At a tiny research facility in freezing-cold Rome, New York, U.S. Air Force scientists cut the ribbon on a supercomputer named the Condor Cluster.
In December 2010, at a tiny research facility in freezing-cold Rome, New York, U.S. Air Force scientists cut the ribbon on a supercomputer named the Condor Cluster. Operating at 500 TFLOPS—trillion floating-point operations per second—it was the fastest supercomputer in the entire Department of Defense and the thirty-third fastest in the world. Contributing to its awesomeness was the price tag—the Air Force only paid 10 percent of what it would cost for a comparable supercomputer. On top of that, the Condor Cluster uses a mere 10 percent of the electricity typically required by similar machines, which means it’s got a smaller carbon footprint and is less expensive to operate. Not a bad day’s work.
One other interesting fact about the Condor Cluster: It was built out of 1,760 Sony PlayStations. True story.
In other parts of the Department of Defense, results were decidedly less awesome. Just three months earlier, in September 2010, the U.S. Supreme Court agreed to hear arguments related to the U.S. Navy’s A-12 Avenger airplane, known to some as the Flying Dorito because of its triangular shape. Begun in 1983 and envisioned as an all-weather, carrier-based stealth bomber/attack jet, the Avenger program was terminated in 1991, at which point the Navy had spent $2 billion but received nothing for its money beyond a really exciting case study in failure for students at the Defense Acquisition University.
After almost 20 years of litigation, the nation’s highest court was finally going to consider whether the government’s decision to cancel this bloated project was justified. Resolution was apparently not to be, and in May 2011 the court returned the case to the lower appeals court instead of deciding, so the saga continued.
We could spend all day looking at similar examples from all the military services, highlighting the ups and downs of defense technology. The Army, Navy, Marine Corps and Air Force all have their own stories of critical new gear being delivered in a matter of weeks, right alongside stories where billions and decades are spent to deliver exactly nothing. Of course, these highs and lows are not limited to the military. NASA has some of the best failure and success stories around. For that matter, the federal government does not have a monopoly here.
Why do some programs deliver their product under budget, while others see their costs expand by orders of magnitude? Why do some deliver ahead of schedule, while others experience endless delay after endless delay? And, most critically, which products work better—the quick and thrifty, or the slow and expensive? Which situation leads to superior equipment?
After a few years of conducting informal research into these questions, I spent 18 months at the Air Force Institute of Technology looking at them more rigorously. The pattern that emerged is this: The most successful project leaders from government and industry alike tend to deliver top-shelf stuff with a skeleton crew, a shoestring budget and a cannonball schedule. In interviews I read and those I personally conducted, project leaders continually echoed one theme: “We had no time and no money. We were just lucky to have a small team of really creative, dedicated people and we got it done.”
In contrast, project leaders who are cursed with large budgets, large teams and long schedules generally have a difficult time delivering even a fraction of the promised capability, an outcome often blamed on an excessively cumbersome process. Interestingly, when faced with cancellation due to severe cost overruns and delays, these leaders typically respond, “If I had a little more time and money, I could fix this.”
Yes, those who had the largest budgets were most likely to ask for more money and least likely to deliver an actual working product. Those with the smallest budgets were most likely to have cash left over after delivering 10 pounds of awesome on a five-pound purse. The faster, cheaper stuff also tends to perform better in actual use than the slower, more expensive stuff. The idea that spending less time and money leads to better outcomes sounds a bit like claiming that moderate amounts of red wine and dark chocolate are good for you. Surely this is too good to be true. And yet, as with the aforementioned health benefits, the data are compelling.
We haven’t said much about complexity yet, so let’s remedy that right now. Successful project leaders tend to place a premium on simplicity in their organizations, processes, documentation and technologies. They tend to view simplicity as a desirable attribute and pursue opportunities to simplify when they are able.
FIRE codifies the practices, principles and tools used by some of the best technology developers in the world—people who sent spacecraft on intercept courses with asteroids or who built fighter planes that dominated the skies of World War II. FIRE also describes the way clever toy designers teach science lessons that are actually fun.
One final note: While improving one’s process might be a fun way to spend a sunny July afternoon, FIRE is emphatically not a process-improvement initiative. Given the modern popularity of process-improvement methodologies, that last sentiment bears repeating: FIRE is not about process improvement.
The primary objective is to improve our objectives and outcomes rather than our processes. There is a tremendous difference between them. Clever project leaders should certainly make an effort to streamline, simplify and accelerate their processes, but the bulk of their attention is rightfully spent on the product itself and on taking care of the people who make it.
Dan Ward served in the U.S. Air Force for more than 20 years, where he specialized in leading high-speed, low-cost technology development programs. In 2012, he received the Bronze Star for his service in Afghanistan. After retiring at the rank of lieutenant colonel, he launched Dan Ward Consulting LLC to help clients reduce the cost, time and complexity of delivering world-class innovations. He is the author of two books. Learn more about Dan at thedanward.com or follow him on Twitter @thedanward.