within our definition of tools. Since they are living things, hunting dogs can chase prey, livestock can forage, and cultivated plants can grow, all without constant human attention or work. Domestication only started about 15-30,000 years ago, while non-living tools have a much longer history. For three million years or so those tools were passive, not doing anything unless people used them. It wasn't until about 7,500 years ago, with the development of sail power, that non-living tools could start to perform active functions to some degree on their own. Wind provided an external energy source to move the sail, and the boat it was attached to. As long as the wind is steady, the boat can continue to move without constant action by people. This differs from rowing or paddling, which do need constant action. Any self-operating tool, whether a trained dog or a machine, needs an outside energy source in order to operate. For passive tools, people and their food are the energy source which makes them work.
Simple tools, like a hammer, need a person's constant attention and muscle power to operate. More complex ones, like an electric drill, replace some or all of the muscle power, but still need our full attention and guidance. A Mill powered by wind or water may not need constant attention to keep turning and grinding, though it still needs labor for other tasks. It is therefore partially self-operating. Automation, robotics, computers and their software, and now artificial intelligence, are recent technologies which further reduce how much attention and intervention by people is required. Collectively we call tools that use these technologies Smart tools. If our tools no longer need our strength, attention, guidance, or control, the need for our labor is reduced or eliminated.
In theory, a set of sufficiently smart tools could be given instructions, then then proceed to build everything else we need and want with little effort on our part. This includes making copies of itself. In practice, we don't yet have tools that smart. The ones we do have, though, are increasingly able to work on their own and need less help from us. Smart tools are the latest in a long series of Productivity Improving Technologies. Like new sources of energy and mechanization, they will lead to a better quality of life, but in the process can disrupt a system based on trading labor for other goods and services.
1.3 - Economics
Prior to the development of tool-use and technology, our ancestors survived the same way as other animals. They were part of biological ecosystems, and obtained their food mainly from other living things. Both primates and their food sources were self-reproducing, and self-expanding to the point of filling their ecological niches. Goods and services, and therefore trade, were limited in scope, consisting of highly local things like sharing food and grooming. Studies of living primates indicate they have a sense of fairness and reciprocity, so our ancestors likely did too. These behaviors enhanced group survival, but they also underlie modern trade.
Early Trade - By the start of the Neolithic, around 12,000 years ago, modern humans had developed a number of technologies and tool types. These included an assortment of stone tools, fire, cooking, clothing, some domesticated species, and ceramics. Prior to the development of agriculture, societies were more mobile, because they had to follow or go to where the food was. This limited the amount of tools and other artifacts they could accumulate to the level of seasonal camps. Whatever they could not carry to the next location had to be left behind and made anew, or if durable enough, stored in caches for later return. Since people at that time were anatomically modern, their social groups could average 150 members, and people could specialize in different tasks. Specialization requires trade for the goods and services made by other specialists. The mobility of hunter-gatherer cultures meant they also encountered other groups in the course of their seasonal movements. This also encouraged trade between specialties among the groups, or trade of items and materials which came from more distant places.
The Neolithic Revolution marks the transition from hunter-gatherer to agriculture and fixed settlements as a way of life. This seems to have happened seven times in separate locations around the world. Domestication of plants and animals increased their quality as food sources. Intentional planting and care of herds greatly increased the food supply relative to the natural state. The increased quality and quantity of food allowed for large increases in population density. Raising plants and animals was an investment of labor with a delayed return. The need to protect the work that went into agriculture led to less mobile populations. Fixed settlements then gained the ability to store food reserves, better protect people and their animals from the elements, and protect them from natural and human predators. At this stage people were still replicating themselves and their food supply through biology, and copying their tools and artifacts through technology. Individual settlements gradually expanded, and their numbers increased by starting new ones in in different geographic regions.
Economic Evolution - The increased work that went into agriculture led to property rights as a way to secure that work. Increased productivity in food production allowed more people to engage in other crafts and services. One of the new specialties was trading items made by others, i.e. merchants. The amount of goods a merchant can handle isn't limited in the same way as a single farmer is limited in the area of crops they can grow. Ownership of the goods can be vested in one person, but they can hire as many people as needed to do the physical work. Similarly, a large area of land can be acquired by conquest or purchase, and then worked by many slaves or tenants. Concentrated ownership made it possible for much larger gains than what one person could do on their own. Another route to unequal wealth was the use of force. Taking what others have made, by pillage or taxation, is also not limited by one person's labor, especially if they hire soldiers or tax collectors.
The development of property rights, division of labor, organized trade, accumulation of wealth, and taxation all came out of the Neolithic Revolution. They produced an economic system with recognizable elements that persist to this day. World population was on the order of 5 million at the start of the agricultural transition. By the time iron production became widespread, about 1000 BC, it had grown to about 50 million, and by the time of Classical Greece to 150 million. The much larger population led to more rapid improvements in technology, including in economic systems. The trend of population growth and accelerated change has continued to the present day.
Modern Economies - From a technical standpoint, modern economies are highly specialized. The US Bureau of Labor Statistics tracks 818 Detailed Occupations, each of which are further specialized. For example, farmers and ranchers are listed as one occupation type, but dairy farms, field crops, and fruit orchards all use different skills and equipment. The US Census Bureau tracks 1,057 different industries under the NAICS Standard, all of which use different tools and skills to produce different goods and services. From an operational standpoint, modern economies are based on a mixture of voluntary trade and action, and involuntary use of force, actual or implied. To the extent people find the functions of government useful, and are willing to pay for them, they are funded voluntarily. There are always some people who must be forced to contribute against their will, so it is not all voluntary.
When the dominant mode of trade and action is voluntary, we call it a Market economy. When that economy also includes large proportions of capital accumulation, wage labor, and trade using money, we call it Capitalism. Modern economies typically include a capitalist sector, smaller scale individual trade, plus a variable sized government sector whose control varies from autocratic to extensive public choice and direction. Trade is now interconnected on a world-wide scale, so there are no truly isolated economies any more. There are many local variations in the scale of government, capitalist, and individual sectors, in the extent of physical and social development, and in the local environments and resources which underlie the economies.
Capitalist economies have an implicit goal of perpetual self-expansion. This is driven by the desire for a better quality of life, and the accumulation of wealth and power. These drives are expressed in monetary form by increasing profits and market values. Perpetual growth for some can be at the expense of others. Over-consumption of resources and discharge of wastes in the pursuit of growth can damage the environment in which we all live. So while modern economies have brought many benefits, they have also created problems, some of which we consider in the next section.
