Introduction: The Myth of the Nuclear Shield
 |
| The Archive of the First Internet |
If you ask the average person why the internet was invented, they will likely tell you a story about the Cold War. They will say it was built by the military to survive a nuclear strike—a decentralized web that could keep operating even if half the United States was turned to radioactive ash.
At First Everything, we deal in archival truth, not convenient myths. While the funding came from the Department of Defense (ARPA), the purpose was far more mundane and human. In the late 1960s, computers were room-sized monoliths that cost millions of dollars. Researchers across the country needed to use these machines, but they couldn't physically travel to them.
The internet wasn't built to survive a bomb; it was built to share a calculator.
This is the exhaustive archive of ARPANET—the primal nervous system of the digital age. It is a story of heavy iron cabinets, telephone lines, and a room in Los Angeles where two letters—"L" and "O"—changed the course of human history on October 29, 1969.
The Visionary: Licklider’s "Intergalactic" Dream
Before there were cables, there was a memo. In 1963, J.C.R. Licklider, a psychologist and computer scientist at ARPA (Advanced Research Projects Agency), wrote a series of memos to his colleagues.
The "Time-Sharing" Problem
In the 1960s, computing was "Batch Processing." You punched cards, handed them to an operator, and waited hours for results. Licklider envisioned "Time-Sharing," where multiple users could interact with a computer simultaneously. But he went further: what if the computers could talk to each other?
The Concept: If a researcher in Utah needed the graphics power of a computer in Utah, but the database of a computer in California, a network could connect them.
The Execution: Licklider left ARPA before he could build it, but his successors, Bob Taylor and Larry Roberts, picked up the torch.
In 1966, they secured a $1 million budget to build the "ARPA Network."
 |
| Intergalactic Computer Network illustration |
The Hardware: The IMP (The First Router)
The biggest technical hurdle was language. Computers in 1968 were isolationists. An IBM mainframe could not speak to a DEC PDP-10. They used different code, different voltage, and different logic.
To fix this, the architects didn't teach the computers to talk to each other; they built a translator. They called it the Interface Message Processor (IMP).
The Honeywell DDP-516
The IMP was the world’s first router.
The Shell: A ruggedized, steel-encased Honeywell DDP-516 minicomputer.
The Specs: It had 12 kilobytes of memory.
The Cost: Roughly $82,200 each (about $600,000 today).
The Role: The IMP sat between the host computer and the network.
It took the data, chopped it up into "packets," and fired it down the phone line. The host computer didn't need to know how the network worked; it just had to talk to the IMP.
Archival Note: The contract to build the IMPs was won by a small consulting firm in Cambridge, Massachusetts, called Bolt, Beranek and Newman (BBN).
The Tech Origins: Packet Switching
Why did the internet succeed where the telephone failed? The answer lies in the First Everything principle of "Packet Switching."
In a traditional phone call (Circuit Switching), a physical wire connection is held open for the entire conversation.
Paul Baran (at RAND Corporation) and Donald Davies (at NPL in the UK) independently invented a better way.
The Logic: Break the data into small chunks called "Packets."
The Method: Label each packet with an address and throw it into the network.
The packets might take different routes. One might go through Chicago, another through Dallas. They meet at the destination and reassemble. The Benefit: If a line is cut (or bombed), the packets just find a different route.
This was the "survivability" aspect, but its real value was efficiency.
The "Day One": October 29, 1969
The First Everything archive pinpoints the exact moment the internet took its first breath.
Location: 3420 Boelter Hall, University of California, Los Angeles (UCLA).
The Room: A windowless room filled with the hum of cooling fans and the smell of ozone.
The Players: * Leonard Kleinrock:
The UCLA professor supervising the experiment. Charley Kline: The 21-year-old grad student sitting at the terminal.
Bill Duvall: The programmer at the receiving end at Stanford Research Institute (SRI), 350 miles away.
The "LO" Event
At 10:30 PM PST, Charley Kline prepared to log into the SRI computer remotely.
Kline: "I'm going to type the L."
Duvall: "Okay."
Kline types L.
Duvall: "Got the L."
Kline: "I'm going to type the O."
Kline types O.
Duvall: "Got the O."
Kline: "I'm going to type the G."
CRASH.
The SRI computer's memory buffer overflowed and crashed the system. The first message ever sent on the internet was not a profound declaration like "What hath God wrought?" (Morse Code) or "Mr. Watson, come here" (Telephone).
