The question that stuck with me for 30 years

Better Developers

My first job out of school was at a company that wrote software for hospitals, and the head of QA was the most feared person in my department.

I know that sounds dramatic, and it was, a bit. As you can imagine, our work had to be reliable, so we had an entire team dedicated to quality assurance. Their job was to run our software for months at a time, with many different inputs, and make sure it didn't cause trouble. If it weren't for them, buggy software could have gone out the door, with catastrophic effects for people being treated in hospitals around the world.

So the QA team ran our lives. The head of QA walked through our department looking for problems, because that was his job. We joked about it, and we were also, quietly, grateful for it.

More than 30 years have passed since I had that job. Since then I've kept teaching Python in a lot of settings, which now includes companies, conferences, books I've written for O'Reilly and Manning, and the newsletter you just subscribed to.

The thing that's stayed with me from the hospital-software days is what the head of QA cared about. He cared about whether the software worked the way we said it would work, and running and compiling and passing the tests was only the start of what mattered. The real question he kept asking was whether the software did what we told people it would do when they depended on it.

That's how I think about teaching Python now. Writing a for loop is only the beginning of what matters, and understanding how Python is running the for loop, and whether your mental model matches the real one, is what I'm teaching. The code runs fine until the day it doesn't, and then you're the person causing trouble.

Better Developers is where I try to close that gap one article at a time.

Reuven