On Digital Cities
No matter how hard I try, my cities fail. They flood, burn, and wither, from a lack of power, plumbing, people, jobs. I embark on multiple pre-built scenarios, from alpine villages to a ferry empire, in the popular city-building game, the modern heir to the legendary SimCity franchise, which boasts an insultingly bland name: Cities: Skylines. Although I’ve played video games since childhood, this one feels inscrutable. My eyes glaze over the hieroglyphic control panels. Every cursor click reveals a pop-up with dense blocks of text disclosing rules, tips, exceptions. For a moment, I wonder what makes a game like this pleasurable to learn and navigate? Is it not another full-time job’s worth of attention spent learning a nonsensical system that takes itself too seriously?
Within a few days, I give up and opt for a cheat code: YouTube. I quickly discover a video tutorial that millions of similarly adrift gamers have found, produced by a friendly voice named Phil who runs an account called “City Planner Plays.” One comment down the page calls him a modern Bob Ross, referring to the cult celebrity who for decades lulled public-access TV audiences with his live landscape paintings. Rather than a watercolor palette, City Planner Phil paints with an array of zoning and utility tools, which chop the landscape into invisible boxes, from which buildings grow like vegetation—that is, spontaneously, with no materials shipped to the site, no toxic fumes or obnoxious noises during construction, no industrial waste carted away upon completion.
There is indeed a meditative quality to the experience. Phil speaks in a calm, assured narration as he gradually converts the coastline into a bustling grid, as lo-fi electro beats roll in the background. Occasionally, he sprinkles his can-do narration with frugal hacks and tricks, like using dirt roads in residential areas. But there are also some obvious and immediate grievances, beyond the immaculate overall depiction of violent land reformation.
To list just a few examples: Every city begins as an off-shoot of a highway that juts into the screen from nowhere. Phil advises new town managers to place their water intake upstream, and their sewage releases downstream—but this logic breaks down when you consider the other town settlements up and down the river. As your population swells, new services and industries randomly become available: garbage becomes an issue only at a population of 420, fire and police become critical at 850, and residents finally get a park when you cross 1,300.
But perhaps the most egregious offense is the game’s very first action, the one you’re required to take before anything else can happen: build single-family homes. The only zoning available in a new town’s toolbox is “low-density residential,” and your town needs these lots before it can collect taxes or attract commercial investments. This detail may sound innocuous, or even desirable: a proto-typical Rockwellian town where every person commands their humble castle. However, zoning exclusively for single-family homes effectively bans any other class of alternative housing: a duplex, triplex, or fourplex, not to mention a multi-unit condo or mid-rise apartment building. Spread your single-family housing policy across all residentially zoned lots and you essentially block a large percentage of the population from ever living there.
If it sounds like I’m reading too much into things, consider how the logical blind spots of city-building games parallel America’s real-world zoning tradition. In recent years, a revived debate around housing and urban development has unearthed the private interests and vile intentions that have long characterized our arcane zoning rules. In his book-length critique of the practice, Arbitrary Lines: How Zoning Broke the American City and How to Fix It, city planner and land-use activist M. Nolan Gray argues, “To this day, the wrong side of the tracks is not merely a saying but a place that is written into law as a zoning district drawn on a zoning map… [it’s] perhaps the most successful segregation mechanism ever devised.”
Amid the early 1900s “progressive” movement, zoning emerged as an attempt to regulate land use and density under a sheen of technocratic governance. The innocent explanation is that the unstoppable rise of the skyscraper triggered these discussions, as well as the cartel of property owners who wanted to cap additional supply so their existing towers could rise in value in perpetuity. While this top-down protectionism is certainly part of the story, it also overlooks the more blatant racial and class biases that fuelled the practice, both in the city and beyond.
In 1910, Baltimore became the first city to implement a racial zoning ordinance, separating blocks into “majority White” and “majority Black” districts. In 1916, New York’s Fifth Avenue Association, of retail magnates and homeowners, banned the use of land for manufacturing, purportedly to mitigate public nuisance but really to keep out Jewish factory workers (who were characterized as “flies”). Later that pivotal year (which also saw the passage of New York’s set-back requirements for high-rises), Berkeley, California became the first town in America to zone residential exclusively for single-family; a letter from the local community group reveals their underlying motivation—keeping out “negroes and Chinese.”
