washington

The US government should provide $3 million over three years to support a database that has been developed to track detailed information on all publicly supported research and development, according to a panel set up by the National Science Foundation (NSF) to assess the future of the database.

Rand Corporation, which set up the database for the government, has threatened to pull the plug on it on 1 March if does not get paid to maintain the system, which costs at least $1 million a year more to run than it generates from user fees.

But as that deadline nears it remains unclear where the government will find the money to run the database, which is known as RaDiUS (for Research and Development in the United States). Some government agencies are said to be reluctant to support a management tool which they fear could be used to criticize their programmes and attack their budgets.

The NSF panel was chaired by Irwin Feller, head of the economics department at Pennsylvania State University. It concludes that “despite its weaknesses, it is extremely important that RaDiUS be continued,” and warns: “If RaDiUS died because of a lack of funding, this would probably also kill any hope of any other such system for a number of years.”

Alarmed at such a prospect, the staff of Representative James Sensenbrenner (Republican, Wisconsin), chair of the House Science Committee, and Senator Jeff Bingaman (Democrat, New Mexico) — two important supporters of science in Congress — have been closely monitoring the fate of RaDiUS.

But a meeting last week between the NSF, the Office of Science and Technology Policy (OSTP) and the Office of Management and Budget (OMB) failed to make much progress in determining who should pay for the database. RaDiUS lost $1.4 million last year after a plan to raise money from users of the system fell flat. According to one official who attended the meeting, NSF is not prepared to pay for the system out of its budget for the current year unless it receives a direct instruction from OSTP to do so.

RaDiUS was developed by Rand with the strong encouragement of Skip Johns, former associate director for technology at OSTP, who wanted the database as a tool to improve coordination of the government's $70 billion research and development programmes. But Johns left OSTP in 1996, and officials there are now said to be unsure of its practical value.

According to Paul Herer, a planning adviser at NSF, RaDiUS lacks strong support from potential users in the government. The system suffers, he says, from receiving incomplete data on the activities of some agencies, and from being “unfamiliar” to possible users. “It's ahead of its time a little bit,” Herer says.

But Holly Gwinn, chief of staff at OSTP, says that while “nobody considers RaDiUS to be a perfect system”, its shortcomings would be “addressable” if the government decided that the database was worth having. “We've been working with OMB to see what the value added is,” she says. “We haven't reached a consensus on this.”