Only SENSORICA logo is copyright. Content on this document is Creative Commons (BY NC SA)
Interfaces between open networks
and classical institutions
the SENSORICA experience
NOTE: This is a work in progress. As the story unfolds, with inputs from different SENSORICA affiliates, please come back from time to time to get the latest. You are invited to comment on this document and to ask questions. We may decide to address your questions in the text.
We are doing this work openly, collaboratively, the SENSORICA way. If you find mistakes, unpolished phrases, a rough style here and there, don’t be sad : ) It will get better with time and with everyone’s input.
One thing is certain, this document will be of great historical importance, because it is among the first to describe in detail and from real life experience economic interfaces between the emergent open world and the classical world.
This is a co-creative document
If you contribute to this doc make sure you respect the Content rules
Table of contents
The Barda case
The Queen’s University case
The IoT for heavy industry case
The Blockchain Access case
To be developed further...
IP, Patents, and Privacy vs. Openness
SENSORICA was officially launched in February 2011, as a for-benefit open collaborative network, focused on open source development of sensors and smart systems. Initial efforts went into the development of the Mosquito sensor, the ambition being to demonstrate that peer production can be applied to material things. Until 2015, the SENSORICA OVN was deploying most of its energy into developing its own products and services, while building infrastructure to sustain its operations.
Sensoricans chose the market of scientific instruments to experiment with peer production, because it is characterized by low volume and high profit margins, with a high degree of customization. To advance the open source scientific instrument projects, sensoricans created strategic partnerships with academic labs. Three ENGAGE grants were obtained, with Montreal Heart Institute, McGill and Ecole Polytechnique de Montreal. Other grants were obtained for commercialization efforts, through an Exchange firm created for this purpose, Tactus Scientific.
In February 2015, sensoricans took on the first service contract with a local architecture firm, to design a high tech observation tower, the Barda periscope project. In May 2015, SENSORICA affiliates arranged a deal with Queen’s University to design an open source scientific instrument for characterizing photovoltaic materials, the PV characterization project. The end of the year 2015 brought an even larger challenge, the development of an open source sensor network, an IoT application for the heavy industry, the Sensor Network project, with a well-established industry leader. At the beginning of 2016, SENSORICA partnered with eVision Inc. on the Blockchain Access project.
Within one year, Sensoricans firmly established that open networks can service classical institutions, in a way that provides great economic benefits on both sides of the deal, as well as for society at large. In other words, SENSORICA demonstrated that open innovation and peer production networks have a place in the present market, while developing the enterprise of tomorrow.
Let’s make sure that we don’t go over this too lightly. SENSORICA is not a corporation, it is not a coop, it is not an non profit, it is not an LLP. It is an open value network. From a legal perspective, it is a non-registered association. It is an open network of freelancers that coordinate and co-manage their work using some IT tools (the NRP-CAS) and some special governance. If you still don’t understand what SENSORICA is and how it operates please don’t panic. It is something new and it takes a bit of time to get accustomed. It is new, but at the same time it is very similar to other new things that have emerged recently, like Bitcoin for example. We can say that SENSORICA is for production what the Bitcoin network is for exchange. So the million dollar question is how do we interface these new creatures with the ones from the classical world? How do you sign a contract with a loose network of individuals? Who is going to show up to do the work? How can we guarantee that a swarm of people converging on tasks from all four corners of the planet via the Internet will deliver on time, and with the required specifications? Who is responsible and accountable?
In our opinion, SENSORICA is the most advanced commons-based peer production network applied to hardware production, using infrastructure and methodologies that have been specifically tailored for open networks. We claim that the OVN model is able to sustain deterministic economic processes and accountability, while preserving the open and fluid nature of networks, while maximizing individual autonomy. This new ability of open innovation and peer production networks to generate predictable outputs, demonstrated by SENSORICA, was the main topic of the meeting with Jenn Gustetic from the White House, in June of 2015.
