At a recent demo day hosted by a Georgia Tech research center, our studio showed a working prototype of the Cartoonist engine for the first time. The whiz kids at UCSC’s Expressive Intelligence Studio have been working overtime on the guts of our system in order to link together our user interface, the tool that converts user input into machine-readable form, the library of action verbs that drive each game, and a playable output.
While still in incredibly rough form, we can now generate a large number of games based on a small amount of information drawn from a current event. (We’ve capped that output at nine games for testing.)
Generating Occupations
One of my first demonstrations of the system was to see whether it could handle the most visible news event of the past few months: the Occupy Wall Street movement. I plotted a fairly simple relationship between protesters, riot police, and public awareness. Protesters were set to “grow” public awareness, police to “attack” protesters, and public awareness to “watch” police. At this time, all of these entities are represented by colored orbs. (We just hired an artist for the project, NY freelancer Rachel Morris.) From this tiny mapping of objects and actions, the system generated nine potential games. While many of them were broken in ways that showed where more coding is needed to flesh out verbs, a few of the results showed the exciting potential of our project.
In one generated game, protesters spawned randomly along an edge of the screen. Players controlled a bubble labeled “riot police,” and could move in four directions while firing projectiles at the protesters. (These could later be skinned to look like gas canisters or rubble bullets.) On the other side of the screen, a “line of sight” marker inside a bubble labeled “public awareness” followed the movements of the player/police. Whenever a new protester appeared, the public awareness bubble became larger; whenever a projectile hit a protester, that protester entity was removed from the game. It was a rudimentary, playable editorial cartoon, and I’d generated eight others with just 15 seconds of work.
Some of the logic desirable for an “OWS: Oakland” game was clearly missing, because only a fraction of our action verbs have been fully articulated in the system. One obvious exemption here is the role of embedded journalists in the documenting and relaying of such events — and, if we’d had a good verb for these, we could easily have more than three actor types on the screen.
Something about the system, about which we had no clue whether it would work or not, stood out to me: Player control is non-arbitrarily assigned to different actor-types in each of the nine game builds. Sometimes I controlled the protesters, dodging projectiles or the tackles of police; sometimes I was the police; and sometimes I simply moved around as public awareness, my line of sight trained on the police.
The View From Everywhere
The fact that we haven’t yet been able to come up with a finalized name for our project testifies to the grayness of the territory that it occupies — for those who haven’t been following the project’s development, we hope to replace the name “Cartoonist” out of respect for the work of political cartoonists.
An important aspect of this platform is that it’s agnostic to ideological bent, the user’s professional status, and the type of journalism that it is used to convey. While it would be possible to use the system for objective, professional reportage, anyone will be free to operate and modify it. The feature noted above — the assignment of player control to different entities in each game build — is actually one of many powerful devices for interrogating the view from nowhere.
Coverage of Occupy Wall Street (and similar events around the world) has foregrounded the rising importance of alternative media and new forms of journalism. Participants and supporters of Occupy are wary of a mainstream media ecology that has either ignored them or subjected them to exaggerated skepticism. A story of the firing of WNYC reporter Caitlin Curran for participating in an OWS rally stands in stark contrast to firsthand coverage of arrests in Oakland by graphic journalist Susie Cagle.
While traditional forms of news media have stuck to an outmoded concept of objectivity, alternative sources have delved into and personalized the Occupy movement to great effect.
There are a number of reasons why videogames about the news lend themselves so well to editorializing. First is the amount of time and expertise required to produce them: In a market where returns on political videogames are sadly minimal, developers tend to make their work an expression of their personal opinions and passions. Second, and slightly more theoretical, is the idea of the simulation gap: It is incredibly difficult to accurately model a real-world system in a playable form, so developers will naturally pick and choose the aspects of, and problems with, that system that they personally see as relevant (consciously or unconsciously). Finally, there’s the strong, expressive power of role-playing.
Because most games work best when the player has a clearly identified role, they lend themselves to explorations of hero and anti-hero viewpoints on a story or issue. So in Occupy: The Game, players control a protester seeking to collect money and supplies while dodging tear gas and garbage, while in Clear the Park the mindset of a greedy “1%er” will be satirically explored through a mix of tycoon sim and improvised siege warfare.
Research isn’t conclusive on which points-of-view and camera angles create the most empathy or opinionated-ness during play, which is why our system is open in this regard. Our expectation is that users of the Cartoonist engine will be pleasantly surprised by control schemes and POVs that they hadn’t predicted when they sat down to generate their games.