The OpenBSD kernel has a set of functions to help detect memory corruption, the poison subroutines. The memory management code uses these functions, but they themselves have a very simple interface, no complicated types or data structures, meaning they’re easy to replace. What if we rewrite the memory corruption detection functions in rust so it’s impossible for them to cause memory corruption?

Interesting article about modern vs classical deps management in software.

Sums up my own thinking on the subject, although I’m leaning towards balancing 1 and 2.