2.0 - Economic Problems and Approaches
Our modern economic system is mostly based on large-scale capitalism. By large-scale, we mean large amounts of capital vested in a smaller number of people and organizations. They then employ larger numbers of people for their specialized labor, and pay them wages in return. For example, in the US, only about 10% of workers do so for themselves rather than others. The capitalists keep a share of the revenues as profits. The majority of people, both wage-earners and capital owners, then use the money they obtain as an intermediate good to trade for the other goods and services they need. The goods and services in turn are mainly produced by the capital owners and their organizations, forming a complete trading system. A system where most people work for wages is a relatively recent historical development. For example, in the United States (US) the majority of workers were agricultural until 1880, and worked for themselves (see Table 2 in Output, Employment, & Productivity in the United States, Brady, NBER, 1966).
The current system is fairly efficient, and has allowed for rapid economic growth over the last few centuries. But it has also caused problems due to the pace of change, job insecurity, and economic inequality. In the future, smart tools may cause wholesale work displacement, and break the cycle of money by which the system operates. The current system has other problems, such as environmental impacts, and the social effects of spending so much time at work and commuting rather than with family. However, here we mainly focus on economic problems, their effects on people, and attempts to deal with them. Large-scale work displacement isn't yet severe, but we will look at proposals to deal with it. It is unclear if these proposals would work on a large scale, so additional solutions are desirable.
2.1 - Economic Problems
Modern economies evolve, and many of their problems can be traced to the effects of change. Other problems can be traced to inequalities of resources and distribution. We group these problems into three current ones: the pace of change, job insecurity, and economic inequality; and one future one: work displacement.
The Pace of Change - The time since the Neolithic Revolution is only 1/300th of the time since the first recognizable tools. Agriculture allowed a large population increase, which then greatly accelerated a feedback loop. The loop consists of (1) more living people are able to invent better technology more quickly, and (2) better technologies then allow more population growth. History since the start of agriculture has therefore been marked by a general acceleration of change. As long as the changes were slow relative to the working life of an individual, they were not a major problem. A given person could continue their current craft, with their existing skills and tools, until they stopped working. They could gradually be replaced by a new generation of people following new methods. So long as most people worked for themselves, which until the last few centuries largely meant farming, they could adopt new tools and methods on their own initiative if they so chose. The capital required for such changes was not yet beyond individual means. For example, the horse collar, horseshoe, and heavy plow increased farm production by 50% and allowed working heavy soils that were previously unsuitable. These improvements were not unreasonably expensive for farmers at the time.
In recent centuries, the pace of change has greatly accelerated. Hand weaving of cloth was replaced by machine looms, and farm tools went from animal to tractor power over time spans of about 50 years. In the late 20th century, electronics repeatedly doubled in performance every few years. The capital required to produce and use the new powered machinery often grew far beyond individual means. For their work to remain economically competitive, people were forced to work for those who had the capital and machines, rather than themselves. Old skills and entire job categories could become obsolete within a working lifetime. People often could not afford to enter new fields, due to time and expense of retraining, or the capital required to start in them. In many cases people had to take whatever work was available. This was despite not being what they wanted to do, or if they were suited to it. In some cases, their old skills were no longer useful, and it was too late in their lives to learn new ones, so they were simply left behind.
Job Insecurity - Although farming has many risks, getting laid off for business reasons is not one of them. Labor is a cost to any business, so owners and management have a constant incentive to reduce it, and thereby increase profits. They generally employ people only when the expected income from those workers exceed their added costs. Shifts in market demand, competition, and new technologies can all make previous jobs uneconomic. Businesses may then resort to layoffs if other worker attrition, from retirement or choice to work elsewhere, isn't fast enough. The level of job insecurity from all causes combined is greater than that caused only by technological change. It includes factors like making bad business decisions or losing market share to competitors. These can happen even if the total market size and the technologies used haven't changed. Changes which affect employment in a given enterprise constantly happen, so a percentage of people are always at risk of losing their job. In the US, the actual rate of involuntary job separations is about 15% per year, with between 3.5 and 10% unemployed at any given time. Lack of work is a problem in a system where most people rely on it as the means to supply their basic needs. Our economic system assumes you trade your labor for money, so that later you can trade the money to others for the goods and services they can supply you.
Economic Inequality - Early humans had to work together and share to ensure survival. Their seasonal movements did not allow for accumulating many physical goods. So economic inequality was limited. Since the development of property rights and gaining from the work of others, the inequalities could grow. For example, ancient kings accumulated far more wealth and power than their slaves or peasants. We have a higher average quality of life today, but these inequalities persist. Stored wealth provides a reserve to withstand setbacks like job loss or illness. Low income inhibits building such reserves, so setbacks can be catastrophic. Lack of resources can result in poor education, which can perpetuate in later generations. Poverty also can result in poor nutrition, health care, and living conditions. Relative poverty can persist even in developed economies, and even while the poor are working.
Work Displacement - New technologies, such as automation, robotics, software, and artificial intelligence, have been developed since the mid-20th Century. They enable Smart Tools, which are increasingly able to replace part or all of the work people do in existing jobs. The tools may still need a person's attention part of the time, but significantly less than when the work is done without them. If fewer people are needed for a given amount of work, then without other changes, those people become unemployed.
The technologies of the Industrial Revolution replaced large amounts of manual labor in the past, but the pace of change was slower. People were able to transition to work in other fields, including new ones enabled by better technology. The problems with smart tools is their ability to be rapidly introduced, and to do jobs so well that humans aren't needed at all in doing those tasks. Demand for new types of goods and services can create new jobs, but some people are not suited for high-skill jobs like programming and designing robots. The number of new jobs may also be much smaller than the ones lost. If new jobs are not created fast enough the unemployment rate may increase in coming years.
Some examples of current and near-term work displacement include:
- Self checkout at supermarkets and card readers at gas station pumps. These have reduced the need for cashiers.
- The mass introduction of warehouse robots by Amazon, which reduces how many workers are needed to fill orders.
- The ongoing development of self-driving vehicles, which may affect all kinds of transportation and delivery jobs. Automated warehouses plus automated delivery puts all kinds of retail jobs at risk, by entirely bypassing retail distribution.
- The development of AI agents like Alexa, Siri, Cortana, and Google Assistant, coupled with augmented/virtual reality devices. They are currently used for home automation and entertainment. But in the future, schools may face pressure to replace some of the teacher's work with interactive systems using these kinds of tools. What happens to those teachers and school systems in that case?
