Graphing Wikipedia
http://matpalm.com/blog/2011/08/13/wikipedia-philosophy/
Internet webcomic XKCD (http://xkcd.com/903/) proclaimed that by clicking on the first link of any Wikipedia article that is not in parentheses or italics, then repeating, you will eventually wind up at the Wikipedia page for philosophy. The posted article takes the statement made by Randall Munroe and analyzes it by following the “all roads lead to philosophy” instructions and recording the number of pages that lead to philosophy and the number that don’t. As it turns out, about 94.5% of Wikipedia articles will actually eventually lead the user to philosophy. This is an interestingly high number, but is far from being an absolute truth. About half of the articles that don’t make it to philosophy are cycles (one that the author pointed out was sand fence -> snow fence -> sand fence) and the other half lead to articles that have been deleted or dead ends.
If we look at this through the lens of the model of basic PageRank update, and restrict the graph directions to just the first link on the page, we would find that 5.5% of the weight of the graph would be distributed across the smaller cycles that don’t lead to philosophy and dead end pages. The 94.5% of the weight would be concentrated in the Philosophy -> Reason -> Natural Science -> Science -> Knowledge -> Fact -> Information -> Sequence -> Mathematics -> Quantity -> Property (philosophy) -> Modern Philosophy -> Philosophy loop. In other words, these 12 pages would be very highly ranked, and nearly everything else would be very lowly ranked. In this method, these 12 pages would be considered the 12 most important pages, and even with restricting the links to just 1 per page, it seems we arrived at a relatively large subset of the most important broad topics in society.
If you wish to try this out for yourself, you can either go to Wikipedia and click through the links by hand, or go here (http://www.xefer.com/wikipedia) and enter the title of a Wikipedia page. It will navigate the pages for you and create a visual history of navigation.