The Unlikely Way Algorithmic Design Could Transform Suburbia

Diverse architecture begets healthier communities, argues architect John Szot.

American architects have masterminded dozens of suburban housing styles. Wright created the Prairie homes, Cliff May had his ranch houses, and developer Joseph Eichler introduced modern design to the middle class. And of course, there’s the McMansion.


But the one factor that’s exerted the most influence over how much of the country lives today is industrialization–and John Szot, a Brooklyn-based architect, isn’t happy about it.

Szot believes that homogeneous, mass-produced, single-family homes have subtly contributed to the political conformity associated with suburban life. “Maintaining diversity in the built environment is critically important to maintaining the diversity of our communities, therefore all manner of perspectives on design are valuable,” he says. “So while the nostalgic vernacular pastiches that currently dominate the suburban market may be trite, such opinions are academic and small compared to the larger problem of how capitalism rewards practices that suppress the architectural diversity needed for a healthy community.”

His proposal for introducing more diverse architecture into the suburbs is on view in Mass Market Alternatives, a new exhibition at the Boston gallery Pinkcomma. The project shows how algorithmic design could make it just as easy and cost-effective to build diverse suburban architecture as it is for developers to design and build boring tract houses.

“It’s already common practice in the industry to rely on algorithmic controls to make our buildings smarter, less expensive, and better suited to serve an information-driven society,” Szot says. “The future of algorithmic design is in ideological applications where it challenges conventions and broadens our spatial and experiential palette. To that aim, my studio’s interest lies in the computer’s capacity to behave like a feral animal, adding creative chaos to the design process and leading us to places that are unusual, counterintuitive, provocative.”

The design and construction of suburban subdivisions is based purely on receiving the most return for each property sold. It’s a game of economies of scale. The more of something you build the less expensive it typically is. While there might be some very subtle variation house to house, they’re mostly the same size, layout, and shape. Their non-offensive, generic aesthetics are designed to appeal to a wide swath of the population. “[Suburbs] have provided us with an extraordinary example of how industrialization and economics shape cultural values through architecture,” Szot says.

With Mass Market Alternatives, Szot holds the economic system driving suburban housing–the industrialized design and the mass production–constant, but uses algorithms to generate more diverse building types. “If such a balance could be found, architecture would gain access to a powerful means for shaping society,” he says.


Szot’s formula for designing his houses involves using a computer program to generate floor plans and layouts, a style “recipe” revolving around materials and how the walls are joined to the roof, and some human intervention to sew up the design. The algorithm generates a box-shaped pattern of different colored blocks, which essentially becomes the floor plan. The program has settings that control the overall proportions of the pattern, like the height and width of each block (which eventually become rooms), the dimensions of the central hole (which is like a courtyard), and the maximum number of colored blocks in each design.

The program created 30 unique patterns for each of the four stylistic recipes. For example, the “precast” series is concrete and has a flat roof, the “patio” series features brick walls and a flat cantilevered roof, the “ranch” house has wood cladding and pitched roof, and the “loft” series has glass walls–for a total of 120 patterns. Each colored block is assigned a specific wall section from each recipe. Then a human designer intervenes and assigns a room function to each block, translating the randomly generated pattern into an actual house.

Like typical suburbs, which are composed of variations on a theme, these algorithmically designed houses share commonalities, but their expression is more diverse in shape and materials than most subdivisions.

The meat of Szot’s argument for more visually diverse architecture in a subdivision is that the houses would then appeal, stylistically, to different people. It’s an idealistic proposal. Architecture isn’t a great predictor of political beliefs. Take Philip Johnson, an influential architect and the progenitor of the modernist glass house. He was also a Nazi sympathizer. Szot is aware of this. “Because the project is focused on the link between cultural values and aesthetics, there is the ever-looming danger of suggesting a direct correlation between style and political affiliation,” he says. “Reality is far more subtle.”

Changing up design might not directly lead to a social revolution; however as Szot shows, the old way of producing suburban tract housing could be upended through algorithmic design, leading to communities that are at least more architecturally diverse. “Architecture isn’t simply about addressing practical concerns,” Szot says. “It can also stimulate personal growth through provocation, and a building is the ideal venue for challenging who we are and what we believe.”

While Mass Market Alternatives is a thought exercise now, perhaps someone in the future might develop the idea of an architecturally diverse suburban subdivision. I asked Szot if he’d live in one of the algorithmically designed houses. “Absolutely,” he says. Demand from at least one person is there. Is supply next?


About the author

Diana Budds is a New York–based writer covering design and the built environment.