Annual and Interim Reports
2014 1st Third Report
- The Seed Factory Project
- 1309 Stroud Ave, Gadsden, AL 35903
LETTER TO PROJECT MEMBERS
This year marks the transition from thinking about and planning the project to developing hardware. We have begun a search for an R&D location near Atlanta. Our current goal is to find a location and move to it in the middle part of the year. We are building an equipment list for the new location by project phases, starting with an inventory of current tools and materials. In the first third of the year we started to get technical help from new members, and hopefully that will continue to increase. We continue to make progress on funding, and work on technical design in parallel to the physical transition. This report covers the first four months (January through April) of 2014. More details follow below. As always, I encourage you to supply feedback, as we are an open project.
Project Conceptual Designer
F.1.0 - Coordinate R&D
Coordination includes external flows, tasking, planning and scheduling, and analyses for the research and development phase of the project.
- Accumulate Funds - During the period, Eder increased personal savings from $9,126 to $14,815 towards the purchase of an office and R&D workshop. This is from a combination of dividend income and increase in the underlying mutual fund values. Bitcoin digital currency acquired by Eder, and contributed by project members are allocated to buying workshop tools, materials, and other shop expenses. Our balance increased from 32.9704 to 33.7367 BTC, while the exchange rate declined significantly, resulting in a value of $14,961 at the end of the quarter. We plan to continue accumulating both types of funds.
- Spreadsheet Data - This includes schedules and project task breakdowns, less empty spreadsheet overhead (8.15 KB). This increased from 159 KB to 236 KB. The current draft of our Equipment List is available for review on Google Docs. This will evolve and be updated in the future.
- Communications - We use multiple communications methods to coordinate project work. We currently use email, Google Hangouts, and telephone for person-to-person. We use Wikibooks for text and image type documents, and have set up domain names, GitHub repositories, and Google Drive for other information types. We made a Reddit category for forum-type discussions and a Twitter account, but those are not active yet.
F.18.104.22.168 - Collect External Technical Information
- Digital Technical Library - In early 2014 we completed the remaining 31% of the conversion/cleanup/optimization of the library and placed it in a maintenance/update mode. We will continue to add to and update the contents as needed. The optimization and merger of fragmented documents reduced the size from 31.1 GB to 24.9 GB, and contents from 2,928 files in 548 folders to 2,374 files in 533 folders. The library has been uploaded to Microsoft OneDrive for backup and use by project members. Unfortunately, since it contains many copyrighted works, we cannot freely distribute the whole library, but we are happy to share what works are unrestriceted. An index of the contents is still pending, so that people can see what we are using as data sources and suggest improvements.e most of the documents. The technical library won't be our only data source. The reliability of online data is often suspect, though, and directly contacting experts can be time-consuming. Our library is intended as a first place to look up the data we need to do our work, but it will not be the only place.
F.22.214.171.124 - Write Conceptual Books and Articles
- Seed Factories Wikibook - Work on the Wikibook slowed in early 2014 in favor of physical work and developing our equipment list. We added about 4 equivalent pages to the contents. We plan to return to the book and other writings once our move to our new Office & R&D location is completed and we get more technical help. Although it is not written up yet, we have been developing the idea of a MakerNet. This is a distributed network of automated production nodes that grow by making items for each other, as well as useful products. Individual nodes or the network as a collective whole can function as seed factories. The importance of this idea is it can rely on already existing equipment that people and businesses have to get started, and only needs to add software and coordination. This makes getting started easier. Our R&D location would then start as a node in the MakerNet, and work with Maker community groups and the wider technical community. It would then grow towards a full factory capacity. We are excited about this idea and plan to develop it further and report on it in the Wikibook and elsewhere.
