Lost in a sea of science data

A free article from the Chronicle of Higher Education describes how librarians are teaming up with scientists to archive and preserve huge amounts of scientific data:

Science is experiencing revolutionary changes thanks to digital technology, with computers generating a flood of valuable data for scientists to interpret.

But that flood could drown science.

Data from experiments conducted as recently as six months ago might be suddenly deemed important, but researchers might never find those numbers — or if they did, might not know what the numbers meant. Lost in some research assistant's computer, the data are often irretrievable or an indecipherable string of digits. That's a scenario increasingly familiar to scholars, says James M. Caruthers, a professor of chemical engineering at Purdue University.

"We are starting to die from data," he says bluntly.

To vet experiments, correct errors, or find new breakthroughs, scientists desperately need better ways to store and retrieve research data, Mr. Caruthers says, or "we are going to be more and more inefficient in the science that we do in the future."

Dealing with the "data deluge," as some researchers have called it, will be among the great challenges for science in the 21st century. Many in the field say that scientists should not be left to manage the data on their own.

Instead, librarians will have to step forward to define, categorize, and archive the voluminous and detailed streams of data generated in experiments. Already, librarians on some campuses — among them Purdue, the Johns Hopkins University, and the University of California at San Diego — are beginning to take on that role.

(read entire article)