A century later, American cities are more segregated than ever, and those with the most restrictive zoning policies—New York, Miami, Los Angeles, San Francisco—suffer the worst housing crises. The latest nail in the coffin was another zoning blunder that targeted the supposed blight of boarding houses and single-room occupancy hotels, affordable waystations that had traditionally served millions of young people, elderly residents, newly married couples, and the recently unemployed or in transition, providing critical relief from a volatile and often cruel housing market and keeping people off the streets.
Homelessness is another hurdle that Cities: Skylines conveniently avoids, maybe because of the bungled way that SimCity handled the issue. While the classic simulation game deserves some credit for its attempt, its low-fidelity treatment of the unhoused sparked some of the internet’s darkest message boards. In the game, each homeless avatar is represented by the flat, yellow outline of a human figure with bags in tow, and their presence algorithmically reduces the city’s property values, which correspondingly sinks the happiness of town residents. Many die-hard fans struggled to manage this dose of messy realism, and in 2015, Italian artist Matteo Bittanti published a collection of the community’s most awful comments, taking as its title the topic of one Reddit form, How to get rid of homeless. Many posts disregard basic grammar and punctuation as they propose their final solutions:
“remove the abandoned buildings and garbage, which they live off”
“lay a trail of raw meat from the mountains to tempt the bears into your city”
“more police to shoot them down”
“drop a meteor on them” ; “drop a tornado on them”
“place a lot of parks around the radiation for them to live and hopefully die in”
“a way to convert them into a fuel source would be nice”
The message board is obviously repugnant, but strip away the despicable language, dress up the claims with technocratic language about safety or zoning, and you have some core tenants of US housing and labor policy over the last half-century. An ever-growing police-industrial complex. Runaway corporate power that decimates unions and makes blue-collar work ever-more precarious. (Recent reports find that Amazon is already running out of its human “fuel source,” or the low-wage contract workers who power its monolithic warehouses.) Finally, malicious zoning rules that fetishize the McMansion and divert lower-income dwellings into toxic buffer zones that separate industrial parks from single-family enclaves. One startling statistic from 2019 found that 80 percent of US incinerators are located in low-income communities or communities of color.
“I don’t want low income sims, since my city is pretty much middle and upper class,” reads another nonsensical comment from the Reddit message board. And yet, as cruel as this statement sounds, it reflects perhaps the most common trait among a certain breed of American social strivers—call them well-to-do, corporate, bourgeois, meritocratic, the “professional managerial class.” Every time one fashions themselves “middle class,” or touts their neighborhood’s “good schools,” we invoke the phantom of some inevitable, permanent underclass.
While the city-building genre has enthralled audiences for decades, it’s obvious these games present a poor proxy of urban life. SimCity’s designer Will Wright noted this soon after the game’s release: “It’s kind of hopeless to approach simulations like that, as predictive endeavors… SimCity was always meant to be a caricature of the way a city works, not a realistic model.” Is it any surprise that the game’s corporate publisher would contradict Wright entirely, and attempt to sell the technology as precisely a real-world simulation?
As SimCity’s popularity took off in the 1990s, the company’s marketing executives at one point used the tagline: “If this game was any more realistic, it’d be illegal to turn it off!” Its MBA-bred leaders even developed a short-lived, in-house agency: “Maxis Business Simulations,” which served as a kind of Millennial McKinsey that develops Sim-like environments for corporate and government clients, including oil refineries and power grids. Rumor has it these tools reached not only Fortune 500 boardrooms, but even the White House.
The notion is always flattering: that such a complex entity as a city can be perfectly knowable, its intricate web of billions of daily connections and interactions mapped, programmed, and optimized. But this fallacy didn’t stop a generation of urban planners and pundits from celebrating the SimCity genre as a kind of collective myth. In a way, it’s baffling how these low-grade simulations could be held up as decent educational tools, that time spent with a digital caricature could be more illuminating than a walk outside to see how people actually move, live, linger, connect.
Today, a bevy of venture-backed startups are resurrecting the SimCity legacy and selling the concept of a “digital twin” for cities. Their specious estimates claim such software can save cities nearly $300 million by 2030, even though there isn’t yet a single working prototype. And, though the software is pitched as a user-friendly, fully simulated view of all city services, it also obscures the massive long-term investments into IoT surveillance sensors, data-sharing architectures, and legal defense bills that would be required to construct a reliable “twin.” Always trained to kill with minimum conscience, these companies are hunting first the densest cities, usually islands, where there is the least room for municipal error. Paying customers already include Los Angeles, Las Vegas, Phoenix, New York, Singapore, Helsinki, and Dubai.
