Last week, we started posting a short series of interview transcripts with Raph Koster of Areae, taken from an epic chat conducted at the beginning of GDC07 in March. Today, we're proud to conclude the series with the final transcript, covering academic gaming programs, prototyping, tools, inspiration and moneyhats.

If you missed part one or part two, you may want to start there first.

f13.net: You mentioned academia earlier, briefly. What's your relationship with the academic community these days?

Raph: It's good? (laughs)

f13.net: I mean, are you working closely to create curricula of game design...

Raph: No, I'm not doing anything like that. The book is on a bunch of curricula of game design already. I try to keep a little list, keeping track. It's in use... USC, UCSD, Georgia Tech, CMU, a lot of the game design programs use it, which is really cool. I still speak occasionally at academic conferences and things like that, just to stay in touch with what's going on on that side, but I don't work closely with any one. I'm an advisor to the program at Indiana. I do jack shit for them, honestly... (laughs)

f13.net: What do you think of those programs? Are they valuable things that we should be fostering, or...

Raph: They're all different, actually. There is no unified curriculum yet. I would say that some of them are very artsy, and some of them are too practical. So, some of the artsy ones really treat games as art objects and that kind of thing. I think the industry sometimes loses patience with those, because they're like, well, "Where's the beef? How do we hire somebody out out of this?" And all the way at the other end are the very practical ones, where all they teach is level building, and it's like... do you actually learn how to do game design from that or come up with anything innovative? And there's a whole bunch of programs that, I think, are successfully charting the middle ground, and some of the game design stuff is actually really, really good. I think the best ones are the ones where you actually have to make games. CMU does that, USC does that.

f13.net: Until we determine the notation, as you were talking about, it's still a very experiential thing, isn't it?

Raph: Yeah. USC has not one but two good game programs. One in engineering and one over on the media side. Actually, I think they have more than two programs, I want to say they have more like four. But, anyway, it's an interesting thing. It's grown by leaps and bounds just in the last few years. And not just in quantity, but the quality has grown by leaps and bounds in the last few years. So I think it's pretty exciting. We've certainly put out calls for interns through those programs.

f13.net: Okay... I had one more question and I'm completely blanking on what it was.

Raph: Moneyhats!

f13.net: Yes, I'm still have my moneyhat, of course. It's kinda my trademark by now.

Raph: What your favorite drink is this year...

f13.net: Still the Black Russian?

Raph: Still the same, I still rely on the Black Russian. My fallback drink.

f13.net: Oh! Okay, I remember what it was now. Sorry. I just want to get an idea of what your work process is when you're creating a new game. Do you start out with a spreadsheet, look at numbers, on paper? Do you go out in the park and stare at the birds for half an hour? It is more of an epiphenomenal thing, more mechanical?

Raph: You like bigger words than I do, that's weird... Um, this is one of those weird things. "Where do you get your ideas? Schenectady." For me, anyway, particularly since the book and the grammar and stuff, I see games almost like lattices, or graphs, or networks, or crystals or something. I remember when we were going through SWG, in my head, I had this picture of a ball, this cool geodesic dome or something, buckyball or whatever. Little pieces connecting and so on. We'd cut a feature and I'd be like, all right, we've removed that, but it still looks like a sphere, so it's okay.

f13.net: Since it was SWG, was it a Death Star?

Raph: (laughs) In my head, I have these... I picture all the moving parts, and to me, they hang together in that kind of way. And certainly walking through stuff like grammar and whatnot pushes you to think even more that way. When I come up with an idea for a game, it's often... Sometimes it's from messing around with prototypes, sometimes something jumps out at me, sometimes it's... Often it's just an image. Last year, I wanted to make games that felt like a kaleidoscope, or I wanted to make a game about flapping, or whatever. There's one I've been wanting to do for a while now involving mice and cats and cheese, where you're trying to guide a mouse from one end to the other through a maze full of cats, and all you can control is the cheese. So, stuff like that. It's often that kind of thing. I don't usually come to it wanting to tell a story...

f13.net: You usually want to convey an experience, almost?

Raph: Yeah, yeah. It's an experience. It's funny because I get yelled at by people like Abalieno, HRose, whatever. Innsmouth, how many more aliases can we come up with...

f13.net: The crazy Italian guy.

Raph: Yeah, because he... he says, "The problem is, you don't approach it from the experience point of view." I think I actually do. But to me, the experience and the underlying model are very closely tied together, and so, I'll have the idea for one, or the other, and it really doesn't matter because the other one follows within like thirty seconds. Which one I start with is kind of a... (shrugs) But my process after that is, I try a prototype. And I'll try it with cards, with glass beads, with a board. I'll often try it that way before doing any code. If I can't do it that way, if it needs code... Like, a lot of puzzle games I can do that way, the bird thing I couldn't. So I'll just throw together some code, but I can do that in an hour. So I'll do that...

f13.net: You prototype primarily in Flash then?

