A Brief History of the Internet

A Brief History of the Internet by Walt Howe
is licensed under a Creative Commons Attribution-
Noncommercial-Share Alike 3.0 United States License. Based on a work at www.walthowe.com.

The modern Internet, as we use it today and based on the TCP/IP protocol, turned 30 on January 1, 2013. In 1981, DARPA began testing the protocol to connect its networks. On January 1, 1983, the lab’s 400 servers, which were based on the NCP protocol, were shut down and the system migrated to TCP/IP, giving rise to today’s Internet. To mark this historic date, I translated this short text, which tells the story of the Internet in an accessible way and focuses not only on the technology but also on the people involved. You can find the original text by Walt Howe here. Here is the translation.

An anecdotal history of the people and communities that brought us the Internet and the Web

The Internet grew out of the visionary thinking of people in the early 1960s, who saw great potential in allowing computers to share information about research and development in scientific and military fields. J.C.R. Licklider of MIT first proposed a global network of computers in 1962, then moved to the Defense Advanced Research Projects Agency (DARPA) in late 1962 to lead the work to develop it. Leonard Kleinrock, of MIT and later UCLA, developed the theory of packet switching, which provided the basis for Internet connections. Lawrence Roberts, of MIT, connected a computer in Massachusetts to one in California in 1965 over dial-up telephone lines. He demonstrated the feasibility of a wide-area network, but also showed that telephone circuits were inadequate. Kleinrock’s packet-switching theory was confirmed. Roberts went to DARPA in 1966 and developed his plan for ARPANET. These visionaries, along with many others not named here, are the true founders of the Internet.

When the late Senator Ted Kennedy heard in 1968 that the pioneering Massachusetts company BBN had won the ARPA contract for an “Interface Message Processor (IMP),” he sent BBN a congratulatory telegram for its ecumenical spirit in winning the contract for an “interfaith message processor.”

The Internet, then known as ARPANET, went live in 1969 under a contract awarded by the renowned Advanced Research Projects Agency (ARPA). It initially connected four main computers at universities in the southwestern United States (UCLA, Stanford Research Institute, UCSB, and the University of Utah). The contract was carried out by BBN of Cambridge, Massachusetts, under Bob Kahn, and the network became operational in December 1969. In June 1970, MIT, Harvard, BBN, and System Development Corporation (SDC) in Santa Monica, California, were added. In January 1971, Stanford, MIT Lincoln Labs, Carnegie Mellon, and Case Western Reserve University were added. In the following months, NASA Ames, Mitre, Burroughs, RAND, and the University of Illinois connected. There’s not much point in continuing the list after that.

Who was the first person to use the Internet?

Charley Kline at UCLA sent the first ARPANET packets when he tried to connect to the Stanford Research Institute on October 29, 1969. The system crashed when he reached the “G” in LOGIN!

The Internet was designed to provide a communications network that would keep working even if some major sites went down. If the most direct route was unavailable, routers would direct traffic around the network along alternative routes.

At first, the Internet was used by computer experts, engineers, scientists, and librarians. It wasn’t user-friendly at all. There were no PCs at home or in the office in those days, and anyone who used the Internet—whether a computer professional, engineer, scientist, or librarian—had to learn to use a very complex system.

Did Al Gore invent the Internet?

According to a transcript of a CNN interview with Wolf Blitzer, Al Gore said, “During my service in the United States Congress, I took the initiative in creating the Internet.” Al Gore was not yet in Congress in 1969, when ARPANET began, or in 1974, when the term Internet first came into use. Gore was elected to Congress in 1976. To be fair, Bob Kahn and Vint Cerf acknowledge in a study titled “Al Gore and the Internet” that Gore probably did more than any other elected official to support the Internet’s growth and development, from the 1970s to the present.

Email was adapted for ARPANET by Ray Tomlinson of BBN in 1972. He chose the @ symbol from those available on his teletype to link the user name to the address. The telnet protocol, which allows you to log in to a remote computer, was published as a Request for Comments (RFC) in 1972. RFCs are a way to share development work across the community. The FTP protocol, which allows files to be transferred between Internet sites, was published as an RFC in 1973. From then on, RFCs were available electronically to anyone who could use FTP.

Libraries began automating and linking their catalogs in the 1960s, independently of ARPA. The visionary Frederick G. Kilgour of the Ohio College Library Center (now OCLC, Inc.) created the Ohio library network during the 1960s and 1970s. By the mid-1970s, more regional consortia from New England, the southwestern states, the Mid-Atlantic, and elsewhere had joined Ohio to form a national, then international, network. Automated catalogs, which were not very user-friendly at first, became available around the world, initially through telnet or IBM’s unusual TN3270 variant, and only many years later through the Web. See the history of OCLC.

Ethernet, a protocol for many local-area networks, emerged in 1974 as a result of Harvard student Bob Metcalfe’s dissertation on “Packet Networks.” The university initially rejected his dissertation for not being analytical enough. It was accepted later after he added more equations.

