Cultivating Distributed Communities: Case Studies and Guidelines from Key Leaders
Participants: Adrian Bowyer, Lenore Edman, Andre Knoerig, David Scheltma. Written and coordinated by Seth Hunter.
After initially building a prototype of a self-replicating 3-D printer, Adrian open-sourced the RepRap* project without the expectation of thousands of people being interested in contributing. They created a core team of contributors, which grew from a small group and expanded to channels in the public sphere.
The whole project works in a mostly hands-off manner, and the community typically is self-sustaining because the community is large enough that if a thread is interesting to the community, it will be addressed by its most active members. Adrian points out that Makerbot and many other 3-D printing companies have been based on the RepRap technologies. However, he believes that the fate of these companies will depend on how open they remain; he labels a closed-source printer “sterile” if it can’t replicate itself.
The Eggbot* community is a much smaller community than the RepRap* community so Lenore typically responds to most forum posts directly from their community. One thing that came out of the community forums was a demand for a new model to support Ostrich* eggs.
The software for the project is part of an open source ecosystem of tools based on Inkscape*. The community is small enough that they communicate across multiple media forms: Thingiverse, the Eggbot forums, GitHub*, Facebook* and e-mail. In more specialized small communities it takes time and dedication to foster their growth, but unexpected opportunities extend from those engagements and from leveraging emerging platforms like Thingiverse.
The ubiquity of Fritzing in many small hardware projects as a documentation tool is in part due to a watermark that they added to the layout tool. This small change is effectively a trademark for a tool they were giving away for free.
Later in the conversation, Andre explains that their primary focus is to try to lower the entry barrier for beginners. He says that what makes a product successful is not merely providing documentation; it is providing an entry point to allow beginners to get started. Their motivation comes from examples that provide a visual means of building a circuit and translating this into a schematic and board layout. Fritzing identified this early in their design by studying how people design their circuits, translating this into a file format to share with others.
What would prevent customers from purchasing an Arduino* or a Sparkfun* product from a copycat product that sells a derivative of your product for less money? Adrian and Andre explain that your trademark is not just an indication of quality but an investment of time into that tool and a vote for the community that creates the product.
A really good tool is not useful unless there is a functioning community behind it. Investing in a community is the same as building trust. By showing support for your initial base of users and engaging with them, you build a community that can start assisting itself.
Make: has a preference for publishing projects with Open Source Hardware because they allow others to hack them, enabling the community to design and build their own electronics ideas. Last year Make: had 131 Maker faires—they are exploding. The reason this movement is taking off is that people feel they are a part of the community.
The community is not oriented toward products but but with its open information is a vehicle for content. Make: gives a platform for contributors to talk about what they are doing and win the trust of their smaller community through recognition of the larger community of subscribers. David says that Make: is really about a an ethos and an approach to sharing knowledge.
One of the challenges with the Intel® Edison and Intel®Galileo projects is that they use a more customizable embedded Linux* build system called the Yocto Project*. It’s extremely powerful for industrial product developers, but difficult for the maker community to adapt to. In this discussion David, Adrian, Stewart Christy (Developer Relations) and Ajay Mungara (Internet of Things Development Toolkit Manager) weigh the pros and cons of the different versions of Linux.
If key individuals (such as Alex and Sergey in the case of the Intel Galileo) with experience using these embedded tools help bridge the gap for makers, it may only take a few dedicated users to help the community leapfrog from prototypes to products. Ajay points out that one issue with international products is FCC certification and security for products that use WIFI. The question is whether we can separate the complexity into layers, so that it is possible for the user to create a product without ever having to understand the OS beyond installing software.
One issue that keeps coming up with electrical engineers is that it is difficult to share board designs between programs (Altium*, ORCAD*, KiCAD*, Eagle*, etc.). At the moment, sharing designs is often constrained by the design tool used. The majority of people in the OSH communities use Eagle files because this is an established norm, and at Intel we use commericial layout programs with a high license fee and a proprietary format.
One option is EDIF, which might be the best attempt to support a neutral netlist format. Sadly, it’s poorly supported by most layout programs. There have been efforts in the past. The more we can share these files across these platforms the easier it will be to innovate between components from different ecosystems.
Adrian Bowyer has a background in mechanical engineering and tribology. He was a senior lecturer in the Bath Department of mechanical engineering and retired in 2012. His main areas of research are geometric modeling and geometric computing in general, the application of computers to manufacturing, and the engineering use of biology, called biomimetics. In biomimetics he works on self-copying and self-assembly in engineering. He is the originator of the worldwide RepRap Project – a project that has created humanity’s first general purpose self-replicating manufacturing machine. He is also a founder and director of RepRapPro Ltd, a company formed to sell RepRap kits and RepRap-related products and services.
Lenore M. Edman is co-founder of Evil Mad Scientist Laboratories, a hobby electronics and robotics business. Lenore’s educational background is in the liberal arts, and she has worked at universities, a library, and a biotech company. Lenore’s interests are at the intersections of papercraft, electronics, cooking, and science.
André Knoerig is an interaction designer and technologist who is passionate about innovation. After earning degrees in computer science and design, he managed and built up the design consultancy IXDS, co-founded and led the open-source project Fritzing, and played a key role in a range of research projects investigating new ways of human-computer interactions.
David Scheltema is the technical editor for Make:. His background is in embedded Linux* systems and engineering. He draws from his previous life as a tennis pro instructor, empathizing with learners to empower individuals in his role as the embedded expert at Make:. He is currently co-host of a podcast called baremetal.fm in which they discuss emerging technologies in open source hardware and the global maker community.