GDC 2006 Day One: Designers Meet Social Scientists

I meant to post these in order, but oh well.

Designers Meet Social Scientists

moderated by Eric Zimmerman

There were three designers on the panel: myself, Heather Kelley of Ubisoft, and Matt Adams of Blast Theory; and three social scientists: Suzanne de Castell of Simon Fraser University in Canada, Mikael Jakobsson of Malmo University in Sweden, and Torill Mortensen of Volda College in Norway.

Eric asked each designer “what design problem is coming to the front of your mind given what you have heard today?

Matt: how to surface the sort of data we’ve heard about from researchers today, and dashboard it for developers.

Heather: Given that the magic circle is dead according to many panelists today, what do designers do when player’s mony spend trumps player skill or player time investment?

Me: The two cultures problem. The academics and the developers can’t seem to quite communicate.

Eric: Raph, that’s not a design problem.

Me: Yes, it is. Uh, “how do developers leverage all this academic knowledge in the development process?”

Eric then asked the social scientists for what key research problems they saw:

Suzanne: Ludic epistemology. (I am going to mangle this one). In what ways do games shape our understanding of learning?

Torill: Understanding fun and pleasure in games.

Mikael: How do we surface process-oriented thinking (what designers do) to the academics and to the general public?

Eric then randomly paired one designer and one social scientist. They were tasked with leading a self-selected groups from the audience in creating one project that solved both problems posed by that pairing. The resulting projects were then pitched to a panel of judges that included Beth Kolko and T. L. Taylor and someone whose name I didn’t catch. The winning group got copies of Boggle and Scrabble.

Heather and Suzanne’s group: a game wherein players advance and gain status by trading knowledge. They were at the other end of the room, so I didn’t always quite make out what was being described. They then decided that the group process they had gone through effectively accomplished that, and declared themselves the winners ad hoc.

Matt and Torill’s: a game with an embedded reputation system where you give other players tokens when they trigger something fun. You classify the sort of fun by the different tokens available (whips and flowers were mentioned as token types — we won’t go there). Then the flow of tokens and types of tokens would be dashboarded. I didn’t quite understand what stopped the tokens from becoming a generalized currency rather than tied to moments of fun, though.

Me and Mikael’s: actually an idea proposed by Dmitri Williams, who was in our group, and then elaborated by everyone. Have social scientists do fairly passive monitoring of a variety of game studios to measure characteristics like how long people sat at desks, how much pizza was eaten, how long crunch was, how many folks were in meetings — basically, gather as much data about development process as possible, Then subject it to regression analysis to identify clusters of behaviors, and try to correlate them to the types of games made and their eventual sales. The goal would be to identify alternate production methodologies that might directly improve production (and thus make the companies who use the academic research more money). Once a model is arrived at, also use it to make a game, Game Dev Tycoon, so that the process can be surfaced to the public.

The reputation system won; afterwards, Dmitri said “we shouldn’t have made our controlling variable be money with a panel of academics as judges.” Me, I concluded that every time I do one of Eric’s game show panels, I always lose. 🙂

There was then Q&A, but I think I already blogged the question that I remember.