I think this analysis is directionally correct, but I will nitpick the “4% of commits to GitHub” stat. Because just anyone can create a GitHub repo, this isn’t really a meaningful statistic. There was a script going around a few years ago to write a message on your frequency graph by creating a bunch of junk commits in an unused repo, which I wouldn’t be surprised if it spiked to 20-30% of all commits in any given day at its height. What’s interesting about software is not its existence, it is instead its use.

Edit: we gotta fix this it-related contraction problem gang.