On the occasion of an email discussion, I asked Alex Rollin to specify his conception on the difference between peer production and ‘open source’. Below is his contribution, which in turn gives me the occasion to clarify my own thoughts.

Alex Rollin:

“Open Source is a number of things at once. When looking at Open Source and Peer Production together, here are a few things that come up.

Open Source is a way of packaging instructions so that others who follow the instructions using comparable materials and efforts will produce the same results. In a certain sense it is a way of writing a recipe or lab instructions. A few of the important indicators for deeming something “Open Source” are: key components are declared. If certain tolerances are required in measurement or consistancy or makeup of materials, or certain skills, if any of these are required in order to insure a successful re-fabrication of the outcome, then these are explicitly declared.

Open Source goods have no rules about them in general except that taking them apart, reverse engineering, or re-engineering them is in no way disallowed.

While many will claim that Open Source means nothing without a community, this is not the case in my experience. Any individual can produce an Open Source good.

A good example of an Open Source product is the following: I just invented a new brownie and peanut butter sandwich. You use brownies as bread, and put peanut butter between 2 slices. Thin chewy overcooked brownies work best. That is an example of Open Source. What it is, how it’s made, and your ability to relate to it (ie disassemble it or make your own.)

Now a quick example of Peer Production. That Brownie sandwich I mentioned: Try it if you like, and if you make it better, let me know! I just learned to make brownies from scratch and they are perfected by my account. If you make peanut butter and want to put together sandwichs with me, that would be great. My neighbor John supplies me with peanut butter and we are now in business together, but he only makes 40 gallons of peanut butter a day which isn’t enough to keep up with my ovens.

Peer Production is about making the Open Source things. It’s not about making them for market, exclusively, but that is used in this example. Peer Production is about several people working together to make a single thing. That thing may or may not be Open Source, but each member of the Peer Production community will bring something to the production process, and they may bring nothing at all to the invention itself.

Open Source covers the invention itself and the relationship that a user has to the invention, denoting, specifically, that that user is also a potential collaborator if they choose to excercise their right to extend, rebuild, disassemble, or re-engineer the invention. Peer Production is the act of 1 or more people interacting together to manufacture a good based on individual skills and access to resources in order to increase production through higher rate of return through specialization in capabilities or resources and decreasing overall costs.

The process of working on an Open Source project might be done using Peer Production, and a Peer Production operation might be building out Open Source projects, but the two are not the same, necessarily.”

Michel Bauwens:

I usually define peer production as consisting of three interrelated processes, forming a coherent mode of production, governance, and property.

The first condition is access to open and free raw material, which may be used without permission, but under certain conditions. Thus, open source , or free software or knowledge or designs, are in certain sense a prerequisite to have peer production. This is so with the exception that a community may start from scratch, and create its own open source in the process of ‘production’ (or produsage, as Axel Bruns would say).

The second aspect is the development process itself, which must follow different criteria, such as equipotential participation, permissionless production, communal validation. Participation itself must be free and voluntary.

Finally, regarding the output, this must be universally available to all, which is usually codified in the use of commons oriented licenses and common property formats. This in turn creates open and free knowledge, software or designs, on which future iterations may be built.

This creates the possibility for all kinds of hybrid or incomplete processes.

For example, it is possible for for-profit companies, or bureaucratic governments, to hierarchically produce ‘open source’ as an output. In this case, there is no open and participatory development, but the outcome does create a commons which others could use for peer production. It is a paradox that a commons can be created in different ways, even as it were, by ‘slaves’.

Another possible variant is to have a open and participative process of development, but that the result is not universally available, either because a privatizing license is used, or because the result of the development is a rival good that needs to be purchased. Cooperatives represent a special and related case, since they produce goods that are not universally available. I do not regard cooperative production as peer production, because work in a cooperative is conditional (no input, no wage), and because the output can be privatized. But both can work together, cooperative production can make use of peer producing communities that deliver their open designs, and can in turn support this commons creation from which they profit. The combination is a desirable goal, as cooperatives integrate more of the values of equity and freedom than classic for-profit production. It would be natural for free software and open design communities to create cooperatives that can operate in the ‘market’ for the creation of ‘scarce’ and rival products.

What if you are paid, but the development process remains open? As I said, we can have companies producing open source code, using salaried workers in a hierarchical relationship: this is not peer production. But we can have companies paying workers, but allowing them open development, in that case the salary functions as a kind of basic income, which essentially enables the voluntary contributions, and therefore, this would count as a variable form of peer production.

A commons can be a hybrid result of a open peer producing community, and paid workers, where the latter can both contribute to the commons, or to marketable privatized added value. Some variants of open source licenses lack the strong defense of the commons, and may lead to privatization of common work (which is impossible with free software and the GPL), which is why the distinction between free software and open source software remains important.

Overall conclusion: one can have open source, which guarantees universal availability of the immaterial non-rival aspect, or at least the how-to instructions for a later physical production, without have peer production. But we cannot have peer production, without the universal availability guaranteed by the open license. And we do not have peer production without open and participative development.

True peer production requires open inputs, open development, and universal availability of what has been produced. Open source only guarantees the latter.