The role of SENSORICA in the service cases enumerated above shaped the network as the interface between the crowd and traditional institutions. In other words, input from the crowd can be structured and channeled towards solving someone’s problem, through SENSORICA’s infrastructure, methodologies and governance. But let’s not get confused, we are not talking about a crowdsourcing platform. So what is the difference? A crowdsourcing platform like Upwork is an intermediary between companies and the crowd: the company posts a problem with a prize for someone who can provide the solution; the platform takes a cut. Taskrabbit is the Uber of cheap labor, connecting people who need chores done with people who can do them, while taking a cut from their transactions. In both cases, the intermediary platform is owned by a company and those who supply the work aren’t organized, they respond individually to demands. In the SENSORICA model, no one in particular really owns the platform. The same holds for Bitcoin, no one in particular owns the blockchain network. Affiliates of SENSORICA organize, they form groups to tackle complex problems that might even require long periods of time. In the cases discussed here, the size of a project is comparable to a startup, reaching up to 10 individuals. The longest duration of a sustained project is over one and a half year, showing the capacity of SENSORICA’s model to sustain long-term crowd engagement. These are the first pilot projects, but the potential is for thousands of individuals per project, which amounts to a large size enterprise, for long-term projects that can take years. SENSORICA is really showing the signs of a new system of production that can operate at large scale. But as an R&D service provider, it can already be seen by classical institutions as R&D on demand, as an adjacent, very cohesive R&D operation open to the crowd, funneling in low cost and rapidly evolving open innovation. Practically the entire revenue generated is split among participants, with only 5% going to maintain and to develop the infrastructure, which is under the total control of participants.
At the third iteration of crowdsourced R&D service offering, the service beneficiary got a fast paced innovation at a quarter of the normal cost. Even more interestingly, the cost cuts aren’t transferred to those who provide the service. They are actual cost savings that result from a heavy use and rapid remix of open source, from the mutualization of resources within the network, from the collaborative nature of activities, from the elimination of bureaucracy, and other inefficiencies that come from lack of motivation. On the contrary, everyone is paid with the same measure, according to the Canadian labor market, no matter where the contributor lives. More precisely, within SENSORICA those who live in Pakistan aren’t paid less. And if that wasn't enough, on top of providing rapid innovation to classical institutions at a fraction of the cost, so that they can maintain jobs, at the same time sensoricans increase the value of the global commons, because everything they do is open source. All the data about the economic activity within SENSORICA is open to the public, we can’t make this stuff up!
This mutually beneficial economic relationship between classical institutions and SENSORICA, as an open innovation and peer production network, can be seen as a bridge between the classical capitalist economy and the p2p economy, as a channel for transfer of resources from the old economy to the new.
The Barda periscope project was the first implementation of a new open project development methodology designed by Fernando, Tiberius and Lynn, in the context of a service provided to a client. This methodology was formalized in SENSORICA’s network resource planning (NRP) software through a concept named Workflow recipes, which are time-dependent and deliverables-dependent bundles of Processes associated with a Project (a context of work). This methodology consists of the following steps: Project initiation, Design considerations, Design, Prototyping, and Product. All the contributions to the Project were logged within this structure.
In order to reduce the perceived risk for the client, the Project was divided into milestones. A cost estimation was produced for the client for every milestone. The agreement was to get paid at the end of every milestone. Every milestone was to be delivered with complete documentation, open source style. The client could stop the process at the end of any milestone and decide to switch to another organization to complete the Project. The documentation provided a guarantee for rapid continuation. The burden was on SENSORICA to provide a good service, at the level of satisfaction of the client, in order to complete all the milestones.
Moreover, the activity logs in the NRP and the associated documentation provided the client with full and real time access to the process. Coordination on different issues and tasks took place in context, directly in the working documents, and the client was invited to provide feedback on the fly.
