23 ms·
In Defense of Systems That Fail (2023)
- nonrandomstring 3y agoNice post that ultimately revolves around cost-benefit analysis. Where I hoped she was going with this was into human value decisions behind those analyses. There isn't such a clear cut between safety critical life and death systems, and ones that can afford to fail now and then as a practical compromise. People's weekend projects get purloined by big corps and used in critical systems they never anticipated or would be comfortable with. Nothing stands alone and seemingly innocuous components in complex systems cause domino effects. That is to say; our intentions as coders, including levels of care, are seldom respected a few steps removed. As always, what bugs me is when practical engineering - learning from failure - crosses boundaries into reckless engineering. The cavalier engineer sees production (other people's lives) as part of their test lab. That is never okay.