F.1.3 - Build R&D Locations
In early 2014 we started our search for a permanent R&D location in the Atlanta metropolitan region. We have arranged with a local real estate agent to begin a property search, and have started to get candidates. The ideal location would serve as a combined home/office/workshop for the project team, so we can use and test our own products and eliminate commuting time and cost. If we cannot find a single location that meets all our needs, then we will get several properties close together. Besides sufficient living space, we also need office and workshop space, or room to build them, and enough land area to build and test items like automated farming and solar energy. There are a number of other requirements for our new location, like good internet access, suitable geology to build on, lack of zoning restrictions for what we want to build, and balancing affordability with location.
As a way to help finance operations and integrate into the local community, we are considering leasing project space and access to our equipment to the Maker community and start-up businesses. Leased space and use of the machinery would supply funding for expansion and put the equipment to better use, and also give us a working relationship with more people when we need help.
In early 2014 I completed cleaning out and organizing existing tools and materials in preparation for putting them in temporary storage. Holding the existing rented house while moving to the new location, or paying a commercial mover for a rapid move are both expensive choices. So we plan to rent storage and move at our own pace as the least expensive option. Side panels for the utility trailer are almost complete as of the end of April 2014, and we can begin moving items once they are done. I also plan to put most of my personal items in storage, and find a temporary residence in the Atlanta area. Property search will go much faster without a 4 hour round trip from my current location in Gadsden, AL.
We expect finishing the property search, buying it, and moving everything over will take the middle part of 2014. As soon as we have our new location we want to start building the workshop, buying equipment, and testing seed machines.
2013 Annual Report
It is with pleasure that I bring you our first annual report, covering the period of March through December 2013. The idea of self-expanding automation goes back to the mid-20th century, and I personally started working on this idea in 2012. However as a distinct project with the intent to build working examples, the current Seed Factory Project can be dated to March 2013, when the Seed Factories Wikibook was started. For people who are new to the project, you might want to start with the Introduction companion page to this report.
We made a good start in 2013. The Seed Factories book started the year as a set of rough notes attached to another book I was writing. That book is about the design of future space systems, for which Seed Factories hold great promise as enablers and cost leverage. The apparent importance of the idea and early Earth uses justified having a book solely dedicated to it. Both books are intended to be engineering level discussions. The rough notes have grown to the equivalent of 230 book pages. There are also about 40 pages worth of related writings here in my User pages. The book is intended to cover the history of the idea, the design concepts and process, and a series of design examples. We currently have four such examples: Personal Factory, Industrial Factory, Distributed Production Network, and Remote Locations. The general design process is the same, but the particular details will be different for each.
We have started developing the examples and an overall research and development program in the form of design notes and system architecture spreadsheets, but they are still fairly preliminary. Building and testing actual hardware will require a place to do it, and funding. I started allocating my personal savings in that direction, and the members this letter is directed to contributed to our first fundraiser in the 2nd quarter of the year. We plan to pursue additional funding and grow our membership in 2014, with a goal of buying an R&D location and building an initial workshop.
We want our designs to be realistic. To that end, from mid-2013 I collected and organized a digital technical library. This is in addition to the paper technical books I have accumulated from my engineering work. The library is being converted to PDF format and optimized for size, so that eventually we can use it on portable devices in the field or workshop. On the physical front, I did extensive maintenance on my truck and utility trailer, and am about half done cleaning and sorting the raw materials (mostly lumber) and tools I have, in preparation for moving to our new location. 2014 is looking to be a transformational year for the project, and I look forward to reporting our progress and getting feedback from all of you. The summary data and notes which follow give more details, and previous reports are appended at the end.
Project Conceptual Designer
- 2013 Summary Data
Starting with this report, we are organizing project data according to our task breakdown structure. Each task or activity gets an identifying name and number for tracking purposes, with subtasks getting additional decimal points and numbers as needed. Thus F.1 is all of Research and Development, while F.1.1 is Conceptual Design under R&D.
F.1.0 - Coordinate R&D
Coordination includes external flows, tasking, planning and scheduling, and analyses for the research and development phase of the project.
