A Chat with Chris Crawford

Author and Screenshots: Mathew Tschirgi
Editing: Cecil Casey
Game Packaging Scans: Bill Loguidice
Online Layout: Mathew Tschirgi and David Torre

Chris Crawford may very well be one of the best game designers you've never heard of. He started working in the game design industry for Atari in 1979 and continued until 1984 when he switched over to computer game design. Many of the games he designed were ahead of their time. Balance of the Planet (1990, DOS) was the first environmental simulation game, managing to both illuminate and entertain players at the same time. Trust and Betrayal: The Legacy of Siboot (1987, DOS) managed to convey a sense of paranoia and empathy through dialogues that primarily consisted of just icons.

Trust and Betrayal Front Cover - A pair of cat-like eyes sit over a outer space background
Trust and Betrayal: The Legacy of Siboot Packaging (Front)
Trust and Betrayal Back Cover - Two small screenshots over a descri<i />ption of the game and its features
Trust and Betrayal: The Legacy of Siboot Packaging (Back)


Impressed by the dialogue system in Trust and Betrayal, I was amazed at how similar it was conveyed to the dialogue system in Will Wright's The Sims (2000, Windows). Both games featured dialogue that was only represented via icons. I asked Crawford if he thought games in the future would reach a point where they focus more on character relationships with Non-Playable Characters over graphical interfaces.

Trust and Betrayal - Dialogue system, featuring several icons and a face outline showing emotional expres<i />sion
Trust and Betrayal - Dialogue system

"I'm sure of it; my own work has demonstrated the feasibility of this. It's no longer a question of hardware or software; it's just a matter of putting the money together and building products. The stuff is complicated and completely different in style from regular games. For example, if you want to do personal relationships, you have to stop thinking in terms of spatial relationships. You don't need maps in a game about relationships; you need data structures that measure and compute emotional relationships, not physical ones," Crawford said.

Crawford certainly has a point. The Sims, for instance, is about relationships in a sense because you have to make your avatar maintain as many relationships as possible with other "Sims" (computer controlled avatars) in order to get hired at better jobs. However, you never quite empathize with the relationships your avatar gets into; they simply speak gibberish to each other while nonsensical graphics pop up in comic-book style word bubbles.

The Sims - A dialogue between two characters
The Sims

Your avatar might hit on the Sim next door, but the Sim never chats about how unromantic her husband was when they went on a date last night. The abstraction of conversation in The Sims allows players to imagine whatever their avatar is talking about, but it also prevents players from truly connecting with their avatar.

Taking care of your Sim is like having a goldfish for a pet-you feed it every once in a while, but don't get any emotional response from it. Conversations are so abstracted that a true connection with other Sims is next to impossible, essentially making the game a spruced-up Tamagotchi.

Crawford's own Trust and Betrayal offers a more intriguing series of character interaction. Set on the moon of Kira, your avatar is an acolyte who is trying to gain a perfect set of auras to become the next Shepherd. To win the game, your avatar must defeat various computer controlled acolytes in mental combat. The majority of gameplay is spent chatting with fellow acolytes, trying to figure out which ones are friends and which ones are foes.

Trust and Betrayal - An encounter with an acolyte
Trust and Betrayal

After a certain number of moves, the mental combat portion comes into play.
The gameplay creates such a sense of paranoia in the player because they don't know which acolytes to trust. Every once in a while there is a random event which fleshes out the personalities of an avatar while offering the player a choice of how their avatar should react to a situation. While the game play takes a bit of getting used to, its unique method of allowing players to have their avatar interact with other acolytes on such a close level allows for a kind of emotional attachment to Non-Player Characters that is unusual for a computer game.

When asked for an example of a game that has come close to being a commercial interactive story project, Crawford points me to Michael Mateas and Andrew Stern's Facade, which has just been released online as a free download via a 800MB torrent. Crawford described an early version of it as, "It's a working interactive storytelling system, which is better than anybody else has achieved."

Facade has received raves from a variety of sources, including MSNBC and Gamespot.

With Crawford having been a veteran of the gaming industry, I decided to ask him for his thoughts on the whole "EAspouse controversy," in which the wife of an Electronic Arts employee posted a lengthy blog entry describing the work conditions her husband faced; this blog entry has influenced the International Game Developers Association to form a Quality of Life committee.

I was a bit surprised to learn that long hours weren't always mandated at gaming companies--or at least at Atari, where Crawford used to work.

"I worked hard on my projects, but only because I wanted to. Nobody ever pressured me to rush the job. Indeed, all through the 80s, the idea of crunch mode was seen with some distaste.

In the first place, crunch mode generates tons of mistakes; when you overwork people, they start to make mistakes. There's a reason why airline pilots are required to meet strict standards for sleep before a flight.

In the second place, crunch mode always demonstrates poor management. A manager should be able to schedule the project so as to bring everything in on deadline and under budget. Any manager who can't bring his projects in on schedule and under budget should be fired. In the games biz, however, crunch time is used as a means of exploiting the youngsters who agree to work at the company.

The system is rather like that used by armies. You maintain a corps of officers who run things, and add a thick layer of enlisted cannon fodder. The whole army is basically a plumbing system that funnels cannon fodder to the front, expends it, and replaces it with more. In the same way, the games biz maintains a cadre of old pros who manage the eager young fools willing to endure this nonsense for a few years. They come in, you work them like slaves until they drop out, then you open the outside door and crack and say, 'We'll take another employee now' and the luck guy who manages to push his way through the crowd elbows in and congratulates himself on his great good fortune," Crawford said.

Probably one of the main reasons for so many companies insisting on longer and longer periods of crunch time is the need to release a game as quickly as possible. This especially applies to sequels and licensed movie spin-offs, where a game title has to come out at the same time (or slightly before) the release of the film itself.

Let's say a gaming company works on a game for a few years called "Deadly Explosions". Much to their surprise, it's a breakout hit, forcing them to come out with a sequel, "Deadly Explosions 2: The More the Merrier!", under a tighter deadline. The longer it takes for a sequel to be released, the greater the potential is for the consumer to forget about the franchise in the first place. Ironically, the quicker a game is developed, the buggier it usually ends up being.

While it's ultimately better to see more original games in the first place, rather than seeing the usual cookie-cutter sequels, it's a hard fact that most of the more successful game companies are based on franchises. If the schedule for developing a game was increased by even six months, perhaps the final quality of the game itself would greatly improved due to extra development time. But these companies feel that this extra development cost would adversly impact their bottom line.

I decided to round out the interview with asking Crawford what he thought of the various level editors that came with recent games as a way for people to learn game design.

"I suppose that a rank beginner would learn something by playing with a level editor, but this just scratches the surface. The real heart of game design lies in designing systems of algorithms.

Those algorithms have to be expressed through a programming language, but the important part is the algorithms. There's no reason why a beginner shouldn't work in Java--it's clearly the language of the future. The performance hit that Java imposes is no longer a deal-killer, and there's nothing better for publicity than putting your work up on the web," Crawford said.

By creating games that try to innovate instead of emulate, Chris Crawford has proven himself as a designer who has been and remains ahead of his time. While this interview has just supplied a peek into his thoughts on current issues in the gaming industry, those who wish to find out more should check out Chris's official web site, Erasmatazz, which features plenty of archived articles covering a variety of topics.