Without question, a large part of JPL's future will be the Red Planet. Elachi has stated the goal of establishing a permanent robotic presence on the surface and in orbit around Mars in preparation for possible human missions. At least one spacecraft will be planned for launch every 26 months, when the relative positions of Earth and Mars are favorable.

Europa, a moon of Jupiter expected to contain a vast ocean that might harbor life, is also on Elachi's agenda, although funds have not been committed to a specific mission.

Other scientists say a Pluto mission must be funded now, because in addition to planning and building a spacecraft, it will take 15 years or more to make the trip, they say. And since the tiny planet is currently moving farther from the Sun in its 248-year-long orbit, it will over the next few decades enter into a lengthy winter, when any possible wisp of an atmosphere will freeze and therefore make the small planet less interesting to explore.

Elachi disagrees. He says JPL and NASA have agreed to further develop an electric propulsion system, known as ion propulsion, used already on the Deep Space 1 craft, so it will be ready for a trip to Pluto whenever additional funding is available from Congress.

"With electric propulsion, you can go to Pluto any year," and make the trip "in about 10 years," he said.

In truth, Elachi does have an overarching goal. All of these missions will be pursued "in the context of shedding light on the evolution of life ... in ours and neighboring solar systems." This primary goal, he says, the understanding of life and its evolution, is to be the glue that holds JPL employees together.

"I think it creates a lot of excitement and a sense of purpose."

And there is another guiding principle in choosing missions. Elachi has no desire to do anything easy, anything that some other technology house could do.

"Our job is to move the frontier," Elachi said. "We are going to pick the [missions] which are really difficult to do, almost at the edge of impossible."

Which means Elachi is willing to take risks. And, likely, he is prepared to accept failure.

"As a rule of thumb, you expect to lose one or two [robotic] missions out of 10," says Howard McCurdy, a NASA historian and professor at American University. McCurdy says NASA has never formalized their expectation for losses, but NASA director Dan Goldin has said that out of a dozen launches of robotic spacecraft, you can expect to lose one or two.

The agency has always known that space exploration is a gamble. Critical missions have been sent out in pairs, as with the Voyager spacecraft (Elachi was a project scientist on Voyager). The idea, McCurdy says, is that if the chance of losing any mission is 10 percent, then the odds go down to 1 percent by launching twins.

So to succeed in rewriting the rulebook and reshaping JPL, Elachi needs to be unafraid of risk, McCurdy said.

"It requires a lot of pyrotechnics and parachutes and solid rocket boosters and altimeters. It's not a benign technology. You don't just pop a bag around a spacecraft and have a smooth landing. The mission can be ruined if they can't deflate the bag. Or if the bag deflates prematurely. Or if there's a problem in the software code. Or if the spacecraft is too heavy -- bags are most effective for light spacecraft."

Krimigis, head of the Applied Physics Laboratory, says Elachi has a "very difficult task in reshaping the laboratory, and I certainly wish him good luck, because his success is going to be the nation's success. We'll all gain from that."

And Krimigis has some advice for any organization trying to restructure and emerge from failure, and any organization that wants to succeed as part of the space program: He says you must be innovative and cost effective, as anyone seeking a NASA contract knows. "And in so doing we have to take prudent risks."