To spur real technological advancement, the Obama Administration should invest in improving how scientific research is analyzed, validated, and distributed
After seven months in office, the Obama Administration is starting to show the first hard evidence of how it will deal with scientific and technological policy. The good news: It looks like the new President will indeed follow his campaign rhetoric and regularly seek advice from scientists and engineers. But the evidence also gives a less certain picture of how the government will support innovative research. If anything, the record shows the Administration is thinking in fairly conventional terms about supporting laboratories and universities, without much consideration of the ways research has changed over the past decade.
The Administration's new science adviser, John Holdren, was quick to assure scientists that President Obama will look to them for help in solving the great issues of our age: the economic recession, the challenges with energy and the environment, the costs and effectiveness of health care, the needs of national security. He speaks of smart grids and new treatments and novel technologies. Indeed, the Administration has backed up those assurances. Consider the staff assembled under Energy Secretary Steven Chu, a Nobel Prize-winning physicist.
Yet taking advice from scientists is not the same thing as supporting research or helping new results move to the marketplace. Alice Gast, chemical engineer and president of Lehigh University, made the point recently in the journal Science. "Transformative change requires a long-term investment in the nation's intellectual infrastructure," she wrote in an issue published in March.
Doubling the NIH Budget
And investment for the long term requires more than simply throwing money at a problem. In the modern research environment, even well-intentioned governments fail to make good investments in scientific and technological infrastructure. At the end of the Clinton Administration, an effort to double the budget of the National Institutes of Health (NIH) created major structural problems in the nation's biomedical research community by deploying funds in ways that encouraged education of new scientists without thinking about where those new researchers would work or how their research would be funded.
As a result, a generation of young biomedical scientists found themselves working at small regional institutions without a history of supporting research. Many of these scientists found they could start a research program only by competing with the large, sophisticated research universities that had educated them. Quickly, grant officers found they could no longer fund all of the promising young scientists who asked for research support and still keep the large educational programs in operation. "NIH research funding is more difficult to get now than it was before the NIH budget doubled," Michael Teitelbaum of the Alfred P. Sloan Foundation observed in Science in August 2008.
The current state of American science cannot entirely be blamed on misguided government policies any more than it can be ascribed to poor planning by the scientists. Many of the problems with scientific research come from the uncertainty of the scientific process itself and the inability of both researchers and policymakers to predict the consequences of policy decisions.
Impact of the Web
And one of the biggest problems facing the research community, the perilous state of scientific publishing, traces its roots to yet another source: the rise of the Internet. The Internet has been highly popular with researchers. During the 1980s, scientists and engineers used networks to gain access to large, high-speed computers supported by the U.S. government. Researchers also began using networks for sharing data, distributing computer programs, and circulating drafts of papers.
More recently, the impact of the Internet has been felt in other ways, not all of them positive. It has changed the nature of popular publication, providing advertisers with an inexpensive way to reach an audience and taking income from traditional magazines and newspapers. At the same time, the Internet has not provided enough revenue to support the editorial activities of such periodicals. Newspapers can't keep foreign correspondents, and magazines cannot hire high-priced authors.
Similarly, the Internet has been undermining the health of scientific publications. These publications are important not only because they broadcast scientific results, but also because they form a community that validates scientific work. Scientific periodicals and the professional societies that support them also form the foundation for future inquiry. They identify the trends of research and the most promising paths that require more investigation. Furthermore, the editors and authors from these journals shape national plans for research. They sit on scientific advisory committees for every branch of government and evaluate grant proposals for the National Science Foundation, the National Institutes of Health, the Defense Dept., and every other agency that funds scientific work.
"Not only are the traditional business models of scientific publication in jeopardy, but also the integrity and very usefulness of scientific journals are becoming problematic," says Sorel Reisman, vice-president for publications at the IEEE Computer Society. The Internet has not only threatened the financial plans of these journals but has also made it far too easy to corrupt the scientific literature with fraudulent papers.
Reisman has seen a steady rise in papers with conclusions based on manipulated data and forged results rather than real research. When unexposed, these papers remain in the scientific record and can confuse a researcher who attempts to build upon their erroneous findings.
What does all this have to do with the Obama Administration? As scientific publication is a fundamental part of the research infrastructure, it deserves the same kind of attention given to laboratories. It could be improved by a wide variety of investments, just as a corporation can benefit from investment in its business processes.
A Specific Proposal
American research would be improved by a central repository for scientific articles, especially those articles supported by government grants. It could be improved by research on how to categorize results, identify the provenance of data, and check the validity of results. It would benefit from more efficient means of distributing results and getting scientists to talk in public about their research.
To most scientists, the Obama Administration has brought a welcome promise for science policy. At the same time, they are waiting to see whether the new President will invest in science and build an infrastructure that advances the cause of scientific research. To actually improve the practice of science, those investments will require careful thought. We cannot merely throw money at science and expect that the result will be good. We will have to think carefully about what we will need for laboratories, for education, for grants, for equipment—and for publication.