In their ongoing efforts to better understand cities, technology companies often seem to destroy what little grasp they have to begin with. A breakthrough technology like Google Maps helped socialize a fairly open-access version of precisely this kind of “digital twin,” allowing anyone with a device and high-powered broadband to zoom into a specific street view across vast parts of the planet—like a highly-detailed, interactive standing globe of old. But, beyond that initial layer of visibility, attempts to track the world more granularly became clunky, incomplete, somewhat creepy. We distract ourselves with multiplying layers of shoddy information: aggregate star ratings, curated reviews, user-generated photos, recent social media posts. Google’s latest and greatest urban initiatives, such as “Immersive View” or its defunct Sidewalks Labs, aim to distill cities into faux-universal components, and ultimately remove spontaneity from urban life, serving merely to optimize routes between home and commerce.
The most promising simulation I’ve stumbled across so far came not from the Bay Area, or a DC think-tank, or any public-private “innovation lab”; rather, it was a group of Drexel students and their advisor who created a simulated model based not on another tourist hub, or central business district, but on Philadelphia’s working-class Mantua neighborhood. Success metrics were based on quality-of-life factors: average rental rates, food distribution, population flight, and gentrification—none of which appear in those virtual city-builders, or possibly even some of those billion-dollar “digital twins.”
I first became interested in world-builders because they seem to be having a cultural moment. Cities: Skylines is one of dozens of video games that mash-up the algorithmic logic of SimCity with fantastical settings or ideological scenarios to fill every conceivable niche. There are space colonies like Rimworld, Dive Legacy (Rimworld, but medieval-themed), and Industries of Titan (factory-slums on a polluted moon). There are eco-parables such as TerraNil (rewilding wastelands), Alba (preserving bird habitats), Lumberjack (eco-fascist bears), and Timberborn (sentient beavers in a post-apocalyptic future). In Buildings Have Feelings Too, architecture itself comes alive, and players must figure out what makes them happy to keep the skyscraper next door from moving towns. And, for the sociopaths out there, Prison Architect lets you design your own supermax prison and play house as a ruthless warden.
I wonder if these games appeal in part because so much of our built world, at least in many swathes of the United States, can feel ossified and decaying. Having lived in major cities like New York and Boston over the last decade, I’ve been both frustrated and baffled by the inaction. Our biggest and most productive cities face increasingly permanent housing crises, building neither market-rate apartments nor dedicated affordable units. Hardly a political campaign goes by in which a mayor or presidential hopeful doesn’t lambast the decrepit state of our public infrastructure, the crumbling bridges, corroded pipes, outdated seaports, and patched-together grids, a shaky foundation of modernity that seems to grow more precarious each year. If Frank Lloyd Wright or Buckminster Fuller were born today, would they, too, sink their imaginations into the metaverse rather than physical landscapes?
In his YouTube tutorials, Phil, the “real-world” city planner, hardly acknowledges the unrealistic speed at which his virtual developments come to life. With one click, construction begins anew without any of the friction caused by pesky local politics, misguided zoning codes, or extensive community and environmental reviews. In the game, no matter how large a city grows, its prices are always rational; there are no bloated, public-private subway or bridge projects that swallow billions of tax dollars and years of squandered potential.
There are pockets of resistance to this learned inertia. For all their drawbacks and completely foreseeable displacement, the fast-rising towers of Zoom Towns like Austin and Bozeman remind us that things can be built fast when there is the will to do so. Efforts to upzone single-family plots have gained traction in progressive strongholds such as Portland and Minneapolis, and college towns like Iowa City and Ann Arbor. Then there is Houston, America’s uniquely “unzoned city,” where one can build up to a seven-story building in a single-family neighborhood with minimal friction, for good or bad.
But even these newly liberated cities overlook some basic building blocks that have defined robust urban communities since time immemorial. It’s not just the need for more diverse housing types, but another simple idea: the mixed-use building; apartments over retail shops, or the triplex with a doctor’s office on the ground floor. Such a radical concept doesn’t exist in the worlds of SimCity or Cities: Skylines, and if you see one in a real-world American city, it’s likely a relic of an earlier era and illegal to replicate today.
Urbanist Jane Jacobs wrote about this planning psychosis and the influence of mid-century starchitects on our overly prescriptive approach to zoning. She writes of Le Corbusier that, “His city was like a wonderful mechanical toy…. It was so orderly, so visible, so easy to understand. It said everything in a flash, like a good advertisement.” By trying to predict the future so precisely, cities actually prevent evolution and transformation from taking place, leading to a zombified reflection of the same office blocks and chain franchises you’d find in any affluent suburb. For all her quirks and blind spots, Jacobs was prescient in her critique of top-down urban planning as the antithesis of healthy urban communities. For example, the organic magic of her 1960s Greenwich Village street, populated with a bookshop, cabinet maker, candy store, pharmacy, tailor, and grocery, a potpourri of uses that could never be predicted nor replicated by zoning theory or any digital city-simulator.
