The Seed Factory Project
An Introduction to the Seed Factory Project
1309 Stroud Ave, Gadsden AL, 35903
2 January 2014
Vision and Goals
Anyone who pays much attention to advances in technology knows that automation and robotics are increasingly replacing human labor. An example is the Google Self-Driving Car project. If implemented, self-driving cars could replace large numbers of truck and taxi drivers. Replacing human labor with technology is not inherently bad, it has enabled the vast majority of people in developed countries to escape being peasant farmers. However, the industrial revolution has played out over centuries, and there are still parts of the world at the peasant farmer level. Newer technologies, like mobile phone subscriptions, went from near zero to equal to the world's population in just 25 years. What would be bad is if automation and robotics replaced people's jobs faster than they could find other things to do or upgrade their skills. To put it in quotable form:
If robots take all our jobs, who will buy the stuff they make?
The key word in that quote is "jobs". Modern economies are based on a large part of the population working for someone else (often a large corporation) for pay. If an employer can save money replacing a human worker with automation, they will tend to do that, leaving the worker without an income. As a society, though, we would like to get the benefits of improved productivity in the form of a higher quality of life. Some people have advocated a "basic income", where you distribute enough for some minimal standard of living regardless if the person works or not. The assumption is that a highly automated society would have enough surplus to redistribute the outputs to everyone. We reject this approach for two reasons. First, it requires taking by force from some people to give to others, which is immoral. Second, it reduces incentives for everyone to be creative and work at something useful. For the unemployed who get their basic income, they have less incentive to do anything at all, and for the more productive population, part of the reward for hard work is taken away, giving them less incentive to work so hard.
Put automation in the hands of the people.
There are two parts to our solution. The first is putting the automation in the hands of the people who will use the outputs. If you own the factories, automation, and robots you are not at risk of laying yourself off. You also have incentive to upgrade because you reap the full rewards of doing so. The second is using part of the production output for parts to build more machinery. The new equipment increases the production rate, so the whole set of machines can grow exponentially, allowing everyone to eventually have them.
Our vision for the future then is an end to material scarcity, where everyone has their physical needs met by automated production. This frees people from having to work to eat and keep a roof over their heads. They can choose to work if they enjoy it, or want extras, or they can choose to relax, like retired or rich people do today. The goal of the Seed Factory Project is to develop the necessary technology and demonstrate working examples of self-expanding automation. These would then be distributed as widely as possible, to enable our future vision.
History and Plans
Origin of the Project:
The Seed Factory Project (SFP) started in early 2013 as an outgrowth of my work on space systems. I worked for many years at Boeing in their Space Systems division, most notably on the International Space Station. I've always been interested in space projects beyond the current ones. After going to work for myself I set about writing a book on the design methods for such future projects. Big space projects run up against the high cost of shipping anything into space. One way to cut down the cost is to use local materials instead of shipping everything from Earth. A few experiments are underway to extract, for example, oxygen from Lunar rock, but all the equipment to do the extraction would still come from Earth. The biggest gain would come from making the equipment as well as the products from local materials. You still need to send some kind of starter kit, but it would be much less than sending everything.
Every big space project has a major part down here on Earth. That includes factories to build the space hardware and rockets, launch centers for the rockets to fly out of, and control centers to monitor things once in orbit. In 2012 I was working on the section of my book that covered all that Earth stuff. Something I knew well from my physics background crossed my mind: the laws of Nature are the same everywhere. If making your own equipment was a good idea in space, maybe it would also be a good idea on Earth. At first I was just thinking about the rocket factories, to make building the rockets cheaper. Pretty soon I had generalized the idea to everything else we make on Earth.
For this century at least, there are a lot more people who need things down here on this planet. Also, the future space projects will be easier to do when we have solved material scarcity and had experience with self-expanding production. So in early 2013 I put aside the space systems book I had been working on, and started a new book about how to design self-expanding production systems. The book is called Seed Factories, because they grow from a starter kit like a tree grows from a seedling, and once mature can produce more starter kits. As an engineer, I know that theory and paper designs can only take you so far. At some point you need to actually build and test them. I also know my own limitations - there is only so much that one person can do. So the Seed Factory Project got started as a way to gather people interested in these ideas, so that we could develop and build them together.
