@grimalkina prob because I work on an environment (ML) where resource changes need to be defined by teams (can be a change, can be a glitch), then team consensus is not always a good measure, and teams open to reconsider issues even (and mostly) if outside of their consensus definition can be (and mostly is) a driver to radical innovation.
Like those stories of scientists bumping into weird results and, instead of discarding them, trying to understand them, resulting in “radical innovation”.