How not to test software
a talk by Stan Prokop
In 1946, Grace Hopper found a bug trapped in a relay and coined the term bug. Except the term “bug” for an error was actually used before, for example in 1878 Thomas Edison in his letter to Theodore Puskas. Moreover, it wasn’t a true bug, it was a moth. And it wasn’t even Hopper who found the moth. Plenty of bugs in the bug’s story.
Software testing is full of myths, mythical silver bullets and “good” and bad practices. In this talk, I’ll debunk myths, share bad practices and explain how good practices can turn out to be not so good when applied in the real world of software engineering. All with focus on our beloved Python.
Examples are unit tests tightly coupled to implementation details, always unstable integration tests and targeting 100 % code coverage.
This talk is suitable for both beginner and advanced Pythonistas.
As a youngster I wanted to become a game developer, but (fortunately?) I happened to become a QA engineer, and I eventually transformed into a software engineer with strong testing background.
My past includes Centrum.cz, GoodData and Skype/Microsoft and I luckily end up between good folks at Accolade (not the game developer which no longer exists, but on-demand healthcare concierge).
I have a wife, one kid, no dogs and no cats. I have no quirks. I just prefer to walk barefoot at home, in the office and sometimes outside.