A problem emerged during this project: very rapidly, the work documents became long and the client’s ability to follow the process was hindered. We spent time formatting the documents to make their content more transparent, but these measures didn’t diminish the time spent by the client to effectively follow the process. The situation was more complex, because this was a three parties relation, between the SENSORICA team, Barda and Parcs Canada, Barda’s client. Information produced by sensoricans had to be reformatted to match Barda’s project management structure and the language used between Barda and Parc Canada. In the end, Barda provided sensoricans with a template for 3-way communication, based on their own open issues and tasks.
The Barda periscope project was a small project, involving only a few contributors (see project in SENSORICA’s NRP-VAS). Coordination was fairly easy at this small scale.
Joshua Pearce is a professor at Queen’s University and Michigan Tech University. He is dedicated to open science and sustainable technologies, and had been following SENSORICA since 2013. He is the author of the Quantifying the Value of Open Source Hardware Development paper. For years, Joshua’s team has designed multiple scientific instruments by building on various open source projects. These instruments have been released under open licenses. In 2015, he decided to take a risk and transfer to the SENSORICA network the task of designing an instrument used in the characterization of photovoltaic materials. This was an important shift in Joshua team’s approach, from in house development with inspiration from open source projects to crowdsourcing development through the SENSORICA distributed network. The main goal was to create an instrument with a community around it, which would increase the speed of innovation, insure continuity of the product, and increase its diffusion rate to universities around the world. At the same time, the PV characterization project was also seen as a pilot project to build an interface between the crowd and a classical institution, Queen’s University, through SENSORICA’s p2p infrastructure, open project methodologies, and governance.
The open science movement is building momentum. It started with open publications, increasing access to scientific knowledge. This initiative became more nuanced, proposing early stage sharing of data and information (prior to the publication), sharing of unpublished past results and even sharing of lessons learned from failed experiments. In parallel with the development on the distribution side, the movement also built infrastructure for data sharing in resource-intensive domains of inquiry, like genomics for example, as well as social networking platforms designed for scientists and scientific projects (like Research Gate). Recently, we have seen initiatives for redesigning scientific instruments that are in tune with the open science philosophy. New instruments are acquiring new characteristics: they become shareable, they facilitate socialization of scientific activities, they become modular and interoperable, as well as easily serviceable and upgradable. Efforts also go into redesigning scientific labs, making them more collaborative, interconnected, accessible through teleproxmity, etc. SENSORICA leads the way to open science, as one can see here. The PV characterization project incorporates many of these new aspects.
This project was started by incorporating all the lessons learned in the Barda periscope project. There was a difference in scale: more individuals contributed to the design and the prototyping of the PV characterization device (11 affiliates and over 200 logged contributions). The requirements for accountability and responsibility were also higher, since we were now dealing with a University. All this put more pressure on our support processes. We created a Project responsible role, to be the interface between the University and the SENSORICA OVN. Financial incentives were attached to it. Moreover, the roles of outreach (find skills), orientation (help new affiliates get accustomed), coordination (make sure that all affiliates are on the same page) and facilitation (make sure that all affiliates get the help they need) became very important. We experimented with new tools for orientation that proved to be more effective. A specific forum was created for the project, in order to focus discussions. The PV project was also more complex, its documentation proliferated faster, which lead to the need of content maps in order to ease the navigation.
During the course of the project we noticed that the outreach function was very important and not so easy to finetune. The answers to our signals propagated on social media were slow to come and the conversion to an active contributor was low. We attributed part of that to a poor general understanding of SENSORICA’s OVN model, including its system of incentives. At the beginning of the project, we grossly underestimated the efforts required for outreach, for generating the content to be broadcasted, for establishing a constant social media presence, for mapping the open source ecosystem, targeting specific pools of talent, and establishing trust relations. The project was run below the critical mass of open projects and therefore required a more centralized form of governance.
NOTE: We cannot publicly disclose the name of our sponsor in the IoT for heavy industry applications case.
