the University of Pennsylvania, returned to the issue of funding balance. He said he had spent 30 years in industry and then 10 years in two major research universities. He said he has a great deal of confidence in the Obama Administration and believes it “gets it” in terms of innovation. “But I am absolutely mystified by the balance of funding,” Dr. Cleare said. “I also believe that good funding of basic research is essential.” The reality in industry, though, is that 1 in 10 inventions gets fully developed.

Dr. Cleare said the federal government should be doing much more than the SBIR program, which supports proof of concept. He said even the SBIR program focuses too little on development, he said. The linear model of innovation championed by Vannever Bush “is broken because of a bloody great gap in it called the Valley of Death.” He said the University of Pennsylvania produces hundreds of inventions, and many discoveries “don’t get the attention they deserve because there is not enough funding.” There is peer-reviewed funding for research, but not for proof of concept to take an idea through the stage of validation “to the point where the private sector feels the risk is justified to take it on. “So why, oh why, don’t we with the stroke of a pen” increase funding for applications? Dr. Cleare asked. What the United Kingdom claims to do better, Mr. Cleare added, is at the proof-of-concept stage. “Nearly all proof-of-concept funding is from the government,” he said. “The government gets it in England.”

Dr. Stevens noted that the American Association of Universities, which represents university presidents, is opposed to the idea of expanding the use of SBIR funds. He said he would prefer to see a new stream of funding so that money for product development is not taken out of the basic research pie, “which is what we have done up to now.”

Mr. Melissaratos said that maybe it is better to get rid of some existing streams of funding. Basic and applied research gets done in about every department of the federal government. “It is not well coordinated across departments,” he said. A presidential science advisor could help the departments of Defense, Energy, and Health and Human Services share their research. Companies with large R&D budgets, like Intel and Microsoft, should work more with federal agencies and fill gaps. “We need to really take advantage of what we’ve got in this country,” Mr. Mellisarratos said. “We need to break down some of these bureaucracies because they exist for their own sake. We need a radical re-do of how this money flows.”

Mark McDougal of Sematech commented that the United States could make more progress working with industry on innovation if there were more technology roadmaps. When corporations like Cisco and Intel “are hitting on all cylinders, they are working on three generations at the same time. They are de-bugging their current product; they are working on next-generation technology; and they are working on research. They are like well-oiled machines,” Mr. McDougal said. “When they look at universities and government, they see bureaucracy, inefficiency, and maybe not something they want to invest with.” He said the United

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement