That's a great principle, but they key to realize its value is actually learning from it, and one of the most important lessons here is that in startups, spectacular successes are driven by luck. If that one lesson doesn't sink in, then all other "learning from failure" is akin to training your machine learning model on random noise.

> If you don't make it running a company, maybe you can make it from writing, just don't stop trying to make it.

Isn't that confusing what someone who fails should do ( do it again/do something else/try another path) with what those who see it should do ( not listen to whoever failed about how to succeed at doing what the failed person did )?