Up early and into work, again struggling to figure out what’s wrong with this software. Instead of working on the original machine which was owned by the test team and I shouldn’t muck with, I switched to my development machine where I can do what I want. The issue is that a completely different set of problems showed up. So now to solve the original problem I felt I had to fix these other problems. Each minute seeking insight caused me two or more minutes of repair work. I was making progress in the wrong direction. The original problem was getting obscured in the mass of secondary problems. I felt I was losing my way.
At different times I would switch back to the test machine trying different things. I keep focusing on why that system could not load a specific Java class. This problem had not shown up on my development machine. I had assumed that the class library was not accessible. Instead what I failed to do was check the contents of this class library. When I did I found it was empty, contents nada, length zero. No wonder the system could not load a specific Java class. The person doing the installation had failed to load the class library properly, that is, it was empty when it shouldn’t have been. So I correctly reinstalled the class library and TA-DA the system started working. I would never have guessed that this was the problem. It was one of those small assumptions that kept me baffled for two days. Morale: Do not assume, instead check your underlying assumptions.