Internet History

The internet wasn't built to connect people — it was built to survive nuclear war. This is your VocaCast briefing on Internet History for Monday, April 27, 2026.

We start with the forces that shaped it: Cold War anxiety, then the visionaries who imagined something unprecedented. The Cold War climate and fears of nuclear attack heightened US government funding for scientific research and development, including packet-switching technology. ^[1] That urgency created space for bold thinking. In the early 1960s, J. C. R. Licklider of MIT envisioned an "Intergalactic Computer Network" for seamless data and resource sharing between computers. ^[2] Licklider wasn't imagining the internet as we know it today — he was thinking far ahead, proposing that remote machines could work together, that a researcher in one location could access resources stored somewhere else.

The vision was radical for its time, but it emerged from a practical problem: the military needed communication systems that could function even if parts of the network were destroyed. That need drove concrete innovation. Paul Baran, a researcher at RAND Corporation, proposed a solution for robust communications networks using redundancy and digital technology in 1962, which later became foundational for packet switching. ^[3] In 1964, Paul Baran published "On Distributed Communications Networks" detailing his work on packet switching. ^[4] These weren't abstract theories — they were architectural blueprints for something that could actually work. A key objective for ARPANET was to create a communication network capable of withstanding potential disruptions, including nuclear attacks.

The need for a decentralized communication system that could withstand a nuclear attack spurred innovation, leading to the development of packet switching. ^[5]

By the late 1960s, the United States Department of Defense's Advanced Research Projects Agency initiated a project to develop a decentralized communication network, aiming to ensure communication resilience even if parts of the network were compromised. ^[6] Those seeds would grow into something no one fully anticipated — not just a military tool, but the foundation for how humans would eventually connect across the planet.

Those early ideas about sharing information across distances found their true test when ARPA set out to build something that had never existed before. The origins of ARPANET can be traced back to the late 1960s, initiated by the US Department of Defense's Advanced Research Projects Agency to create a robust communication network. ^[7] The project began as a way to share computers among academic and research institutions funded by ARPA. ^[4] This was not about connecting people to people — it was about letting researchers at different universities access expensive machines that sat miles apart. The world's first wide-area packet-switched network had its first node installed at UCLA in 1969.

That single installation became the birthplace of the internet as we know it. ^[8] On October 29, 1969, a graduate student at UCLA typed the first letters of the word "login" into the system — just an 'l' and an 'o' — before the entire network crashed. ^[8] It was hardly a triumphant beginning, but it proved the basic concept could work. Within days, the system recovered. The first successful message exchange on ARPANET was implemented in October 1969 between UCLA and Stanford Research Institute. ^[9] These weren't long philosophical transmissions — they were test signals proving that data could travel across a network without being destroyed or lost.

The Network Control Program, one of the foundational network protocols developed for ARPANET, made this kind of reliable communication possible. ^[10] What started as a military research project had just opened a door that would reshape how humans exchange information forever.

As ARPANET grew through the late 1960s, a fundamental challenge emerged: how to make different networks talk to each other reliably. The answer came through a concept called packet switching. Donald Davies developed a store-and-forward packet switching methodology at the UK National Physics Laboratories in 1966, creating a framework where information could be broken into small units and routed independently across a network. ^[4] This became the technical foundation that would allow multiple separate networks to communicate as one unified system. The real breakthrough came when researchers tackled the specific problem of reliable communication across diverse networks.

Vint Cerf and Bob Kahn began their research on reliable data communications across packet radio networks in the spring of 1973, which led to the creation of TCP/IP. ^[5] The TCP/IP protocol suite, a set of rules for communication between networks and devices, was designed in the 1970s by the same pair. ^[11] Think of protocols as a shared language—without agreed-upon rules for how data should be formatted, addressed, and transmitted, networks remained isolated islands. Testing the system came next, and it revealed whether the theory actually worked in practice. The first test of TCP/IP was carried out in 1975, with successive testing through versions 1 through 4 between 1975 and 1983.

The Internetting project developed the Internet system of networks and the TCP/IP Protocol Suite, named after the Transmission Control Protocol and Internet Protocol themselves. ^{[12] [13]} By the time those tests were complete, the technical foundation for global connectivity had been solidified.