Except that they didn't even quite get that right. The article seems to contrast git and hg's versioning schemes, when in reality they're both hashes of the co

Except that they didn't even quite get that right. The article seems to contrast git and hg's versioning schemes, when in reality they're both hashes of the content (the non-unique hg version numbers referred to are merely local aliases).