If the main goal is hardening existing code bases then this Clang proposal is the right approach IMHO. I think integrating slices "properly" would touch quite

If the main goal is hardening existing code bases then this Clang proposal is the right approach IMHO. I think integrating slices "properly" would touch quite a few other areas of the language, which require opinionated design decision by the committee. And committees aren't exactly known for coming up with sensible language design decisions (and looking at C++, I'm fine with the C committee being very conservative). Basically, if I want an actually "better C" for new code, I use Zig. But C is still a valid option for low-level libraries that need to work everywhere (and of course for tapping into the existing library ecosystem).