The Wizards of Infocom

For most people, one of the biggest considerations about a game is "How good are the graphics?" But one company, Infocom, creates games that are all-text -- no pictures whatsoever. Nevertheless, Infocom's incredible adventures have zoomed into the top of the Softsel list, which tracks computer game sales. They actually nailed down half the Top Ten on one chart last summer, a feat that qualifies them as the "Beatles of computer games." We visited Infocom's secluded headquarters north of Boston to find out how they develop the richly detailed "universes" that make their all-text adventures such rarified interactive entertainment.

Marc Blank (Deadline and Zork I) was one of the group's pioneers, along with Stu Galley (The Witness), who had been a physicist until he found computers to be more fun. Steve Meretzky started as a game-tester and has now written his first game, Planetfall. Science fiction novelist Michael Berlyn (Suspended) wrote adventure games such as Cyborg for his own outfit, Sentient Software, before joining forces with one of the most formidable bands of gaming programmers on the planet.

Computer Games: Why do you write computer games without graphics?

Berlyn: Home computer graphics right now can't create a picture good enough to convey the kinds of scenes and situations that we need to convey in these stories. They can't draw a scene as well as we can describe it in a paragraph. And when you read the text, you put yourself in the story.

CG: That doesn't explain why you seem to trash video games in your ads.

Blank: The ads are a satire of the video game ads. They make a point of saying that sure, we like video games, but that there's something else, there's more than joysticks.

Berlyn: Our attitude toward video games is that they're great. The president of our company (Joel Berez) has a video game console in his home. Marc and I go to the arcades all the time.

CG: What kind of games do you play?

Berlyn: Q*Bert, Pac-Man, Ms. Pac-Man, Tempest. We like a lot of different games. The point is that if people think that that's all there is to computer games, they're missing the boat. We're not necessarily better. We're just saying that there is something different out there. A lot of people don't even know about our games, because they end up buying Zaxxon or Frogger or Froglifter or whatever is hot or the salesman knows about.

CG: How did you first get involved in writing prose adventures?

Blank: Everyone who was originally with the company started at MIT, and the original Zork was written at MIT in 1977 on a mainframe, the DEC System 10. It was written in response to the original "adventure" game, which we all loved to play. And that game turned into all of Zork, about two-thirds of Zork II and a little piece of Zork III. This wasn't at all a commercial thing -- the players were people on computer networks around the country: high school and college students, grad students and even some faculty types who had access.

CG: Why were you studying programming to begin with?

Blank: I never studied programming. I was a life sciences major. That would be biology anywhere else, but at MIT nothing's normal. I happened to work for a term at the lab for computer science, doing a research project. Stu, Al Vezza (Infocom chairman), Joel Berez and Dave Lebling, who wrote the original Zork -- these people all came from that lab. And my meeting them and working on that really had nothing to do with my under-graduate career. In fact, no one who's writing games for us is trained as a programmer or a computer scientist.

CG: How many people are on staff now, just programming and designing games?

Blank: Six or seven. They're not all writing, I mean, there are different levels of programming for the games.

Galley: And we have consultants and advisors and directors...

Berlyn: Dave Lebling (Zork II, Starcross and half of Enchanter) functions as a consultant rather than a full-time employee.

CG: What kind of computer do you use?

Blank: All the games are written on our DEC System 20.

CG: And then they're all converted to the other home computers?

Berlyn: We don't do conversions, which is one advantage we have over most software houses. Our games are written in a language that is understood by all the machines. All the machines have an interpreter that understands that language. So there's no conversion.

CG: What is so unique about your "Proprietary programming tools"?

Berlyn: If we told you, they wouldn't be proprietary.

Galley: The issue is one of complexity. It would be impossible to develop programs this complex using BASIC or assembly language, because they don't have enough structure, enough features, to enable you to handle the problems in parts, to be able to break them down and tackle them one at a time. That's what our proprietary language does.

Berlyn: But what's really interesting is that our games can understand complete sentences.

Blank: And it's even deeper than that... the entire level of complexity. I don't think our games are better because of the parser individually, or that they understand more words individually, or that the problems are more complicated individually. I think that it's the whole of it -- and it's all made possible by the language we use. I mean, just imagine if the language that you use, instead of having FORs and NEXTs (BASIC keywords), had operations that dealt with things like people and objects. And it handled that in a very high-level sense, so you didn't have to spend all your time chasing bits around, but could deal in high-level abstractions. And dealing in those terms is much easier, I mean, there's so many things going on in a game like Deadline or Witness or actually any of the new games, Planetfall and Suspended, that you couldn't keep track of them all unless you had a high-level language that was doing it all for you. It would take three to four years to write Deadline in machine language, and then you'd only have it running on one computer.

Berlyn: There's just too much detail to keep track of. I mean, it would be a nightmare.

CG: Is it as much fun to write these games as it is to play them?

Meretzky: I think it's more fun writing them than playing them.

Galley: I'd rather do both.

Blank: The actual writing is a lot of fun. For me it's a lot more fun than playing them.

Berlyn: It's a lot more work to write than it is to play them, but you can enjoy making something happen. Like getting Mrs. Robner to stand up and walk around a room, you know, and walk out the door. That can be very exciting. That kind of excitement can transcend any kind of excitement you can get by playing one of these games. Because it's a different kind of satisfaction.

CG: What's the most fun?

Meretzky: The first time you see someone playing your game.

Berlyn: I agree. That's the most fun.

Blank: It's always a big thrill for me to see it run on a micro as opposed to our mainframe.

