Fun project, but buggy data makes for a bit of a crappy user experience. Doesn't GH classify languages pretty well using it's Linguist project? The project with ~6,700 stars is correctly classified as [ Ruby 97.5% | HTML 2.5% ] (the 2.5% seems to be from documentation and templates, which in itself is already weird)
I understand that many people find gamification fun, but I know a lot of people who don't like ranked leaderboards, based on stupid things like "how many people bookmarked your project", it leads to unnecessary egotistical competition, and driving more barriers between us than we need. (Example, what benefit would we get from comparing stars on vim, and neovim?… so we could argue to one group of people that their work is less important than another group?)