This story appeared on Network World at

http://www.networkworld.com/columnists/2005/041105bradner.html

'Net Insider

Forgetting the public future

By Scott Bradner, Network World, 04/11/05

Scott Bradner

One of the historically most important forces in technological innovation apparently has decided to focus on the short term at the expense of the future. The New York Times felt that this change was important enough for a front-page mention and to dedicate a quarter of the first page of the business section to the story.

The Defense Advanced Research Projects Agency (DARPA), according to the enabling memo, was established "to engage in such advanced projects, essential to the Defense Department" responsibilities in the field of basic applied research and development which pertain to weapon systems and military requirements." While its name has alternated between ARPA and DARPA a few times, it has been a steadfast supporter of cutting-edge (and beyond) research in a variety of fields since being established in 1958. Most importantly, DARPA program managers have maintained a flexible definition of what might pertain to weapon systems and military requirements.

I first heard about DARPA in the early 1960s when I was working at Information International and have been the recipient of a very small amount of the more than $65 billion that DARPA has spent over the years. The DARPA-funded research that might be most familiar to the readers of this column is in the area of packet networks, including the ARPANet, generally considered to be the start of the Internet. But DARPA funds research in many other areas, some of which are mentioned in the 1997 "Transitions" document, which covers research that has transitioned from research to actual use. It would not be reaching much to say that most of what we currently recognize in the areas of computing and data networking is a result, at least in part, of DARPA-funded research.

But things are changing. The Times report, headlined "A blow to computer science research," notes that DARPA has been reducing the amount of money it is spending on "blue sky" research at universities and shifting those funds to focus on short-term and classified work with a closer tie to "military requirements." DARPA says that the shift is a result of events such as the Sept. 11 terrorist attacks and an increased reliance on corporate research. DARPA also has instituted requirements that research projects only employ U.S. citizens.

These changes are, at best, very shortsighted. It's fine to spend money to have corporations do classified, U.S. citizen-only research, but if DARPA continues to cut off support for open basic research at universities, our technical and economic future is threatened. And by "our" I mean the U.S., not just universities. Alternate funding for such research is limited. For example, funding for the National Science Foundation has been essentially flat, and NSF's grant review process doesn't encourage support for out-of-the-ordinary ideas.

This is not a good time for people who are worried about maintaining the U.S. edge in technological development. Fewer students are interested in these areas (for example, only 1% of the students accepted into Harvard for next year said they were interested in computer science). And with less federal grant money, there will be fewer funds to support graduate students. That will result in less research and fewer advanced degrees, which will result in fewer teachers for tomorrow's students.

Governments are the only ones able to fund research that might fail. Start-ups cannot afford to do so. If only safe research gets funded, we will retreat from the leading edge.

Disclaimer: Harvard is basically an experiment for most incoming students, and few fail. While others at Harvard share my opinion, the above lament was created without any input from the university.