The "Hard Boiled" School of Software

Eric Grevstad

You might pick up an Infocom game for its distinctive packaging — Starcross comes in a plastic flying saucer; The Witness, a 1930s private-eye thriller, includes a newspaper and matchbook as well as a floppy disk. But the Cambridge. Massachusetts, company is best known for unparalleled programming. And for publishing games that use absolutely no graphics, just text.

Zork, written by MIT graduates Marc Blank and Dave Lebling, revolutionized the adventure-game genre by letting players use complex sentences instead of two-word commands. Other games that use Zork's 600-word vocabulary and sophisticated parser (the part of an adventure game program that interprets input), such as the interactive mystery Deadline and sci-fi comedy Planetfall, have made Infocom the leading adventure publisher, and a more than $6-million-per-year business enterprise.

FAMILY COMPUTING went to Infocom's offices to interview Blank, now Infocom's vice president for product development, and Michael Berlyn. author of several novels and Infocom's futuristic smash, Suspended. The firm had just released its ninth and tenth games — Enchanter, the first in a trilogy of fantasies in the Zork tradition, and Berlyn's Infidel, a tale of drama set in the Egyptian desert.

FC: How did Zork come about? I know Colossal Caves, Will Crowthers and Dan Woods' original adventure, was sweeping through computer labs in the mid-'70s. Did it sweep through MIT's?

Blank: Yes. Colossal Caves was the only adventure around. So after we played it and liked it a lot, we asked. "Why don't we do some thing like it?" So Zork was first written at MIT on a big mainframe computer.

The old Zork was very small, with just a two-word parser. We realized that was really very limiting, so we expanded the parser [so it could understand longer sentences] and invented new problems. Eventually Zork, between 1977 and 1979, got really, really large. It was just a hodgepodge of different sorts of things; when ever anyone had a good idea for a problem, we'd add it.

When we put the games on micros, especially Zork II and III, we tried to give them a plot, which the original didn't have at all.

FC: So now plot comes before problem.

Blank: That's exactly right. There's a plot, and then we say, "Well, what are the problems going to be?" In Deadline, for instance, we said, "Okay, the story is this, and this guy's dead, and these people are going to do this and that. Now, what's the game, what are the pieces of evidence and what can you do?" With Suspended, I think the characters came even before the plot. You had these different robots with different abilities....

Berlyn: The setting, the underground complex...

Blank: And then the question was: What are they doing?

Berlyn: Always a good question.

Blank: But you can see that the newer games are really all much better thought out. People that play them, I'm sure, will notice that they hold together much better. And it's because, even though they're changed a lot after the initial design, they're designed at one time — as opposed to over years by a number of different people with different interests.

FC: How does the design process work? What's the first step?

Blank: Someone who's finished with one project says, "I have an idea for this story." If it's something that we all think sounds reasonable, then they can come up with a synopsis and see if we like that; then they can go off and start. And as they're doing it, there's a lot of feedback from the other authors.

FC: There seems to be as much interaction here in the office as in the games. What's it like to work here? Is it a bunch of off-the-wall loonies playing around the water cooler?

Berlyn: For the most part, yes.

Blank: I think we're actually a bunch of relatively well-behaved loonies. I mean, we're loony enough to do what we do, but not so loony that we don't get things done. We understand that we're in a business.

Berlyn: Were professional loonies.

FC: I'm getting the impression of fun-loving game writers who are also serious business men. Who's your competition?

Berlyn: Hemingway?

Blank: Actually, we're our own competition at the moment, because we're internally driven to do something better or new each time.

"AGATHA CHRISTIE WROTE ALL THE MYSTERIES THAT ANYBODY NEEDED TO WRITE."

Berlyn: All the loonies who write the games are easily bored, and no one wants to sit down and recreate what's already been created.

FC: People say the market's changing. Who's your audience? Is it male, female, professional, young, old?

Blank: Whenever anyone docs a survey, you can look at it and say, "Oh, we have a very limited audience — young, college-educated, male, with incomes over $40.000." Then you look at the type of people who own computers and they're the same. You're really not getting any extra information: the question is. who owns computers?

