Airbus believed that larger airplanes were needed to connect congested airports in the world's largest cities. Boeing executives weren't so sure.
They believed airline passengers would pay a premium to avoid those same congested hubs with long nonstop flights between smaller cities. Now they just needed to develop a plane that would somehow make such trips economical.
It had been 13 years since Boeing started development of a new plane, the 777. The company had recently scrapped two other major projects: a larger version of the 747 and the Sonic Cruiser, a plane that would fly close to the speed of sound.
A development team with a knack for assigning new planes code names based on national parks had just the thing: Project Yellowstone.
The plane — eventually rechristened the Dreamliner after a naming contest — was unlike anything else previously proposed.
Half of its structure would be made of plastics reinforced with carbon fiber, a composite material that is both lighter and stronger than aluminum. In another first, the plane would rely on rechargeable lithium-ion batteries to start its auxiliary power unit, which provides power on the ground or if the main engines quit.
While other planes divert hot air from the engines through internal ducts to power some functions, the 787 uses electricity. Getting rid of that air-duct system is one thing that makes the plane more fuel efficient.
There were also benefits for passengers. The plane's extra strength allowed for larger windows and a more comfortable cabin pressure. Because composites can't corrode like aluminum, the humidity in the cabin could be as much as 16 percent, double that of a typical aircraft. That meant fewer dry throats and stuffy noses.
Before a single aircraft was built, the plane was an instant hit, becoming the fastest-selling new jet in history. Advance orders were placed for more than 800 planes. Boeing seemed to be on its way back.
"Employees knew this was going to be a game changer, and they were stoked that the company was taking the risk to do something big," said Michael Cook, who spent 17 years as a computer developer at Boeing.
But this was no longer the trailblazing, risk-taking Boeing of a generation earlier. The company had acquired rival McDonnell Douglas in 1997. Many McDonnell Douglas executives held leadership positions in the new company. The joke was that McDonnell Douglas used Boeing's money to buy Boeing.
The 707 and 747 were blockbuster bets that nearly ruined the company before paying off. McDonnell Douglas executives didn't have the same appetite for gambling.
It was a tough sales job for Alan Mulally, then head of Boeing's commercial airplanes division and current CEO of Ford. The only way the board of directors would sign off on the Dreamliner was to spread the risk among a global chain of suppliers. In December 2003, they agreed to take on half of the estimated $10 billion development cost.
The plan backfired as production problems quickly surfaced.
"I saw total chaos. Boeing bit off more than it could chew," said Larry Caracciolo, an engineer who spent three years managing 787 supplier quality.
First, there were problems with the molding of the new plastics. Then parts made by different suppliers didn't fit properly. For instance, the nose-and-cockpit section was out of alignment with the rest of the plane, leaving a 0.3-inch gap.
By giving up control of its supply chain, Boeing had lost the ability to oversee each step of production. Problems sometimes weren't discovered until the parts came together at its Everett, Wash., plant.
Fixes weren't easy, and cultures among the suppliers often clashed.
"It seemed like the Italians only worked three days a week. They were always on vacation. And the Japanese, they worked six days a week," said Jack Al-Kahwati, a former Boeing structural weight engineer.