Deming said, “Every system is perfectly designed to get the results it gets.”Â
Back in 2001, I presented a paper titled “Better Testing, Worse Quality” at the SM/ASM Conference.Â
In it, I demonstrated how a heavy investment in building up a QA group ironically caused overall quality to go down. We designed the system to improve quality, but it got worse. How could that possibly be?Â
The answer is that all the elements of a system influence each other, sometimes in surprising ways. When we increased the investment in QA, the developers decreased their investment in testing. It makes sense once you find the hidden connection, but until then it is maddeningly counterintuitive. That’s how systems are.Â
In this talk, Elisabeth Hendrickson examines the nature of software development systems with thinking tools like causal models and Shewhart charts, as well as frameworks like the Theory of Constraints and Queueing Theory.Â
You’ll learn how to reason about the interactions and (sometimes hidden) relationships between different aspects of the software development system. By the end of the talk, you’ll be more equipped to understand why things are the way they are, and help others see what you see.
Comments