Note that the wealthy, those with sufficient retirement income, and beneficiaries of social programs are not at immediate risk of becoming unemployed. Their needs are met in other ways than through jobs. Also a significant number of people are dependents of workers, so in total only about half the US population is currently working or between jobs, and directly at risk of displacement. However, large-scale replacement of work has broader effects on the entire economic system than just the people who are out of work. It can eventually affect everyone.
Without earned income, the unemployed don't pay income and payroll taxes. They don't buy as many goods, so sales taxes also go down. Therefore governments will lose a share of the revenue they need to function. Lower government revenue means social programs and government employment also decrease, further lowering economic activity. The businesses the unemployed no longer patronize also lose income and employment, even if they use smart tools. The owners of such businesses lose a share of their income too. Lack of enough paid work is then a systemic problem for a capitalist economy which depends on it as an integral part. Such spirals of contracting economic activity are called recessions or depressions, and cause large-scale unhappiness and even civil unrest. These are undesirable results. If the existing system can't handle this problem, then what sort of system should replace it? If changes are required, how do we manage the transition?
2.2 - Previous and Recent Approaches
Modern humans are Social Animals. We constantly interact with other people and depend on them, rather than operate as lone individuals. Some things we can do for ourselves, such as cooking meals at home. But for many other things, such as the electricity and natural gas for the cooking, the production of the appliances and utensils we use, and the production and supply of the food to cook, we rely on other people. There have always been members of society who are disadvantaged in some way, or can't fully contribute to the functions of society. Examples include the very young or old, and those who are injured or sick. Because we function as a society, we care for others and realize we may become disadvantaged ourselves at some point. Care for others is carried out at the level of family and friends, through voluntary contributions and organizations, and through involuntary programs funded by taxes and operated by governments. The care has one of two goals. The first is solve the person's disadvantages if that is possible. So we raise and educate children, and heal the sick and injured, so they can function as part of society. If the problems are unsolvable, such as the ones that happen to very old people, our goal is to give them at least a decent quality of life.
The economic problems noted in the previous section affect those who are able to work, but disadvantaged by their circumstances. Their skills may be obsolete due to technological change, or their employer no longer finds their work useful enough to cover their wages. They then are left without income buy the things they need. They may not have the capital, or access to it, to purchase improved equipment. They may also lack the time, funds, or ability to learn new skills. These disadvantages leave them unable to enter new fields and continue working. In the future, so much work may be displaced by smart tools that there simply isn't enough left for the people who want or need it.
People have dealt with past economic problems the same ways we deal with other problems in society. They approach them as individuals, through family and friends, voluntary organizations, or involuntary programs. Previously used methods have had varying levels of success, and none have yet had to handle wholesale replacement of work. Proposals to meet the challenges of smart tools may be inadequate. To the extent they are not full solutions, or we are uncertain of their future capacity as such, there is a need for additional new approaches.
Previous Approaches - Everyone is born unable to care for themselves, and without the knowledge and skills to function as part of society. The general approach to solving this is care and education of the young by their parents, and some combination of others, including interacting with other children. Childhood education supplies the basic skills for self-care, language, and other areas that are assumed to be needed by everyone in society. More advanced education and training then supplies specialized skills. These enable trading work for income. Some people follow their interests rather than skills for maximum income. This is only possible in societies with enough surplus to support optional activities beyond those needed to live. Generally childhood care and education is provided by parents first, then by government programs funded by taxes. The more advanced education is provided in a variety of ways: by government programs, by private sources and institutions who have sufficient funds, and by students borrowing against their future income. Some retraining programs exist for people in later life who choose or are forced to change fields. The education system generally works, although there are some problems with funding, personal cost, and access.
For people who are disadvantaged in other ways besides being young, another general approach is wealth redistribution. This is mainly via taxes and social programs, with some voluntary contributions and organizations. Wealth normally flows upwards towards those who already have some. They profit from the work of those they employ or lend capital to, in addition to whatever work they do themselves. The profits are unlimited, unlike their own work, so their wealth can rapidly accumulate. Redistribution can then have two purposes. One is to avoid extreme accumulation by the wealthy at the expense of everyone else. The other is to obtain funding from those who are more able to give, for the general protection and improvement of society as a whole, or directed to the less fortunate or able.
Redistribution happens when the incidence of funding sources is different than the social distributions they provide. When a billionaire is taxed to fund a public school system, the net distribution is downwards. But this is not always true of taxes and government programs. For example, government contracts with companies that have highly paid workers and wealthy corporate owners may move funds higher on the wealth scale than the people taxed to pay for them. Social programs generally intend to help those who need it via redistribution. One such program to deal with job insecurity is unemployment insurance. Employers pay a tax to fund this program, which then temporarily pays people who can't find work. Another set are social insurance and medical programs funded by employers and the workers themselves. People eventually grow too old to work, and need more medical care. They can also become disabled or die when younger. The programs then provide income and medical coverage for former workers and their dependents. The redistribution effect comes from benefits being capped while funding has higher limits, and from minimum benefits being supplied even if lower income doesn't cover them. A third set of programs are aimed at re-education and retraining for people affected by economic changes.
In addition to voluntary and government programs, people are encouraged to save from their current wages, to cover adverse conditions like job loss or health problems, and for eventual retirement. They can also pursue self-training and education. People helping themselves is useful, but not always possible given their circumstances. If their income is low, there may not be enough surplus to save. Their work and family care may not leave enough time to gain new skills. Overall, the set of approaches used to deal with economic problems have worked well enough that people haven't yet demanded large changes to them.
Recent Proposals - The rapid development of smart tools has the potential to cause large-scale displacement of work, without creating enough new work in other areas to replace it. The concern is for the social and economic disruption this can cause in a system based on trading work for money. The most commonly proposed approach involves increased government redistribution from those who have, to those who now don't.
One version is called Basic Income, or sometimes Negative Income Tax. It is intended partly to replace other existing social programs with a simpler method, and also to address permanent work displacement. The concept is to provide payments to every adult, or in some versions every individual, regardless of need or income level. It would provide a minimum floor to living standards. People and businesses with sufficient wages and income would be taxed to fund this system, much as they do existing programs. An alternate version is to increase government employment and funded work programs, such as a Federal Job Guarantee This may be more acceptable than simply giving money to everyone, since the recipients are seen as working for what they get, and the work itself can be useful.