The Internet matured in the 1970s as a result of the TCP/IP architecture, first proposed by Bob Kahn of BBN and developed by Kahn and Vint Cerf at Stanford, along with others, throughout the decade. The Department of Defense adopted it in 1980, replacing the earlier Network Control Protocol (NCP), and it was universally adopted in 1983.

The Unix to Unix Copy Protocol (UUCP) was invented at Bell Labs in 1978. Usenet began in 1979, based on UUCP. Newsgroups, discussion groups focused on a particular topic, followed, providing a way to exchange information worldwide. Although Usenet is not considered part of the Internet because it does not use TCP/IP, Unix systems were connected around the world, and many Internet sites took advantage of newsgroups. They were an important part of the community-building that took place across the networks.

Similarly, BITNET (Because It’s Time Network) connected IBM mainframes across the educational community and around the world, offering email services starting in 1981. The Listserv software was developed for this network and others later on. Gateways were developed to connect BITNET to the Internet and enabled email exchanges, especially for discussion mailing lists. These mailing lists and other forms of email discussion lists became another important element in the community that was taking shape.

In the old days, it was fascinating to watch a BITNET message being sent, pausing and moving along until it reached its destination. We would see it arrive at one site, then be passed on to the next, then the next, and the next. Life moved more slowly then!

In 1986, the National Science Foundation funded NSFNet as a 56 Kbps cross-country backbone for the Internet. It maintained its sponsorship for nearly a decade, establishing rules for its noncommercial use and research applications.

As the commands for email, FTP, and telnet became standardized, it became much easier for nontechnical people to learn how to use the networks. It was by no means easy by today’s standards, but it opened up the Internet to many people at universities. Departments beyond libraries, computer science, physics, and engineering found ways to make good use of the networks—to communicate with colleagues around the world and share files and resources.

Although there were few sites on the Internet, it was easy to keep track of the resources of interest that were available. But as more and more universities and organizations—and their libraries—connected, the Internet became increasingly difficult to navigate. More tools were needed to index the available resources.

The first effort to index the Internet beyond library catalogs was created in 1989. Peter Deutsch and Alan Emtage, students at McGill University in Montreal, created an archive for FTP sites called Archie. The software periodically checked all known and publicly accessible FTP sites, listed their files, and built a searchable index of the software. Archie’s search commands were Unix commands, so using it to its full potential required some knowledge of Unix.

McGill University, which hosted the first Archie, discovered one day that half of the Internet traffic coming to Canada from the United States was accessing Archie. Administrators were concerned that the university was subsidizing so much traffic and closed Archie to external access. Fortunately, by then there were many more Archie servers available.

Around the same time, Brewster Kahle, then at Thinking Machines Corp., developed the Wide Area Information Server (WAIS), which indexed the full text of files in a database and allowed those files to be searched. There were several versions with different levels of complexity and development capability, but the simplest was made available to everyone over the networks. At its peak, Thinking Machines maintained pointers to more than 600 databases around the world that had been indexed by WAIS. These included things like the complete set of Frequently Asked Questions files, the full text of papers such as RFCs written by those developing Internet standards, and much more. Like Archie, its interface was far from intuitive, and it took some effort to learn to use it well.

Peter Scott of the University of Saskatchewan recognized the need to bring together information about all library catalogs accessible via telnet, make it available on the Web, and include other telnet resources. In 1990, he made his Hytelnet catalog available. It provided one place to find information about library catalogs and other telnet resources and how to use them. He maintained it for years and added HyWebCat in 1997 to provide information about Web-based catalogs.

In 1991, the first genuinely user-friendly Internet interface was developed at the University of Minnesota. The university wanted to develop a simple menu system for accessing files and information on campus through its local network. A debate ensued between mainframe advocates and those who believed in smaller, client-server systems. The mainframe advocates “won” the debate at first, but when the client-server advocates said they could make a prototype available quickly, they were given the go-ahead to build a demonstration system. The demo was called gopher, after the University of Minnesota’s mascot—the Golden Gopher. Gopher proved very successful, and within a few years there were more than 10,000 gophers around the world. You didn’t need to know anything about Unix or computer architecture to use it. In a gopher system, you typed or clicked a number to select the menu option you wanted.

Gopher became much easier to use when the University of Nevada, Reno, developed VERONICA, a searchable index of Gopher menus. The name was supposed to be an acronym for Very Easy Rodent-Oriented Netwide Index to Computerized Archives. It was a search robot that indexed Gopher menus around the world, collecting links and updating them in the index. It was so popular that it was very difficult to connect, even with a number of other VERONICA sites set up to make access easier. Similar indexing software was developed for individual sites and called JUGHEAD (Jonzy’s Universal Gopher Hierarchy Excavation And Display).

Peter Deutsch, who developed Archie, always insisted that Archie was short for “archiver” and had nothing to do with the comic strip. He was appalled when VERONICA and JUGHEAD appeared.

