Logistically, did you have to go literally word-by-word...

JS: To find the words? Yes.

I guess you had to put them into groups? When I wrote "gouda", I just got a Swiss cheese wedge. Did you do something like, "Okay, all these words are under cheese."

JS: Yeah, definitely. There are definitely synonyms. It just depends how important it is. Is gouda a big thing? What you're going to get back in terms of the art, it's still going to be cheese. Gouda is the same thing. Maybe like Limburger is slightly different. It makes people run away because they're afraid of it because it's stinky.

Other than that, if it's really basic, like a box and a crate, it's the same thing. People aren't going to go, "Where's my box? Where's my crate?" That doesn't make sense. When things really make sense, we have android, robot, and cyborg. We thought those were different enough so we made them different. We made sure that they're all completely different, acted different, look different. It really just depends.

Obviously, we have a limited development schedule, so you can only fit so much in. We just try to make sure that everything... We have like brown bear, polar bear, and black bear. They're different bears. They look different, and they act different.

How many different behaviors do you have?

JS: It depends. Here's another thing that's going to sound PR-y and like fake but it's true -- we had a QA plan that was set up during the middle of development, and we were wondering how we were going to QA this. We found we can't. It's not possible. No human can ever interact with everything. It's not possible. We just kind of hope that it all works and that it doesn't crash and that it doesn't break.

We just make the system and make sure it works. Then we check every object and say, "Oh, does it work?" We have no idea. If you freeze your airplane, take it back in time on a time machine, put an old man on it, come back, and set it on fire, what's going to happen? I don't know. You can't test that. How do you test that? It's impossible.

Do you have to do that stuff all on a case-by-case basis, or is there stuff where with brown bear or black bear, or brown cat and black cat, we're just going to do palette swaps on these or something like that?

JS: Every object has been tweaked. There's a system that takes the hierarchy of everything and says, "Okay, where does this go? Where does that go?" And then we tweak them. It fills out the basic. This is AI, so we know that an AI is going to walk around. And then you can insert what is it afraid of, what it likes, how many hit points it has, can it swim, will it fly, it will drown, does it like fire, does it hate fire, is it going to die in fire? All these things you can tweak, and then you can get really, really nitty-gritty with everything.

So, those are all tweaked by hand.

JS: Yes, yes.

A lot of projects are programmer-led or artist-led. There's a lot of art in this game, but it seems in a way that this game requires almost as many designers as artists to tweak that stuff.

JS: Yeah. I think our company is very designer-led. If you look at our games, all through our games, Drawn to Life, Lock's Quest, Drawn to Life 2 even, and Scribblenauts -- Drawn to Life 2 not as much -- but they're all very, very different. They're all on the same platform, that's about it. That's the only thing that's the same between them. Everything else is different.

Lock's Quest is isometric, completely revamped everything. It's normal to program things like, "Oh, we're going to make an FPS engine, and then we're going to make a boring FPS with a boring story and random enemies." Who cares about that stuff? We're just like, "What's cool? What can we do? What's the next great idea? Let's do it. I don't know how we're going to do it. We'll figure it out as we go."

That's Marius's job. He's really on board, and so are all the programmers. They're really on board with following the design and saying, "Look, we're doing something completely different. We're not taking a racing engine and then making another racing game, and then taking that and making the sequel to that racing game." All our games are completely different from each other. There are not many companies that do that, to be honest. That's kind of an indie spirit. A lot of independent developers will just be like, "I want to make this. This is cool. I want to make that. That's cool."