GDC: Jason Rohrer Talks Diamond Trust's 'Knowledge Chains'

Independent designer Jason Rohrer is one of the gaming industry's few radicals. It makes sense that he fears people will assume his next project, a DS title ostensibly about the diamond trade in Angola, will be a quintessential "serious game," a grave lecture on the brutalities of that business. "It's not a finger-wagging game at all," a cheerful Rohrer said at GDC, where he introduced Diamond Trust of London as part of the Serious Games Summit. But he recognizes it may seem strange for an indie who's always given away his games -- source code and all -- for download to develop a DS game for Majesco. Certainly, it's not surprising the accomplished and widely-acclaimed designer has been approached by major publishers before: "I turned down interest from Microsoft in the past for an XBLA game," he says, intimidated by the idea of working with a game that runs on a full-sized television. "It's really scary to fill that screen with a lot of art assets," he adds. But the DS, with its more compact display and its enormous addressable audience, felt like something he could do, he says. Rohrer was interested in exploring multiplayer experiences like Between, which requires no more and no less than two players. He also had a particular interest in turn-based strategy, inspired by an interest in German board gaming -- people who don't enjoy video games enjoy playing them with him, whereas they may be more hesitant to share video gaming with him. That interest of his raises an issue: "If you make turn-based strategy games for multiple players, why are you trying to force them onto computer screens?" He poses. In other words, he's been searching the game design space for something unique that gaming can offer turn-based strategy besides simply scale. One thing about the screen environment that pen and paper can't offer is a unique view of the world for each player, says Rohrer -- in a board game both individuals are looking at the same world, which makes something like spy mechanics difficult to pull off: "One thing you can't do [with a physical game] is send a spy, unbeknownst to your opponent, into your opponent's secret area," for example. Spy mechanics are really hard to pull off in a board game. The opportunity for one player to do things that the other is unaware of led Rohrer to explore what he calls "knowledge chains" with Diamond Trust; he defines it as "how one player knows what another player knows about what they know." Rumors, gossip, the secrets people keep from each other and the assumptions people make about what others are and aren't aware of create social dynamics that fascinate Rohrer. "Every tier that gets added evolves the social dynamic and changes behavior," he says. His first idea was actually a game about cheating spouses who have to gather evidence to build their divorce case against one another, where neither party knows what transgressions the other had discovered. Naturally, that concept might not have been such a good fit for the platform. "Majesco rejected the idea because it was too sensitive a subject for the DS market," he says. So in aiming to theme it in a way that wasn't so touchy, he thought about resource-gathering corporate agents whose loyalty was for sale. Diamond-trading in Africa would be a good setting for this, he reasoned, as it employs the same kind of knowledge mechanics about bribes and counter-bribes. "This was not going to be a game where there's a bunch of little kids wearing AK47s, making you feel bad about wearing blood diamonds on your finger," he said on his decision not to make a kids' game. It needed an E rating, but he still wanted the kinds of deep mechanics that would appeal to mature players. Rohrer first built a prototype on paper -- although he wanted to use games to do things that couldn't be done with physical objects, he still developed the underlying prototype on paper. Throughout the course of testing it that way, he came upon the importance of intransitive relationships that make it difficult to predict what the other person is going to do -- this solved the problem of Rohrer and his wife settling into strategic ruts by settling on strategies that would beat each other no matter what. In the game design, he achieved this variable factor by creating a penalty for selling too many diamonds at once: the player who sells zero during his turn, for example, can beat the player who sells two -- diamond companies don't like too many diamonds on the market. He avoided creating a straightforward "value computing machine" by taking inspiration from how the system works in real life.