I have a folder called _graveyard in my home directory. It has 34 subdirectories.
Most of them are projects I started with genuine intention and then quietly abandoned somewhere between “this is working!” and “okay but what’s the actual point of this.” There’s a CLI tool for managing dotfiles. A static site generator I wrote to avoid using static site generators. A thing that was supposed to track my reading but mostly just tracked my guilt about not reading.
I used to feel bad about this. Now I mostly don’t, but I’m not sure if that’s growth or just lowered expectations.
The thing nobody says clearly enough: most projects are for the making, not the having.
The dotfiles manager taught me something about POSIX file permissions I would not have learned any other way. The static site generator made me understand why the existing ones are designed the way they are. The reading tracker is a perfect archaeological record of two weeks in 2024 when I thought I had my life together.
None of them shipped. All of them were worth doing.
I think the “finish what you start” advice is well-intentioned and mostly wrong. It optimizes for completion as a virtue independent of outcome. But completion is just one possible good result — sometimes the good result is what you figured out halfway through, and the finished product would’ve been a distraction from that.
The unfinished projects aren’t failures. They’re just a different shape of done.
That said, the reading tracker really should have been a spreadsheet. Some lessons are expensive.