Ask HN: What's a side project you built to make money that hasn't?

Sorry, I meant hosting. In hindsight, the domain was actually my best asset (it’s a PR 4 domain), so I should’ve kept that :( )

I built itrackmine.com (http://mashable.com/2008/12/26/itrackmine/). And its killer recommendation engine (books, movies, and music) -- we knew what you owned from all stores, not just what you //bought// at Amazon (or single store) so ours were extremely accurate. Along with a "user A is this similar to user B" system...and the whole tracking, sharing, mobile app, barcode-scan, manage-your-stuff-package.

Made $10 over the ~8 years it was up...from one donation. Yey.

My first attempt was a music site called "Rhythmscore" (no longer own the domain) written mostly in bad PHP. It was awesome, 100% ad-supported, and it was obviously going to make me filthy rich. Small, indy artists would upload their music, and visitors to the site would listen to songs (in random shuffle-order) and give them a numerical rating. The songs with the highest "Rhythmscore" would be displayed on a leaderboard, etc.

One thing I discovered, which surprised me, was that it was very easy to convince artists to get involved and upload their stuff (free exposure I guess). Getting regular users/music fans to care was not. I think I made a grand total of ~$40 in ad-revenue over about a year before I shut it down. Lesson: Ad-revenue doesn't work unless you're getting truly massive amounts of traffic.

Next attempt was a homeschool tracker-type application written in Ruby-on-Rails. The thing shuffled along, zombie-style, for 2 years with about a dozen paid subscribers before I conceded that it wasn't going to be successful and shut it down. Lesson: Stay away from the homeschool market. Egads, those folks are cheap (I should have known: we homeschool our son, and I'm cheap).

I'm currently preparing to launch my latest attempt, a productivity (intended to be a B2B SaaS, though could be useful for personal productivity as well) application written mostly in Clojure on the backend and ES6/mithril on the frontend. It's certainly my best work (from a technical standpoint) so far. I wrote it to use myself for a few things... We'll see if anyone else finds it useful. I certainly learned a ton building it.

Arcade game developer here.

I loved the game concept of ABA's "Defeat Me": http://wonderfl.net/c/9ykQ . SHMUP where each enemy is a copy of you. Round 3 you fight a clone of Round 1, Round 2, mirrored.

but I disagreed with several major design decisions:

1) Rounds were too difficult (a deadline came in very fast, the bullets were nearly the size of the ship) 2) Player movement & weapons changed rapidly (speed & number of shots per spread was a modulo on the level number 3) Losing meant you retried the same level.. at that point in the game, you basically be stuck. I wanted the game to restart.

I improved on the game with pixel-perfect collision, weapon animations and other design decisions like letting the player screen-wrap around the edge but not the clones, regaining time on the deadline when you killed a clone and more.

I even experimented with a co-op, 2 variants of 1 Versus 1 mode and a Replay site before removing them for the core experience.

I made the game because I wanted to play it. Making money was a secondary goal but I'm proud of the game and find it way more fun than most games I've played. How many games do you know that have an AI procedurally determined by your movements?

Revenue earned: 3 * $10 on "pay what you want" from itch.io. 1 from a friend's friend who overhead on a Skype chat that I made a game, 1 from an indie developer at a games meetup and 1 from a friend when I sent them a link.

+ I think it's cool that $10 was the amount chosen by 3 people who didn't know each other as a valuation for the game. However it's disappointing that all 3 were people who knew me.

I sent out ~100 emails to game reviewers with download links and got 1 brief article but no revenue.

Why: Discovering games is hard. Marketing to people who would like this kind of puzzle-y arcade game is difficult to reach without looking like a spammer or spending $.

Differently: Not try so many weird experiments with gameplay and focus on the core relentlessly + Playtest more.

Iterations: 2 prototypes from TigJam. 2 more iOS versions then another re-write for an App Store launch. 2 Javascript versions (the latter being launched) + 1 seperate version for multiplayer.

Time: Hundreds of hours. Maybe a thousand? A lot of this was done earlier on when I was less experienced with programming + design. I playtested a lot

Lessons for Game Developers: Playtest often, stick to the core relentlessly. Carve away crap that doesn't fit no matter how cool it is or fun to make. I spent too much time iterating on mobile interfaces when gamepad controller support was far superior and passed "the bar test" (give to strangers at a bar and watch them cheer other strangers on as they pass the controller around).

Play [1], An old trailer [2]

[1] http://QuantumPilot.me (redirects to itch.io) [2] https://www.youtube.com/watch?v=oy_GBLM_6Qs

I built a personal data self tracking app (quantified-self stuff) and I still use it but no one was nerdy in quite the way I was to want to track things similarly. https://uyday.com/