Doubtless, my Gentle Reader has heard some form of the adage, "The definition of insanity is doing the same thing over and over while expecting different results." It's not a bad rule for nearly all situations, actually.
Problem is, it doesn't necessarily work that way in Software Development, particularly during debugging.
See, when you're trying to reproduce a problem, you want to see the same results after doing the same thing over and over. Anything else spells extra time and resources. Second-to-worst case scenario is you'll end up essentially creating a VirtualBox-type simulacrum of the production environment so you can replicate the live conditions as closely as possible. Maybe you'll end up dorking with the system date, or writing custom scripts to roll back test data to a specific point in time...repeatedly. For sure, you'll be spelunking in the database, likely scribbling down ID numbers and combing through matrices of data.
Worst-case scenario, though, is that you're never able to reproduce the error, EVER. Nope, no matter how faithfully you re-create the conditions from the bug report, the gremlin never reappears. That, mes amis, is The Short Road to Crazytown.
So if you've ever wondered why some programmers (and other I/T folks) have a slightly skewed perspective on the world, this is one of the reasons. For us, madness lies in the exception, not the rule. In more than one way.
