There are too many shiny objects and it is killing me

In a way, this is where it can help to have a developer day job with a salary or at least that mentality. Where commercial pressures and tight deadlines can light a fire under you to just get things done.

In some cases these types of motivators can cause you to hack out something that only barely works at first, before circling back with test writing, best practices, etc. But maybe that can be better than not completing anything at all?

You can always refactor your way to an awesome codebase if it starts out a bit quick and dirty at first. Sometimes just having the minimum amount of friction to complete the requirements to get going can be helpful, as starting is sometimes the hardest part.

Maybe you just need to set deadlines as if you are the CEO of your company and not the developer where by certain dates you need the MVP or new feature and it has to happen no matter what.

Maybe build an MVP such that you can start charging people and get customers, which can also be a motivating force to continue on and continuously improve it because you feel the obligation to your paying customers.

The guy should just read what he just wrote. Anything you want to create, you can probably do with the list of technologies that you already know well and have experience with. The real problem is that he doesn't want to actually build anything. Building something is hard, putting out there for people to see is hard, realizing that it might not work or isn't perfect or wasn't what you set out to build is hard. Setting up project after project, is easy. It's masturbation with Hacker News. If he was interested in building something, he'd build it with the fastest, crappiest stack that he knows the best and gets it infront of as many people as possible so they can tell him how much it sucks and what's it's missing. This guy just wants to fiddle with terminal colors more than building anything.

Thank you for giving it a name. I do believe you nailed it with FOMO. Interesting.

There's a concept called (in various circles) a spike, proof of concept, etc. You don't write tests, plans, requirements. You just code.

Its a terrible way to write production software. Its a rare thing in my life to spike out an idea and put it unmodified in production. But when i find myself doing what you're doing (which does happen from time to time) I find its a crucial part of breaking myself out of the cycle.

And not for nothing, but you sound like you're isolated. Working with other developers as part of a team (or better yet, a pair) can also do a lot for breaking out of this kind of cycle. ADD drugs will help, but they're not all of the solution. Stop "preparing" to write software and just write some.

I wouldn't say that I'm the exact opposite of the OP because I do constantly add to my tools, but I tend to be about 1 year behind and prefer to wait until things settle before I choose technologies.

I would say you suffer from a combination of analysis paralysis and simply just fucking around too much! One thing that helps, I feel, is to just allow yourself some time for experimenting and deciding upon your development stack. Give yourself that time guilt-free, but set a limit. Then when you've decided, just force yourself to stop dicking around and get to work.

It's important to realize that you will probably regret some decisions and celebrate others, but that there is almost no way you will know which is which until later. This is just the way it goes. There will always be more shiny things to play with on the next project!