Cartel

A Large-Scale Simulation of Entrepreneurship in Cities

Based on my interest in market urbanism, I built a large-scale simulation of entrepreneurship in cities. We built a deep prototype that simulated supply and demand market competition, externalities, crime, infrastructure, and an elaborate construction system.

Here's what happened.

What if the classic game SimCity was turned upside down?

Cartel was an attempt to invert SimCity and create a city-building game from an entrepreneur's prospective. We wanted to model the evolution of real cities, rather than the sandbox of SimCity. To do that, we needed to simulate supply and demand, market competition, externalities, crime, infrastructure, build a construction system, and much more.

Before we ran out of money, Cartel was accepted to the Google/Improbable SpatialOS Games Innovation Program.

The Lessons that Games Teach

A whole generation learned about cities through Maxis' SimCity franchise. I loved these games as a kid and was sad to learn that Maxis had been shut down after the disappointing reboot of SimCity a few years ago.

Maxis had also pioneered the amazing evolution simulator Spore, which was a beautiful game for learning about science and life without being explicitly 'educational' (i.e. boring).

The sad decline of Maxis gives an entrepreneur the chance to make something fundamentally new in the genre.

But SimCity got so much wrong. The biggest problem with SimCity was that it ignored the complex, evolutionary nature of cities. It gives players far too much control over what happens within their city. The player-mayor is almost like a demigod, who literally hovers over the game world using command-and-control tools like zoning to shape the urban space.

Anyone who has spent any time in cities, especially growing and developing cities outside the United States, is well-aware that this perspective is nonsense.

Cities are crazy, dynamic, bottom-up ecosystems that no one can even fully understand, much less control.

While in Central America, I learned that some neighborhoods there are so autonomous that the police don't enter them. These neighborhoods build and change according to their own internal logic. They're almost like informal corporations or cooperatives, run by live-in entrepreneurs.

These neighborhoods have a fractal beauty, even if they don't always seem clean or safe. They're at the center of the human experience for billions of people.

)

Many of the most successful projects in fast-growing cities, like Colombia's efforts at slum-upgrading, work with the evolution of the city, not against it. SimCity's mayor-as-demigod is nowhere to be found.

But we can't really blame the makers of SimCity or similar games for getting this wrong. The technology just wasn't available to allow the complex and large-scale simulation necessary to properly model a city.

Prototyping Cartel

Cartel started as a piece of construction paper and a few cut foam pieces...

As you can surely guess, the name Cartel is a nod to probably the most famous (infamous) business game of all time: Monopoly by Parker Brothers. Monopoly focuses on the individual land-holder and is pretty unflattering about the role of the entrepreneur in society.

Aside from building houses or hotels, the players in Monopoly profit through luck. They don't add value. They can't do much to differentiate themselves. They're modeled as the cliché exploitative landlord, which is exactly what the original designer of Monopoly intended.

I wondered whether it might be possible to make a game that focused on the positive aspects of entrepreneurship in cities while keeping the edgy, competitive elements that attract people to games like Monopoly or Risk.

Cartel is about the positive transformation of cities through entrepreneurship. Our website used a before and after image created from an in-game screenshot to show the transformation of buildings from blight to prosperity.

The core of Cartel's design is about how value is created or destroyed spatially. For example, an ugly, polluting factory will drag down the value of everything nearby. A beautiful park can add a huge premium to land values nearby. Neighborhoods are rising and falling in value everywhere, always.

Players can create or destroy value through how they buy, sell, build, and decorate properties throughout the city. The dependency of properties on surrounding properties connects the fate of the city to the dispersed decisions of all player-entrepreneurs.

Cartel turns SimCity's on its head. Everything occurs from the bottom-up. You walk around the city. Your knowledge is incomplete.

You are but one force struggling among many. This is the logical inverse of SimCity's mayor-demigod.

Player-entrepreneurs may work together for common cause or sabotage each other. The game creates complex opportunities for selfishness but also for cooperative behavior. You never know what someone else will choose.

Detroit House by Kaspars Pavlovskis on Sketchfab

One of the most fun aspects of the design was working with a 3D artist to create abandoned and developed versions of buildings. We used tragic photos of Detroit Ruins as dark inspiration.

We also kept a bit of cheekiness in the design. Some buildings are outright sleazy and we magnified how the player company's brand gets shown off on almost every building. This is a bit of tongue-in-cheek about 'company towns', but also serves an important gameplay purpose – a strong visual cue of the success of player companies.

Strip Club by Kaspars Pavlovskis on Sketchfab

Modular Factory by Kaspars Pavlovskis on Sketchfab

From Construction Paper To Crisis to SpatialOS

After the initial construction paper models, I partnered with a friend to prototype Cartel using Unity3D. For several months I sat at a kitchen table after work and coded with him to build our basic version.

In late 2016 I learned of a new networking technology called SpatialOS, which simplifies and lowers the cost of building large-scale simulations. Until recently, the technology just wasn't available to allow the complex simulation necessary to model a city properly. The amazing thing about SpatialOS is that the simulation is persitently running in the cloud.

Cartel running on SpatialOS

Even if no players are logged in, the city continues to evolve based on the rules of the simulation. Since the simulation works across servers, thousands of player-entrepreneurs can be in the same world (called a 'single shard'). I was captivated by the possibilities of this new technology.

Unfortunately, just as I was getting excited, my partner started to lose interest.

So I bought his share of the project and tried to figure out how to keep it going. There was no way I could do the design, marketing, art, programming, and everything else Cartel would need by myself.

My answer was to assemble a distributed team from Europe and South East Asia under the name Bold Conjectures.

After all, what was a giant simulation but a Bold Conjecture about the world?

After months of prototyping and working on design documents to improve the game in my free time, Cartel was accepted to the Google/Improbable Games Innovation Program. We continued work to make a city-building experience like no other.

Unfortunately, I had run out of money to finance the project by 2017. SpatialOS was still maturing as a technology, and progress was much slower than I had anticipated. However, I still view Cartel as a great investment, especially since I learned how to build software as a result of working on the project. I would like to take another run at Cartel in the next few years.

Wind Turbine by Kaspars Pavlovskis on Sketchfab

Renewable energy for your city!