100% this. I have become better about not nixing projects I am working on when I hit technical walls recently. My strategy has basically become to just do the bare minimum to test whether an idea is good or not and if it sucks I will at least have fulfilled my intellectual curiosity and then I can stop being distracted by that idea.

For example, I am currently working on a project to send personalized slack messages in bulk. But I then became really interested in the idea of automatic podcast transcription using GPT-3. So I ended up just one night banging out a prototype on jupyter notebook and it turns out the auto transcription was pretty bad and expensive, but at least I know now. So now I don't feel as bad and I can focus on the boring aspects of building this slack plugin, like accepting payments.

This is exactly me. It is like I get bored as soon as I realize that I could do something, and move on to the next thing.

The book "Refuse to choose" was really enlightening to me, it talks about "scanners", people that have this attribute. I felt very seen and it taught me that this style of working can be an incredible strength.

https://www.amazon.ca/Refuse-Choose-Interests-Passions-Hobbi...

I once had an artist friend who was world class in his medium. I mentioned to him one day that I have a lot of unfinished code and his response has stuck with for over 20 years.

He said great artists have one thing in common, they finish projects. It's OK to have a workshop full of experiments, but you must finish projects.

I've always taken it to mean that forcing your self to do that last 20% of a project is the only way to learn how to make something great.

I do this too, but the larger issue I face is that none of my ideas are even the slightest bit marketable.

I suppose I naturally gravitate to problem domains with little market potential. I don't know why.

I go into more detail here: https://news.ycombinator.com/item?id=31217221