- Accumulate Funds - Income from Eder's personal savings is being assigned towards purchase of an office and R&D workshop. This was zero at the star of the period, and at the end of 2013 stands at $9,126. Bitcoin digital currency mined and purchased by Eder, and contributed by project members directly and from a crowdsourced fundraiser are allocated to workshop tools, materials, and other expenses. The balance allocated to the project was zero at the start of the period and was 32.9704 BTC at the end of 2013, with a value of $24,040. The value per bitcoin has greatly increased since they were contributed earlier in the year. Therefore we are changing our contribution policy to valuing them at the time converted or spent, rather than time donated, as that is more fair to the contributors. The contributed value will determine future output assignment percentages from the equipment we buy and build.
- Spreadsheet Data - This includes schedules and project task breakdowns, less empty spreadsheet overhead (8.15 KB). We started tracking them in the 4th Quarter, and they amount to 159 KB or 54 equivalent text pages. They are not yet posted online.
- Online Media - Our main method to coordinate is via online media. We use email to communicate directly with members and interested people, and Google Hangouts when live video is helpful. We use Wikibooks for text and image type documents, and have set up domain names and GitHub repositories for other information types. We made a Reddit category for forum-type discussions and a Twitter account, but those were placeholders in 2013 as we didn't have enough activity to use them yet.
F.126.96.36.199 - Collect External Technical Information
- Digital Technical Library - In doing R&D you want to know the current state of technology so you don't repeat past work, and know what areas need improvement. For doing designs and estimates you also want accurate input data. Eder has a paper library from his work as an engineer, but it is not very searchable and parts are out of date. In 2013 we started developing a digital library for the project. The first step was gathering the documents and organizing them by library subject categories. This resulted in 49.8 GB of files, and was complete by early 3rd Quarter. The next steps include converting all the files to a consistent format (we are using Adobe PDF 9.0 since there is a free reader for it), merging multiple files into single documents and cleaning up bookmarks when needed, and applying compression and optimization to reduce their size. We want the end result to be usable for designers at their workstation, and portable enough to reference on tablets in the shop or field.
- By the end of 2013 we had completed 69% of the conversion/cleanup/optimization steps (we do all of them at once on each document). As of the end of 2013 the library was 31.1 GB in size (37.5% smaller), and consisted of 2,928 files in 548 folders. We expect to finish the remaining documents in the 1st Quarter of 2014. Once the initial version of the library is complete, we intend to publish a contents list so that members and other people can see what sources we are using and suggest improvements. Unfortunately due to copyright laws we cannot distribute most of the documents. However, some are government works or open source and can be distributed. We can use the library internally at our R&D location. We are obviously aware of the Internet and other data resources, such as contacting suppliers and technical experts. The reliability of online data is often suspect, though, and directly contacting people can be time-consuming. Our library will be the first place to look up the data we need to do our work, but it will not be the only place.
F.188.8.131.52 - Write Conceptual Books and Articles
- Seed Factories Wikibook - As of the end of the 1st Quarter the book was 135.6 equivalent pages, and at the end of the 4th Quarter it was 232.5, an increase of 96.9 pages (71.5%). We anticipate it will be about 500 pages in size when all the sections are complete, on that basis it is about 43% complete. We use 3,000 bytes or 500 words to calculate an equivalent book page, since wiki pages can be variable size and much longer. We can also export the book in PDF format, where it will actually be formatted as book pages. We have not done so yet because some of the sections are too incomplete.
- Other Writings - Currently these reside on local user pages, and include this report. We started tracking them in the 4th quarter, and they amount to about 40 pages.
F.1.3 - Build R&D Locations
The research and development work includes creating new component technologies and software, then building and testing physical prototypes. As a distributed open-source project, some of the work can be done in many locations by project members. A complete Seed Factory, though, will need a dedicated and relatively large location to collect renewable energy, grow organics, and to house the indoor production equipment. A dedicated location would include office, workshop, and test areas.
