It began with a question: “Why is Uncomfortable Situation Seal different than Socially Awkward Penguin?”
From November 2014 until early 2016, I was collecting data on Internet Meme (image macros) usage, predominantly via Reddit.com. I wrote a script to pull down memes that had been transcribed with an OCR script (written by someone else), stored them in a database, and
cron‘d this to check nightly. Over 15 months, I collected in excess of 27,000 unique samples.
This project also involved writing a Ruby on Rails application to serve the dataset, allow me to better maintain and sanitize the data using the
ActiveAdmin gem, and also introduce some “games” to better understand how intuitive some of these were. The application uses Facebook OAuth for authentication, but anonymized all user’s names (for privacy). Statistics are compiled using some fancy
In early 2016, I submitted a proposal to the PSU Elements conference to present my findings, and was accepted. The talk was about 45 minutes long, and was titled Fantastic Memes and How to Use Them. The slides are available online (see link below).
This presentation has appeared in: