Learning from mistakes versus learning from feedback

On the knowledge management for development list there was a discussion about learning from mistakes. Matt Moore pointed to the mistakebank ning platform. I decided to join because I was curious to understand whether you can really learn from mistakes. Some of the examples I felt were rather blunders. For instance, I bumped into another cyclist today- that was clearly my mistake or rather blunder but nothing that you can really learn from, unless maybe to watch out more carefully but hey this is life... so you just say sorry.

And can we only learn from our own mistakes or also other people's mistakes? Through the mistakebank I found a blogpost citing Eleonore Roosevelt who said "learn from the mistakes of others, you can't live long enough to make them all yourself" which was contrasted with the idea that you can only learn from your own mistakes. That would imply we can observe others, their mistakes, and incorporate that in our own practices.

However, most mistakes are not as obvious as bumping into another cyclist. Most mistakes may go unnoticed, may not be recognised like mistakes by the mistaker (did I hear someone mention Bush??) or only by some and not by others. Or the effect of the 'mistake' may take long to become visible. I went back to Kenya where I worked and found out the whole irrigation scheme we built was flooded again because the water levels had risen. And at times it is easy to label something as mistake in retrospect.

In terms of changing practices, learning from honest feedback may be more important, in the form of spontaneous feedback or in organised processes like After Action Reviews. Today I got positive feedback about my blog by being mentioned in this list of knowledge management bloggers. I also got positive feedback after a meeting I conducted "may I give you a huge compliment for this meeting" someone said spontaneously. Since I became freelancer I got much more positive feedback than when I was an employee. I have the impression it is much harder to create a culture of feedback in organisations, after all, it is quite sensitive and may alter your relationships. I also got feedback recently I don't recognize. Hence, it's important the receiver recognises the feedback in order to act upon it, by giving clear examples.

Maybe we need to distinguish levels and types of mistakes? Or rather talk about experiments? Or is it more worthwhile to improve feedback? And how do we learn from mistakes as organisations?