The concept of data communication – transmitting data between two different places through an electromagnetic medium such as radio or an electric wire – pre-dates the introduction of the first computers. Such communication systems were typically limited to point to point communication between two end devices. Semaphore lines, telegraph systems and telex machines can be considered early precursors of this kind of communication. The telegraph in the late 19th century was the first fully digital communication system.

Fundamental theoretical work in data transmission and information theory was developed by Claude Shannon, Harry Nyquist, and Ralph Hartley in the early 20th century. Information theory, as enunciated by Shannon in 1948, provided a firm theoretical underpinning to understand the trade-offs between signal-to-noise ratio, bandwidth, and error-free transmission in the presence of noise, in telecommunications technology.

With limited exceptions, the earliest computers were connected directly to terminals used by individual users, typically in the same building or site. Wide area networks (WANs) emerged during the 1950s and became established during the 1960s.

Christopher Strachey, who became Oxford University's first professor of computation, filed a patent application for time-sharing in February 1959.

The development of transistor technology was fundamental to a new generation of electronic devices that later effected almost every aspect of the human experience. The long-sought realization of the field-effect transistor, in form of the MOS transistor (MOSFET), by Mohamed Atalla and Dawon Kahng at Bell Labs in 1959, brought new opportunities for miniaturization and mass-production for a wide range of uses.

Christopher Strachey passed the concept on to J. C. R. Licklider at a UNESCO-sponsored conference on Information Processing in Paris that year.

Licklider, Vice President at Bolt Beranek and Newman, Inc., discussed a computer network in his January 1960 paper Man-Computer Symbiosis: A network of such centers, connected to one another by wide-band communication lines [...] the functions of present-day libraries together with anticipated advances in information storage and retrieval and symbiotic functions suggested earlier in this paper.

The issue of connecting separate physical networks to form one logical network was the first of many problems. Early networks used message switched systems that required rigid routing structures prone to single point of failure. In the 1960s, Paul Baran of the RAND Corporation produced a study of survivable networks for the U.S. military in the event of nuclear war. Information transmitted across Baran's network would be divided into what he called "message blocks". Independently, Donald Davies (National Physical Laboratory, UK), proposed and was the first to put into practice a local area network based on what he called packet switching, the term that would ultimately be adopted. Larry Roberts applied Davies' concepts of packet switching for the ARPANET wide area network, and sought input from Paul Baran. Kleinrock subsequently developed the mathematical theory behind the performance of this technology building on his earlier work on queueing theory.

In August 1962, Licklider and Welden Clark published the paper "On-Line Man-Computer Communication" which was one of the first descriptions of a networked future.

In October 1962, Licklider was hired by Jack Ruina as director of the newly established Information Processing Techniques Office (IPTO) within DARPA, with a mandate to interconnect the United States Department of Defense's main computers at Cheyenne Mountain, the Pentagon, and SAC HQ.

InUnited States Department of Defense, Licklider formed an informal group within DARPA to further computer research. He began by writing memos in 1963 describing a distributed network to the IPTO staff, whom he called "Members and Affiliates of the Intergalactic Computer Network".

Although Licklider left the IPTO in 1964, five years before the ARPANET went live, it was his vision of universal networking that provided the impetus for one of his successors, Robert Taylor, to initiate the ARPANET development. Licklider later returned to lead the IPTO in 1973 for two years.

Following discussions with J. C. R. Licklider in 1965, Donald Davies became interested in data communications for computer networks. Later that year, at the National Physical Laboratory (United Kingdom), Davies designed and proposed a national data network based on packet switching. The following year, he described the use of an "Interface computer" to act as a router. The proposal was not taken up nationally but by 1967, a pilot experiment had demonstrated the feasibility of packet switched networks. He and his team were the first to use the term 'protocol' in a data-commutation context in 1967.

The Merit Network was formed in 1966 as the Michigan Educational Research Information Triad to explore computer networking between three of Michigan's public universities as a means to help the state's educational and economic development. With initial support from the State of Michigan and the National Science Foundation (NSF), the packet-switched network was first demonstrated in December 1971 when an interactive host to host connection was made between the IBM mainframe computer systems at the University of Michigan in Ann Arbor and Wayne State University in Detroit. In October 1972 connections to the CDC mainframe at Michigan State University in East Lansing completed the triad. Over the next several years in addition to host to host interactive connections the network was enhanced to support terminal to host connections, host to host batch connections (remote job submission, remote printing, batch file transfer), interactive file transfer, gateways to the Tymnet and Telenet public data networks, X.25 host attachments, gateways to X.25 data networks, Ethernet attached hosts, and eventually TCP/IP and additional public universities in Michigan join the network. All of this set the stage for Merit's role in the NSFNET project starting in the mid-1980s.

