In 1961, New York City had an idea. Give developers extra floor area if they include a publicly accessible plaza. Build more sky, give back some ground. By the mid-1970s, 166 plazas had been built under this program. Most of them were empty most of the time.
Nobody could figure out why.
The design professionals had theories. The plazas were too exposed, or too windy, or too far from pedestrian flow. They needed more amenities. More trees. Better lighting. Everyone had an opinion about what hypothetical people would want. Nobody had actually watched real people use a space and built backward from that.
In 1970, a journalist decided to watch.
William Whyte had been one of America's most influential writers on organizational sociology. His 1956 book The Organization Man sold over two million copies. He had also, a decade earlier, commissioned the essay that would change urbanism forever. He invited Jane Jacobs, then an editor at Architectural Forum, to write a piece for Fortune about cities. The Rockefeller Foundation read it and gave Jacobs the grant to expand it into The Death and Life of Great American Cities (1961). Jacobs built the theory. Now Whyte was going to build the data.
The Street Life Project ran from 1970 to 1975. The method was simple and radical: deploy time-lapse cameras on New York plazas, watch what people actually do, and build conclusions from that instead of from assumptions. A team with still cameras, movie cameras, and notebooks stationed themselves in the city's public spaces for five years.
What they found contradicted almost everything urban designers believed.
People do not spread evenly across available space. Given a wide open plaza, people cluster. They cluster near entrances. They cluster in the sun. They cluster near food. They cluster where other people already are. The modernist plaza — generous, open, symmetrical — had solved a design problem that actual human beings do not have. People seek density, not dispersal.
Sunlight is the most predictive single variable. Shaded corners stay empty regardless of other amenities. Sunny spots fill up. People track the sun seasonally: seeking it in cool months, seeking shade in hot ones. This sounds obvious. It was not obvious to the people who built glazed towers that cast permanent shadows across their own plazas.
Movable chairs outperform fixed benches. The act of moving a chair, even a few inches, gives a person a sense of agency over the space. Fixed benches are furniture. Movable chairs are an invitation. Many plazas had benches bolted in place or ledges sized at the wrong height. "The human backside," Whyte wrote, "is a dimension architects seem to have forgotten."
Food is infrastructure. Plazas with food vendors or cafes activated surrounding space. Plazas without food stayed inert. This was not a cultural preference. It was a social mechanism. Food gave people a reason to stop, which gave other people a reason to gather, which gave the space a reason to exist. "If you want to seed a place with activity, put in food."
People attract people. The most counterintuitive finding: people choose locations that are already occupied, not empty ones. Given a choice between an empty corner and a busy one, people go where the activity is. This is self-reinforcing. The busier a space gets, the busier it gets. The corollary is damning for grand civic design. A large empty plaza does not fill up on its own. It stays empty, which makes it emptier.
Women are the diagnostic. Whyte observed that women were disproportionately absent from failed plazas and present in successful ones. He used this as a metric. If women feel comfortable in a space, the space works. If they avoid it, something is wrong that no amount of additional design will fix. Their presence or absence was more diagnostic than any planning survey.
Whyte called the tendency for strangers to start talking because of a shared external stimulus "triangulation." A food cart. A fountain. A street performer. An unusual piece of sculpture. The external element gives two strangers a common reference point without requiring either to take social risk. Without it, people in a space remain atomized. With it, the space becomes social. Most incentive-zoning plazas had nothing to triangulate around.
The deeper problem was methodological. Urban planners in the postwar decades worked from models, surveys, and professional intuition. They designed spaces for abstract users — people who would behave rationally, prefer orderly dispersal, respond to good design by using it appropriately. They had never watched anyone.
This is the same failure that produced factory assembly lines designed without asking workers how the work actually happened. The same failure that produced hospital layouts designed without watching how patients moved through them. The same failure that now produces software interfaces built without user research. In each domain, the designers knew their field deeply and their users not at all. The failure mode is identical.
In Denmark, Jan Gehl had been running a parallel inquiry since 1966. Working from architecture rather than journalism, he arrived at the same conclusions independently: edge conditions were where social life concentrated, sunlight was primary, walking pace was the diagnostic of whether space worked. His book appeared in Danish in 1971 and in English in 1987. Neither Gehl nor Whyte appears to have coordinated with the other. Two investigators, different methodologies, different continents, describing the same thing.
That convergence matters. It means the findings aren't cultural or contextual. They're structural. Human beings in public space behave in consistent, observable, predictable ways that bear little resemblance to what designers assumed they would do.
The results eventually forced specific changes. Whyte was commissioned in the late 1970s to write recommendations for Bryant Park, then a dangerous and largely abandoned space behind the New York Public Library. His prescriptions: lower the surrounding plantings so the interior was visible from the street, add movable chairs, add food. The 1992 restoration followed this prescription. Bryant Park now draws roughly 12 million visitors a year.
The Times Square pedestrianization pilot in 2009 applied the same logic to the street. Turning Broadway into a pedestrian space between 42nd and 47th Streets reduced pedestrian injuries by 35 percent. Motorist and passenger injuries dropped 63 percent. The street, given to pedestrians, worked the way streets do when pedestrians actually occupy them.
Whyte's most quoted line: "It is difficult to design a space that will not attract people. What is remarkable is how often this has been accomplished."
The failure he documented was not stupidity. The people designing bad plazas were technically competent and often well-intentioned. The failure was the assumption that design knowledge transfers to human behavior without observation. That you can derive what people will do from first principles. That the people in the space are a secondary consideration, to be accommodated after the design is settled.
The cameras told a different story. People move toward food and sunlight and other people. They sit where they can adjust their position. They stop at corners and talk. They make invisible social calculations about whether a space feels safe. None of this required innovation to discover. It required watching.
We still mostly design spaces for people we have never observed. The cameras are still running.