GIFs are the second-most sophisticated form of expression on the Internet, after linking to Paul Krugman columns. Which is why you should always have a robust collection of them on hand. Now you need look no further than GIFGIF, a vast database of GIFs sorted by emotion.
The site, launched by MIT Media Lab students Travis Rich and Kevin Hu, is an interactive project that collects data about a visitor’s emotional reactions to GIFs and then uses that information to add to and categorize its growing collection of animated images. The result is a fantastic digital tome of GIFs that range in category from contempt to satisfaction to pride.
The data is collected quite simply. Whenever you visit the site, you’re presented with a question (for example, “Which better expresses satisfaction?”) and two GIFs. You must then make a quick selection. Sometimes there’s a clear winner. Sometimes neither applies. But once you’ve voted, Rich and Hu use that data to add to their detailed collection.
When I say detailed, I’m not joking. If you’re looking for a GIF that expresses an intersection of four or five different emotions, you can use a search feature that allows you to rank the expressional parameters of the feeling you’re trying to convey. In other words, you can find the perfect cross-section between, say, pleasure, shame, pride and relief.
Currently the site is made up of 1,024 GIFs that Rich and Hu refer to as “naturally grown, free-range gifs.” They were plucked from the popular GIF site Giphy, but Rich and Hu plan to add new clips to the site soon.
Though GIFGIF is still in its early stages, the pair hope they can use it to shed some light on how GIFs depict human expression.
“We’d love to find some data measurements that show this variation quantitatively,” they told BuzzFeed. “We would also like to answer questions like, how are angry GIFs different from sad GIFs? Why are some GIFs ambiguous while others are clear?”
You can contribute to the project right here.