The problem with these approaches is they can't be sustained in the face of large-scale replacement of work. As we noted above in section 2.1, the systemic secondary effects of work displacement reduce many of the other sources for taxation. There would not be enough tax revenue left to make the payments from. Very high tax rates on the remaining sources encourage people and businesses to move elsewhere to avoid them. Unless the new taxes are applied everywhere at the same time, the incentives to move away will be strong. We see this already with corporations and the wealthy moving funds "offshore" - to another country or jurisdiction) to avoid taxes. The alternative of simply creating money to fund the payments would cause inflation, which has other undesirable side effects.
2.3 - Composing a New Solution
Previous and recent approaches have not fully solved the economic problems which already exist, and may fare worse with large-scale displacement of work by smart tools. We therefore think additional solutions are needed. We also think these solutions need to be developed now, so they can be tried and refined before future problems become too severe. We have seen how self-expansion, replication, and tool use have long existed in biology and among our pre-human ancestors. We continue to use them today to grow our food and make all the artifacts of civilization. What seems obvious to us is using these same processes, and the tools and legal structures which cause the problems, to build a solution to those very problems:
- From biology we can draw on the ideas of growth, replication, storing and modifying instructions, evolution, and ecosystems with energy and material flows.
- From technology we incorporate already accumulated knowledge and its transfer, and the accumulation and sequential making of better tools with time. Instead of seeing smart tools and other productivity-improving technologies as a problem to be fought, we can use them as part of the solution.
- Concentrated ownership and unfair terms of trade, backed by the force of governments, are underlying causes of many economic problems. People lose their livelihoods due to change or business instability. Due to unequal resources they are unable to withstand these losses, and those who have more are unwilling to help. Alternate ownership structures like cooperatives and membership societies can level these inequalities.
By combining all these ideas, we can create a new approach, one which allows people to help themselves. This kind of approach can can still work even in the face of continuing large-scale change in the future.
3.0 - The Bootstrapping Approach
Our proposed solution to the problems noted above are self-expanding production networks. These systems are built by groups of owner-operators, who bootstrap from smaller and simpler starter sets of tools. They use these tools to make finished items for themselves, for each other, and to supply products and services to other people. They also use them to make better and larger tools to upgrade their capacity. This would eventually grow to include smart tools, and making new starter sets for new groups. Since the smart tools work for the owners, they are not at risk from job insecurity or work displacement to themselves. As owners, they still benefit from what their tools produce, no matter how advanced or automated they become. The expanded network would supply all of their basic needs, such as food, shelter, and utilities, and some luxury goods. The more of their needs they can meet within the network, the less they are affected by outside economic problems. Since the network makes things, we can call it a "Maker Network" or MakerNet.
Modern society is complex. No single person can acquire all the knowledge and skills to maintain a high standard of living. Buying all the necessary tools and equipment would also be too expensive. Working in groups within a network makes it possible and affordable to live well. The needed knowledge and tools are distributed across a large number of people. Making many of their own tools and products greatly reduces the costs involved. Through numbers, they can accomplish projects too complicated or physically large for individuals. Smart tools should be usable by average people despite their complexity. The average person can't build complex devices like desktop computers or smartphones, but they can certainly use them. Similar methods should work for smart tools. They would come with instructions for installation and operation. Where needed, training and support from specialists would be available, the same way they are for wireless devices and computers today.
3.1 - Network Operation
The network as a whole is self-directed and self-improving, rather than being directed from above. Members follow their interests and abilities. They teach, learn, and improve their skills when possible. They help each other to acquire new equipment and complete larger projects. They trade products and services within the network, and sell to other people to cover things the network can't provide.
Getting Started - Founding a network requires an initial supply of labor, knowledge, funding, tools, materials, and power sources. Depending on the resources of the founding members, they may need time to accumulate enough of these items. Once several self-expanding systems are in place, their experience, resources, and production capacity can be used to help set up additional systems. A new group forming a network may include the currently employed. They can participate as a spare-time activity that supplements conventional jobs. Reasons to begin include better use of their existing resources, hobby interests, and having backup work when conventional jobs are insufficient. The under- or unemployed have more time available, but fewer resources. They may participate out of need for basic goods, to work their way out of poverty, or for self-improvement. Finally there are the retired and the wealthy who would not work themselves, but can contribute funding, work space, and other resources. These people can combine their efforts to help each other and get more done.
Not everyone will start with the skills and experience to use the different tools and equipment. They can supply basic labor or other resources to start participating. They can later gain skills and experience by working with other members. The network can also supply formal training for those who want to learn new skills, or people can learn on their own from publicly available sources, then apply that knowledge to network projects. New people may start with no tools or equipment of their own. They can accumulate them by trading labor or making their own within the network, or building up an ownership share that lets them use larger collections.
Growth - At first, the network can only make a limited range of items. It doesn't meet all the members personal needs, or have all the equipment needed to make upgraded tools. So they make a surplus of the products and services they can make with the abilities and tools they have. These are sold or traded for the parts, materials, and equipment to upgrade. Surplus production can also be sold or traded to satisfy some of their personal needs. At this stage the network can partly solve the job uncertainty problem. Whenever other work is not available, they can employ themselves more or instead to compensate. As more people join the network, and their skills and equipment improve, they can do more for themselves and need less supplied from outside. Over time, members can build or acquire their own smart tools. This lets them make things very efficiently, with less of their labor needed. At this level they can replace conventional jobs entirely. Members can transition from working for others to working for themselves.
Continuing Operation - Working members of the network can later become disadvantaged by age or misfortune, and eventually pass away. They frequently will have non-working dependents. These people are unable to operate the tools and equipment themselves. They can still benefit from an ownership share in the equipment, while new people take up the necessary work, and acquire their own share of the network. Total productivity with advanced tools should be high enough that it can support current workers plus the beneficiaries of previous ones. With very advanced tools, the machines could mainly run themselves, with occasional maintenance in exchange for a share of the outputs.
Some people may have other goals than meeting their personal needs from within the network. For example, by working within a network, they could bootstrap an independent business without having to supply all the starting capital and equipment themselves, or going to outside financing. The network allows them to get started and be productive, accumulate these items and outside customers, then set out on their own when ready.
3.2 - Seed Factories
A starter set suited for this type of self-expansion is called a Seed Factory, in the sense it is the seed from which a larger factory grows. However, a collection of simple tools cannot grow by themselves. They require other resources such as raw materials, labor, and energy to operate. Knowledge is also needed - in the form of plans and instructions for what to build, and as skills and experience of the operators so they know how to carry out the work. All of the inputs of Tools, Resources, Energy, and Knowledge are needed to make a complete system, capable of growing and copying itself. To help remember these components, we call this the "TREK Principle", after the fictional replicators of the Star Trek universe.
