> Personally, I like to put a single song on repeat for hours and hours, days even. It helps me zone in.<p>I'm pretty sure this is a form of torture. It would be for me, at least.<p>I think this advice is okay, as long as you've defined parameters for what "finished" even means. If your goal is to make an app, publish it to app stores, and maintain it in perpetuity as a personal income stream, then great. Go for it.<p>Bringing a piece of software to production is almost never my goal with a side project. Usually, my goal is to explore a new concept. Maybe I want to get a feel for how "problem X" may be solved in the real world, whether or not my program is one of those real-world solutions. Maybe I'm revisiting an old project with new wisdom.<p>Take this for example: I have created no fewer than 4 visual novel engines, none of which are production-ready. A couple years separate each one, and all of them are wildly different under the hood. One of them is made entirely out of Nim macros. Another one uses Nim to do the drawing, animation, "game-enginey" stuff, but embeds the entire Ruby MRI with a custom DSL that the developer uses to write their game code. Another one is built on pixi.js, and the first one was just an enormous abstraction on top of ren'py.<p>All of these approaches are wildly different. All of them taught me new software design principles. None of them needed to be "finished" in order to do that.<p>The goal of side projects doesn't have to be "publish a production version" or "turn a profit." Cut it out with that hustle culture nonsense. Go make art.