You weren't richer but the worl might be. Sometimes clean sheets and first principles discover assumptions that have changed. Either in how people use software, how the domain itself works, or the constraints of computing.

I think that depends a lot on whether the failed startup produced any code, papers, conference talks, even podcast episodes, etc. If not, and everything was just buried in the person's head, then no, the world is not richer for it.

This kind of thing is why I'm so grateful to the founders of startups like Starsky, who are obviously still immensely passionate about the domain they were in. That even in the wake of the company failing, they've continued to advance the state of the art by sharing what they learned, especially where it's wisdom that goes against the current dogma (eg, that teleop is a hugely important part of a self-driving vehicle strategy, but that investors of the late 2010s were turned off by that focus because they wanted to see companies trying to shoot the moon on L4/L5 autonomy): https://medium.com/starsky-robotics-blog/the-end-of-starsky-...

My startup idea was to recognize objects by their shape. I got it to the point where I was able to setup an elasticsearch index that looked at point clouds uploaded and performed feature extraction on what was ingested. I haven't open sourced by code because I probably had committed API keys (but I should make it public now since all those keys stopped existing). The biggest issue was a large amount of debt and lost opportunity cost.

But it gave me a new perspective and at times it was really fun learning so much. I think it has made me a better developer because I made so many complicated mistakes and now I can simplify things.