What the seed grows to become is called a factory because it produces items in quantity by coordinated effort. With modern transport and communications, the tools and machines don't all have to be located in the same building or have a single owner, like traditional factories. The owner/operators don't even have to be in the same place as the equipment if they can control smart tools remotely. The physical arrangement can then be more like the modern Internet: concentrated activity in data centers, and local activity in people's homes and work places. A modern Distributed Production System can have multiple machines in one place when it is more efficient to do so. But it can also have local equipment in or close to homes and workplaces when that is preferred. What makes the system as a whole function like a factory is that all of the work is coordinated by people and their Information Technology. The result is the right amount of finished items are delivered when and where they are wanted.
Design - There is no fixed design for an ideal starter set. What you need to start with depends on a number of factors. These include the economic resources and skills of the first network members, the finished products they want to start making for themselves, local environment conditions, and what supplies of tools, materials and parts are locally available. As progress is made in smart tools and other technologies, the preferred composition of a starter set will also change. The shared features that make up a seed factory are that it enables self-expansion and upgrade, can grow to to meet the owners needs, resolves their economic problems, and can partly or fully replicate the starter set.
A minimal starting point may not have any ability to produce better tools by itself. For example, a pickup truck and utility trailer can't be used upgrade themselves. However, they can be used to supply moving services. The income from that can be used to buy other tools and materials, and the vehicle used to transport them to where a workshop will be built. Production can start after enough of the workshop is finished. Eventually members can fabricate improvements, such as an enclosed box for the trailer, completing the upgrade cycle. Providing services may be a better route to get started for those with limited resources, since it can be done with limited equipment. With a larger group of people or more funding, they can shorten this fairly long path to growth, and immediately set up a working space with more advanced tools. A third approach is a new network pooling their savings until they accumulate enough for a starter set.
Whatever starting point is chosen, it needs an effective growth path. One way to do this is to select flexible tools for the starter set. For example, a single solar furnace can be used to make bricks, cast metal parts, or dry lumber, all by using suitable attachments and accessories. Another way is modular design, so the expansions can be done in smaller steps. A production building can be added to one structural bay at a time. A generic electric vehicle chassis can have robot arms or farm implements added a piece at a time.
&emspp;Designing everything at once is likely too hard for a small group just starting out. The design effort can be shared the way open-source software is developed. New designs and plans would be shared across all the groups and networks building self-expanding type systems. In fact, for smart tools part of the design will be software, for which there is a lot of experience in open-source methods. We don't know in advance the resources and skills of a group getting started, or what they want to make first. A way to handle this uncertainty is to build up a library of growth paths, and individual item plans and instructions. They can then choose their own path as needed. Previous groups can also help new ones by offering packaged starter sets, training, and other help.
Cost - A new modern Machine Tool (Figure 2) may start at $45,000, and an industrial robot can start at $30,000. To use the most advanced production methods you generally need several machines in this price range, plus a medium-sized building to house them. This is too expensive for someone of modest means. We use several approaches to bring the cost within reach. The first is to pool the resources of a group. Farm and electric cooperatives, and credit unions have long demonstrated that ordinary people can do larger projects when working together on a part-time basis. The cost of more expensive or less frequently used tools can be divided in a similar way among network members. More expensive equipment can be individually owned, then loaned or access provided to others, or the products traded within the network. This way the whole group can benefit from their use. Some areas have community workshops called by various names like Hackerspaces or makerspaces. These already have collections of tools and equipment that can be used by members or the public. They can serve as starting points to grow from.
Second, smaller and simpler versions of most tools exist, some in kit form to save even more. Industrial-grade machines are designed to run continuously. Home and hobbyist-grade ones are more lightly built and have smaller motors, so they cost less. They can be adequate in the early growth stages. Third is the bootstrapping approach. The first set of tools are used to supply products and services. The income is then used to buy additional/larger/better ones. The current set of tools can also be used to directly make new tools, in a planned series of expansions and upgrades. To the extent tools can be made, rather than bought, it lowers their initial cost by substituting the group's own work.
Feasibility - We know starter sets in general are possible. Civilization as a whole grew from smaller and simpler sets of tools. Settlers in new areas historically brought along a set of tools to start with. They also typically imported more tools for a while, until they could make their own. Our approach repeats this pattern. Modern homesteading and do-it-yourself projects show people can partially make what they want using their own tools. Factories that make robots and machine tools already use their own products to make more of the same types. This shows replication of complex tools is possible. What we don't know are the optimal starter sets and growth paths to reach particular goals, given finite resources, modern technology, available designs, and what is likely to be possible in the near future. More work is needed to identify the best paths, prove them by building working examples, and build up experience with this approach.
Although we don't yet know everything about self-expanding systems, we do have some beliefs. One is that self-expanding production networks cannot function in isolation unless they are very advanced. They will have some level of interaction and trade with the rest of society. We also don't think bootstrapping can effectively be done by one person, because of the range of skills and amount of time required. But these are beliefs, and not yet proven.
Necessary Inputs - We have noted that an initial supply of tools, resources, energy, and knowledge are needed, before people can effectively provide products and services. In developed countries such as the US, a basic supply of these items is widely available at low cost. Many households already have some basic tools, especially if they are home-owners. If they follow a hobby they typically have better or more specialized ones. People who work in mechanical and construction trades tend to have better tools for their jobs, and people who work in manufacturing have access to even better ones. Used tools can be obtained from online markets and exchanges such as eBay or Craigslist. They can also be found locally from thrift and pawn shops, flea markets, and yard sales. New tools are more expensive, but widely available locally and online. Getting enough tools to get started should be feasible even for low income people.
Raw materials and energy are also widely available at low cost, even though many people don't realize it. A typical acre of land (multiply by 2.5 for hectares) contains 100,000 tons of soil and rock in the top ten meters, worth about a million dollars at bulk quarry prices. It receives about 875 MWh/year of net usable solar energy, worth more than $40,000/year at wholesale rates. Outside large cities, US cropland is available for $1,000-13,000/acre ( USDA, 2017 ), and undeveloped land for even less. These are far less than their resource value. What is needed is an effective way to use these resources, which self-expanding networks can provide. In developed areas, materials and energy can of course be purchased. But surprising amounts of materials are available second-hand, as scrap, or as discards. Self-built solar power can be very low cost, and free to use. Finally, essentially all the world's knowledge is now available online, in addition to traditional sources like books in libraries. These can usually be accessed for free. Many people already have skills and experience they can share in a network with people who don't.
