Hi DH fans! I have a question for you — when you think about information on the Internet, what benchmarks might come to mind that would reveal a certain topic as having thorough, accessible information online?
I, for one, immediately think about Wikipedia. I’m not alone in this — here at the DHC we’ve had multiple Wikipedia edit-a-thon’s focused on topics like women’s reproductive health and Algorithms of Oppression, aimed at growing the representation of certain topics and identities on what is now undoubtedly the world’s most-used encyclopedia. Another way of thinking about information online, however, has to do with the content of those Wikipedia pages themselves — how long are they? How many sources do they cite? How many sections do they comprise? How densly do they inter-link with other, related pages? And, althought you might not immediately flag it as a representative figure, there’s also how many images the page has. Enter the Wikipedia Cross-Lingual Image Analysis tool, a tool that looks at Wikipedia pages across languages and compares how many and which images a given topic has in any of the languages it has a page in.
When you enter the url of a Wikipedia page into the tool, it gives you a list of languages with links out to their respective pages, and a set of thumbnails of the images embedded in each page. To explore the tool, I looked up the wikipedia page of the early 20th century German film star Marlene Dietrich — I was curious whether her page would be more thoroughly built up and image-d out in German or English, the two languages in which she made films. Sure enough, while the majority of languages had two or three images on her page, German and English were among the most densly image-populated — the German page has 12 images and the English page 19. Not entirely surprising, given that the super-star phase of Dietrich’s career kicked off once she got to Hollywood. What did surprise me, however, is that the French page had them both beat cold, with 34 pictures uploaded.
Obviously, my example is a bit of a silly one, but one can imagine the use of a tool such as this in research that considers the differing biases and perspectives and therefore differing avaiablity of information on any given subject across cultural and national lines. Give it a shot!