In December 2015, sensoricans were contacted by a Montreal-based company to help develop an IoT solution for applications in heavy industry. They wanted to make their product “smart” and able to predict its life expectancy. The requirements consisted of a mesh network of sensors that send data to a cloud for analysis, in order to predict failure. The race to be first to market set the pace for fast innovation and low cost. The company crafted a business model based on services, not on selling the hardware, which is fully compatible with the open source development that SENSORICA can offer. The agreement was that everything that SENSORICA develops can be released under an open source licence, with no restrictions for Sensoricans to remix this technology in other projects, including commercial ones.
Thus, the company became the sponsor of an open source IoT applications development project. CAKE, the custodian of the SENSORICA OVN takes in financial contributions from the company and distributes them to network affiliates, as a reward for their involvement to the project, as fiscal sponsorship. The company is not a client of CAKE, since this a three party relationship, between the company, CAKE and the world, the later benefiting from the open source IoT applications design, and not simply a one-to-one service exchange between two organizations, even if the company can draw a direct benefit from this relationship.
The Sensor Network project started almost in free form. The first tacit agreement was that the sponsor informs development based on their knowledge about these applications. Decisions on development were to be made during scrum meetings between Sensoricans and employees of the sponsor, Sensoricans would work on tasks, log their time contributions and get some financial compensation every two weeks, relative to their efforts. As the project unfolded, we felt the need for better planning and cost estimation. The first improvement was to manually create a map of content generated by SENSORICA’s R&D activities. This brought the idea of being able to generate dynamic content maps, either from the NRP-VAS (every development process has R&D documents as deliverables) or from our CRM (content management system), which is not yet implemented. In order to allow the sponsor of the project to follow almost in real time metrics about the project, we created an experimental dashboard. In the end, we realized that we needed to synchronize the sponsor’s ERP with SENSORICA’s NRP. We crafted a shared language and project development structure, and the agreement was to keep track of work in both places. This brings the need to create interfaces between the two management systems, which hasn't yet been implemented. Moreover, we also decided to produce cost estimates for future tasks, to allow the sponsor to better plan its budget. All these measures had a positive impact on our relationship by making our activities much more predictable and auditable, and by increasing the level of reliability of the network.
As the value created during this project increased, the project sponsor realized that his first-to-market advantage was in potential danger if the technical work was put in the context of their direct business interest in a public way. This sparked an interesting debate on openness (access to participation) and transparency (access to information). We drew on SENSORICA’s past experience with a project that was sensitive to transparency, and implemented an open and semi-transparent project model. In more concrete terms, anyone can join the project, which preserves the openness aspect. Most of the technical information generated is public from the start, but some documents that contain information about how different components can be used in an application similar to the business case of our sponsor were made non-public. Project affiliates need to sign a non-publication agreement for these documents clustered into a separate folder, but there is no restriction related to the use of this information in any other project. All these non-public documents have a date for publication, which is related to the sponsor’s market deployment strategy and pace. We believe that in through this arrangement we preserved the nature of the SENSORICA OVN, while mitigating the risks perceived by the sponsor, which led a stronger synergy between the two entities.
At the beginning of 2016 we were contacted by Living Labs Montreal and eVision to contribute to a digital access management project (smart locks). Our role was to explore the pertinence of blockchain technology in this field - see more on the project.
The main value perceived by Living Labs and eVision was the ability of SENSORICA to gather rare talent, in this particular case people with blockchain skills. The interface between SENSORICA and its partners was modeled as the Sensor Network IoT project. Most of the development work was performed during the summer of 2016.
During the Blockchain Access project sensoricans learned the hard way that culture is very important when interfaces between open and classical organizations are established. eVision’s corporate culture and Living Labs’ non-transparent and competitive culture clashed with SENSORIC’s culture of open sharing and collaboration. Sensoricans ended up doing more work than they initially bargained for, had to catch up with delays introduced by these companies, suffered from lack of transparency and had to endure a corporate/naif project management style. The experience has been recorded in this document.
