That's true for product development, but it's not true for mathy libraries. Perfect is achievable. For a released software that humans will decide to use or not, rapid iteration is great. But also: https://randomascii.wordpress.com/2014/01/27/theres-only-fou...

Precision and exactitude and formally proven correct software can exist in some problem domains, and it's kind of silly to not achieve that when it's achievable.

But in this case, C is not "good". It is more like "abysmal". "Good" is just producing a correct result or error, with no ambiguity which case applied and no UB. "Perfect" is arguing over the most usable and elegant API for it.

Once a program is available over the internet, hackers are the enemy of merely good programs that don't perfectly validate their input.

"You have to get lucky every time. We only have to get lucky once".