The Experimental Gameplay Workshop

Inspiration

The EGW was created to support the idea that there's a group of developers who strive for originality, and who intentionally try to create new kinds of gameplay.

What exactly this means be a difficult thing to talk about - gameplay is a vague and elusive thing. It's easier to show gameplay than to talk about it. That's what the EGW does, once a year, at the GDC: show new gameplay.

We know we're doing something right. Both previous years we've held the EGW, the game development community has responded very enthusiastically. Not only to the cool games that are being shown, but to the very idea that new things are being attempted. Attendees leave the session in a pumped-up state, excited to see that the industry is not just about rehashing old stuff and thinking about what experimental things they can do.

Background

Through the 1990s, as game budgets grew, publishers became more risk-averse. Correspondingly, publishers placed more pressure on developers to create games that were very similar to prior successes. Too much originality became deemed as risky. Now, the publishers aren't completely anti-originality, they know that too little originality is also a formula for failure. So they seek to inject some originality into things like character, story, and setting, which are relatively un-risky areas. Messing with core gameplay is extremely risky, so they don't want to do that. They do want to improve the gameplay of existing games - they're not trying to make bad games, after all. But they want to do that by making small, incremental changes to established successes.

By the year 2000 or so, this effect had become quite obtrusive to developers and many of us in the industry wanted to see more originality in gameplay. It's the publishers' job to be risk-averse, and that's something of a law of nature, as money is a large and powerful force. So the idea of the Experimental Gameplay Workshop is: rather than trying to change the mainstream industry, perhaps we can accept it, yet augment it to create a richer overall system.

Indie Development

Publisher conservativeness is a well-known phenomenon in many art forms - movies and music are two prominent examples. Both art forms have indie movements. Usually these indie works appeal to a niche audience, much smaller (and less lucrative) than mainstream audiences. Nevertheless, the mainstream admires the work of the indies. Whenever an indie does something particularly successful or powerful, it influences the mainstream - some elements of style will even be co-opted directly. This flow of ideas from the indie into the mainstream helps keep the mainstream fresh. It prevents the mainstream from iterating too much on proven successes and ultimately stagnating.

The mainstream does come up with its own original ideas, but the indies are just going to try a lot more unusual things. The indies will be hit-and-miss, they will mostly miss, but due to the large number of attempts, the number of good discoveries made by the indies will be significant.


Looking Forward

These two pieces - the mainstream and the indie - form a cohesive system that propels the art form. Interestingly, computer scientists can see that this system is effective, because it parallels a common family of algorithms used for solving hard problems. The idea is if you're trying to maximize something, you devote most of your resources (budget) toward searching in areas you already know to be good, so that you can climb higher in that direction. But you want to inject some elements of randomness into the search, to help you find paths you would not have seen. Still, you want to devote only small budgets to these random directions (like indie budgets) since most of them will not lead you in the best direction. So we have these low-probability branching choices that correspond to the indie projects, and high-probability choices that correspond to the actions of the mainstream.

To ensure that the art of computer games is being propelled via this sort of effective search, we need both pieces of the algorithm: the indie and the mainstream. The mainstream has happened naturally, via the profit motive. The indie side, however, has not sufficiently developed yet. What has developed naturally so far is that independent developers are mostly people trying to make traditional games, but with a low budget. Mostly they are amateurs trying to break into the mainstream industry, or industry veterans who are sick of big-company politics and want to kick-start their own businesses. This kind of indie is good for providing career paths and feeding the industry, but it doesn't do much to foster the innovation we need.

These excerpts are from an interview with Jonathan Blow; the complete transcript will be printed in the upcoming GDC 2004 Guide, available by mail or online in January 2004.

General Information