Patrick Ewing infamously proclaimed, in his role as the leader of the player’s union during the heated negotiations with NBA basketball team owners circa 1999, that the strike was an issue of putting food on the table for his children. Ewing and his offspring may not have faced any real threat on the food and shelter front, but it could be argued that it is indeed a question that dominates (and determines) the lives of most human beings. In previous posts ([1], [2]) I offered some responses to Dan Jalkut and John Gruber’s criticism of the GPL and Matt Mullenweg’s defence of it. Both Dan’s criticisms and my responses were centred around his sense that the GPL stifles participation. But the other 800-pound question is: is a GPL model sustainable? And importantly, are the alternative Open Source ones any more so?
The GPL for better or worse (and I believe mostly for the better) is associated with Richard Stallman and it is not unreasonable to wonder if outside of enclaves like MIT (where Stallman resides) and without the spare time hobbyist pursuits of supergeeks, a viable model can exist for sustainable free software development as envisaged by GNU and the GPL. If talented coders are spending time developing complicated applications for free, how do they put food on the table? Is it not by leveraging their skills to either gain an income elsewhere (i.e., a giant closed-source non-free corporation) or gain research, academic or equivalent funding for pursuing such projects? How well do such models even scale?
Stallman addresses these very issues in his essay Why Software Should Be Free. His response is along two lines. The first is the simpler one: as in the case of my Ewing analogy, the issue is really not one of putting food on the table but the pursuit of enormous wealth, and once the wealth factor is discounted, it will be found that such development is not so costly an affair after all. The second line is the more interesting and important one: Stallman points out that software is a public good and is as such best funded using public funds.
Along the way RMS also outlines why, as his title notes, software should be free.
Open Source proponents, in particular Linus Torvalds and Eric Raymond, have argued for a “pragmatic” approach that Jalkut (if I read him right) echoes. If corporations can use the code, thus increasing its usage, why not? While I am not certain that they make such an argument (it has been a while since read Raymond), one can be made that in turn such corporations (e.g: Google, RedHat) can be patrons of open source development, thus solving the funding problem. And therein lies the problem. There are differences between patronage and libre and one of them is that while the latter might require scaling down one’s expectation of wealth (as outlined by Stallman) the former is the one whose sustainability rests on the charity (to put it a bit bluntly) of others.
An aside: do they get it?
In papers such as An empirical analysis of economic returns to open source participation, a strange sort of hand-wringing occurs as a result of the seeming paradox of why someone might write software for free. That the paradox may be a result of the theoretical commitments of the researcher is best left to Amartya Sen to explain (sorry, pay site). RMS explains it in simpler terms in his essay as well, in terms that should be eminently sensible to anyone but a Cartesian sceptic!
An apology: this post was meant to be a much more long-winded one that attempted to explain and defend my thoughts on this issue in a more rigourous manner. But perhaps you are relieved rather than in need of an apology?
