Comment by drivers99
6 years ago
Sounds like one of the principles in Nat Greene's book "Stop Guessing". It has 9 principles of problem solving and one of them is called "Smell the problem." He has an example in the book of a machine which was wrapping packages of toilet paper and they could never run it above a certain speed and that was limiting the output of their factory. Eventually they had to look in the machine while it was running faster than it normally could to see a screw would extend out into the path of the plastic sheet only when it was running above a certain speed. So they were able to fix it for free, when the alternative workarounds were to buy a whole new machine.
I use that a lot at work and often it involves packet captures and other low level tools to observe hard problems at the lowest available level to see what's happening, instead of guessing and coming up with possibly expensive workarounds based on guessing.
A thousand times this. The better I get at debugging, the less inclined I am to guess and check. It’s almost always worthwhile to learn or write a tool that will show you what is actually happening. This can feel like getting sidetracked, but it isn’t. The answer almost always becomes clear.
I have seen a case where so many people wanted to go and see part of a company’s operation that it started putting on a traveling stage production of itself, setting up in an HQ conference room every few months. I joined early enough to see it in situ and found that the touring version was nothing like the real thing.