It was the Saturday before Labor Day, 1969. Len Kleinrock and a throng of shaggy-haired grad students were pacing a loading dock on the campus of the University of California at Los Angeles, eagerly awaiting the arrival of their new baby. One student cradled a bottle of champagne.
Before long it came: a 900-pound crate fresh off the plane from Boston. The group gingerly unpacked its contents, revealing a gunmetal gray computer the size of a Coke machine. It had four steel eyebolts welded to its lid so it could be lifted by crane or helicopter. A nameplate on the machine read: "Interface Message Processor." The group dubbed it "IMP."
If not for an unexpected turn of events, what happened next might have been celebrated only by the handful of engineers in attendance. Instead, 30 years ago this past week, it quietly made history.
Kleinrock and his team wheeled IMP to the third floor of the UCLA engineering department at Boelter Hall and cabled it to a nearby computer. Within an hour, bits of data were trickling back and forth.
"That was the first breath of life the Internet ever took," says Kleinrock, then as now a UCLA professor of computer science and one of the network's early pioneers.
If you want to pinpoint the birthplace of the Internet, Room 3400 of UCLA's Boelter Hall is as good a place as any to start. The IMP became the first link of an experimental computer network being developed by the Defense Department's Advance Research Projects Agency, or ARPA. Over the years this network -- known as ARPANET -- gradually evolved from an obscure resource known only to a handful of military and university researchers into a network of networks used by millions around the world.
But that was still many years in the future. In 1969, Len Kleinrock's labors on IMP went virtually unnoticed.
"No one knew it was going to be important," says historian Janet Abbate of the Unversity of Maryland, College Park, and author of the new book, "Inventing the Internet" (MIT Press).
Before IMP arrived, UCLA decided to mark the historic moment by circulating a press release: "UCLA To Be First Station In Nationwide Computer Network." Not one reporter called. "We didn't even have a camera," laments Kleinrock.
But a month after UCLA broke ground on ARPANET, researchers hooked up a second computer a few hundred miles up the California coast at the Stanford Research Institute and attempted to send the first real message across cyberspace. Posterity once again was far from their minds. Rather than tapping out, "What hath god wrought?" or something else equally poetic, the researchers simply tapped: "L-O-G-I-N."
In any case, the message never made it. Upon receiving "G" the Stanford computer choked and died.
Eventually researchers worked out the kinks. By 1972, the ARPANET was installed at 15 universities and research institutes around the country, stretching from UCLA to the Massachusetts Institute of Technology in Boston.
At first, nobody knew quite what to do with the network -- or how to work it. Terminals were hard to find. There was little in the way of instruction manuals or technical support. As a result, few people outside a small band of computer scientists even bothered with it.
One of the first non-computerniks to gain access in those years was Michael Hart, who in 1971 posted the Declaration of Independence to the ARPANET from a terminal at the University of Illinois.
"You have to realize how few people were on the Net before the '80s," Hart recalls in "Inventing the Internet." "It was boring."
Gradually, however, interesting things began to show up online: an air-traffic simulator, a chess game, a system for displaying Chinese characters and a computer psychiatrist named "Eliza."
But it wouldn't be until 1972 when a quiet engineer named Ray Tomlinson wrote the software that would cement the future of the fledgling network: e-mail.
The software spread like wildfire. A year after e-mail was invented, a study of the ARPANET showed it made up 75 percent of traffic on the network. It was being used for jokes, love letters -- there were even rumors it had figured in a drug deal or two.
That's when it began to dawn on a few ARPANET researchers that their network was destined for bigger and better things.
"I realized it was the people-to-people communication and not the machine-to-machine communication that really mattered," says Kleinrock.
The history of the Internet also has its share of missed opportunities. In the early 1970s, for instance, AT&T; passed up a chance to buy the computer network from the government, who worried that managing the network would grow too taxing. In the 1980s the first personal computers hit the scene and traffic on the ARPANET began to explode. Other local computer networks were springing up like mad and being linked to the ARPANET like bulbs on a string of Christmas tree lights. By the late 1980s the ARPANET was showing signs of strain. When Tim Berners-Lee, a shy British computer scientist working at the CERN particle physics laboratory in Switzerland, invented the World Wide Web in 1990, it sealed the fate of the rickety old network. In February of that year, the ARPANET was formally decommissioned.
Len Kleinrock, however, is still right where he was in 1969. He still has an office at Boelter Hall; on his desktop is the successor to the network he helped build 30 years ago. Just down the hall from him you can also find IMP -- now a permanent museum piece.
Recently, Kleinrock says he dug out a copy of the 1969 UCLA press release announcing the ARPANET. In it, he was surprised to find he had spoken these words:
"As of now, computer networks are still in their infancy. But as they grow up and become more sophisticated, we will probably see the spread of computer utilities, which like present electric and telephone utilities, will service individual homes and offices around the country."
Three decades later, that extravagant prediction seems almost timid.