If you are making the point that human made creations will never be perfect, I'm in 100% agreement. But with some (minor? major? moderate?) effort we could reduce the incidence and severity of bugs by factors of thousands if not millions. In my view, that seems like a goal worth striving for. Along with capabilities, and memory management, we can employ better
testing strategies and static typing and other lightweight formal methods. Here, I'll mention some of Les Hatton's
publications. I'm only railing against the notion that we're already doing a good job at security. We're not. We need all the help we can get. And better languages will make it a lot easier.