An old-fashioned magnetic tape recorder on board the doomed space shuttle Columbia is helping NASA fill in critical gaps about what caused the spacecraft to break up on re-entry on the 1 February.
The Orbiter Experiments Recorder (OEX), which was not designed to have the durability of the flight data recorders used on commercial aircraft, survived the searing heat of re-entry because it was shielded by a larger section of the shuttle's interior, which was recovered by search teams on 19 March.
For IT leaders, the discovery of the outdated, yet reliable recorder and its accompanying data is a reminder of the value of older, proven technology in a fast-paced world of change.
"It's a holdover from the old technology of the 1980s," said Allard Beautel, a spokesman at NASA.
The recorder, which weighs 58lbs and was equipped with 9,400ft of 1in magnetic tape, was installed by NASA on Columbia in 1981 to provide extra instrumentation for the vehicle's maiden flight.
Columbia was the first shuttle to fly, and engineers viewed the craft as an experiment in progress, so they wanted to collect as much flight data as possible to ensure that it performed as expected.
The OEX was itself an experiment and was installed only on Columbia, the first of the five shuttles built for NASA.
"It wasn't designed as an accident 'black box', but to provide information after landing," Beautel said.
The device remained on the shuttle since its first launch, providing information on each flight from as many as 800 different sensors that monitored control surfaces, temperatures, pressure, vibration, acceleration and other variables. It was switched on before take-off and turned off once the shuttle was in orbit, then was turned back on before re-entry and shut down after landing.
The low-tech device may have yielded the best information so far about what occurred on board during Columbia's last mission, because key parts of the modern radio telemetry being beamed back to Earth from the shuttle ended when the ship was torn apart.
"This is a wonderful example that the old, reliable stuff counts. We definitely rely on new technology without having tested it properly," said Peter Neumann, principal scientist in the computer science laboratory at SRI International.
Henry Petroski, a professor of civil engineering and history, who specialises in engineering failure analysis, said NASA is fortunate to have had the recorder on board Columbia.
"New technology often promises more than it can really live up to," Petroski said.
The Columbia Accident Investigation Board, which was created to search for the cause of the accident, announced that "significant data" had been found on the recorder's tape from about 420 temperature and pressure sensors, and more data extraction is expected.