How to Work with Stupid People

And the HN submission:

http://news.ycombinator.com/item?id=1299764

I usually assume the same thing and it usually works. But it hurt me once, pretty badly. I spent about a month (!) trying to wrangle good color performance out of a particular scanner for an imaging system. The unit was pre-chosen, I was to make it work. I was assuming the whole time that I was doing something wrong, or that I didn't know something. The real answer was that the scanner manufacturer had fired all of its color scientists and the scanner it question had hardware flaws that were effectively non-correctable.

In this case, it would have been far better to be more aggressive. I could have started screaming as soon as things started to look strange and thus saved myself several weeks. But usually the heuristic is a good one.