When taken to its extremes, these misguided utopias are not only middling or counter- productive, but potentially nightmarish. In a recent satire of zoning and city-building games, the experimental project “Magnasanti” aims to conquer the unwinnable SimCity by optimizing population density on a 12×12 grid. Artist Vincent Ocasla spent years of trial and error developing a theoretical working design, and the result is a crime-free city of six million “happy” sims, with zero abandoned buildings, no water pollution, and no traffic congestion; it’s also a city with a massive police force (the only way in the game to curb violence) where few residents live beyond fifty, and there are no hospitals, schools, or fire stations. From a birds-eye view, the city looks like an island consumed by skyscrapers, not unlike parts of Manhattan, Hong Kong, or Singapore.
Does Magnascanti represent heaven or hell? Even the most radical, green-urbanism activists would concede there’s such a thing as too much density. In earlier eras, vertically-inclined artists and satirists imagined towering apartments bolted onto bridges in London or Rome; one memorable cartoon from 1909 imagines a several-acre sized plot stacked 84 times, each level containing a single-family home and lawn—an absurd way of preserving the suburban dream. A century later, this ridiculous concept is not too dissimilar from the supertall pencil towers that border Central Park South, where each floor is its own penthouse and urban geographer Samuel Stein has critiqued the “vertical sprawl” of “sixty gigantic condos… [that] could fit 442 normal-sized units.”
The origin story of sprawl is its own cautionary tale. In the mid 1800s, Frederick Law Olmsted, the father of landscape architecture, designed countless iconic green spaces including Central Park, the Niagara Falls reservation, and the great necklace parks of Boston and Milwaukee; in 1869, he pioneered another experimental concept: the planned community of Riverside, Illinois, a relatively modest single-family suburb that he envisioned as a romantic return to nature, and which provided the blueprint for the McMansions and minimum lot sizes that would later characterize “middle-class” development across not just the United States. but affluent suburbs around the world, from Argentina to Malawi. The irony, of course, is that this naturalist’s dream first had to destroy and remake the land in its own vision. The biological concrete of crabgrass and pavement. The noxious trails of car exhaust and roadkill. The very notion of a “lawn,” which is today the country’s most irrigated crop.
In the more than 30 years since SimCity was released, the world’s population has become majority urban for the first time in human history, and much of that development follows the same metropolitan sprawl that the west pioneered long ago. But, even today’s breadth of world-building games continues to overlook another core tenet of recent urban development: almost all of it has occurred outside the west, primarily across Southeast Asia and the Middle East, and on a scale never seen before. China alone has dozens of cities with over 5 million people, multiple bullet trains, and more skyscrapers than New York. What remains to be seen is whether the pandemic has at all challenged this conviction, or if we are already locked into development patterns via path resistance and sunk costs.
The reality is that architects and city planners can rarely predict the actual impact of their dreamy sketches. Even with good intentions, our utopian visions can wreck devastating, unintended consequences. So, we must look instead to looser guardrails within which development and community can emerge more naturally and equitably. When President Hoover kicked the single-family engine into hyperdrive with his Standard State Zoning Enabling Act, the legislation was arguably as radical, expansive, and top-down as today’s prospect of a Green New Deal, which might alternately—and never unilaterally—encourage people to live in multi-family or even high-rise dwellings. Federal, single-family mortgage guarantees are as communist as today’s Red Deal plan for free sustainable housing and public transportation, an amendment of the GND recently advanced by Indian and Indigenous activists.
Sometimes during visits home, to the sleepy Illinois suburb where I grew up, I tease my parents and relatives about what it’d be like to upzone their neighborhoods. We could turn a four-bedroom single-family house into a four-story apartment complex, with a ground-floor grocery for good measure. Wouldn’t the neighbors enjoy the convenience? Often, the blank look in their eyes resembles terror, or maybe confusion. But the scarier, more confounding prospect is a housing model that doesn’t evolve beyond the status quo; one that continues to uphold car-dependent castles as the universal ideal, rather than looking for creative ways to densify, especially near major commuter rails; a broken framework that maximizes property values above all else, under the guise of a shared dream, which in reality appreciates to the detriment of the environment, social cohesion, and our basic sense of empathy.