3.3 - Advantages
The bootstrapping approach has a number of advantages over other solutions. It leverages a wider range of inputs than money-based income transfer programs. People who are able, but not fully employed, have a reserve stock of their own labor. This approach can put their available time to good use. Highly automated production can turn widely available resources into useful items at low cost. In purely economic terms, this approach may have higher rates of return than conventional investments. At the same time it can provide more economic security, a higher quality of life, and fewer environmental impacts than current methods. It does not change the fundamentals of trade and private ownership, so it can be implemented within the current economic and legal framework. If most of people's basic needs can be met with highly automated production, it also frees up their time to pursue their interests, rather than a job they may not like.
We think this approach is better than alternatives like Basic Income transfer programs. Those can't scale to a future where most jobs are replaced by smart tools, because there won't be enough funding sources left. Government-run transfer programs are subject to political meddling, and therefore uncertainty. There will always be some who resent being taxed to give "free money" to others. Whoever is taxed to fund such programs will likely object and seek ways to avoid them. Self-reliance is likely better in terms of general acceptance, and people's own motivations and psychology. A sector of society will always need outside help, so the need for support programs will not go away. But we should consider bootstrapping as an approach to provide that help rather than traditional cash transfers. For example, public or charitable programs could loan out starter sets or individual tools until a new group or network is self-supporting.
3.4 - Some Example Systems
We don't yet know everything about this approach. However, we know enough to present some examples of starter sets and how they can grow. Specific examples can also clarify some of the ideas presented above. The first example begins with a single product or service category, then expands to other categories over time. This is more suited to smaller groups with fewer skills and resources. The second one begins with a more complete starter set of equipment. It is more suited to a larger group with more skills and resources.
Single Product Starter
This route can begin with a single person or a small group. They start with a single category of product or service. Examples are sewing, carpentry/woodworking, or light household moving for hire. These don't require a lot of tools, and many people already have a sewing machine, some hand and portable power tools, or a pickup truck to get started with. Sales outside the network are used to buy better tools and materials in their chosen category, then later used to expand to other categories as more people participate. Members also trade within the network and help each other with projects. People are unlikely to fully support themselves when starting this way. They will need conventional jobs or other outside support until the network has grown enough.
We can list starter categories, and the later expansions, according to the type of materials or services provided. Within each category they would start with projects that need the fewest tools and skills, and are nearest to finished products. As tools accumulate and skills improve, they can start working back towards lower cost materials and larger scale of operation. Using woodworking as an example, buying finished furniture from a store doesn't require any skills or tools, but has the highest cost. A shelf unit kit may require assembly, but all the parts are supplied. You only need simple tools to complete it. Making a set of shelves from bought lumber requires more tools and skills, and shaping and assembling more complex furniture would need even more. Finally, cutting trees, milling the resulting logs into lumber, and drying the wood requires the most tools and skills, and presumes a high volume of products. But sourcing wood directly from trees has the lowest cost of materials.
New people joining the network can follow their interests. They can start in other categories and follow similar upgrade paths. As more categories are added, members can start to support each other. For example, people doing concrete and metalworking can help expand a woodworking shop, and in return the woodworkers can supply furniture and cabinetry to the others. If a network can grow to cover all the main categories, it can supply a major fraction of people's needs, and further self-expand in an effective way. Our example list of categories, with some starting points and upgrade steps includes:
- Basic Tools - These include basic hand tools, like screwdrivers, hammers, and wrenches, and portable power tools like an electric drill, circular saw, or reciprocating saw. Besides household maintenance, basic tools are used to maintain other classes of equipment, so this category is listed first.
- Light Crafts - This includes small-scale crafts like jewelry, candle & soap making, and fine arts, which don't require large equipment or work space. They are distinguished from basic tools by needing specialized ones for a given craft.
- Carpentry & Woodworking - These use some of the same tools as the first category, but differ in the size and complexity of the projects. Additional tools like ladders, ropes, sawhorses, chisels, routers, table saws, and woodworking benches can be bought, found used, or self-made, and added one item at a time.
- Stone & Concrete - This can start with hand-laying of found stones, which needs very few tools, and using bagged pre-mix concrete that needs water and a few tools to use. Larger projects can add tools for shaping stones, portable mixers, making forms and casting blocks, and shaping and placing concrete reinforcement.
- Metalworking - Basic hand and portable tools can shape and drill thinner metal pieces. Heavier items require a furnace to heat and soften the metal, then hammers, stakes, anvils, or presses to shape. Casting requires higher temperatures, crucibles, and molds. Much of these can be self-made. More advanced machining needs special tools, that can be added one at a time.
- Glass & Ceramics - Cutting glass can be done with simple tools. Shaping glass and firing ceramics requires high temperatures, starting with devices like gas torches and firepits, moving up to more advanced furnaces.
- Polymers - Small plastic items can be molded with light craft tools, or shaped with woodworking equipment. Complex items need a 3D printer or laser cutter. These are expensive/complex for starting out, but commercial services or community makerspaces can be used to make individual items.
- Electrical & Electronics - Electrical may start with simple repair, like replacing motors and cords, and work up to larger wiring projects. Electronics projects can begin with connecting pre-made items, like assembling an entertainment system or desktop computer, then moving to more complex ones involving custom circuit boards and programming.
- Coatings & Printing - Household painting and finishing furniture can start with just a few tools and supplies. More complex screen, spray, and paper printing methods can be added later.
- Fiber & Fabric - Hand sewing can be done with almost no investment in tools. Consumer sewing machines are widely available at low cost, and upgrades to industrial machines, cutting & sewing tables, and other equipment can be incremental. Working with other materials, such as leather or fiber-reinforced plastic, can also be added later.
- Assembly & Construction - Mechanical assembly and welding can be started on a small scale, as can installation of items made by others, such as cabinets and appliances. Larger construction projects would go beyond carpentry and concrete to include earth-moving, utilities, and crane work. These require larger equipment, so they are normally supplied by specialist companies, but smaller machines are available to start from.
- Agriculture & Forestry - This can start with small-scale home gardening and animal-raising. Community gardens and farms require more land and tools, and people's time to get started, but can be expanded in steps. In suitable climates, growing trees may not require any work, just allowing them to grow. Harvesting can start with smaller bushes, saplings, and branches, and don't require a lot of tools, then gradually moving to larger size trees and quantities.