As Kleinrock later put it: "Lo and behold." It was accidental poetry.
The First Four Nodes: The Square Network
 |
| The Square Network |
UCLA (Node 1): The Network Measurement Center (where the traffic was analyzed).
SRI (Node 2): Stanford Research Institute (home of Douglas Engelbart’s "Mother of All Demos").
UCSB (Node 3): University of California, Santa Barbara (specializing in interactive graphics).
University of Utah (Node 4): Specializing in 3D rendering.
This was the "Big Bang" epoch. If you were a hacker in 1969, these were the only four doors you could knock on.
The Evolution: Email and the @ Symbol
The ARPANET grew quickly, but it was mostly used for remote login (Telnet) and file transfer (FTP).
Ray Tomlinson, an engineer at BBN, wrote a program to send messages between users on different computers.
The Innovation: He needed a separator between the user's name and the machine's name. He looked at his keyboard (a Teletype Model 33) and chose the symbol that meant "at": @.
The Result:
user@hostbecame the standard syntax of digital identity. Suddenly, 75% of all ARPANET traffic was email. The network became social.
The Pivot: The Flag Day of 1983
Technically, the "Internet" as we know it didn't start in 1969. ARPANET used a protocol called NCP (Network Control Program).
Vint Cerf and Bob Kahn invented a new language: TCP/IP (Transmission Control Protocol/Internet Protocol).
January 1, 1983
This is the second "Day One" in our archive. The Department of Defense mandated a "Flag Day." On this date, every machine on the ARPANET had to switch from NCP to TCP/IP or be cut off.
The Risk: It was a hot swap of the planet's nervous system.
The Outcome: It worked. ARPANET became the backbone of the "Internet"—a network of networks.
The Decommissioning: The End of ARPANET
By 1990, the ARPANET was obsolete.
On February 28, 1990, the ARPANET was officially decommissioned. The original IMPs were unplugged. The "First Internet" died, but its child, the modern Internet, had already eaten the world.
Timeline of the First Internet
1962: J.C.R. Licklider writes the "Intergalactic Computer Network" memos.
1968: BBN wins the contract to build the Interface Message Processor (IMP).
Aug 30, 1969: The first IMP is delivered to UCLA (Node 1).
Oct 29, 1969: Day One.
The first host-to-host message ("LO") is sent from UCLA to SRI. Dec 1969: The 4-Node network is complete (UCLA, SRI, UCSB, Utah).
1971: Ray Tomlinson invents email and chooses the @ symbol.
1972: First public demo of ARPANET at the ICCC conference in Washington D.C.
1974: Vint Cerf and Bob Kahn publish "A Protocol for Packet Network Intercommunication" (the birth of TCP).
Jan 1, 1983: Flag Day.
The network switches to TCP/IP. The modern "Internet" is born. 1990: ARPANET is decommissioned.
1991: Tim Berners-Lee releases the World Wide Web (WWW) to the public (a software running on the internet).
Conclusion: The Archival Legacy
The ARPANET archive teaches us that the digital world is fragile. It started with a crash. It started with a glitch.
When Charley Kline typed "LO," he wasn't trying to build a trillion-dollar economy or a social media empire. He was just trying to do his homework on a better computer. The First Everything philosophy is validated here: massive revolutions rarely start with a bang; they start with a practical solution to a boring problem.
Today, the internet is invisible air we breathe. But in 1969, it was 350 miles of copper wire and a room full of bearded academics praying the system wouldn't crash. They built the road; we just drive on it.
Archivist's Note: The "Log" Book
The most sacred artifact in internet history is a simple, green spiral-bound notebook kept at UCLA. On the page dated October 29, 1969, in handwritten pen, is the entry:
22:30 Talked to SRI Host to Host
That scrawled line is the birth certificate of the Information Age. It currently resides in the UCLA archives, a physical testament to a virtual miracle.
Archival References
Hafner, K., & Lyon, M. (1996). Where Wizards Stay Up Late: The Origins of the Internet. Simon & Schuster. (The definitive history).
Kleinrock, L. (2010). The History of the Internet: The First IMP. UCLA Archives.
Abbate, J. (1999). Inventing the Internet. MIT Press.
Segaller, S. (1998). Nerds 2.0.1: A Brief History of the Internet. TV Books.
Request for Comments (RFC) 1. (April 1969).
"Host Software." authored by Steve Crocker. (The very first technical document of the internet). Computer History Museum. Internet History Program Archives. Mountain View, CA.