Street Scores: Composing the Social Life of Public Space

Authors

Jingrong Zhang (Massachusetts Institute of Technology), Fábio Duarte (Massachusetts Institute of Technology)

Presentation

Session

VISAP Paper 1

Time

Wednesday, Nov 11, 08:48 – 09:00 (US/Eastern) · session 08:00 – 09:30

Location

Hall Essex south

Keywords

Pedestrian behavior, urban public space, data visualization, sonification, computer vision, live performance, choreography, art-science methodology.

Abstract

In 1980, urbanist William H. Whyte filmed pedestrians in four American public spaces and, by hand, frame by frame, mapped how people gathered, lingered, and moved. His central finding was elegantly simple: what attracts people most, it would appear, is other people. Thirty years later, researchers returned to the same sites and compared this archival footage to newly filmed contemporary footage, this time analyzed with computer vision, and found that public space had quietly changed: pedestrians now walk 15% faster, lingering has been cut in half, and while people still walk alone at the same rate as before, they encounter each other in groups far less often. Street Scores takes this finding, and the decades of observation and footage behind it, and asks what form is adequate to carry it, not a chart, but a score. Individuals become rectangles, groups become segments, and the resulting composition was translated first into an interactive visualization, then, as one exploratory step, into a live musical and choreographic performance staged in a public plaza. This paper reflects on that translation, examining what a score makes newly available relative to more conventional representations, and what is gained and lost in the process, and arguing that the choice of form is not neutral but itself an argument about what data means. We treat the live performance as a first experiment in an open, ongoing inquiry into which forms can honor what public space still owes the people who pass through it. More broadly, we frame this translation, from science to concept to object and back into new questions, as one instance of a recurring art-science methodology.