Matt Soar discusses the invaluable aspects of failure over on Design Observer:

For example, Jonathan Hoefler says, “Increasingly I think about the work that I do not so much as a directed effort, but as the ability to recognize accidents and interpret them productively. Even failures have their place, since without them there’s no progress: anything that’s truly ‘experimental’ has to run the risk of failure.” Hoefler describes these moments as “happy accidents:” “Several times a day, some misstep on the computer produces an unexpected result, and sometimes these results are fetching, intriguing, even provocative.”