Random
Waiting to fix known problem
I just spent a good hour debugging a problem I knew was there when I first implemented this feature. At the time, I thought to myself that I would get to the solution for this problem when I was done with the feature. So I finished the feature and, of course, completely forgot to resolve that known issue (it was a corner case that was not all that obvious of a problem). Two weeks later (today) I'm having this problem I can't figure out. An hour later I remembered (well, debugged until I "remembered") the problem and quickly fixed it. Had I fixed it at the time I would have saved myself an hour of frustration two weeks later.
It's those kinds of lessons that are *very* hard to communicate to others. It seems like you have to not only experience it yourself to understand but you have to experience and realize that you are experiencing it.
