“The definition of insanity is doing the same thing over and over and expecting different results”
~  Author Unknown

Recently read an article from Joe Kraus of Google VenturesIt’s premise was to reject the commonly held notion that people learn best from their mistakes.   His position is that we tell ourselves that to simply console ourselves; that people gain real insight from their successes.  He makes the argument that in experiencing failure we simply have identified one avenue that does not work out of countless possibilities; but when you find success you have found the path.
I think that is unsound reasoning.  As engineers we are constantly striving to find the “best” solution to a particular problem.  That problem is bounded by certain constraints.  If we were to find success in one project, to assert that we could simply repeat the identical steps to solve a different problem is simply put, asinine.  Solutions are, unfortunately perhaps, highly tailored both to known constraints but also unknown factors that contribute to our success or failures.  Call it fate, destiny, luck or what have you, but success and failure is not always a matter of having the best solution to a problem.  There are fudge factors, be it timing or some subtle nuance in design that contribute to our successes and failure independent of our choices and simply beyond our control.
So my contention is that both failures and successes can be false indicators of future successes or failures.  Just because something worked this time for this particular problem doesn’t mean you can simply repeat the process and be guaranteed success on the next project.  Likewise, a product or tactic that failed previously could be exactly what the current problem needs as a solution.  My point is simply this, do not become jaded one way or the other by your failures or your successes.  Learn from both, remember the context in which the worked or failed as well.  And do not assume that just because something worked last time that it will be guaranteed recipe for success this time.  Unless you simply solving the exact same problem under the exact same conditions.  Even so, as I contend earlier, it is impossible to know all the intangible, beyond your control factors that previously contributed to your success or failure.  So take your successes and your failures, put them in your toolbox but don’t blindly trust or distrust either.