Current Progress and Plans:
We are barely a year old, so the project is still evolving. I refer to "current plans", because we expect them to evolve over the next couple of years. I still spend most of my time thinking and writing. I'm also gathering a technical library to help fill in wherever the project doesn't have an actual expert. The physical research, construction, and testing will need places to do it, so I started saving up my own funds and soliciting contributions. Taking a page from open-source software development, we want to start as a distributed effort, with skills and tools people already have. We would help each other upgrade towards more automated and self-producing equipment, eventually leading to complete starter kits and integrated factories. More details about our recent progress and plans can be found in our Reports page.
The Seed Factory Project plans to use several operational entities. The first is an open-source non-profit research organization. Technical developments and designs created or contributed by them will be distributed openly to whoever wants it. To that end, the Seed Factories book is hosted on Wikibooks, a sister site to Wikipedia, and we have set up repositories on GitHub to host other kinds of data than text. Second is a network of people and companies who want to build and operate specific designs and equipment for their own use or for profit. This "Distributed Production Network" would have open membership. People can join by listing themselves, along with their location, interests and capabilities. This allows people to find each other and trade items or work on an individual basis. At a higher level of activity, members can collaborate on larger projects. They can contribute to Network-level projects, like complete production nodes, or self-organize specific items or locations to work on together. The general principle is project contributors get shares of the resulting output in proportion to their contributions, whether it was cash, labor, materials, or tools. Once functioning, output shares can be assigned or sold like any other property.
Design and Technology
A linked set of machines that make parts for each other and also end products for users will be complex. Open-source development lets each person work on whatever part of a project they want, but building hardware has higher costs than software for making constant revisions. So we would like to coordinate the work to minimize the number of hardware versions that need to be built. Experience has shown the Systems Engineering method is effective for complex projects. We plan to adapt it to our project in the form of Open-Source Systems Engineering. This means applying the systems engineering methods in an open-source way.
The Systems Engineering (SE) method is interdisciplinary, and follows an organized path from initial goals and user desires to a final design that meets those goals. It includes breaking down a complex design into simpler parts, to the point that each one can be worked on by an individual or small design team. It also includes making mathematical models and simulations of the parts and how they interact with each other and the outside world. This allows understanding and optimizing the design before committing to physical hardware. The systems approach considers the whole life of the project, from initial concept to final disposal or recycling. It also considers all the factors that matter to people inside and outside the project, not just performance and cost. Finally, information about the state of the design is communicated to everyone who needs it, so that they all are working from the most up-to-date versions.
Traditionally the SE method has been used by large government and corporate projects with a top-down organization structure. To adapt it to an open-source approach we propose to use the well-developed tools from software development, such as code repositories, but expand it to include other data types like documents, simulations, and design drawings. These would be collected into design packages for different parts of of the system. To make sure multiple pieces will work together there will be "integration" packages that draw on the data and people from multiple design packages.
The first stage of a project life cycle is "Conceptual Design". In this stage we establish design goals and requirements, then proceed to a completed "System Concept". This includes defining the main elements that make up the project, their design features, how they will be operated and maintained, and cost and schedule estimates. At present we have two projects in the conceptual design stage: a Personal Factory designed to meet the needs of a local group of owner/operators, and a "Distributed Production Network", which is non-local and has multiple owners. These projects are not entirely distinct, a Personal Factory can be one location within the distributed network. They also would share automation hardware and software. We have two other projects in the idea stage: an "Industrial Factory" designed for large scale production, and "Remote Locations" such as a mining operation in a difficult climate. The point of having different projects is to see what areas self-expanding production can be applied to, and what technologies and designs need to be developed for them. Individual hardware elements, like a solar furnace, can also be developed as separate products. The conceptual design work is being documented in the Seed Factories wikibook, and other data types will be placed in online repositories when sufficiently developed.
Later stages of the projects will cover preliminary and detailed design, fabrication, assembly, test, installation, operation, and finally disposal.