From the History of the Internet

The seeds of the Internet were planted in 1969, when the Advanced Research Projects Agency (ARPA) of the U.S. Department of Defense began connecting computers at different universities and defense contractors. The goal of this early project was to create a large computer network with multiple paths – in the form of telephone lines – that could survive a nuclear attack or other disaster. If one part of the network were destroyed, other parts of the network would remain functional because data could continue to flow through the surviving lines. ARPA also wanted users in remote locations to be able to share scarce computing resources.

Soon after the first links in ARPANET (as this early system was called) were in place, the engineers and scientists who had access to this system began exchanging messages and data that were beyond the scope of the Defense Department's original objectives. People also discovered that they could play long-distance games and socialize with other people who shared their interests. The users convinced ARPA that these unofficial uses were helping to test the network's capacity.

At first, ARPANET was basically a wide area network serving only a handful of users, but it expanded rapidly. Initially, the network included four primary host computers. A host is like a network server, providing services to other computers that connect to it. ARPANET's host computers (like those on today's Internet) provided file transfer and communications services and gave connected systems access to the network's high-speed data lines. The system grew quickly and spread widely as the number of hosts grew.

The network jumped across the Atlantic to Norway and England in 1973, and it never stopped growing. In the mid-1980s, another federal agency, the National Science Foundation (NSF), joined the project after the Defense Department dropped its funding. NSF established five "supercomputing centers" that were available to anyone who wanted to use them for academic research purposes.

The NSF expected the supercomputers' users to use ARPANET to obtain access, but the agency quickly discovered that the existing network could not handle the load. In response, the NSF created a new, higher capacity network, called NSFnet, to complement the older and by then overloaded ARPANET. The link between ARPANET, NSFnet, and other networks was called the Internet. (The process of connecting separate networks is called internetworking. A collection of "networked networks" is described as being internet worked, which is where the Internet – a worldwide network of networks – gets its name.)

NSFnet made Internet connections widely available for academic research, but the NSF did not permit users to conduct private business over the system. Therefore, several private telecommunications companies built their own network backbones that used the same set of networking protocols as NSFnet. Like a tree's trunk or an animal's spine, a network backbone is the central structure that connects other elements of the network. These private portions of the Internet were not limited by NSFnet's "appropriate use" restrictions, so it became possible to use the Internet to distribute business and commercial information.

Interconnections (known as gateways) between NSFnet and the private backbones allowed a user on any one of them to exchange data with all the others. Other gateways were created between the Internet and other networks, large and small, including some that used completely different networking protocols.

The original ARPANET was shut down in 1990, and government funding for NSFnet was discontinued in 1995, but the commercial Internet backbone services have easily replaced them. By the early 1990s, interest in the Internet began to expand dramatically. The system that had been created as a tool for surviving a nuclear war found its way into businesses and homes. Now, advertisements for movies are far more common online than collaborations on physics research.