Berlyn: We get letters and warranty cards from 10-year-olds, 50-year-olds, male, female, married, single. There's no way for us to say that no women like our games, no children can play our games....

Blank: Actually, it looks like a relatively high percentage of women play our games. The male/female ratio of computer owners is some thing like 10 to 1 or 15 to 1. For our games, it's more like 6 to 1.

FC: That brings up another point. You've been careful lo have generic or nonsexist characters — the role you play in Deadline, for instance, is identified only as "The Inspector." But now, with Infidel, the protagonist's not only a man, but an obnoxious, unscrupulous one.

Blank: We didn't really have a choice on that, although all things being equal we wouldn't have wanted to do it.

Berlyn: The character had to be drawn the way it was in order for the story to work properly. We discussed at great length whether or not having a male character would interfere with any female player. The thing we kept coming back to is that women watch TV shows where men are the main characters, they read books where men are the main characters, and men do the opposite.

FC: What do you think your customers want? What are your most and least successful games?

Blank: Zork and Deadline sell well. They're often the first games like this [text adventures] people play. Starcross doesn't sell as well as Zork; is that because of the package, or because it's science fiction?

Berlyn: Or because of the weasel aliens? There's no way of knowing.

Blank: But we hear the same sorts of things from people that like all the games. They like the interactive nature of it. they like the stories and the puzzles, they like the attention to detail.

But, it's like owning a bookstore — you don't take science fiction off the shelves because people buy other kinds of books.

"PLANETFALL MADE SOMEONE WHO WAS PLAYING IT CRY. DID ZORK DO THAT?"

FC: One of your new games, Enchanter, is a fantasy like Zork, but one in which you use magic and cast spells. Is it fair to say you're taking on Wizardry (the role-playing adventure from Sir-tech Software]?

Berlyn: Not in any way, shape, or form. It's nothing like Wizardry. They don't look alike, they don't act alike...

Blank: Wizardry is for multiple players, multiple characters, the role-playing thing. It's mostly fighting and gathering treasures. It's similar because they're both fantasies.

Berlyn: They're similar in the way that Pac-Man is similar to Zork. They're both games. And they're a little more similar in that they're both fantasies. But when you're a publisher of this kind of fiction, do you say "We shouldn't publish a fantasy, especially if there's spells in it. because that's been done in Wizardry?" We certainly don't. If we did, we'd never have done Deadline or Witness. Agatha Christie wrote all the mysteries that anybody needed to write.

Blank: That's true. It's a different view of magic, too, I think. We're trying to build a progression. At the end of Zork III you become a magic user of sorts, so we thought it'd be nice to change the theme a little. So instead of being an adventurer, who just wandered into this crazy place, you've just graduated from magicians' school. And in Sorcerer, the sequel to Enchanter, you're a member of the circle of enchanters, and something's happened to the leader, and you go off and look for him.

FC: You've stressed the progression from game to game within a series, but $50 is a lot to pay for a game you play once.

Blank: If you can play the game only once, but play it for 30 hours and enjoy it, that's better than a game you play for two hours and then get sick of.

Berlyn: Books are expensive. How often do you reread a book? Also, some of the games you can play again and try to do better.

Blank: My guess is the average person who finishes a game has seen about half of what's there.

FC: That brings up the question of detail. You can read all the bottles in the medicine cabinet in Deadline; you can jump on the bed and it says WHEE! How do you second-guess people as to what they'll try to do?

Blank: Mostly, we write for ourselves, but we do spend a lot of time with testers seeing what they do. We definitely try to anticipate things: if there's a bathroom, you say, "Hey, someone's going to try to flush the toilet." That's more a frill, not really important to the story. But it's important to making the world we're creating a little more real.

Berlyn: You begin Infidel in your army cot in your tent; you can fold up the army cot and carry it around with you. That came as a result of a tester saying. "It's an army cot. Army cots can be folded up." Now, some people say they want to be able to strip the cot of the canvas, boil it in water, and eat the canvas. But that's irrelevant to the story and unreal.

Blank: But something like waving at the plane may be relevant to the story, because you may want to be rescued.

FC: Even though it just dips its wings in reply and flies off.

Berlyn: Oh, you tried that?

FC: Some graphic adventures get more detail by using multiple disks. Would you consider doing that?

