All models of reality make assumptions about reality. The better sorts of models try to make those assumptions explicit and, best of all, changeable. More worrisome are the models which hide the assumptions within swanky graphics and animations. Many of us here greatly enjoy SimCity, the well-known and highly-regarded urban planning simulation from Maxis. We're not alone -- SimCity is now in its fourth iteration (Windows users can even play the original SimCity online for free.), and continues to be a steady selling game. Unfortunately, SimCity is often seen as more than a game: SimCity, in all of its versions, shows up in classrooms, research papers, and (rumor has it) planning offices around the country. And that has some troubling implications.
Daniel G. Lobo and Larry Schooler, in October's The Next American City, have a terrific article about the history, use, and model assumptions of SimCity. "Playing With Urban Life: How SimCity Influences Planning Culture" walks us through the ways in which models and simulations alter the way planners think about cities.
As mayor, the player operates in “God Mode,” with absolute power to build, demolish, tax, and spend. Unwieldy growth and megalomaniacal, destructive behavior are the two poles of city operation and the player’s most likely courses of action. Thus the heart of the game is much less a universal vision of city design than it is a reflection of the most extreme tendencies of development in America, found in the few areas in which one person has total control over a large parcel of land—whether a powerful mayor pursuing an urban renewal project, or a developer creating a massive planned community in the middle of desert or farmland. But the many parts of urban planning and development that do not reflect this model of total control over virgin territory get short shrift. SimCity’s narrow lens only tells half of the story of urban development. But aspiring and practicing urban planners have been looking through this lens for fifteen years, with influential results.
While some of Lobo & Schooler's complaints arise from the fact that SimCity is built as a game -- the "God Mode," for example -- most derive from inability to modify the underlying model, whether to include mixed-use development (the ground-floor commercial/upper-floor residential buildings which help to make dense urban environments livable), to vary the demand ratings for various services, to make pedestrian travel more acceptable, or to alter the efficiency and availability of renewable power generation. As a result, some models of urban development, such as the "New Urbanism" movement of the mid-late 1990s, fall outside the scope of the simulation, and become invisible to developers-in-training. While a free/open-source version of the software would be the ideal (if highly unlikely) solution, format information and tools for altering the model would be sufficient. They have tools for changing the appearance of buildings and props, why not tools for the parts of the sim that really matter?
Simulation games like SimCity are valuable because they give a peek at the complex relationships between cause and effect in big systems such as cities. They're a chance to play at the edges of complexity, to see "what happens if I do this?" in both an iterated and replicable fashion. They can be wonderfully seductive digital sirens leading to unexpectedly staying up to 3:30 AM. But to be good educational tools, the models have to be transparent and changeable. We should be able to play with the system itself, not just the system's effects.
Here's something I wrote some time ago after noticing a webpage with a Electronic Arts and Maxis joint project while looking at the capabilities of SimCity 4. It's a bit random as it started as I was trying to figure out if SimCity 4 would utilize my dual processor dual monitor setup. I've reproduced it as is:
what kind of a world would maxis create if given simcity 4?
What kind of a world would EA?
can i have organic farms in simcity 4?
Strawbale houses? BioDiesel?
Certified organic biodiesel?
is certified organic biodiesel sustainable?
i want certified organic biodiesel in my simcity?
Light rail? Bike paths?
Its a common problem that can be though of in many games. Because games simplify things to make a reality entertaining it introduces some problems. I think of it when I am playing a war simulation for example, its very easy to become involved in the game and forget exactly what it is that i am doing. I take actions that would have consequences in life that i would never want to face. I see my disconnection from my actions, thinking of them as a game, and I wonder is this how atrocities happen in real life? Is it so difficult to fully access all the consequences of our actions and then make decisions upon them. But maybe this isn't so much a problem with the game itself, more in how we see the world. I never really thought of it this way when I've played sim city... it always caused me in effect more to think of the days as a child playing legos, then the little people in those houses and in those communites that were dealing with my mayoring on a daily basis in the game.
SimCity definitely feels to me like it contains the author's biases. It doesn't really matter what the light rail / bus / car tradeoffs are in real life; the author prefers certain modes of transportation and penalizes others in the game. There are assumptions that industries have to be polluting, that cars and highways are bad, etc. So it scares me to hear that people are using this for real life planning!
In SimCity, the best city is the least convenient one. You get people to use public transportation by making the roads winding and unweildy, leaving them just connected enough to let a fire truck through if you're desparate. To some extent, I think some urban planners have tried to do stuff like this: c.f. Portland, Oregon. =D
Simcity 4 was, in many people's opinion, the worst in a very long time. The original creator of the Simcity series, Will Wright, didn't have nearly as much to do (directly) with Simcity 4. There are so many little details that seem to have been forgotten about or glossed over in pursuit of sexy graphics.
