Research money crunch in the U.S.

At a symposium here last week put on by the Marconi Society, technology researchers and scientists gathered to honor two of colleagues: Gordon Moore, co-founder of Intel, and Claude Berrou, co-inventor of turbo codes, which are used in 3G mobile telephone standards.

At the two-day event, attendees voiced concern over the state of technology research in the United States.

"I think we are in trouble," said Leonard Kleinrock, professor of computer science at the University of California at Los Angeles and creator of the basic principle of packet switching. "Years ago, people took a long-range view to research. There was high-risk research with the potential for big payoffs. That's no longer the case."

Videos:

Federico Faggin's
chip shot
Building the world's first microprocessor.

The Web we fund?
Leonard Kleinrock on following the money.

Putting research funds to work
Inventor Robert Lucky on 21st century realities.

What's driving R&D?
Inventor Claude Berrou on market influences.

For much of the 20th century, major breakthroughs in technology came from large research laboratories like AT&T's Bell Laboratories, Xerox's Palo Alto Research Center (PARC) , and IBM's Watson Research Center.

These research facilities operated much like national laboratories, making their discoveries and innovations available to anyone for modest license fees. Many of the inventions and discoveries at Bell Labs, for example, were first used commercially outside the Bell system and benefited the nation as a whole.

The labs are still around, but some experts say the labs conduct basic research on a much smaller scale than they used to.

In the early 1960s, the U.S. government started pouring money into information technology and communications research. It formed the Defense Advanced Research Projects Agency to fund high-risk, high-reward research. One of the greatest developments to come out of this research was the Internet, which started out as a research project to develop a communications network for the U.S. military.

For years following the creation of DARPA, innovation in U.S. communications technology grew substantially under government-funded research. Today, DARPA and the National Science Foundation fund a large portion of the academic IT research in this country, say research experts.

Flat U.S. investment
But some of the engineering legends attending the Marconi Society event say the nation is facing a crisis, as industry-run research withers and government spending is slashed. Much of today's most important and long-range research is moving to Asia and Europe, which could have a significant impact on the U.S. economy as well as national security, they argue. The reason appears simple: The U.S. government is not increasing its investments in science.

Federal spending on scientific research has remained flat for several years. President Bush's budget request for 2006 proposes spending $132 billion on scientific research, which is roughly the same as the previous year, according to the nonprofit American Association for the Advancement of Science. A large proportion of the available funds have been allocated to science related to defense and counterterrorism, the AAAS has said.

"The government isn't stepping up to the plate," said Robert Lucky, Marconi Society chairman and former director of Bell Laboratories. "We're eating our seed corn."

Others in the technology community have echoed these sentiments. Last month, the National Academies, a group of institutions established to provide Congress with advice on science and health policy, urged the U.S. government

Many in the research community also believe that the research being conducted today is too focused on short-term, market-oriented results. The current DARPA policy, which mandates 12-month "go, no go" research milestones for information technology, has shortened deadlines, thus discouraging long-term research. And with more research focused on national security, programs formerly open to academics are now classified. DARPA has also slashed spending on academic research.

"Traditionally funding in computer sciences has come from the U.S. government," Kleinrock said. "And it's contributed to some remarkable advances, such as the Internet and artificial intelligence. They (the government) used to step back and with some direction let you go develop something new. But that's not the case today. And DARPA is no longer thinking long-range."

More competition, fewer dollars
The effects have been significant. In the last five years, IT proposals to the National Science Foundation jumped from 2,000 to 6,500, forcing the agency to leave many proposals unfunded. Other agencies, such as NASA, have also reduced spending on communications research. Since most government funding comes only from these two sources, researchers are flocking toward the NSF as DARPA cuts back or changes its priorities.

"There is much more competition for far less money," Kleinrock said. "And the result is researchers are spending more time writing proposals that never get funded...We're losing our leadership in engineering. And that will have an effect on the economy."

Private industry isn't expected to pick up the slack where the government falls short, experts say. For the past several decades, the high-tech industry has become increasingly dependent on government-funded research partnerships with academic institutions to spur innovations. Companies such as Cisco Systems partner with researchers in academia, rather than operating their own large-scale research labs focused on long-range issues. Only a handful, like IBM, Sun Microsystems and Microsoft, maintain large and growing research arms.

"The competitive nature of these industries means they can no longer fund research that offers benefits for humanity," when the research has no immediate return in investment for the company, Lucky said.

But some researchers say they understand the tough decisions the companies and politicians have to make in choosing how much they allocate to science and how they actually spend their research money.

"There are trade-offs," said Robert Kahn, a co-developer of the TCP/IP protocols used to transmit traffic across the Internet, and CEO of the Corporation for National Research Initiatives. "Do you spend one penny on R&D, or a dollar on national security? These are the tough choices that have to be made."