One of the more arcane challenges for the foundation creating the 9/11 memorial downtown was how to arrange the names of the more than three thousand victims of the tragedy.
It was complicated enough to fit them all within a taxonomy of their location at the time of the attacks, but grew more complex with hundreds of requests for certain names to be placed adjacent to loved ones.
So the foundation turned to computer science, hoping an algorithm could help them sort the problem. According to a wonderful piece in the New Yorker by Nick Paumgarten, more than one of the initial experts contacted for the task declared that it was an impossible. “The reasons for these requests were varied. Sometimes the victims were cohorts, or best friends. In other cases, the families knew, from last phone calls, whom their loved ones had been with in the end—in an elevator, on a ledge—and wanted those people listed together.”
Luckily Jake Barton, principle at Midtown design firm Local Projects, didn’t know about the doubters when he accepted the challenge.
“They had decided on a structure before they had the capability of making those designs a reality,” Barton told Betabeat by phone today. “They had imagined you could design an arrangment where all the names were spread evenly, but underneath this seemingly random set of names, would be this incredibly complex lattice work of meaning.”
Barton and data artist Jer Thorpe used a classic algorithm designed to solve the “knapsack problem”, a way of arranging irregular objects in the most efficient way possible. “A computer did all the heavy lifting and number crunching,” says Barton. “But we made the tool adaptive enough so that the foundation could come in afterwards and make changes, and the system would respond dynamically. Efficiency is important, but you also need that human touch.”