Commons-Based Peer Production – Theory

With the enormous success of free software, many people started to ask themselves if their philosophy or modus operandi could be applied to other disciplines. That’s the thing about success, everybody likes it. Non-software companies wanted to know if they could apply the same “tactics” in order to triumph in their sector. Software companies wanted to understand what had hit them. Everybody turned their heads towards Richard Stallman, the founding father of free software. Stallman’s answer was “I don’t know”. Clearly, that answer did not satisfy anyone. An answer was needed.

Four years ago, Yochai Benkler, a Law Professor at Yale University, wrote an essay providing an answer to that question. The essay was called Coase’s Penguin, or, Linux and The Nature of the Firm and it captured many people’s imagination, including mine. The answer was “Yes”.

Production modes

Until now, we knew of two ways on how to organize production: markets and companies:

Markets are formed by individuals and coordinate production based on a price system. In other words, individual A says “if you want me to do that, you’ll have to pay me so much” and individual B responds “that’s too expensive, I can find someone cheaper“. In the end, an agreement is reached so that A produces (or not, and another task is sought). Markets suffer from high “transaction costs“: ask freelancers how long and how much money they use to define and ensure that each client meets their commitment.

Companies are formed by employees and coordinate production based on an order system. The manager says “I’ll pay you a salary. Do this” and the employee says “OK“. There was already an agreement at the time of hiring, the employee produces (or not, and seeks another job). Companies suffer from high “organization costs” as a result of coordinating several people in a hierarchical, centralized way.

The third way

However, the success of open source cannot be explained using either of those models. Open source projects produce the highest quality code that can make monopolies tremble. Yet they do not coordinate based on a price system or a hierarchical managerial structure. This indicates the existence of a new production method which Benkler calls “Commons-Based Peer Production” (CBPP), of which open source is one example (the most visible one) but not the only one (wikipedia, digg, blogosphere, etc.).

Benkler goes even further and states that CBPP is a better system than the other two when (a) information or culture is to be produced in large-scale projects, and (b) the necessary resources (computers, telecommunications) are widely distributed. If those two requirements are met, CBPP is superior for two reasons:

1. Lower transaction costs: Even if you allow a large number of contributors to collaborate, transaction costs are minimal. By eliminating property, no contracts are needed.
2. Lower organization costs: The costs of allocating human capital to production processes are lower because people “self-allocate” and choose where they want to contribute (what wikipedia article to improve, what news item in Digg to vote, what post to write, etc.).

Doubts and mistrust

What production method can we choose? Easy: the one that, after subtracting costs, provides greater profits. If you set up a company that provides services via the Internet, you’re producing information (software). If you do this at a large scale and believe what Benkler says, the answer is clear: CBPP is the way.

What is CBPP’s weakness? Although it’s true that, by eliminating property and hierarchy, transaction and organization costs fall, it is also true that, if profits cannot be allocated to anyone, there won’t be any motivation to participate, and that if, noone coordinates, there won’t be any organization and the system will fail. Benkler’s response is that motivation is very complex, many people participate because of social and/or psychological motivations unrelated to money. Regarding organization, the secret lies in establishing microtasks that require little motivation for completion (granularity), where each module can be completed independently, incrementally and asynchronously (modularity), and where there’s a cheap mechanism to integrate the competent contributions (a quality code that does not violate property laws).

Can CBPP be harmonized and make money? I mean real money. The same as, or more than, any other competitor that uses the company’s traditional method. And, if possible, outside the open source production for the stack (operating systems such as Red Hat, databases such as MySQL). This is the one million dollar question (I couldn’t have said it better) which Negonation will try to answer in practice.

In my next post, I will comment on our experience and difficulties in implementing CBPP. If this is not clear, I am using and will use free software and open source indistinctly. I have also deliberately left out any type of philosophic or moral consideration when commenting on the various systems. My aim is to assess and demonstrate the usefulness of CBPP in “business” terms because this is the only way that “business” people will take this seriously.

By David Blanco
Saved in: Business, Entrepreneurs | 1 comment » | 30 August 2006