The internet's ancestor sent its first message in 1969. It was supposed to say "LOGIN." The system crashed after two letters, so the first word online was "LO."
At 10:30 p.m. Pacific time on October 29, 1969, UCLA student programmer Charley Kline sat at a computer in Boelter Hall and tried to log in to a computer at the Stanford Research Institute. It was the first host-to-host connection ever made on the ARPANET.
Kline began typing "login," but the SRI computer crashed after he typed just two characters. The first message ever sent over the network was "lo." About an hour later, after SRI programmer Bill Duvall made adjustments, Kline logged in successfully.
The project began in 1966 with frustration. ARPA official Robert Taylor had three separate terminals in his office, each wired to a different research computer with its own commands, and concluded "there ought to be one terminal that goes anywhere you want to go."
The popular story that the ARPANET was built to survive a nuclear attack is false. ARPA director Charles Herzfeld said it was not started for that purpose. The real reason was that only a few powerful research computers existed and the scientists who needed them were scattered across the country.
In January 1969, ARPA hired the firm Bolt Beranek & Newman to build the network's routers, called Interface Message Processors. Each was a ruggedized Honeywell DDP-516 with 24 kilobytes of memory, connected over 50-kilobit phone lines. A seven-person team designed and installed the system in just nine months.
By December 5, 1969, the network had four nodes: UCLA, where Leonard Kleinrock ran a network measurement center; the Stanford Research Institute, home of Douglas Engelbart's hypertext system; UC Santa Barbara; and the University of Utah.
Ray Tomlinson sent the first network email in 1971, and by 1973 three-quarters of ARPANET traffic was email. The network moved to TCP/IP on January 1, 1983, and was decommissioned in 1990, having proved the technology that became the internet.