CG: When you're first developing a game, do you know exactly what is going to happen from start to finish?

Berlyn: It's closer to writing a novel than a script. You develop characters, you develop an environment, and the problems and the solutions come out of that. But you can't plot a linear path through a complex interactive story. You can certainly plot out a linear path through a puzzle. You may not know exactly when a character's going to get up and walk out of the room and go down to the will reading (in Deadline).

Blank: And the will reading happens when they've all arrived. But there's no single piece of code that does that. Each of the characters is to some extent self-motivated. And even I don't know when they're going to get there.

Berlyn: So because of that, how can you plot that out in fine detail or even linearly? You can't.

Blank: You can decide what the rules are. It's more like setting up rules than saying exactly what happens.

Meretzky: When Mike wrote Suspended, we had no idea whether it was too hard, or even impossible, under the conditions. And after he started playing with it, tweaking the various levels of how bad conditions on the planet began getting and how soon...

Berlyn: Up until it had been played ten or twenty times by our game-testers, we had no idea what the best score was. When I play the game to this day, I can finish it. And I can get a decent score, but I'm from the best score.

Blank: And when you've gotten as good as you can, you can effectively jumble all the parameters and make it even harder on yourself. Like say, "Could I solve this with two robots tied behind my back?"

Meretzky: In fact, the ability to configure the robots was one of the last things that was added.

Galley: We also think up puzzles or scenes, elements that could eventually go into a game, and we keep track of them on the backs of envelopes and so on. And when we get serious about developing an actual product, then we have to decide what the goal is, how we expect a player to get there, what kinds of difficulties, what the whole feeling will be like.

Blank: It's not as if someone sits down at the beginning and says, "This is the way everything fits in, and I think it works better that way."

Berlyn: Nobody can sit down and write one of these as if they were writing a novel. They are too interactive and they are too complex. Take it from someone who's done both. It can't be done.

Blank: And it's because the main character in the story you're writing is not you -- it's the people who play it. And they have to tell you what makes it a better story.

CG: Have you ever written any games that were never released?

Meretzky: We have so many ideas for potential games that we sort of do the weeding out before we start working.

Blank: It's not like everyone here is in a vacuum, where someone submits something and we say, well, that's okay.

Berlyn: That's the critical difference -- we're not publishers. We're not here waiting for someone in a basement to sit down on their home computer and come up with the next Zaxxon look-alike, to send it to us so we have something to publish, so while we're waiting for that kid to write that game we have to publish a bunch of so-so games to stay in business.

Blank: And we have many more good ideas than we know what to do with.

CG: Where does the sense of humor in all the games come from? (Everyone points at everyone else.)

Galley: A lot of the humor is generated because our testers say, "Gee, can I try to 'froo the mumble'?" And they think the program should do something special for that -- it's better than just giving the standard response, "You can't do that."

Blank: Like in Deadline, there's a toilet. Someone's going to say, "Flush the toilet." Or "Look in the toilet" or "Jump in the toilet."

Meretzky: When you say "Look in the toilet," it says, "You've stooped to a new low."

Blank: Or something about the tidy-bowl man.

Meretzky: If you say "Follow the cat" in Witness, it says, "It's sad to see a former great detective now stalking cats."

CG: So a lot of the humor emerges from the game-testing?

Blank: Absolutely. That's a very, very important part of the game.

Berlyn: Some of the best problems are added to the games after they've been played by our testers and recommended by our testers.

CG: What about the effect of interactive prose adventures on literature in general?

Blank: (laughs) I think we'll kill it.

Berlyn: Yeah, it's dead. I think we'll put Hemingway right out of business.

Blank: That's true. In fact, I think he's dead, isn't he?

Berlyn: Is he still dead?

CG: Tell me about Enchanter.

Blank: We sent it off for duplication today. So unless the plane is hijacked, we expect to see it on the shelves soon. It's in the Zorkish tradition. But instead of being the bumbling adventurer, you're a novice who's graduated from Magician's School. You have a spell book. It's a problem-solving game like Zork. It's all magic, it's not underground. And it's intended as the start of a series.

CG: When will Zork IV be out?

Blank: There will not be any more Zorks. This is Zork IV for anyone who really cares.

Berlyn: For anyone who wants Zork IV, this is it. For anyone who doesn't want it to be Zork IV, this is it.

CG: Are there any other types of fiction you'll do in the future?

Berlyn: "Tales of adventure." Our first one, Infidel...

Meretzky: In general these are realistic stories of treasure-seeking...

Blank: In this one there's a pyramid, the next one might be a...

Meretzky: Safari, or a mountain climber...

Blank: A spy thing, anything that would be a present-day thriller.

Berlyn: Yeah, they're thrillers, tales of adventure in which you are a swashbuckling adventurer...

Blank: In different settings. Infidel's the first of a series. We're also going to have some games aimed at kids. It's early to talk about that, but we're interested.

CG: What about the future of computer games in general, say five years from now?

Blank: Spreadsheet wars.

Berlyn: They're all going to be made into spinach salad and eventually we'll eat them all.

Meretzky: There'll be a lot more variety, there'll be kinds of games that you can't even imagine.

Berlyn: I can't imagine that. But we are working on the future.

Blank: Yes, we're building one in our backyard.

Left to right: Stu Galley, Marc Blank, Steve Meretzky and Michael Berlyn in Infocom's underground computer complex. All their games are written on the massive mainframe behind them.

Thanks to André St-Aubin for transcribing and donating this article.