I have a pet peeve whenever people bring up the old "Billion dollar mistake" quote. Calling it a mistake means that it could have been avoided. But we couldn't have avoided null pointer errors (or some other name for the same thing) any more than humans could have avoided the bronze age. In the 60s and 70s we didn't have the "technology" to avoid null errors. By technology I mean production-ready languages with a static typing layer sophisticated enough to implement a Maybe/Optional type. Those languages weren't production ready until maybe around the 80s. And most of those compilers were themselves implemented in unsafe-null languages like C.
But anyway, that's history, now we have nice languages that have safe nulls, so the interesting questions moving forward are: why are people still creating new languages that have unsafe nulls (looking at you Go), and why are people still choosing to use languages with unsafe nulls?
But anyway, that's history, now we have nice languages that have safe nulls, so the interesting questions moving forward are: why are people still creating new languages that have unsafe nulls (looking at you Go), and why are people still choosing to use languages with unsafe nulls?