Troubleshoot
I'm finding that a lot of technical troubleshooting is comparing a thing that works with a thing that doesn't. Troubleshooting interrogates these two things with a series of questions. What makes the thing that works different than the thing that doesn't? What are the steps needed to close the gap between those differences?
That little heuristic has saved my hide many times within the first few weeks at my new gig and continues to be a useful tool of thought. Funny how its underpinnings come from subjects that years of schooling have only left scant traces.
Control groups sound vaguely familiar from science labs. I can see where the thing that works acts as a control group for my troubleshooting experiments. Then there are independent and dependent variables. Perhaps troubleshooting is the understanding of what the dependent variables are for something to work and removing the independent variables from the picture.
It's fun to think about those things again — countless subjects that come to the surface long after you've learned about them. Life as a game of remembering what you learned long ago.