- R&D Office - In 2013 Eder started working from home on the project, as an extension of the Tirion Designs engineering and design business. Since I already had computers and software, nothing extra was needed during the year aside from a couple of hard drives for extra storage. My intent is to work full time on the Seed Factory Project, but since I'm currently in a rented house on a small lot, it can't support the large scale work. Therefore I started accumulating funds under task F.1.0 above towards buying a new location with at least 1 hectare (2-3 acres) of land, and enough living space for myself and office space for the Project. I'm currently about 2 hours away from Atlanta, which is too far away to make use of existing maker workshops and talent in that area, so I will be looking for someplace closer in early 2014. Existing equipment is sufficient for now. If we get more project members working regularly at the location, or need to house servers and control stations for the R&D work, we would need more desks, IT hardware, and software, but we have not identified specific needs at this point. As a distributed project, we want to be able to support live video and remote operations if possible, so we will pay particular attention to internet bandwidth.
- R&D Workshop - Ideally our dedicated location would already have a large workshop where we build components and prototypes. At a minimum we will require enough space to build such a workshop. Until such a workshop is built and populated with equipment, our plan for 2014 is to work with Atlanta area maker groups who already have community workshops. We will also explore other options, such as collaborating with non-profit or educational institutions or local industrial development agencies. A complete R&D facility is likely to exceed my personal funds, so we also expect to do more fundraising, and business collaboration for post-R&D projects.
- I already have an inventory of portable tools and raw materials from my past hobbyist and small scale construction activity (sheds and small workshops). That can serve as a starting point to build and outfit a dedicated workshop, but we will need a number of larger stationary machines. The $24,000 in bitcoins we have in hand, and future fundraising and collaborations will go towards buying them.
- R&D Transport - In addition to stationary buildings and equipment, the R&D work will require transportation to move equipment and supplies, and to move production outputs once we have them. In 2013 I restored to working condition the 2000 model truck and utility trailer which had been sitting unused. Due to their age the truck needed new tires and engine maintenance, and the trailer needed painting and replacement of the wooden deck, and some wiring repairs. They are now both functional. In early 2014 (weather permitting) I plan to build side panels for the trailer so smaller items can be kept from falling off. At some point in the future we expect to build or buy the equivalent of a farm tractor/construction backhoe for heavy duty work, but we don't have a schedule for that yet.
Third Quarter 2013
In the 3rd Quarter of 2013 we made progress in all areas of the project:
As an open-source collaboration, we intend to make the project data available online, and open to contributions by anyone with sufficient knowledge. Our data so far includes:
- Seed Factories Wikibook - This intended as a textbook-style introduction to the concept of Seed Factories, how they are designed, and several design examples. In this quarter the book has grown from 186 to 210 equivalent pages (13%). Most of the additions this quarter have been in section 5.0, which describes the Personal Factory example, and a start at updating the early sections.
- Project Websites - We obtained two domain names: www.seed-factory.org and seed-factory.info for public information, and have started to build their websites.
- Online Data - We set up an account on GitHub, currently with two repositories. The "data" repository is intended to host project code and other data types such as design files. Wikibooks is a good collaboration tool, but it mainly takes text and image files. Therefore other file types need a different storage location. The "seed-factory-project.github.io" repository holds the public information website code.
Developing a technology from a new idea to practical working examples has many steps, and we are still in the early stages of doing this. In addition to expanding on the ideas in the Wikibook, we also worked on the following:
- Technical Library - In developing new technology, we want our designs to be based on the current engineering knowledge, and not to repeat work others have already done. Our technical library will help in that goal, although not sufficient in itself. We will also need people skilled in various fields and other sources of information.
- In the 3rd quarter we completed cataloging the reference data already in hand, and began optimizing the pdf files to make them easier to back up and use. The electronic library is now 39.1 GB in size, organized by the Library of Congress system. This is significantly smaller than the 49.8 GB reported in the last quarter because of elimination of duplications and optimizing. Currently 10.07 GB of the files have been optimized. We intend to make the distributable portion of the data available once we finish optimizing it, and provide a list of the copyrighted works so that others can find them. We intend to update the library over time, and suggestions for additions or better sources are welcome.