In 1989, another significant event helped make networks easier to use. Tim Berners-Lee and others at the European Particle Physics Laboratory, better known as CERN, proposed a new protocol for distributing information. This protocol, which became the World Wide Web in 1991, was based on hypertext—a system of embedding links in text to connect to other text, just as you have done whenever you clicked a text link while reading these pages. Although it began before Gopher, it developed more slowly.

The 1993 development of the graphical Mosaic browser by __Marc Andreessen__ and his team at the National Center for Supercomputing Applications (NCSA) gave the protocol a major boost. Andreessen later went on to become the brains behind Netscape Corp., which produced the most successful graphical browser and server until Microsoft declared war and developed Microsoft Internet Explorer.

Soon after the Mosaic graphical browser was introduced, the Library of Congress made some wonderful, colorful images of the Vatican available. With the slow connections of the time, it took 20 minutes to load a single page. We would start the download, take a coffee break, and come back to marvel at the image that had filled our screen.

Because the Internet was initially funded by the government, its use was originally limited to research, education, and government work. Commercial use was prohibited unless it directly supported research and education goals. This policy continued until the early 1990s, when independent commercial networks began to grow. It then became possible to route traffic across the country from one commercial site to another without going through the government-funded Internet backbone (NSFNet).

Delphi was the first national commercial online service to offer its subscribers Internet access. It opened an email connection in July 1992 and an Internet service in November 1992. All pretense of restrictions on commercial use disappeared in May 1995, when the National Science Foundation ended its sponsorship of the Internet backbone and all traffic began to pass over commercial networks. AOL, Prodigy, and CompuServe were available. Since commercial use was so widespread by then, and educational institutions had been paying for their own use for some time, the loss of NSF funding had no significant effect on costs.

MICHAEL Dertouzos (1936–2001)

The early days of the Web were a confusing period in which many developers tried to put their personal stamp on the way the Web should develop. The Web was threatened with becoming a mass of independent protocols that would require different software for different applications. The visionary Michael Dertouzos of the MIT Laboratory for Computer Science persuaded Tim Berners-Lee and others to form the World Wide Web Consortium (W3C) in 1994 to promote and develop standards for the Web. Proprietary plug-ins still abound on the Web, but the consortium ensured that common standards would be present in every browser. Read Tim Berners-Lee’s tribute to Michael Dertouzos.

In 1986, the NSF moved from supporting the backbone to building K–12 and local public-library access on one hand, and high-volume connections for research on the other.

Microsoft’s large-scale entry into the browser, server, and Internet service provider markets completed the major shift toward a more commercial Internet. The launch of Windows 98 in June 1998, with Microsoft’s browser closely integrated into the desktop, showed __Bill Gates__’s determination to capitalize on the Internet’s enormous growth. Microsoft’s success in recent years brought legal challenges to its dominance. We’ll leave it to you to decide whether these battles belong in the courts or the marketplace.

During this period of enormous growth, companies entered the Internet arena looking for business models that would generate returns. Free services supported by advertising temporarily removed some of the direct costs for consumers. Services such as Delphi offered free web pages, chat rooms, and message boards for building communities. Online sales grew rapidly for products such as books, music CDs, and computers, but profit margins were small because price comparisons were so easy and public confidence in online security was still shaky. Business models that worked well included portals, which tried to offer everything to everyone, and live auctions. AOL’s acquisition of Time Warner was the largest merger in history at the time and showed the enormous growth of the Internet business. The stock market had a rocky ride, with ups and downs for new technology companies; the dot-coms produced both good and bad news. Falling advertising revenue spelled disaster for many dot-com companies, and survivors went through a major shake-up in search of better business models.

The current trend with major implications for the future is the growth of high-speed connections. 56K modems and the providers that supported them hung on for a while, but they are now at the low end. 56K is not fast enough to deliver multimedia, such as sound and video, except at low quality. But much faster new technologies, such as cable modems and digital subscriber lines (DSL), are now widespread.

Wireless networks have grown rapidly in recent years, and travelers look for wireless areas where they can connect when away from home or the office. Many airports, cafés, hotels, and motels now routinely offer these services, sometimes for a fee and sometimes for free.

The next area of growth is the major expansion of universal wireless access, where almost everywhere is a “hot spot.” Municipal or citywide Wi-Fi, WiMAX access offering broader coverage than Wi-Fi, EV-DO, 4G, LTE, and other formats will compete for dominance in the United States in the coming years. The battle is economic as well as political.

Another trend quickly affecting web designers is the growth of small devices that connect to the Internet. Small tablets, pocket PCs, smartphones, e-readers, video game devices, and GPS devices can now access the Web anywhere, and many web pages are not designed to work at that scale.

As the Internet has become ubiquitous, faster, and increasingly accessible to nontechnical people, communities, social networks, and collaboration services have grown rapidly, allowing people to communicate and share interests in many different ways. Sites such as Facebook, Twitter, LinkedIn, YouTube, Flickr, Second Life, Delicious, blogs, wikis, and many others let people of all ages quickly share their current interests with others anywhere.

As Heraclitus said in the 4th century B.C., “Nothing is permanent except change!”

For more information about the history of the Internet, visit these sites: