… bit dated and I haven’t been able to find any recent study about the efficacy of regression tests, but so much of what we’ve been conditioned to believe to be good software engineering practices and disciplines is built upon the primacy of automated testing that any form of doubt cast on its efficacy sounds sacrilegious to many. If there’s anything I’ve learned in the last few years of witnessing how services fail, it’s that …
It’s a good idea to be able to describe what is testing (and automation). A good description is the ‘search for unknown unknowns’. You can think of automation as scaffolding. This makes automation quite different from the search for unknowns.
You can use automation for the search for unknowns (however, that is not relevant for starting the discussion).
These ideas have been proposed by the likes of ‘Lessons Learned…’ for a long time.
I think the bigger issue is that organizations aren’t aware of the problems users/customers face.