All of us encounter the same types of problems over and over again -- a user not adopting a system, a project running longer than anticipated, an application that just can't do what you need it to. But how many times do we take a step back to assess if there's a way to address the issue that might just be better than the way we've dealt with it before? Innovation should never be an end in itself, but solving problems in better ways than we have in the past should be.
As a knowledge manager, part of my role is to make sure that the practitioners at my firm have access to all the ways other practitioners at my firm have handled similar issues before. But I'd hate for that to be the end of the line. As Toby recently pointed out, modeling future choices on past behavior is not always possible, nor is it even necessarily reliable. Don't get me wrong, I'm not arguing that we should never rely on the past; when like equals like, reinventing the wheel is the epitome of foolishness. But relying on the past solely because it was done in the past is not good enough. Each time we face a challenge or a puzzle to solve, we need to evaluate not just the problem but also the past solutions, and assess if there's a better way.
I remember when a good friend of mine got the latest toy on the market for her birthday -- she immediately put the headphones on, found her favorite cassette to put into the yellow box clipped to her belt, and began dancing around the living room while no one else could hear the music she was playing. I'll always have fond memories of that yellow box (which, to be sure, was itself a true innovation). But something tells me there's an even better way.