Why I prefer making useless stuff

"I think the #1 project every programmer should try is a project that has customers." - jfoster on HN

"Don't waste your time on homework problems." - anonytrary on HN

Update 5/24: See the discussion of this post on Hacker News and Reddit.

I get a lot of responses like this about spending time on useless projects. But you know what?

There is nothing I find more liberating than to spend a Saturday afternoon coding on some toy project.

There is no expectation or obligation to ever become more. It is just for fun, learning, and curiousity. I can add whatever features I want. I can use whatever technologies I want. I can throw it all away if I want. These are the projects I yearn for and look forward to when I'm busy with work.

Perhaps George Bernard Shaw was on to something: "We don't stop playing because we grow old; we grow old because we stop playing."

Funny enough, Titus Barik published a paper that contributes a qualitative analysis on the sentiment of programming and play on Hacker News. The themes he found include play as artistry, catalyst, fun, playgrounds, spontaneity, tinkering, and anti-work. There is a quote that I particularly like, "The joy of programming for programming's sake is something you do in your own time."

It all makes sense to me, because I have spent a lot of time building useless stuff. Mostly compilers and game engines. To name a few projects:

• Nash, a low-resolution HTML5 game engine with no dependencies
• Wiggleface, a very low-resolution game engine using Pygame
• HarvestMemory, a game where you code a bot in an assembly-like language
• Knox, a compiler for my own language that overemphasizes "explicitness"
• cosine, an implementation of cosine from scratch
• SirDrCaptain, yet another static website generator that no one uses
• goforth, a bad implementation of the Forth language
• CHIP-8 virtual machines, I think I have made a dozen of them
• Darkest Days, a remake of my favorite Flash game

And if those aren't enough, I have several more useless projects planned! I can't wait to get started on them:

• A typed Lisp compiler with a focus on performance
• An interpreter for my own language that feels like old-school PHP plus dataflow programming
• A "better" CHIP-8 that supports networking and sound
• Two different todo list apps for iOS

Do you need ideas for useless projects? Check out my two lists:

• Challenging projects every programmer should try
• More challenging projects every programmer should try

Whatever you do, don't forget Bob Dylan's wise words about the things you enjoy doing: "you're gonna have to serve somebody".