The threat was not from below, as Ingle saw it, but from above, from the biggest player in the digital game, Microsoft. In 1996, Microsoft had launched Sidewalk, a network of city guides that, Ingle told the naa gathering, would eat into markets that newspaper websites had made their own.

He was not alone in that fear. In 1995, Knight Ridder had joined with the leading news organizations in the country—The New York Times Company, Tribune Company, and The Washington Post Company among them—in an alliance against the online threat called New Century Network. By 1997, however, the group was falling apart, riven by conflicting agendas. Ingle remained convinced that Microsoft had to be stopped, or at least stalled. If New Century Network could not find a way to work together, he reasoned, Knight Ridder would go it alone.

As it happened, Kathy Yates had come up with a plan to leverage the chain’s reach into a potential market for national advertising. The project would involve all of Knight Ridder’s papers. Microsoft may have had the national reach. But each Knight Ridder paper, the thinking went, knew—and, in effect, owned—its town.

“Our mission was to try to create a defensive property to help protect the core of the newspapers,” Yates would later say. “Real Cities was about a way to standardize what we were doing on the web.”

The Real Cities network would be Knight Ridder’s attempt to push back in a broad and ambitious way against the forces of disruption—forces that, as Ingle and Yates would soon discover, could turn smart and devoted people against each other.

David Demilo arrived at Knight Ridder Digital in 1997, recruited by Chris Jennewein to build something altogether alien at the company—a technological development unit that would operate not in the chaotic “daily miracle” of a newsroom, but in accordance with the strict rules of order that, the jeans-and-sneakers dress code aside, prevailed in the digital world.

Jennewein, who had helped build the early version of Mercury Center, had been dispatched by Ingle to those Knight Ridder newspapers that still needed to build websites—the first step in creating the new Real Cities network. DeMilo, who had worked on his college paper at Harvard and who had interned at the Herald when Ingle was still there—and who remembered what terrific holiday parties he threw—had been at Primedia, where he helped build booksonline.com, the firm’s first e-commerce site.

He was an outlier among his new colleagues. DeMilo liked newspapers, but had been away from them long enough to recognize that the qualities that made newspaper people most proud were regarded as secondary in the digital world. Chief among them was content. Newspaper people, he heard time and again, insisted that content was king. But in DeMilo’s world, content was not king. The platform was king. It was king because the platform reflected not the sensibilities of the people who created the content, but rather the people who used it. If it worked for them, they would use it.

Knight Ridder’s technology staff was a collection of consultants, many still in their twenties. DeMilo’s job was to create an organization that could standardize all the company’s websites so that information could be streamed into databases in a way that made them searchable; no national advertiser, the reasoning went, would be interested in a network in which each ad had to be formatted to fit an array of sites. This meant imposing clear and immutable lines from inception to launch but which, in the eyes of DeMilo’s new colleagues, sapped the creative process of its spontaneity. But last-minute changes of mind and direction were difficult to accomplish in the intricate, numerical language in which DeMilo worked: code.

Code, in DeMilo’s eyes, possessed an elegance that could bring digital ideas to life on a screen. He tried to explain how in the language of code, a decision on, say, a color or format could not be changed without additional costs and delays. DeMilo could only work within the parameters of the prevailing technology—JavaScript was still relatively primitive; XML did not yet exist—limitations that his colleagues could not seem to fathom: “They would say, ‘You’re being stubborn.’ ”

Michael Shapiro is a contributing editor to CJR and teaches at Columbia's Graduate School of Journalism. His most recent book is Bottom of the Ninth: Branch Rickey, Casey Stengel, and the Daring Scheme to Save Baseball From Itself.