Processing my process. Or: How to get better at game design.

I’ve sent my alpha version of Strangers out to some folks for feedback.  Most of the feedback boiled down to: ‘This is kinda complicated’ and ‘Where’s the fun?’

After getting the third return with the same feedback, I started to do some serious introspection.  While I don’t talk about it here on my gaming blog (heck, I don’t talk about much here at the moment) – I also do web design and programming.  Over the last few months, I’ve been lucky enough to gather a nice stable of clients and steady coding work.  It’s taken up a lot of my ‘free’ time, which is why I haven’t had much to say here.

Since I’ve been programming a lot more lately, I’ve started really honing my process for development, at least when it comes to web projects.  What I’ve discovered is that I’m: iterative, atomic and utile.  I’m no rockstar when it comes to coding – in fact, my programming blog is called ‘Groping for Code’, which it sometimes feels like I’m doing – looking at an end result that I want to achieve, and slowly building a scaffold to get me there, refining it along the way.

So:

Iterative: I make a lot of small changes, seeing what the result is along the way, and usually breaking shit.

Atomic: I work on one feature at a time, making it work, then integrating it with the whole.

Utile: I look for the best, cleanest way to do what I want to do, regardless of any sort of overarching programming principle.

I realize these things may even hold me back from being a good coder, but I’m working on improving – and more importantly, they’re working.  I’m getting things done.  Projects are moving.

So let’s contrast this with my game design process.  First of all, I don’t really HAVE a process, as such.  I just sort of jump into a design with both feet and see what happens.  Part of this is an artifact of how my completed (or mostly completed) projects to date have come about – they’re products of things like 24-hour RPG challenges or Game Chef, which imposes either a one- or two-week deadline.  Soup to nuts, make everything at least look like it hangs together.

But it’s not entirely that.  I have a sense of myself as ‘game designer’, so to some extent I’ve pre-judged the end result of my efforts: it will work, because this is hat I’m passionate about.  My designs tend to be all-or-nothing thunderclaps of effort.  This is just about as opposite an approach to process as you can get from my programming.  

I’m not iterative: Any changes that I make, I propagate them through the design without testing their viability, or the relative merits of the original version.

I’m not atomic: Games come out of me whole-cloth.  Any one section of the rules may or may not stand on its own.

Utile: I work from more formalist principles (the desire to create games that push certain envelopes, like integrating board game components, or exploring some facet of play) as opposed to looking for what works, then throwing mechanics at it until it sticks.

And the results are perfectly obvious: my programming gets done.  My games don’t.  So the question now is: how do I apply my programming process to my game design?

Iterative: Test, test, test.  Start with a framework for the game, then start hanging components on it, testing each component as I go.  If it doesn’t work, figure out what’s wrong and change it.  Question everything: is the form factor wrong?  Is the theme wrong? Is the approach wrong? IS IT FUN?

Atomic: Work with the minigame model – write up  mechanics for each aspect of a game, keeping in mind the overall shape of the project.  Test each minigame, then integrate and test again.  Obviously, some sort of testing mechanism is going to be crucial, here.

Utile: If principles are getting in the way, eject the principles for now.  If the principle is important, integrate it into the process at a lower level – does this facet support the principle AND the game, or just the principle alone?

Obviously, there are implementation problems that are independent of the process problem.  It’s a lot easier to iterate my web projects when all I have to do is alter a few lines of code and sync my localhost web directory.  But I like the shape of this, and I should invest brain cycles in trying to figure out a) how to make it work, and b) applying it to Strangers and Terrible World.