I was making a point about what factors influence clone performance in general, not trying to contribute to the tedious Git vs. Mercurial debate. If I had found

I was making a point about what factors influence clone performance in general, not trying to contribute to the tedious Git vs. Mercurial debate. If I had found any data about, say, Gnu Arch vs. Codeville (or some other abandoned codebase), I would have used that instead.