The short version
The internet began as an American research project in the late 1960s and became a worldwide public network over the following three decades. Its origin can be told in three stages: an idea, a first network, and a way to connect networks.
Stage one: the idea
In the early 1960s, electronic computers were rare and expensive, and each one tended to be used by people who were physically near it. Time-sharing, a technique allowing several people to use one machine at once, led some researchers to think about sharing computers across distance. J.C.R. Licklider, who headed computing research at the Advanced Research Projects Agency (ARPA) from 1962, wrote about a network that would let researchers use each other’s programs and data. At the same time, other engineers worked out a method of sending data in small labeled pieces, called packets, rather than reserving a whole telephone line for the duration of a connection.
Stage two: the first network
ARPA decided to build a network linking the computers at research universities, since each site had specialized equipment that others wanted to use. A company in Massachusetts, Bolt Beranek and Newman, built the hardware. The first message was sent between the University of California, Los Angeles, and the Stanford Research Institute in October 1969. The system was called ARPANET, and it had four nodes by the end of that year. By the middle of the 1970s, it connected dozens of sites and carried electronic mail, which had not been part of the original plan.
Stage three: connecting networks
By then, other networks existed too, using radio, satellites or other rules. A way to make them cooperate was needed. Vinton Cerf and Robert Kahn proposed a protocol design in 1974, which led to TCP/IP. After ARPANET switched to it in January 1983, any network that followed the same rules could join. That step is what turned a single network into the internet, a name for a connected group of networks.
A common misunderstanding
A popular story says the internet was designed to survive a nuclear war. The truth is more mixed. Paul Baran at the RAND Corporation did study how communication systems could survive attack, and his ideas fed into the broader development of packet networks. But the people who built ARPANET gave a more everyday reason: they wanted to let scientists share scarce computing resources. Both threads existed, and sources still disagree about how much weight to give each. It is safest to say that survivability influenced some of the early thinking while resource sharing was the stated purpose of the network.
Why it spread
Several practical reasons help explain why this network, and not one of its rivals, became the global standard. Its protocols were published openly, so anyone could write software for them. The designs were simple and ran on many kinds of hardware, from big research computers to small workstations. Universities adopted them widely in the 1980s, and a generation of students carried the habit of using them into business. Competing systems run by telephone companies and standards bodies were more elaborate and were slower to reach working products, which gave the simpler internet approach time to take hold.
What came after
During the 1980s, universities joined through networks funded by the National Science Foundation, and the old ARPANET was closed in 1990. The World Wide Web arrived in 1990 and 1991, and the first widely used graphical web browsers in 1993. Commercial service providers then offered access to homes and businesses. Each of these stages is a larger story, but this three-part outline holds up as a first picture.
Why the origin matters
Knowing the origins helps explain features that still shape the network today. It is organized as many separate networks rather than one system, it relies on public technical documents that anyone can read, and it was built by researchers who valued open sharing. The openness is part of why different people could build email, the web and many later services on top of it without asking anyone’s permission. It also explains some of the weaknesses: the early network was built among a small, trusting community, and security was not a major concern in the first designs, a gap that later developments in cybersecurity have had to address.