Blank: We'll have larger games. If in the future we're trying to do something ambitious, the "next thing," we may need to go to multiple disks. That's something that we're working on.

Berlyn: But it's not going to be a Zork with a glandular problem. We won't make a game bigger for the sake of making it bigger.

Blank: The larger the game is, the more it has to go to disk, the more disks you have to swap — the more tedious the experience.

Berlyn: You're not going to want to play an exciting, interactive game where your next move may be life or death, and have to wait 15 seconds for swapping disks.

Blank: The idea is to make the computer invisible. You don't want to know you're dealing with a computer. You want to be interacting with the story, and anything that gets in the way, having to swap disks or communicate in two words, detracts from the experience. The full-sentence parser doesn't improve the story; all it does is make it easier for people to get through to the story and interact with it. It's just a device, like a nice keyboard as opposed to a crummy keyboard.

FC: Here's a hostile question. I think Zork and Deadline and Suspended were like nothing anyone had ever seen before, great leaps or innovations, and I'm not sure I get that feeling from your new games. You have another trilogy like Zork [Enchanter], another mystery [The Witness], another science-fiction game [Planetfall], and so on. Aren't you playing it a little more conservatively?

Berlyn: Not at all. Infidel has things in it that no other game has ever had.

Blank: It's got an ending that...

Berlyn: Don't mention the ending.

Blank: Well, I'll say something about it. It's the only game I know of where the ending is not a happy one, and in a way, you should have expected that from who you were, from your character. I think its really revolutionary, because when people are done with it, they're going to think a lot about what's been happening. That's a little unusual. You know, it's not like you've gotten the last treasure and won the game.

Berlyn: It's the first game that's ever drawn the character clearly and in great detail, and then set that character out on a mission, and had an ending that is more consistent with literary conventions and reality than with a game. And for that reason it's less a game and more a piece of fiction. So I disagree that Infidel doesn't make any breakthroughs.

And look at Planetfall. Planetfall is the first story where I saw someone who was playing it cry. There are people who are really touched by that game. Did Zork I do that? And there are other things in Planetfall, too. Doing different things takes different amounts of time.

Blank: Which we were playing with. I mean, that's an experiment — the fact that taking something is quicker than walking from one room to the next, and that walking down a long corridor takes more time than walking down a short one.

Berlyn: And there are fewer suspects in The Witness because there's more that you can talk to them about. More interactions with the characters are possible. Those kinds of things may not seem as radically different as Zork did. but there are still minor breakthroughs with every game.

FC: Are there any major breakthroughs left?

Blank: I think in a year we're going to see something that is to Deadline what Deadline was to Zork.

"THE LARGER THE GAME IS, THE MORE IT HAS TO GO TO DISK, THE MORE TEDIOUS THE EXPERIENCE."

Berlyn: The new game changes the type of game; it doesn't just improve on an existing genre. What Zork was was Adventure, done as full-blown as it could be done — a full-sentence parser, and a big vocabulary. And Deadline changed what detective games were like — with characters that move around independently, whose lives you can affect, with time flowing through the story and with things happening whether you're there or not. The new game will be out early next year. Our kids' game [due in May] will also be very different.

FC: Time magazine featured you in the December 5 [1983] issue, and said a lot of nice things about adventures and Zork I's being the best-selling piece of recreational software [sales have now topped 250.000 copies]. Then they said, "By literary standards, Infocom's stories are crude. The characters are two-dimensional, plots are forever clunking to a halt, and the writing tends to be sophomoric."

Berlyn: That's very true. By "literary stan dards," 99 percent of the fiction published in the United States is crude, it's sophomoric. I won't argue with the Time quote. It's entirely accurate — and very unfair. Unfair, in that to be compared to "literature" at this point shows their total lack of understanding.

Blank: It's flattering that they think our work is close enough to literature to say that. I mean. I don't think they would bother comparing Wizard and the Princess to literature.

Berlyn: We're not upset by it.

Blank: We're going to put stickers on our packages that say "Crude! Clumsy!" — Time magazine.

---

---

---

These historical, out-of-print articles and literary works have been GNUSTOed onto InvisiClues.org for academic and research purposes.