Besides all that, though, the base premise of Simcity has never been set in reality -- it's just a game -- and cities don't pay out of their own pocket for probably 90% of the things that a Simcity needs to pay for.
There are a lot of really fun, great ideas in Simcity 4, but I don't think it's anywhere near reality, and perhaps now that the game looks so good, it's begun to fall into the uncanny valley of simulation: "It looks like a real city, so why isn't it acting like one?"
I bought it as soon as it was released, of course.
A German company, Glamus GmbH, together with the Weimar Bauhaus University and the traffic research department of DaimlerChrysler, created a game called 'Mobility', which does the simcity thing but really focuses on, well, mobility.
It doesn't frustrate me nearly as much.
Nobody seems to have commented on the author's main point, that the simulation should be -transparent- and changeable. That's a large obstacle. Games strive to make their systems magical and proprietary. Even with games that have sophisticated developer SDKs freely available, there is always a final wall you can't pass through where the real magic takes place. ( Unreal Tournament is a notable exception to that along one axis, as it exposes the innerworkings of it's AI which is quite fascinating. )
To ask that SimCity open it's guts wide open therefore is a tall order. That's like giving all your trade secrets away. I honestly believe the only viable solution would then be to garner support and resources to further develop an open source SimCity clone, such as LinCity ( http://lincity.sourceforge.net/ )
Very interesting article. Definately has some very valid points indeed. The game will essentially behave the way it was designed (to) based on the variables involved.
Some parts of the games' (SC4) simulator are actually exposed to some degree (like being able to change various Traffic Simulator variables for example), however it's not of the raw kind like a lot of us would like (especially us modders of the game).
I personally think the game itself represents 'some' aspects of real-world environments rather accurately. Most of it though is essentially 'just a game' as so many would put (it).
Getting Maxis (or EA for that matter) to have anything they make open-source though, well that's like asking someone to count how many stars there are in the sky; it just ain't going to happen.
It's a real pitty Will Wright didn't have more (hands-on) involvement with the latest incarnation; it really shows. Too much emphasis was put on the visual aspect of the game.
Fact is Amit, there are only an exceedingly small handful of industries that don't contribute to pollution, either directly or indirectly in terms of waste production and heat pollution. (Electronic in particular are the worst offenders in adding to the heavy metal varieties of toxic waste.) I've lived all my life in urban areas, but I've also seen what has happened when light industry moves out to former rural zones. As far the influence of automobiles, it's no accident that Los Angeles, and New York, cities with far more cars then there is practical room for are smog capitals of their respective coasts.
And highways do impact on their surrounding neighborhoods. In recent decades New Jersey has finally gotten to put up sound shields in the most heavily impacted areas of the New Jersey Turnpike.
The author/coders do have biases no doubt, but that does not mean that they are totally unbounded in fact.
To be clear, as much as I would love to see a free/open source version of SimCity come to fruition, my point was that the underlying models weren't open to be adjusted. It is entirely possible to produce a closed-code version of a simulation which nonetheless allows for modification of core components. Civilization III, to cite another game, is far more modifiable than is SimCity, but is equally proprietary code.
And: Mike Gillis, your comment about an "uncanny valley of simulation" is brilliant!
will wright did say back in the summmer his ultimate dream is ot have simcity "more customable"
Which to me means having the ability to adjust and manipulate as much as we can.
I think he realizes what we want, we'll see i suppose if it happens.
Interesting article. I've recently started to work on my Ph.D., which also involves modeling complex processes (crystal growth, in my case), and I've become fascinated by all the possibilities of computer simulation and I'm trying to expand my horizon beyond computer models of material science. Just for a lark, I've decided yesterday to write a java applet that models the growth of populations of fantasy races across a map - you can see the (very simple) application here:
with a short explanation of the program here:
This is, of course, just a game without any serious applications. But I guess games could be used to teach real-world processes - and for serious researchers, such programs that allow for more customization could be more than just games.
Maybe I should give some more thought how to turn real-life processes into games - if only so that our students don't get bored... ;-)
Jürgen -- good to see you here! (Mr. Hubert was one of the contributors to Toxic Memes.)
I have found that the use of simulation games -- even in abstracted, non-computer, "role playing" form -- is a surprisingly powerful pedagogical method, whether the students are political science undergrads or directors of global Fortune 500 corporations. The goal isn't to show precisely how a system works, but to give the students a visceral sense of component interaction in a system, and to allow them to experiment with choices.