This project, titled “Wikiviz” by Chris Harrison, is a visualization of the links within Wikipedia articles. There are around 1.5 million wikipedia articles out there, and Harrison wanted show all of them, and how they connect to each other. Wikipedia is an interesting system of data, and by studying the links in the articles, you can get an understanding of how we (as humans) organize information and make connections. This visualization reminded me of a game I used to play with friends using Wikipedia— We would decide on a particular topic (e.g. a specific political figure, animal, food, country…etc.), hit Wikipedia’s “Random Article” button, and then see how many clicks it would take to get to the decided upon topic. The person who could do it in the fewest clicks won. It was often surprising how quickly you could jump from topic to topic, especially considering how diverse the articles on Wikipedia are. I think this phenomenon has interesting implications on the interconnected nature of our world— nothing exists in isolation.
To deal with this massive data set, Harrison used Java to create a his own software (called WikiViz) that renders nodes on a graph layout with a spring-based technique. Connected vertices attract each other, and draw nodes together. What’s cool is that it can use many sources— It doesn’t just have to be data from Wikipedia. In another project, Harrison used the software to analyze connections between books on Amazon.
The result is a beautifully complex web of connections. Because there’s so much information, it’s hard to digest, but it communicates the sheer amount of information out there, which is awe-inspiring.