The internet was not invented by one person — it grew from decades of work by computer scientists, engineers, and government agencies building on each other's ideas
The internet started as a military research project in the 1960s called ARPANET, funded by the U.S. Department of Defense's Advanced Research Projects Agency (ARPA). The goal was to create a communication system that could survive if parts of it were destroyed. Computer scientists at universities and research labs — including UCLA, Stanford, UC Santa Barbara, and the University of Utah — connected their computers together to test whether the system would work. By 1969, the first message traveled across ARPANET between UCLA and Stanford Research Institute. That was the internet's first moment.
But ARPANET was not the whole story. Dozens of researchers contributed the pieces that made modern internet communication possible. Vint Cerf and Bob Kahn designed TCP/IP, the set of rules that let different networks talk to each other — without those rules, your computer could not reach a server across the world. Jon Postel managed the early system of domain names and IP addresses. Ray Tomlinson wrote the first network email program in 1971. Robert Metcalfe invented Ethernet, the technology that lets computers in the same building connect. None of them "invented the internet" alone, but each solved a problem that made the whole thing work.
Key Takeaways
- ARPANET, funded by the U.S. Department of Defense in 1969, was the first working network that connected computers across different locations.
- Vint Cerf and Bob Kahn created TCP/IP, the fundamental rules that allow different networks to communicate with each other.
- The internet grew from contributions by many researchers at universities and labs, not from a single invention or inventor.
- The World Wide Web, created by Tim Berners-Lee in 1989, was a separate system that made the internet straightforward for regular people to use.
- Government funding, university research, and open sharing of ideas were essential to building the early internet.
How ARPANET became the foundation
In 1966, ARPA program manager Bob Taylor wanted to connect the computers at different universities so researchers could share computing power and talk to each other without traveling. At the time, each computer was isolated — if you wanted to use a computer at another university, you had to go there or mail punch cards. Taylor asked Larry Roberts to design a network that could link them together.
Roberts and his team built ARPANET by connecting four computers in 1969. The first connection went from UCLA to Stanford Research Institute on September 2, 1969. The second connection added UC Santa Barbara and the University of Utah a few months later. The network used a system called packet switching, which broke messages into small pieces, sent them separately, and reassembled them at the destination — this meant the network could keep working even if one connection failed, which was the whole point of the original military design.
ARPANET grew slowly through the 1970s. By 1977, it had about 100 computers connected. Universities and research labs shared the network because ARPA funded it and required that the technology be open — anyone could see how it worked and build on it. That openness became one of the internet's defining features.
The protocols that made networks talk to each other
ARPANET used a protocol called NCP (Network Control Protocol) in its early years, but NCP only worked if all the computers were part of the same network. In the mid-1970s, Vint Cerf and Bob Kahn realized that the real problem was bigger: how do you connect many separate networks together? They designed TCP/IP to solve that problem.
TCP/IP is actually two protocols working together. TCP (Transmission Control Protocol) breaks messages into packets, numbers them, and makes sure they all arrive in the right order. IP (Internet Protocol) addresses each packet with a destination address and routes it through the network to get there. A computer receiving the packets uses TCP to reassemble them into the original message. This system works whether the packets travel through one network or ten different networks — it is the reason your email can reach someone on a completely different internet service provider.
On January 1, 1983, ARPANET switched from NCP to TCP/IP. That date marks the moment when the "internet" as a connected system of networks officially began, because TCP/IP made it possible to connect networks that were not originally designed to work together. After that, other networks could plug into ARPANET and talk to everything already connected.
The domain name system and early infrastructure
As the network grew, researchers needed a way to remember where computers were. Early on, they used a straightforward text file called HOSTS.TXT that listed every computer's name and its IP address — a number like 192.168.1.1. But as the network grew, updating that file on every computer became impossible.
In 1983, Paul Mockapetris designed the Domain Name System (DNS), which let computers ask a central server "what is the IP address for stanford.edu?" instead of looking it up in a local file. DNS made it possible for the network to grow without breaking — you could add a new computer and register its name without updating thousands of files. Jon Postel, who ran the early internet's central registry, managed which domain names existed and which IP addresses belonged to which organizations.
