Quote:
Originally Posted by IKE
I am a firm believer in the high value of Success after failure.
All too often the "learn more from failure" seems like the nice thing to say after something pretty bad goes on. From a value perspective, I would actually say, that you learn more from 1 success than 1 failure. IE, if I try something that works, I have found something that works (which is valuable piece of knowledge). If I try only one thing that fails, then I only know that one failure point.
Now, if I get the chance to dig in, and learn why something failed, and then turn it into a success... that has a lot of value in my mind.
Of course, learning how to "do soemthing the right way" that leads to success... that can be priceless.
|
For some reason your post made me think of something that popped up on my Facebook news feed a while ago:
http://www.nytimes.com/interactive/2...m-solving.html
I recommend giving it a try before reading on, or the whole thing will be spoiled
Basically, it deals with a question of confirmation bias. In trying to find out the rule the number sequence follows, a surprisingly large number of people will only out in sequences they think follow the rule - get a bunch of "yes" and your assumption at least fits the rule, although it may not match the rule.
For me, I went almost the exact opposite when I tried it - I tried to get failures. I put in number sequences specifically to rule out possibilities, and the more failures I got the more I knew about the parameters of the rule.
Taking this to the team... Failure and iteration can help you to explore many different paths and define the full parameters of success for a robot. Building something that happens to be successful right off the bat is nice, but can lead to more narrow thinking about what success looks like, as you haven't tried other things that went in other directions.