What happens if you take a supercomputer and the entire contents of Wikipedia, then analyse every single article to see how the world's sentiment has changed over time and space? This wonderful video is what you get.
Put together by Kalev H. Leetaru, this visualisation shows how opinions on situations across the world have changed — at least, based on Wikipedia articles about places and events — as a function of time and space. In Leetaru's words:
"See the positive or negative sentiments unfold through Wikipedia through space and time. Each location is plotted against the date referenced and cross referenced when mentioned with other locations. The sentiment of the reference is expressed from red to green to reflect negative to positive."
For the most part, sentiments tend to stay positive — with the exception of major wars. The visualisation works amazingly well up to around 2000, but then it kind of becomes one massive green blob. That's because there are fewer articles on Wikipedia which are tied to events stretching back historically. So while the data for the 19th century is interesting and might allow us extra insight, without some form of filtering the modern data isn't much use. Still, it's fascinating to watch. [YouTube via Flowing Data]