How a two-letter, accidentally poetic message launched the network that connects the world.
The short answer
The first word transmitted over what became the Internet was “lo”. It was sent on the evening of 29 October 1969 from a computer at the University of California, Los Angeles (UCLA) to a computer at the Stanford Research Institute (SRI) in Menlo Park, California, over the ARPANET, the experimental network that was the direct ancestor of today’s Internet.
The message was meant to be “login”. The system crashed after the first two letters, so only “lo” arrived.
A note on terms
Strictly speaking, the Internet did not exist in 1969. The ARPANET was a research network funded by the US Department of Defense’s Advanced Research Projects Agency (ARPA). The modern Internet, a “network of networks” built on the TCP/IP protocols, emerged in the 1970s and 1980s, and the ARPANET switched to TCP/IP on 1 January 1983. When people ask about the first message “on the Internet”, they almost always mean this first ARPANET transmission.
Why the network was built
In the 1960s, computers were large, expensive and scarce. Researchers at different universities often could not share each other’s machines or data. ARPA wanted a way to connect its funded research sites so that computing resources could be shared across distance.
The key idea was packet switching: breaking data into small chunks (packets), sending them independently across the network, and reassembling them at the destination. The concept had been developed by researchers including Paul Baran in the United States and Donald Davies in the United Kingdom, and Leonard Kleinrock at UCLA had contributed important mathematical work on queueing theory for data networks. Packet switching was more efficient and more resilient than the circuit-switched model used by telephone networks.
ARPA contracted the company Bolt, Beranek and Newman (BBN) to build the network’s switching computers, called Interface Message Processors (IMPs). Each IMP sat between a site’s host computer and the long-distance telephone lines leased from AT&T.
The night of 29 October 1969
The setup
The first IMP was installed at UCLA in early September 1969, in Leonard Kleinrock’s Network Measurement Center. The second was installed at SRI about a month later. The two sites were connected by a leased telephone line, and the host machines were very different from each other: UCLA’s was an SDS Sigma 7, and SRI’s was an SDS 940.
The attempt
UCLA student programmer Charley Kline sat at the UCLA terminal. At SRI, researcher Bill Duvall was at the other end. The two were also connected by an ordinary telephone call so they could confirm what was happening.
The plan was simple: Kline would type the word LOGIN, one letter at a time, and Duvall would confirm that each letter appeared on the SRI machine. The SRI system was set up to complete the word automatically once it recognised the opening letters.
The crash
Kline typed L, and Duvall confirmed it arrived. Kline typed O, and that arrived too. When he typed G, the SRI system crashed. The message that had been successfully delivered was just two letters long: “LO”.
The transmission took place at around 10:30 pm Pacific time. SRI’s system was repaired and restarted within about an hour, and the team then completed a full login, so the connection was working properly before the night was over.
“Lo and behold”
The accidental message turned out to be perfectly worded. “Lo” is an old English exclamation meaning “look” or “behold”, as in “lo and behold”. Leonard Kleinrock has often told the story with that wordplay, noting that the first message sent across the network was, by chance, a kind of announcement. Nobody planned it, and no grand statement had been prepared for the occasion. Compared with Samuel Morse’s “What hath God wrought” or Alexander Graham Bell’s call to his assistant, the first network message was modest, but it has become one of computing’s best-loved stories.
What happened next
- November to December 1969: Two more sites joined. The University of California, Santa Barbara, and the University of Utah were connected, giving the ARPANET four nodes by December.
- 1971: Ray Tomlinson at BBN sent the first network email and introduced the use of the @ symbol to separate a user’s name from their host computer. Email quickly became the network’s most popular application.
- 1972: The ARPANET was shown publicly at an international conference, and the number of connected sites grew steadily.
- 1973 to 1974: Vint Cerf and Bob Kahn developed the TCP/IP protocol suite, designed to link different networks together.
- 1 January 1983: The ARPANET adopted TCP/IP, often considered the birth of the Internet as we know it.
- Late 1980s to early 1990s: Tim Berners-Lee at CERN invented the World Wide Web, with the first website going live in 1991. The web made the Internet usable by the general public.
Common confusions
Was “lo” the first message on the World Wide Web? No. The web arrived more than twenty years later. “Lo” was sent over the ARPANET, the underlying network.
Was the first message really “login”? The intended word was “login”, but the word that actually arrived was “lo”. Accounts differ slightly on small details, such as the exact time, but the “lo” outcome is well documented, including in a UCLA log entry from that night.
What about “QWERTYUIOP”? Some people claim the first email carried this string of keyboard letters. Ray Tomlinson himself said he did not remember the exact contents of his early test messages. This concerns the first email in 1971, not the first network message in 1969.
Legacy
UCLA’s room where the transmission took place has been recognised as a historic site, and the event is widely commemorated as the Internet’s “birth” moment. A network that began with two computers and two letters now connects billions of people and devices, carrying messages, video, financial transactions and much of modern life.
The story also carries a useful lesson: world-changing technologies often begin with a small, imperfect test. The first transmission failed to complete its word, yet it proved that the idea worked.
Quick facts
| Detail | Information |
|---|---|
| First message | “lo” (intended: “login”) |
| Date | 29 October 1969 |
| From | UCLA (SDS Sigma 7 host) |
| To | Stanford Research Institute (SDS 940 host) |
| Sender | Charley Kline |
| Receiver | Bill Duvall |
| Network | ARPANET |
| Switching hardware | IMPs built by BBN |





