[Letters from a New York Indie is a diary written by friend of Hookshot Inc, Kevin Cancienne, one of the creators of Drop7, who recently left his position as Director of Game Development at area/code to pursue life as an independent game-maker. The series documents life as a one-man indie as Cancienne explores what happens next.]

I think I know how to make videogames, though I do have my doubts sometimes.

One of the things I think I know about making games has to do with how you approach a new idea, and try to bring it into the world.

Fela Kuti was famous for the “endless groove.” In “No Agreement,” he introduces one instrument at a time, building slowly over 15 minutes until the song is a richly textured, seething ball of rhythmic and melodic complexity. In my mind, the ideal game development process would feel like a Fela Kuti song.

I prefer to approach a new game with what I’ve come to think of as an additive process: start with a small, quickly prototype-able idea, then layer on new concepts and game mechanics one-by-one, testing and confirming as you go, so you can be sure what you have is working as expected.

New York-based game designer, academic, and gadfly Eric Zimmerman is a playtesting and prototyping fundamentalist. Recent tweets sketch his design philosophy. On playtesting: “[...] if you’re not playtesting the living shit out of a game you’re developing, something is wrong with your design process.” On prototyping: “A prototype is a cane designers use to feel through the pitch dark space of design possibility.”

I agree with Eric’s methodology, even if I sometimes lack the discipline and thoroughness with which he applies it. But why bother painstakingly, cautiously adding new pieces to your design in bite-sized chunks? Because making — and knowing you’re making — a good game can a slippery prospect.

The artform of games is about uncertainty at its core, and that uncertainty rears its head from the very start of the design process. Because, as Michael Brough explained recently, good games tend to sit on intrinsically unstable ground, and your unfinished house of cards always threatens to collapse.

Different games seem to have different “critical mass,” from a design perspective. That critical mass is the bare minimum you need to prototype before you can see your game in action, and develop a sense for whether you’re barking up the right tree. If you’re making a platformer with a twist, first you need a platformer.

If you’re making a first person shooter with a twist, as I did a month ago with Plus Gun, first you need a basic first person shooter… If you don’t happen to have one just sitting around, you end up burning a lot of time just getting to square one, before you get to experience any of your fancy new design ideas. And before you start to wonder whether those ideas were as compelling as you thought they’d be.

As I described in my first column, this is what’s great about building small games. Small games push you, the designer, to shit or get off the pot.

Why be in such a hurry, though? Why insist on rushing to some kind of critical conclusion about a game you’re working on? Why not allow your idea, and your ambition, to flow outward and fill as much space as they want? For some types of games, and for certain designer-auteurs, this is exactly the thing to do.

But for most small teams or solo developers, these decisions are crucial. Because time, for each of us, is finite. Because the universe, cruelly, seems to be infinite, and it’s hard to not to suspect there might be some other, more fruitful game idea lurking just around the corner. And because your job has only just begun once you finish a game. Marketing, promoting, and tweaking await, and threaten to take more time than you spent building the thing in the first place.

And the prospect of spending even more of your finite time beating the drum for a thing that might not even be any good is downright terrifying. One of the biggest challenges for an indie developer isn’t knowing what to do, but when to stop doing it.

So I’ve decided to set Plus Gun aside for now and start something new.

Already I worry my new project might be suffering from the lack of a sufficiently small or defined critical mass. How long will I need to poke through Zimmerman’s “pitch dark space” before I see some light? How much stumbling is enough, how much prototyping is enough, before I know whether this idea is good enough or whether I should pull the plug and move onto the next thing?

Creeping around the edges of this topic is the eternal question: what does “good” mean? Playtesting and prototyping help, to be sure. If “good” simply means “not broken,” dutifully testing your game in progress should get you there. Or does “good” mean “interesting?” Or does it mean popular? And if so, how popular? Among whom? Or does “good” mean profitable?

My friend Charles Pratt, a games researcher at NYU, reminds me to keep my belly-aching in check. If you don’t appreciate how lucky you are to be making games in the first place, and manage to derive satisfaction from that, maybe you should look elsewhere. He’s got a point, I guess. For now, I haven’t had enough.