Open source and the commons

Much of the software running the modern world is written by people who are not paid for it and given away to anyone who wants it. To anyone trained to expect self-interest, this should not work. Understanding why it does is a useful corrective to a narrow view of motivation.

The economic puzzle is straightforward. Code is non-rival: my using it does not stop you using it. Traditionally, such goods are under-produced, because a producer cannot easily charge for them. Yet the Linux kernel, the browsers almost everyone uses, and the libraries underneath most websites are all free in this sense, and they are among the most carefully maintained artefacts in the industry.

Several mechanisms explain the result. Contributors gain reputation that converts into employment. Companies contribute because a shared foundation is cheaper than a private one, and because they would rather influence a standard than follow it. Licences do real work: a copyleft licence requires that improvements be shared back, converting generosity into a rule rather than a hope. And the cost of contribution is low, because the artefact is infinitely copyable and the work can be done in fragments.

The model has well-documented weaknesses. Critical projects are sometimes maintained by a single exhausted volunteer; the people who write the code are not representative of the people who use it; and an open licence offers no guarantee that anyone will respond when something breaks. Security researchers have found serious flaws that sat unnoticed for years in code everyone assumed someone else was reading.

A school project sits inside this tradition, modestly. Publishing the source of a student platform is not a claim that the code is excellent. It means that the method can be inspected, copied and corrected by the next school — which is, in the end, the only form of durability a volunteer project can honestly promise.