Mann Library has been awarded a three-year, $400,000 grant by the National Science Foundation to make sharing digital data among researchers easier. Cornell librarians will develop a set of services and electronic tools to document and archive digital scientific data to make it available to other academic and government repositories.

When researchers publish their findings in scholarly journals they seldom have space to include the underlying data. Publishing the data makes it possible for others to reanalyze and verify the results, or to use the data in new ways or even in different disciplines.

"This is a very exciting time to be working in an academic library. We're exploring new ways of managing information and interacting with Cornell researchers," said Gail Steinhart, research data and environmental sciences librarian, who will lead the project. "Some Cornell faculty members have been quick to recognize the potential value of these activities, collaborating with us to include sections on data management and archival plans in their grant proposals.

NSF, she noted, is among several funding agencies that have been or are considering including data-sharing requirements in their grants, and the project will focus on "small science," where such sharing is not yet routine. "It's our hope that planning for these activities at the proposal-writing stage will both make proposals more attractive to funders and make the process as simple as possible for researchers when the time comes to prepare and document data for archiving and distribution," Steinhart said.

A "staging" repository will be created at Mann Library, where researchers can store their data sets and make them available to a limited group of collaborators, then later transmit them to established discipline-specific repositories, or where no such repository exists, to Cornell's own digital repository, eCommons. Mann Library expects to set up two new servers for the purpose, each with about 20 terabytes of storage, Steinhart said. The project will create new software tools to make it easy for researchers to add data sets without having to learn the sometimes confusing and inconsistent cataloging systems employed by many different discipline-specific repositories.

But the system also will provide "human beings to help make this happen," Steinhart added. "We will do whatever it takes to help them get their data published," she said.