Raph: No, actually, I use BlitzMax. And I do that because it's cross-platform, Linux, PC and Mac, so I can give it to my wife, she uses a Mac. It gives me access to... basically, it's 2D and 3D OpenGL, so I can do stuff like alpha and rotating and all that, which makes prototyping much easier. It still runs plenty fast. But it's a basic syntax and I don't have to deal with all the goddamn overhead of Windows libraries and all that crap. Because, honestly, for the last few years, that's actually been my barrier. C is not the barrier, I can do C and C++. Visual Studio is my barrier.

f13.net: Well, the hardest part in any real programming these days is dealing with the platform, the libraries...

Raph: It's the platform, the OS, and the libraries. I'm like, how do I open a window, just give me a window, oh...

f13.net: "Give me a canvas and let me paint."

Raph: Let me paint, exactly. It's really frustrating. Rod Humble, actually, is the guy who turned me on to doing stuff with that, so I've been really enjoying that, and we're actually using that in the office. I'll prototype it in that. We know that the whole prototype is not going to be final code at all and we're not going to ship any BlitzMax, but...

f13.net: You're willing to "make one to throw away".

Raph: Oh yeah, we actually make five or six to throw away, right?

f13.net: That's the point of a rapid prototyping tool.

Raph: Absolutely. It's a lot of fun to do it that way. I keep thinking I should go learn Flash, but I keep getting scared off by the fact that, apparently, it's a completely different metaphor, with the whole timeline and stuff. I'm used to the old school, because that's how I learned to do it.

f13.net: Object-oriented, graphics...

Raph: Older than that, I'm a procedural guy. I learned to program in the original BASICs with line numbers! So I'm used to GOSUB. (chuckles)

f13.net: I'm sure your engineering staff loves that.

Raph: Oh no, there's no GOSUBs any more, there's no... But I'm more of a functional programmer than I am an object-oriented programmer. I've found that I use OO only for... Aw, let me give you an example. If I go make a shoot-em-up, I'll absolutely use types for all of the enemies and the bullets and the explosions and the particles. I probably wouldn't for the player, because I'm like, there's only one of him! And I definitely wouldn't for the screen! In full OO, you'd do it for the screen, you'd do it for everything.

f13.net: Give the score an object!

Raph: Yeah, I guess I'm just not really there. I'm like, if it's only one number, then what the fuck? Just make it a AddScore(). So yeah, that makes me sloppy and bad, but whatever, I don't care, because it's a prototype! As long as we understand what's going on... It's too easy to get hung up on the methodology. The point is to get the fun on the screen.

f13.net: So do you use your family and your friends and your blog a lot to test your prototypes out, or ideas that you've had?

Raph: I haven't used the blog that much. The only stuff that I've posted to that is actually the bird thing. I use family and friends. Make a game to give away to them for Christmas. A few years ago, before there was XBox Live... So, in college, it was a Mac campus, that meant we played Crystal Quest, a LOT! And, I got a hankering a few years ago, like three years ago, I wanted to play Crystal Quest, dammit! So I went looking and well, there was a PocketPC version like four years ago, and that was it.

f13.net: So you made your own?

Raph: So I made my own, and I gave it to people for Christmas. And then, of course, Live launched a year later with Crystal Quest on it. Not the one I have on my laptop, because, you know, you can't get the Live one on your laptop. So it's just one of those things, you miss it. And that one... that was the first game I made when I decided "I want to get back into doing this". So that was my learn-BlitzMax game, was cloning Crystal Quest...

f13.net: So, XBox Live, especially here at GDC this year, they're pushing XNA and TorqueX and their rapid development, rapid prototyping tools. Have you looked at those? What do you think of them?

Raph: I haven't, I keep meaning to, because I'm pretty tool-agnostic, you know, so I was awfully tempted. Oh, let's go port Andean Bird over to XNA and C# and see! Because it'd be awfully fun to play it on an XBox, that'd be cool. But then I go, yeah, if I what I really want is to try the triggers, I'll just plug in the controller to the PC. But, learning a whole new language just to prototype the triggers...

f13.net: It'd be a good learning experience.

Raph: Yeah, it'd be a good excuse. I think it's really cool, I still think it seems... I don't know if it's aimed broadly enough yet. The fact that you have to pay to consume, not just to produce...

f13.net: Yeah, that's the biggest hangup that people seem to have, to pay $99 a year just to play other people's stuff that Microsoft isn't even making for you, or even to play your own stuff on your XBox.

Raph: Yeah, and you compare that to the Youtube For Games stuff that's out there now. There's so much, where you just go and you play stuff. And it's like, well, I'm not going to give Microsoft more money for indie titles when indie titles are practically trying to find you all the time, yeah. So it's kind of a weird tradeoff. I think they'd have to open it more to get that kind of thing. A lot of the Youtube For Games projects out there are just awesome, and there's some fantastic stuff out there.

f13.net: What's your favorite one?

Raph: I like Kongregate, but there's so many! There's just tons of 'em.

f13.net: All right, well, I know I have to get back to work, I don't know if you have to do something right now, but...

Raph: Well, I need to finish my slides...

f13.net: Thank you Raph! I appreciate it.

Raph: Sure!
[discuss]