SYSTEMS AND FAILURE: Why We’re Still Learning The Lessons Of The Titanic.

Major disasters often occur after such long, uneventful stretches. Before the partial meltdown of the reactor at Three Mile Island in 1979, no U.S. nuclear plant had experienced a serious accident for 25 years. Similarly, before the blowout of the BP Macondo Prospect well in April 2010, the Deepwater Horizon rig had gone seven years without a serious mishap while drilling some of the deepest wells on the planet. “When you think you have a robust system, you tend to relax,” Henry Petroski, a professor of civil engineering at Duke University, tells Popular Mechanics. Over time BP and its contractors began to cut corners: Alarms that would have warned of a gas leak were silenced, safety checks canceled. The blowout preventer—a last-ditch device intended to shut off a runaway well—was only partly functional. And workers were constantly urged to drill faster. That kind of culture invites trouble.

This phenomenon was the subject of an excellent article by my colleague Greg Stein recently.