More is less

Once again, the guys at 37signals just get it right. People often think that the best system is the one with the most options, the most flexibility. Absolutely, scientifically-provably wrong:

Offering shoppers samples of six items yields more sales than offering samples of 24, students who are offered six extra credit topics are more likely to write a paper than students who are offered 30, etc. In some cases, just one additional choice can produce outright analysis paralysis. People wind up frozen by indecision.

There’s a direct application to government IT projects here. We talk all the time about ‘stakeholders’, capturing requirements from all and sundry, and (inevitably) designing by committee. This is precisely why we’re wrong to do so.
Don’t take it from me… take it from Professor Barry Schwartz. He wrote the book on it. See him describe his thinking at Google.