The graphs indicate the number of various words in c, h and S source files in the linux kernel, updated monthly or whenever I hear of a new release. It's inspired by the linux kernel fuck count, which unfortunatly is very outdated.

As is apparent from the graphs, the number of swear words has increased a lot, starting around version 2.4. However, the size of the code has increased even more, so there is in fact less swearing per line now.

A little about the counting method used: A word is counted if it appears in any context, even if part of another word. Originally it was smarter, but there turned out to be a really nasty memory leak in FreeBSD awk's regex engine at the time, and I never updated it since. A word will can be counted multiple times per line, in case some developer had a really, really bad day.

Back to other stuff Vidar Holen's made.

You can get the script here, but it's pretty horrible and I wouldn't recommend it.