- Permanent Location - Our long-term goal is to build working examples of seed factories, to prove that our ideas work in the real world. Over the next year we will work towards a permanent location where we can start to do this. The location would include design space, a conventional workshop to make and assemble parts, and areas to test the prototype equipment. We plan to work with the Atlanta-area Maker community, such as Freeside Atlanta, who already have a high tech do-it-yourself workshop. Demonstrating factory processes like solar energy and food growth will require extensive land area. Thus our location will be near Atlanta, but sufficiently far from the central metro to where land is affordable. Since we are currently in Gadsden, Alabama, we have begun preparing to put personal items in storage before looking for the permanent location.
- Workshop Equipment - This quarter we started preparing our first major equipment items for the conventional workshop. These are a 2000 model Chevy truck and same aged 16 foot utility trailer. They will be used for general transportation of equipment and materials. Due to their age, both required significant maintenance. This quarter we purchased new tires for the truck, and began to repaint the frame and replace the deck on the trailer. In the 4th quarter we expect to finish the maintenance and have the vehicles ready for use. We have a fair amount of basic shop tools and raw materials (mostly lumber) which will be inventoried and later moved to the permanent location. We expect to acquire a large amount of additional tools and materials to outfit the workshop, so this is a first step.
Conceptual design only requires computers, which project members already have. As we move into physical hardware significant additional funding will be needed. We expect to meet those needs from a combination of member contributions and outside fundraising. Open source development does not preclude making money. Our plan is to operate this Seed Factory Project as a non-profit entity for doing the early research and technology. When we reach the point of building an operating factory, that would be set up as an independent business entity. For now, all activity is aimed at the non-profit open-source research.
The project's founder, Dani Eder, is a retired aerospace engineer. Payments from the Boeing Company and personal savings make about $1500/month available towards project expenses, which are the sole regular source of funds at present.
Between a previous fundraiser, and personal "mining" and purchases, our holdings of the bitcoin virtual currency are now 32.23 units, and further purchases are planned. At the current market value of $123/unit this has a value of $3,965. We have been moving these funds to an offline "cold wallet" to protect them from theft. We are holding bitcoins for several reasons. It was an efficient way to do fundraising, and we think their value will go up in the long term. In addition, one of our design examples, the "World Wide Factory", is a distributed peer-to-peer production system. We envision uses a virtual currency to settle payments among owners, so bitcoin gives us a working example to implement that.
We have started developing a project financial report. It is incomplete, but total project-dedicated assets are estimated at $13,000 as of this quarter. It consists mostly of bitcoin units and the two utility vehicles. We do not currently place an asset value on technical data developed within the project, as it is intended to be open source. This might change when we get competent accounting advice.
Besides Dani Eder, who started the project, this quarter we started getting help from a graphics/web designer, and got several expressions of interest from other technical contributors. We will need more project members to contribute to the technical work, and will continue to expand our website and publicity efforts towards that goal. The 11 people who contributed to the bitcoin fundraiser are considered project members, and will get a share of future factory outputs when we reach that point.
Second Quarter 2013
- Seed Factories Wikibook
Historically, civilization has grown from smaller and simpler production capacity, especially when settling new areas. The practice of using tools to make more tools is much older than that, even. The idea of machine replication began to be investigated in the mid-20th century, and NASA briefly considered placing an automated factory that could copy itself on the Moon in a 1980 study. Computers and communications were not up to the task at that time, and NASA never considered applying the idea to Earth, therefore they did not pursue it further. In the course of working on a space engineering textbook, author Eder realized that automation and robotics had advanced dramatically since the 1980 study, and the idea of self-expanding factories applies just as well on Earth as in space. What started as notes on the concept in the textbook evolved into a separate book and a project to implement the idea for Earth applications.
