By Stephen Barr
Monday, April 7, 2008
The best program managers are "freewheeling zealots" with big ideas. The staff has been called "100 geniuses connected by a travel agent." And the boss describes his agency as a home for "radical innovation."
It's DARPA -- the Defense Advanced Research Projects Agency, which is celebrating its 50th anniversary with a dinner for 1,700 alumni, friends and partners Thursday night in Washington.
From its beginning, the Defense Department agency has looked worldwide for fundamental scientific and technology discoveries ready for conversion into a blockbuster asset for the military.
"DARPA will take a chance on an idea with no data. We'll put up the money to go get the data and see if the idea holds," said Anthony J. Tether, the agency director.
"That is the highest-risk type of research you can have."
Small and secretive, DARPA has compiled a number of impressive achievements in the past 50 years. It pulled together researchers who created the blueprint for the Internet. It sponsored the inventor of the computer mouse (the first was carved from wood and had one button).
It developed the Saturn rocket engine program that allowed the nation to go to the moon. It came up with the technologies that have made possible stealth fighters and bombers, precision munitions and the pilot-less Predator planes used in Iraq and Afghanistan.
Like many government initiatives, DARPA was born out of a crisis.
The Soviets launched the satellite Sputnik in 1957, beating the United States into space. At the direction of President Dwight D. Eisenhower, DARPA opened for business the next year, focused on helping guard the nation against technological surprises. The agency's mission has been evolving ever since, and today DARPA also works to create its own technological surprises that permit the U.S. military to overwhelm adversaries.
Unlike most federal agencies, DARPA operates with little red tape. It has only two management layers, encouraging the rapid flow of ideas and decisions.
About 240 people work at DARPA, and 120 of them are program managers and office directors on appointments of four to six years. The agency does not own or operate labs, but sponsors research carried out by industry and universities.
By rotating technical professionals every few years, DARPA has "a constant freshness of people and energy," Tether said. "Everything else we do stems from that."
One of those short-term managers returning for Thursday's anniversary dinner is Lawrence G. Roberts, who led a DARPA team that designed a network that evolved into the Internet. He made some of the key decisions in 1967, when he was 30. As Roberts described it, "Putting A and B together and getting Z. Taking obscure things and seeing there is an intersection there."
He hopes that DARPA will always be able to focus on innovation -- "working on something that should change the country and generate the economy shift that the Internet did."
Some of DARPA's current projects may hold that potential.
Researchers are working on a two-way speech translation system that would permit soldiers to go anywhere in the world and understand the people around them. The idea, Tether said, is to create a miniature headset that would immediately translate a foreign language into English and feed it to an earpiece. In turn, a reply by an English speaker would be converted into the appropriate language and broadcast from small speakers on the headset.
When the technology is perfected, "the world will become a safer place. People will be able to talk to one another and understand one another," Tether said.
Another project looks for ways to restore severely injured soldiers. Researchers are trying to develop a prosthetic arm and hand that can be directly controlled by the brain and used as a natural limb, with dexterity and sensations. Prototypes are in development, Tether said, and hold promise that disabled soldiers can stay in the military "and contribute as before" rather than be discharged.
DARPA conducts research in almost every field -- biology, microelectronics, satellites, unmanned cars and aircraft. "We are extraordinarily broad. If you can think of it, we're doing it," Tether said.
Of course, numerous projects are classified because they may have a useful military application or because DARPA does not want the world to know everything it is doing.
The government always will need a place to test and finance big ideas, Tether said. "The 50 years of history proves it has been well worth it, and I have to believe that in the next 50 years DARPA will come out with technological advances that will stagger even my imagination."
Stephen Barr's e-mail address firstname.lastname@example.org.