The headlines say it all.

"$85-million U.S. missile test goes nowhere," trumpeted the Detroit Free Press. "Intercept missile test fails at launch," added The Associated Press. It's enough to make a long-time supporter of missile defense think twice, right?

Wrong.

Obviously, the scrubbed test on Dec. 15 wasn't good news. But it certainly wasn't the end of the program. Nor should it be. The truth is, we can learn as much when things don't go as planned as when they do.

Franklin Roosevelt could have been speaking about the early stages of missile defense when he said of his banking policies, "I have no expectation of making a hit every time I come to bat." Or as Thomas Edison said of his efforts to invent the light bulb, "I never failed once. It just happened to be a 2,000-step process." Well, building a missile defense system is at least a 2,000-step process. And the recent test actually was one of the early steps.

In fact, this particular test wasn't necessarily supposed to demonstrate that a test missile could hit a target missile. We already know that can be done.

In five out of eight tests, an interceptor missile has knocked down a test missile. In the most recent successful test, the target destroyed was some 140 miles above Earth and the missiles collided at about 15,000 miles per hour. At least 17 similar tests are planned over the next few years.

The main goal of the recent failed test was to collect information about the interceptor missile. We'll learn from the failure: Why did the computer shut down the interceptor? What can we do to make sure it will launch next time? Experts already are working on these questions.

And there will be a next time. Because the interceptor missile never launched, it can be re-used. So scientists will learn from studying what happened this time, and then learn more from observing the next test launch.

Also, let's keep in mind that these "mid-phase intercepts" are among the most difficult challenges any missile-defense system will have to face. Scientists are attempting to bring down an attack missile that has already left the planet's atmosphere. By that time, it's at full speed and isn't generating much heat, so it's difficult to target. To destroy it, scientists must literally "hit a bullet with a bullet."

In order to make sure our homeland is as safe as possible, Congress also should press ahead with research into what's called "boost-phase" programs. These programs aim to shoot down an enemy missile at the beginning of its flight, rather than waiting until it's on the way back down. The boost phase is generally the best time to attack a missile, because it's moving fairly slowly and generating plenty of heat as it blasts off.

The best way to keep us safe from missile attack is to put in place a layered missile defense, where the ground-based inceptors already in place at Fort Greely, Alaska are combined with boost-phase inceptors on naval ships and end-phase ("terminal") defenses. That will give our military several chances to hit an enemy missile.

To be honest, we'd all like to see every test go perfectly as planned. But in the real world, that never happens. Even years from now, when we have a complete system up and running, there's a slight possibility that a missile will somehow get through the shield. But at least we'll have done everything possible to shoot it down.

Building a missile defense shield is a long-term project, one that's going to involve many successes and occasional setbacks. But the goal is critical: to protect our cities and our people from a potential attack.

So let's ignore the negative headlines -- and press on to success.

Ed Feulner, president of the Heritage Foundation (heritage.org), a conservative think tank based in Washington.