Ask HN: How do I get out of the “new project” loop?

The same happens to me to a lesser extent (there are a dozen half-baked chess engines buried in my Eclipse workspace). Maybe you are just looking for a distraction because your current task bores you - then there's probably no happy ending to those project skeletons. But maybe it's part of your creative process? I've procrastinated several ideas for decades, always starting and abandoning an implementation in a matter of minutes, but eventually I saw (some of) them through. Maybe that is just your way of "thinking out loud". If that indeed bothers you, then maybe tweak the parameters a little, eg. think more about the project before coding the first line or draw schematics on a paper or discuss them with somebody first?

To get out of a loop, break unconditionally. Literally. At any of the steps you have described, just don't follow it.

So, got an idea? Don't build a skeleton. Just leave it there. I would suggest you do write it down somewhere, maybe take a few extra notes, but that's it. Leave it there for a while. Another idea? Same thing.

Not only will you have gotten out of the loop you're in, but in a short time you will probably have a bunch of ideas written down. Now you can review, compare, relate them. And still don't start building on them yet. Just sort them, prune them, refine them, discard many, remix a few, join others, modify a few more.

Avoid getting started on any idea too soon. Only after this process, when you do find one idea which particularly calls your attention and it's fleshed out enough to convince you that you are actually interested in following through, only then, I repeat, start working on it.

And still, don't just jump into building it. Again, before building: think, plan, find the hard parts, find the easy parts, ponder, evaluate, refine it further.

The thing is... implementing the idea is not that creative. It is, but only to a certain point. The creative part is the idea itself, shaping it, refining it, making it consistent, solid, interesting, useful... But if you start building it right away you don't leave room for the idea to grow, to take it's correct shape, to become actually interesting or be revealed as the opposite.

It sounds like you need to be told what to do.

A solver with no real problem is a miserable thing indeed.

Someday you may have it in you to “make” yourself finish something.

Until then, you will need some external master demanding passable results or you will dwindle in idle infinite possibilities to the end of futility.

As someone who gives up on ideas after a week of work ( not after 30 minutes ), I must say the New-Shiny-Idea-Monster is pretty kind for you ;-)

Maybe try to write down each of your ideas, give it a +1 vote each you think about it. After a month select the one that had the most votes and try to implement it. Writing to down the idea might be enough to give you strength to postpone executing it immediately.