- Materials Processing - Most people are familiar with cooking food, which follows steps like mixing and heating, and they already have the equipment for it. Other materials can also be processed by following a series of steps. These convert a source material into a desired finished material by mechanical, thermal, chemical, or electrical means. A very simple example is air-drying of wood from the natural state, which is too wet, until it is dry enough for construction or furniture. This category can start with such simple steps, adding new ones individually. The tools and equipment for each step are acquired or made as needed.
Each of these categories will need other resources besides tools. These include:
- Knowledge & Plans - Most people start with a level of general knowledge gained from early education. Books, online information, classes, and working with more experienced people can add specialized knowledge as needed. For anything beyond the simplest projects, plans and instructions for how to accomplish them are very helpful. These can be found for individual items, but custom projects and putting them in the right order is needed when a group is attempting larger and more complex tasks.
- Work & Storage Space - This can start with a spare room, garage, or basement. Later additions can include out-buildings added to current homes, leased or purchased commercial buildings, or undeveloped land which is then improved. Storage equipment can start with small tool boxes, drawers, and shelf units, and upgraded over time.
- Power & Lighting - These can start with conventional wall outlets and natural and artificial lighting. Portable power can be supplied from devices plugged into vehicles, portable generators, batteries, and dedicated engines. Solar and wind can provide more power for dedicated locations.
- Transportation - Most people have access to some sort of transportation, and can share the cost of larger transport within a network. Boxes and containers can often be found for free, or self-made for specific needs.
- Parts & Materials Supply - These can be purchased at retail to start with, and often found used or for free. As activity and tools increase, the network can obtain items earlier in the supply chain, eventually sourcing from raw materials.
Full Starter Set
A larger group, or one with more resources, can begin with a more complete starter set. They are able to immediately start making a significant percentage of new attachments and accessories for the original set, and making or buying complete new tools to upgrade their capacity. A full starter set is planned as a unit, but may not be acquired all at once. Our example includes eight main item types, with suitable accessories and attachments. The starter set would include one of each type, and add more units and different types with time. This list is intended for general-purpose industry, but it not unique. Other starter sets can be designed to meet different needs.
- (1) Building and Support Equipment - Most tools and machines need some protection from the weather, especially the electronics of smart tools. People also benefit from sheltered work and design areas with good lighting. Storage areas are generally needed in addition to the workshop spaces. To accommodate self-expansion, the building can be a modular design. A simple example is to use shipping containers (Figure 3) as a basic module. They are weather-proof, and can be used directly as work-spaces. They can later be arranged and stacked to form the walls of a larger building which has a roof spanning between them. Another approach is a more conventional steel-clad industrial building, where one end is a temporary wall. Additional structural bays can be added one at a time to lengthen the building, moving the temporary wall as each bay is completed. Support for the main tools and machines would include secondary ones for grinding, sharpening, and measuring, and portable and hand tools for maintenance.
- (2) Modular Robots - This includes both stationary and mobile robots, driven by electric or hydraulic power. The robots are modular in design, so their configurations can change according to different needs. The stationary version uses jointed arms (Figure 4) and end tools, mounted on a fixed base or rails. The mobile version starts with a chassis that functions like a farm tractor, providing basic power and motion. A variety of attachments can be mounted on the chassis or pulled behind it, such as wagons, lifting arms and buckets, jointed robot arms, and specialty end tools and implements. Mobile versions can be used internally within a building, or outdoors for construction or farming work. One attachment would be a "manual control module", to allow people to control the machine directly. This is useful for unique tasks when programming the robot isn't worth doing.
- (3) Solar Furnace - Powered tools and machines require energy to function, and many processes require heat. A solar furnace can supply both. A collection of mirrors are mounted on an axis to follow the Sun (Figure 5) and focus the energy on a stationary point. Replaceable targets can be mounted at the focus for tasks like making bricks, melting metals, producing steam to generate electricity, or lower intensity heat for other processes. Concentrator-type solar cells at the focus can produce electricity. The cells are not likely to be made internally, but they only represent a small part of the furnace, so that may be acceptable. Energy can be stored by directing the heat to an insulated rock bed, then extracted later as needed. Multiple copies of the furnace make it modular, and can then be assigned different tasks. Later versions can be built larger than the first one.
- A solar furnace can be mainly built from metal and glass parts. It can be designed to melt scrap of both types to cast new parts. So it can be largely self-reproducing from low cost materials. Large amounts of energy are needed for many industrial processes, so this may be a key path for self-expanding systems.
- (4) Bridge Mill - This is one of two basic Machine Tools which cut or shape metals and other rigid materials. The Horizontal Bridge moves vertically on fixed posts, and the tool mount moves horizontally along the bridge. A sliding table moves underneath to provide the third axis of motion. In our version, there are four tool mounts, two on the front and two on the back sides of the bridge. This allows mounting up to four robot arms or various other tools at the same time. The bridge extends past the posts and table to tool-changers at the sides, so they can be changed automatically. The rails that support the sliding table can be made as long as needed. New tables can be brought in with new parts to work on, or multiple tables can support very long work-pieces, sliding them under the bridge to be worked on. The tool mounts and tables can provide additional rotation or angled motion axes for more complex parts.
- (5) Horizontal Lathe - The bridge mill is better suited to rectangular or irregular parts. A lathe is better suited to round or symmetric parts, so we include it as the second basic machine tool. The simplest version of a lathe has a rotating spindle and drive motor (Figure 6) with a chuck to hold one end of a work-piece. A tailstock mounted on rails supports the other end of the piece. Cutting tools are applied as the piece rotates to generate round parts. Halting the rotation at known angles and moving a cutter along a plane can generate faceted sides. In our version, the lathe would have four rails, such as the Weiler V-Series. The second pair of rails allows multiple cutting heads carrying different tools to move independently into place as needed. Tool changers are mounted above the cutting heads to swap out different tools or replace worn ones.
- (6) Hydraulic Press - A hydraulic press can perform tasks like pressing shapes using dies and molds, shearing, rolling, bending, and shaping. One or more hydraulic pistons force the upper plate down, while several posts keep it parallel to the lower plate (Figure 7). The upper and lower plates are slotted to accept different inserts, and side supports can feed long objects for shearing, rolling, or bending. Thin or soft materials can be pressed at room temperature, while thick metal can be pressed hot, a process called Forging. In some cases the inserts would be heated, but they must use a higher melting material than the piece being worked.
