> It used to be you could email a patch to someone and that patch would get reviewed and committed. Yes, but 1990 is also when this study came out: https://ft

> It used to be you could email a patch to someone and that patch would get reviewed and committed. Yes, but 1990 is also when this study came out: https://ftp.cs.wisc.edu/par-distr-sys/technical_papers/fuzz.pdf https://ftp.cs.wisc.edu/par-distr-sys/technical_papers/fuzz.... They find that a lot of UNIX tools could be easily crashed by generating random inputs. The quality of our most popular open source software has dramatically improved over the last 30 years.