But the web’s growing power was not limited to speed. Mercury Center’s audience grew, though how much it was hard to say: the company claimed in its 1995 annual report that traffic had doubled, but did not provide numbers; it was not until 1997 that it would report 1.2 million monthly visitors for Mercury Center. And Ryan had come to understand that numbers, both in traffic and in revenue, were flexible; it was not difficult, say, to claim a percentage of ad revenue for online “upsells.”

Mercury Center’s reputation was growing. In 1996, Editor & Publisher named Mercury Center the nation’s best newspaper on the web. Ryan, and the entire news organization, would soon learn just how powerful a tool they had—for better or worse.

In August 1996, The Mercury News published the first of three stories on an alleged connection between the US-backed Nicaraguan contras and the devastation wrought by the crack trade on the inner city. Gary Webb’s series, “Dark Alliance,” charged that drug money was funding the insurgency. Much as Bob Ingle had envisioned in 1990, the publication of the series was a joint effort by the print and digital sides—Mercury Center posted documents, as well as audio recordings from wiretaps and hearings. The response was electric; the story became the subject of talk radio, and of every conspiracy theorist who believed that the government was secretly behind the crack trade. “You don’t have to be The New York Times or The Washington Post to bust a national story anymore,” Webb said. Traffic boomed—Ryan estimated a 15 percent boost in readership. There were days when over a million people visited the site. “For us, it has certainly answered the question: Is there anyone out there?”

But the early excitement soon gave way to growing suspicions about the story, not necessarily the facts but the extent to which Webb had extrapolated upon what he had learned. The same national press whose ranks the Merc believed it had joined began running stories rebutting Webb’s charges, and suggesting, none too kindly, that the upstart Merc had overreached. At the paper, there was growing anger—not necessarily at Webb, but at the editors who had run a story whose core premise, on a closer reading, felt vastly oversold. Webb was reassigned; Ceppos, who’d been on medical leave, was compelled to address his restive staff and later write an explanatory public letter, backing away from the series.

Lost in the fallout and recriminations was how widely and quickly Mercury Center had spread the story. The experiment dismissed as “Bob Ingle’s train set” had displayed a power unlike anyone at the paper had ever seen. No one was then using the word “viral” because it had never happened before.

In 1995 and 1996, Knight Ridder began to jettison properties, among them its remaining cable television systems, as well as its Information Design Laboratory in Boulder, Colorado, a $900,000 annual investment that in 1992 began developing an early version of a tablet newspaper. The laboratory, in the view of its director, Roger Fidler, had been a Jim Batten project, an enthusiasm that Tony Ridder—and Bob Ingle—did not share. Far larger, and noticeable to investors, was the decision to sell Knight Ridder Financial, the business news service that was fighting with Reuters, Dow Jones, and, more recently, Bloomberg LP, to retain its share of the market. The service, which employed about 1,100 people, averaged about $200 million in annual revenues. But its profits, Ridder told the Times, were modest. Wall Street greeted the news by boosting the value of Knight Ridder stock by $3.38 a share. Ridder announced that the proceeds would go toward paying down debt, as well as further boosting the value of the stock with a buy-back program. The company would also buy more newspapers.

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.