- (7) Process Plant - A variety of materials are typically needed in complex machines. Such materials are typically not found in the needed form in nature. They can be bought from others, but that increases the costs and makes a system less able to self-expand. A large number of Production Processes are available to convert raw materials to the desired form. According to what is needed for later production, individual process steps, known as Unit Operations, are used to form complete process flows (Figure 8}. All the process flows taken together, some of which may share steps, make up the process plant. An example flow for wood would include cutting the trees, bringing the logs to the sawmill, cutting the logs into pieces, and stacking them to allow drying. Each unit operation needs its own equipment to carry it out, and often energy and other inputs. For general industry, the process plant might take in scrap metal, use the solar furnace to melt it, cast basic shapes, then trim and grind as needed to feed the Mill, Lathe, and Press.
- (8) Electrical Shop - Machines and tools that use power, especially smart tools, need a variety of electrical or electronic parts. This includes wires, insulators, motors, generators, batteries, switches, relays, transformers, resistors, capacitors, inductors, filaments, circuit boards, and microelectronics. The Electrical Shop produces and assembles as many of these items as is feasible. It accepts some parts and materials made elsewhere in the factory. Some items too complicated or difficult to make internally, such as modern computer chips. These items are purchased, but they make up a relatively small part of the factory, and used equipment or items people already own may be sufficient.
4.0 - Building the Future
The examples in section 3 are aimed at solving existing economic problems of individuals, which may be made worse by increasingly smart tools. Since they are intended to meet personal needs we call it Personal Production. But this isn't the only way such systems can be used. Self-expanding networks that bootstrap from starter sets can be applied at all levels of complexity and scale, just like living things range from microscopic bacteria to entire ecosystems. Unlike biology, a network that begins with people and their knowledge can do more than just copy their tools. People are the original "smart tools", and can change and adapt. Once a network is put on a growth path, it can add new items not in the original plans, upgrade to higher levels of complexity, and expand to larger scales of operation. It can also change course and make an entirely different set of products and services, if that is what the owners want. Finally, existing networks can also produce new starter sets with different purposes than the original, and spawn whole new growth patterns.
In this section we will look at some of the other ways these kinds of networks can be used. As worthy and interesting as these other projects may be, we don't think we are ready to jump in and start building more advanced versions of such networks yet. We think more basic research and development is needed, and gaining experience with smaller and simpler examples first. We finish the section with some final thoughts of a more philosophical than engineering nature, and some ideas that need more work.
4.1 - Beyond Economic Problems
Increasing Scales - A network of people working together can keep upgrading and expanding their equipment, and adding more people. This can take them beyond personal goods and services, into the small business and commercial scales of operation. This goes beyond meeting their own needs, to meeting the needs of people and businesses outside their network. The larger network will tend to include people in multiple locations, so we call it Distributed Operations. The final growth step is to industrial scale, reaching more widespread markets and at the most efficient levels of operation. As activity grows to larger scales, it will tend to become more specialized. This is because it gets harder to find places large enough, or with all the right conditions, for multiple industrial-scale operations. Although physically separate, the various locations can still support and supply each other as part of a network. Just because part of a network has grown to larger scales, it does not mean the whole network has to. It can include a mix of activity ranging from small to large sizes.
Harder Environments - All of the previous examples assume moderate environments with some level of existing development. These are the kinds of places where most people already live and work. But these places only encompass about 13% of the Earth's surface, and on average only a small distance above and below the surface. Highly automated and low cost production could make it affordable to live and work in harder environments, and use more of the material and energy resources found there. Difficult environments include open waters on oceans and lakes, very cold or ice-covered areas, hot and dry desert regions, very wet and swampy areas, high altitudes, depths underwater and underground, and places which are currently too remote and undeveloped to use. Different starter sets would be needed for all these different conditions, but they could be produced by previous generation automated factories in more moderate locations.
The hardest environments are beyond the Earth in outer space. In addition to other difficulties encountered on Earth, you add problems like lack of gravity, air to breath, and high radiation levels. The difficulty, cost, and distance involved in space projects is in fact the reason bootstrapping from an automated seed factory was first considered. But since there isn't yet experience in such projects, we think it makes sense to start on Earth, where conditions are easier, and leave space for last. Like harder environments on Earth, the starter sets for space locations would be produced in previous factories in easier places, then shipped to where they need to be. Once delivered, there are vast amounts of local energy and raw materials available to supply their growth. As with places on Earth, some items are too hard to make locally, or use rare ingredients. These would continue to be supplied from elsewhere.
A Better Future - Future civilization can develop along utopian or dystopian lines. We believe where we end up is a choice, not an inevitability. If we use them wisely, smart tools and self-expanding systems can let us build the kind of future we want to live in. First, by supplying basic needs like food, shelter, and utilities, people would no longer be subject to the kinds of economic problems and uncertainties we listed above. Highly efficient production would free up people's time, and allow them to do what is interesting, rather than what is necessary to pay the bills. Low cost production can enable high levels of renewable energy and recycling. This would reduce or reverse environmental problems. People would have more choices of where and how to live. The development of space would greatly increase available energy and materials. It could offload intensive industries from our planet, further improving the environment. I personally see this as a much more optimistic future than our present world.
4.2 - Some Final Thoughts
Self-Reliance and Personal Contact - Modern humans are a migratory species. We originated in Africa and have spread to all parts of the world with reasonable climates, and even some that are unreasonable. During our migrations we necessarily had to use local resources, and carry or make starter sets of tools to make a life in these new places. We relied on ourselves and the network of people around us to do these things. With the advent of industrialization, large cities, and large organizations, we have gotten out of the habit of relying on ourselves and people we know. We have become dependent on distant farmers and factories to supply our daily goods. Shareholders we have never met own the companies where we work to afford those goods. This puts us at the mercy of forces beyond our control. We can be told "sorry, we don't make that product any more", or "you've been replaced by a robot", with no recourse from those decisions. It is easier to treat workers as disposable, and make things they don't really need or want, if you don't know them. We think a return to some level of working for ourselves and dealing with people we know would be an improvement.
Artificial Life - The combination of smart tools and self-expanding systems can be considered a form of artificial life. In principle, a collection of sufficiently smart tools can grow unassisted by adding to their number, and reproduce by copying the original collection. Self-growth and reproduction are properties usually associated with life. If the smart tools require people to do some of the tasks, we can consider it assisted growth and reproduction. Alternately we can consider the people plus their tools as a higher level ecosystem which can grow and copy itself. The individual tools and machines are then analogous to the organelles of Eukaryote cells. Uncontrolled growth and reproduction can cause problems similar to biological plagues and cancers. Some thought should go into avoiding such problems, and how to deal with them.