I'm a bit confused on the meta level. If it actually works well in practice, why aren't there quite a few memory-safe, single owner-based systems programming l

I'm a bit confused on the meta level. If it actually works well in practice, why aren't there quite a few memory-safe, single owner-based systems programming languages out there? Is it such a new idea? (The author argues that it's not, C programmers have been doing it without compiler support for ages). Or is only a part of it new? If yes, which part?