I have this pet theory, rather grand, and falls into the category of what you believe is true even though you cannot prove it. That open source will realize the end of history.

In 1989 Francis Fukuyama wrote the celebrated and controversial book, The End of History, which posited that the collapse of the Soviet Union was a Hegelian triumph of liberal democracy as the last remaining form of government and political philosophy. Fukuyama went on to explore issues of social capital and tyhmos, “desire for recognition” that drives free-market economics. His critics were manifold, particularly those on the wrong side of history. Marxist criticism centered less on liberal politics than liberal economics — particularly market failure. The classic debate over the role of government centers on what economists call market failure: when the market fails to provide social goods.

Similar to how Doc says the demand side is supplying itself, with open source and open content social goods are produced through peer production. Let’s explore one aspect that is less about code and more about social dynamics triumphing over economics, language. For a small country like Rwanda, a localized version of Office would never be supplied, so they do it themselves. Some vendors are open sourcing their localization in recognition of unevenly distributed demand. While more research is required, some patterns emerge with stories behind them when comparing language support by markets and peers:

World population and internet population are gauges of demand. Web content is supplied by both markets and peers. Wikipedia is produced by peers, although the stories behind the community distort the current outcome. LISA.org (Localization Industry Standards Association) is a measure of market production for localization.

Wikipedia isn’t a perfect gauge of peer supply when markets fail, because it is a community with rich stories of how it evolves. Perhaps over time and at greater scales the rise of the Swedish version would be a signal of bottom-up fulfillment, but today it may very well be preferential attachment spawned by early adoption and there is also a high level of market-based translation effort. The Polish exception may well be the same, but there is an interesting story here.

Wikipedia has had two forks in its history, both by language based communities when commercialization was a potential threat. The Polish fork was resolved and re-integrated. This explains why Spanish Wikipedia is low in its ranking relative to online population:

Arle Lommel from LISA was kind enough to gather this data for me (perhaps a benefit of Socialtext’s membership), and also provide some analysis which I encouraged to share openly. Beyond the tabled measures of translation in volume, he provides analysis of strategic languages that are off the chart:

One generalization is that Wikipedia lags behind all others in Chinese translation because its relatively centralized and censorable.

I had coffee with Hong Kong University Researcher Andrew Lih today and I will pass on some of his research on regional language use in Wikipedia in an update later. But he made a significant point that second languages are a primary determinant of development in wikipedia. For example, users in India and the Philippines have such a high rate of English as a second language that their own languages have yet to develop within Wikipedia.

But I would end with this thought of the Polish exception. A polish online encyclopedia at the scale of the wikipedia version would not have been developed with market and contractual signals alone. Social signals are driving this production and producing a social good. The story behind it is an exceptional community, but an exception that could very well become the norm as we march towards the End of History.