During the course of this project (IoT) I suffered from impaired self-determination, a psychological process that leads to the well known “crowding out” effect:
When individuals perceive an external intervention to reduce their self-determination, they substitute intrinsic motivation by extrinsic control. Following Rotter (1966), the locus of control shifts from the inside to the outside of the person affected. Individuals who are forced to behave in a specific way by outside intervention, feel overjustified if they maintained their intrinsic motivation.
This was due to the company’s project management putting a choke on exploration to limit their costs. This didn’t affect some individuals but affected me, a core contributor. The other main contributor was motivated to find a better interface to manage these effects, so his motivation to the project was moderately affected and the other was a new contributor, who wasn’t part of the initial project terms.
This lead me to a few realizations:
I hope this makes some sense :-)
I am playing with the notion of abstracting these external events in the future. Usually when two unbalanced forces are joined together there is a need of a buffer to provide a balance.
SENSORICA service systems
Read Open Value Networks: a framework for many-to-many innovation
Traditional intellectual property law evolved to protect the livelihood of inventors. This makes a lot of sense, because it is much easier to reproduce a technology than it is to invent one. Invention takes many months or years of thinking, striving, experimenting and struggling, before an invention sees fruition. The patent system protects inventors by giving them the exclusive rights to market their inventions, thereby giving them an incentive to invent things in the first place. Without a patent system, an inventor who spent years developing a technology would be powerless to stop a wealthier man from stealing his methods and leveraging his capital to beat the inventor out of the market. With a patent, the inventor keeps this right for up to 25 years to profit from his technology before the patent expires and anyone can market the technology freely.
However, owning a patent comes at a cost. The fees to get your patent globally recognised can accumulate in the tens of thousands of dollars per patent. Also since the nature of a patent is that you gain intellectual property at the cost of publicly revealing your ideas their methods, it becomes tempting for others to use your methods and violate your patents. Even though you have the right to a patent, defending a patent in court can cost hundreds of thousands or in some instances even millions of dollars in legal fees which is often enough to bankrupt a small company before the court date.
A traditional alternative to patents is to keep trade secrets. With trade secrets, you do not reveal your methods publicly, so this is often used in cases where it would be difficult to reverse engineer your methods. The drawback of this method is that you become limited to the intellect within your organization as no one outside your organization can develop your technology since they don’t know it exists.
Today, we live in the world of lean startups, venture capital, and rapidly developing technologies. 25 years in an eternity in the high tech sector and a patent can completely block a market from developing. There is a case to be made for open innovation, as technology would develop more rapidly if everyone shared their technology with others.
However, from the perspective of a “traditional” institution, opening their technology up to the public represents a loss in competitiveness, since large investments into R&D can be easily taken and reproduced by competitors. In game theory terms, this is a tragedy. Since every traditional company would be better off if they held some trade secrets, and absorbed the open source technologies of others, it leads to an in-ideal situation for the market where information is not shared openly.
One solution would be to enforce temporary non-disclosure agreements, or 6-month holds on the release of information, source code, or designs, to allow a company to get a head start in the market, while still allowing the open source community to develop and improve base technologies. Another options would be to invoke government support for open technologies, such as tax breaks for open innovation. A third option would be the legal development of shorter term patents, that better reflect the current high-tech market conditions. For example cheap 5-year term that a start-up could employ to protect their IP, which better reflecting the pace at which the industry is moving.
At the end of the day, if an open source community wants to interface with a “traditional” company, they need to strike a balance between the open source’s need for transparency and the protection of research investment by the traditional company. There is a need to develop social, legal, and market frameworks that move the Nash equilibrium from the in-ideal closed market tactics of today, towards the open innovation of tomorrow. Humanity stands to benefit from it.