First day of the Peer production seminar in Nottingham university is over (though I'm posting this the next morning). I have to say it exceeded my expectations, so I'm really glad I was able to attend. There are several characters with a strong Free Software background here mingling with academics who are sociologists or whatnot. It creates a good mix of people on the one hand being able to give very concrete examples of how Free Software projects do stuff and on the other hand the academic minded people being very intelligent about analysing and generalising the concepts in action.
Andreas Wittel and Michael Bauwens introduced the day. Michael clearly has been thinking about this stuff for a long time, most things that we end up talking about he is already aware of and can immediately give very detailed insigths to them.
Michael drew up a good topology of various peer production economies:
|Sharing (Youtube, early Linux)|
|use value||exchange value|
|community production||foundations provide infrastructure||commercialisation|
|probabilistic community production vs
command and control outsourcing
The first category of peer production communities are those that simply emerge to facilitate sharing. Youtube is a good example. It is a commercial site (exchange value), but the actual activity of producing and sharing videos is driven by volunteer work (use value) and the division of these two aspects is characteristic. One crucial thing to understand is that the non-commercial nature of the volunteer driven part is a good thing. The volunteers have various motivations for there participation, but money is not one of them. Time and time again there are ideas to create a better sharing system where part of the revenue on the commercial side (advertising revenue, say) is shared with the volunteers (youtube video producers, say), but these attempts always fail. They are either unattractive to begin with, or if they are shoehorned onto an already existing community they might end up destroying the ecology that was initially non-commercial.
The commons model is very popular in mature and established projects: Apache, Linux, Mozilla... The idea is to have a foundation that can finance the necessary infrastructure like servers, perhaps some seminars, etc. Yet, the actual production stays volunteer driven and the foundation stays in the background. (Note that in the case Free Software developers are employed, the employing company can be seen as a volunteer participant in the project.) Also in this model there are commercial ventures that monetize the product, and obviously these may be a crucial financer of the foundation in the middle.
The final model is much like the previous ones in my opinion - perhaps from an engineering perspective - but the difference is that the initiator is the commercial company, who wishes to incorporate the peer produced product into its own products. Michael had a good slogan for this model that he called "probabilistic peer production" in contrast to a traditional command and control outsourcing model. It is like "you make it and we see if there is a market for it". I would like to add that the community is often quite good at such probabilistic production, it is probable that a community will produce products it finds usefeul. Since this is not a given in a company setting (I have first hand knowledge, believe me), this probability is probably one reason community production is actually very efficient.
Ok, that was just the first of many very high quality talks, but now it's time to post this and continue with the next days talks (including my own...)