In a letter (pdf) to John Marburger, Director of the White House Office of Science and Technology Policy, the Bay Area Science and Innovation Consortium — a group that includes representatives from IBM, HP, SIA, Lockheed-Martin and representatives from Bay Area universities and federal labs — urged the Adminstration to address concerns about federal support for fundamental research in IT. The letter makes a case that should be very familiar to readers of this blog — namely, that “at a time when the U.S. faces enormous challenges to its scientific and technological leadership, U.S. policy is headed in the wrong direction.”

For example, the Defense Advanced Research Projects Agency is reducing university participation by: (1) classifying research, even in broad, enabling areas such as embedded software for wireless networks; (2) focusing more on shorter-term deliverables, and dramatically reducing its traditional levels of investment in high-risk, high-return research; and (3) evaluating success of projects on one-year time-scales. Between 1999 and 2004, DARPA’s research funding at the top-ranked computer science departments (Berkeley, Carnegie Mellon University, MIT, and Stanford) declined by 38-54 percent. These trends are not limited to IT research, but are evident in a broad range of fields.

In fact, beyond just the top schools, the overall DARPA investment in university-led IT research has declined precipitously since FY 2001, falling from $199 million to $108 million in FY 2004 (in constant dollars).
The letter goes on to point out the burden placed on NSF as a result of DARPA’s “retrenchment,” noting the precipitous fall of proposal success rates and the impact that has on the peer-review process — it becomes more conservative, resulting in proposals that tend not to be as high-risk and potentially high-return as we need to be supporting to keep the U.S. at the cutting-edge of technological innovation.
BASIC makes two specific recommendations:

1. DARPA should be given a clear mandate to dramatically increase its support of high-risk, unclassified, university-based research.
…
2. The National Science Foundation should be given additional funding in the Administration’s FY 2007 budget for a “Pioneer Award” for IT research.

These ~$500k awards would be for “individual scientists of exceptional creativity who propose pioneering approaches to major contemporary challenges.” The coalition urges an immediate funding increase for NSF to fund at least 25-50 of these pioneer awards, with an eventual “steady state” of 100-150 awards.
It’s an interesting approach, and it makes essentially the same case we’ve been making about IT research — and many other groups have been making about the physical sciences and engineering generally. But the more groups that make this case — especially groups with significant industry membership like BASIC and the Task Force on the Future of American Innovation and the Council on Competitiveness and the American Electronics Association and the Telecommunication Industry Association and the Business Roundtable and many others, the harder it is for the Administration to ignore the message.
You can read the full letter here (pdf).