At the start of the 2nd quarter, the accumulated notes had just been moved to the Seed Factories Wikibook and the book outline was at an early stage. Wiki pages are variable size, and usually closer to chapter length than traditional book pages, thus we track progress in terms of paper book equivalents of 3000 characters or 500 words. The notes and text started the quarter at 135.6 pages and finished at 186.1 pages, an increase of 37.2%. The book is still far from complete, but we think we have incorporated the main ideas and methods for designing Seed Factories, and have made a start on one of the four example designs, the Personal Factory. The intent is for the book to become a reference manual for designers and builders. Since working Seed Factories don't exist yet, the book will include how to design them, and also multiple completed design examples.
We expect to continue work on the book for the remainder of 2013 and into next year. We are using Wikibooks because it is open source and makes it easy for other people to collaborate. As the example designs move from concept level to more detailed hardware and software, we expect to start using other open-source locations that are better suited for those types of data.
- Technology Development Plan
The technology needed for building a functional Seed Factory is not entirely new. Many of the industrial processes it will use already exist, and computers, automation, and robotics have reached a high state of development. Nevertheless we are putting them together in a new way, and there are no experience-based engineering handbooks to refer to yet for these kinds of projects. We expect to use a considerable amount of custom hardware and software, and need to test some new technologies and reduce uncertainties before anyone will commit to an operational factory project. We devised a Technology Development Plan and intend to follow it and update it as needed.
During the 2nd Quarter we continued to build a science and technology reference library. Mr. Eder's original personal library, accumulated over the past 30 years, is on paper. This is hard to search and parts are very out of date. We want to base our designs on the latest available existing technology, and not repeat work others have already done. Therefore we have been accumulating reference data. At the end of the second quarter this amounts to 38.8 GB of cataloged data, and 11.0 GB still to be cataloged. We plan to continue to organize this data over the next few quarters so that it will be ready to use by the time we start doing detailed design.
The plan includes continuing the conceptual work, documented in the Wikibook, and making progress towards preliminary design and component research. Eventually this leads to building prototypes of factory hardware and software and testing them to prove their performance. We hope that more people will join our project and collaborate in the development. The early stage work (conceptual and preliminary design) can be done by anyone with the technical skills and a computer. When we get to physical hardware, some of that might be built and tested by remote teams, especially in small scale. But at some point a dedicated location will be needed to test things like robotic farming and large solar furnaces. We will investigate collaborating with existing non-profits and research institutions such as universities, but we think it more likely the project will need its own location when it gets to building integrated factory prototypes.
Whatever physical location is used for testing, funding will be needed for equipment, parts, and materials. In the second quarter we ran a fundraiser on the Bitcoinstarter crowdfunding website. That fundraiser reached its goal, and a private individual made a matching contribution direct to us after finding the fundraiser, so effectively we reached twice the goal. In addition to the bitcoins held by Eder already, this amounts to around $3000 in virtual currency. We plan to hold these funds until they are needed for project purchases. We plan to do more fundraising in the future, and project founder Dani Eder plans to apply his personal savings and income above living expenses to the project.
Although the technology developed in this project will be open-sourced, physical hardware built using it will belong to project members who contribute to its development and construction. Any products such equipment produces will also be proportionally shared among project members.
- Dani Eder - Is the project founder and currently sole regular technical contributor. He has over 30 years experience in aerospace engineering, most of it with the Boeing Company. As of 01 August 2013 he will reach retirement age and be able to devote full time to this project.
- Financial Contributors - A total of 11 people have contributed funding as a result of the Bitcoinstarter fundraiser. These people will own a share of the eventual hardware and factory, and the products and sales they produce.
- Seed Factories Wikibook - The main source of information about the concept and designs.
- Fast Company Article - by Dani Eder about the project.
- Project Notes - These are unconsolidated notes with additional information related to this status report.