Here is the final version of our paper for Wikimania, Network Analisis for Wikipedia.
The aim of our simple experiment is twofold: to gain some understanding of the high level structure of Wikipedia, and to get some insights about its content, and in particular on its hidden cultural biases. Each user usually browse a (relatively) small set of entries during the normal usage of an encyclopedia; and such small sample is more representative of the enquiring user world view, rather that the whole encyclopedia. As a consequence, nobody is really able to have a mile-high fisheye view of Wikipedia’s content; of course it is possible to perform some basic statistic analysis (like counting the entries, or measuring the rate of growth) but these are purely syntactic measures. We maintain that network analysis offers a simple way to have some more "semantic" measures, since it formally analyzes an intrinsically semantic human-generated type of content: the use of terms to define other terms.See also: our previous work on network analysis and Wikipedia.