Skip to content

"LO": The First Message Sent on the ARPANET

Abstract

The first message transmitted on the ARPANET on October 29, 1969, was “LO.” Charley Kline, a student programmer at UCLA, was attempting to type “LOGIN” to connect to the SRI computer in Menlo Park. The system at SRI crashed after receiving two characters. Kline called the SRI operators on the telephone to tell them the system had gone down; the phone call is documented in the log. The complete LOGIN message was sent successfully about an hour later. The inadvertent first word (“LO”) has since been noted as perhaps the most poetic accident in computing history: “Lo, the internet.”

The Two-Node Network

On the night of the first message, the ARPANET had exactly two nodes. UCLA’s IMP had been installed at the end of August 1969 and SRI’s arrived on October 1; the other two sites came later that year, UCSB in November and the University of Utah in December, completing the famous four-node map by December 5, 1969:

  1. UCLA (University of California, Los Angeles): node 1, September 1969
  2. SRI (Stanford Research Institute): node 2, October 1969
  3. UCSB (University of California, Santa Barbara): node 3, November 1969
  4. University of Utah: node 4, December 1969

Each node had an Interface Message Processor (IMP), a Honeywell DDP-516 minicomputer built by BBN Technologies that handled packet routing and error detection. The host computers at each site connected to their IMP, and the IMPs connected to each other over leased telephone lines.

The ARPANET project was managed by Larry Roberts at ARPA (renamed DARPA in 1972) and originated in the theoretical work of J.C.R. Licklider. The full engineering story of the network’s construction is in ARPANET: Building the Network.

The Login Attempt

On October 29, 1969, at 10:30 PM, programmer Charley Kline at UCLA’s Boelter Hall attempted to log in from UCLA’s SDS Sigma 7 to SRI’s SDS 940. The Network/IMP software was experimental; this was the first test of an actual host-to-host connection.

The SRI system’s software required the client to send the word “LOGIN” to initiate a session. Kline typed “L.” The SRI system received it. He typed “O.” The SRI system received it. The SRI system crashed.

Kline phoned Bill Duvall at SRI to report the failure. Duvall brought the system back. About an hour after the crash, Kline sent “LOGIN” successfully and established the first host-to-host connection on the ARPANET.

The event is documented in a log kept by Leonard Kleinrock’s group at UCLA, which shows the timestamp, the two-character transmission, and a note about the phone call to SRI.

What the Network Was For

The ARPANET’s official purpose was resource sharing: allowing researchers at one institution to use computational resources at another. UCLA had a powerful computer; SRI had different software and data. The ability to log in remotely meant a researcher at UCLA could use SRI’s resources and vice versa without traveling.

What the network became in practice was different. Ray Tomlinson added networked email informally in 1971, with no central planning, and by 1973 an ARPA study found it accounted for three quarters of all ARPANET traffic. The resource-sharing vision was technically realized but eclipsed by the communication use case almost immediately.

The “LO” message is sometimes cited as evidence that the internet was named by accident, or that its most poetic description came from a system crash. The claim is post-hoc: nobody at the time interpreted two letters of a failed login attempt as a greeting. The poetic reading emerged decades later, in retrospective accounts of the internet’s history. But it persists because it has the structure of a good story: the machine spoke two letters, the machine broke, and those two letters turned out to be sufficient.

The Phone Call

The UCLA log records that Kline called SRI to report the crash. The existence of the phone call is significant for a reason beyond the immediate: it illustrates that in 1969, the telephone was the backup communication system for a network designed to route around failures. The network could not yet sustain itself; it needed human operators with telephones to recover from crashes. That dependency ended within years as the network became reliable enough that phone calls to report outages became exceptional rather than routine.

📚 Sources