These systems — TCP/IP, DNS, and the IP address system — are still the foundation of how the internet works today. They were designed to be flexible and to work even if parts of the network failed, which is why they have lasted for forty years.
Email and the first internet applications
While the network infrastructure was being built, researchers were also creating the first applications that made the network useful. Ray Tomlinson wrote the first network email program in 1971, sending a message from one computer to another on ARPANET. He chose the @ symbol to separate the user name from the computer name — that is why email addresses still use @ today.
Email became the first "killer app" that made people want to connect to the network. Before email, the network was mostly used for file transfer and remote login — researchers could access a computer at another university, but it was not something most people needed. Email was different. Once email worked, researchers realized they could collaborate without being in the same room.
Other early applications included FTP (File Transfer Protocol) for moving files between computers, and Telnet for logging into a remote computer as if you were sitting at its keyboard. These tools were powerful for researchers and engineers, but they required learning commands and typing addresses — they were not something a regular person could use without training.
The World Wide Web made the internet accessible
For the first twenty years, the internet was mostly used by computer scientists, engineers, and researchers at universities. It was powerful but difficult. In 1989, Tim Berners-Lee, a physicist working at CERN (a research lab in Switzerland), created the World Wide Web — a system of documents linked together with hyperlinks that you could navigate by clicking.
Berners-Lee wrote three things: HTML (the language for writing web pages), HTTP (the protocol for requesting and sending web pages), and the first web browser, which let you view pages and click links. He released all of it for free and did not patent it. By the early 1990s, web browsers became straightforward enough for regular people to use, and the internet stopped being a tool only for specialists.
The web and the internet are not the same thing — the internet is the network of connected computers, and the web is one process that runs on top of it. Email, video calls, online games, and streaming video all use the internet but not the web. But the web is what made the internet something most people wanted to use, which is why the internet became what it is today.
How government funding and open standards shaped the internet
The internet could have been built very differently. It could have been owned by a single company, with proprietary protocols that only worked with that company's equipment. Instead, it was funded by government research agencies and built by universities that shared their work openly.
ARPA required that the technology be open — researchers had to publish their work and make it available to other labs. This meant that if you had a good idea, you could build on what others had done instead of starting from scratch. When TCP/IP was designed, it was published as an open standard that anyone could implement. When DNS was created, the specifications were public. This openness meant that the internet could grow faster than any single company could have built it, because thousands of people could contribute.
Government funding also meant that the internet was not built to make money — at least not at first. ARPA funded research because they wanted a communication system that would survive a nuclear war, and universities participated because they wanted to share computing resources and collaborate. That mission shaped the internet's design in ways that still matter today.
Frequently Asked Questions
Did the military invent the internet?
The military funded ARPANET through the Defense Department's research agency, but the actual work was done by computer scientists at universities and research labs. The goal was military — building a communication system that could survive damage — but the people who built it were researchers, not soldiers. After TCP/IP was created, the internet became open to any network that wanted to connect, and it quickly grew beyond military use.
Why did it take so long for regular people to use the internet?
The early internet required learning commands and typing IP addresses or computer names. It was designed for researchers who understood computers. The World Wide Web, created in 1989, made it possible to navigate by clicking links instead of typing commands. Even then, web browsers did not become straightforward to use until the mid-1990s, and internet access required a phone line and a modem that was slow by today's standards.
Could the internet have been invented differently?
Yes. If a single company had built it, it might have been faster in some ways but closed to competitors. If it had been designed for profit from the start, it might have required payment for basic services. The fact that it was built openly and funded by government research meant that anyone could contribute ideas and build on what others had done, which is probably why it grew so quickly.
Who owns the internet now?
No single person or company owns the internet. It is made up of thousands of networks owned by different organizations — internet service providers, universities, companies, and governments. They all connect to each other using the same protocols (TCP/IP and DNS) that were designed in the 1970s and 1980s. Organizations like ICANN manage the domain name system and IP address allocation, but they do not own the internet itself.
Is the internet still being invented?
The basic infrastructure has been stable for decades, but new technologies are always being added on top of it. IPv6 is a newer version of the IP protocol that provides more addresses. New security protocols protect data in transit. But the core idea — breaking messages into packets, routing them through many networks, and reassembling them at the destination — is the same as it was in 1983.