Yes, I'm puzzled by your posts in this thread. It looks like you are alternating beween pushing a Nirvana-fallacy strawman (where "secure" is a binary variable

Yes, I'm puzzled by your posts in this thread. It looks like you are alternating beween pushing a Nirvana-fallacy strawman (where "secure" is a binary variable and everything not perfectly secure is trash) and treating "secure" as a spectrum... One can't defend against unknown unknown, but mitigating known issues (input validation) and delegating to tools that know better than you (rustc) are both valid strategies to produce software that is more secure upfront (and I know you know it, eh)...