Robert Taylor was promoted to the head of the information processing office at Defense Advanced Research Projects Agency (DARPA) in June 1966. He intended to realize Licklider's ideas of an interconnected networking system. As part of the information processing office's role, three network terminals had been installed: one for System Development Corporation in Santa Monica, one for Project Genie at University of California, Berkeley, and one for the Compatible Time-Sharing System project at Massachusetts Institute of Technology (MIT). Taylor's identified need for networking became obvious from the waste of resources apparent to him.

Bringing in Larry Roberts from MIT, he initiated a project to build such a network. Roberts and Thomas Merrill had been researching wide area networking for computer time-sharing. At the first ACM Symposium on Operating Systems Principles in October 1967, Roberts presented a proposal for the "ARPA net", a distributed network using Interface Message Processors to create a message switching network. At the conference, Roger Scantlebury presented Donald Davies' work on packet switching and mentioned the work of Paul Baran at RAND. Roberts incorporated their concepts into the ARPANET design and upgraded the proposed communications speed to be used from 2.4 kbps to 50 kbps.

ARPANET development was centered around the Request for Comments (RFC) process, still used today for proposing and distributing Internet Protocols and Systems. RFC 1, entitled "Host Software", was written by Steve Crocker from the University of California, Los Angeles, and published on April 7, 1969. These early years were documented in the 1972 film Computer Networks: The Heralds of Resource Sharing.

ARPA awarded the contract to build the network to Bolt Beranek & Newman and the first ARPANET link was established between the University of California, Los Angeles (UCLA) and the Stanford Research Institute at 22:30 hours on October 29, 1969. "We set up a telephone connection between us and the guys at SRI ...", Kleinrock ... said in an interview: "We typed the L and we asked on the phone, "Do you see the L?" "Yes, we see the L," came the response. We typed the O, and we asked, "Do you see the O." "Yes, we see the O." Then we typed the G, and the system crashed ... Yet a revolution had begun" ....

By December 5, 1969, a 4-node network was connected by adding the University of Utah and the University of California, Santa Barbara. Building on ideas developed in ALOHAnet, the ARPANET grew rapidly.

The software for establishing links between network sites in the ARPANET was the Network Control Program (NCP), completed in c. 1970. Further development in the early 1970s by Robert E. Kahn and Vint Cerf let to the formulation of the Transmission Control Program, and its specification in December 1974 in RFC 675. This work also coined the terms catenet (concatenated network) and internet as a contraction of inter-networking, which describe the interconnection of multiple networks. This software was monolithic in design using two simplex communication channels for each user session.

Early international collaborations on ARPANET were sparse. Connections were made in 1973 to the Norwegian Seismic Array (NORSAR), via a satellite link at the Tanum Earth Station in Sweden, and to Peter Kirstein's research group at University College London.

The CYCLADES packet switching network was a French research network designed and directed by Louis Pouzin. First demonstrated in 1973, it was developed to explore alternatives to the early ARPANET design and to support network research generally. It was the first network to make the hosts responsible for reliable delivery of data, rather than the network itself, using unreliable datagrams and associated end-to-end protocol mechanisms. Concepts of this network influenced later ARPANET architecture.

With so many different network methods, something was needed to unify them. Robert E. Kahn of DARPA and ARPANET recruited Vinton Cerf of Stanford University to work with him on the problem. By 1973, they had worked out a fundamental reformulation, where the differences between network protocols were hidden by using a common internetwork protocol, and instead of the network being responsible for reliability, as in the ARPANET, the hosts became responsible. Cerf credits Hubert Zimmermann and Louis Pouzin (designer of the CYCLADES network), and his graduate students Judy Estrin, Richard Karp, Yogen Dalal and Carl Sunshine with important work on this design. Concurrently, an International Networking Working Group formed in 1972, led by Cerf; other active members included Alex ­McKenzie, ­Donald Davies, Roger ­Scantlebury, Louis Pouzin and Hubert Zimmermann.

Based on international research initiatives, particularly the contributions of Rémi Després, packet switching network standards were developed by the International Telegraph and Telephone Consultative Committee (ITU-T) in the form of X.25 and related standards. X.25 is built on the concept of virtual circuits emulating traditional telephone connections. In 1974, X.25 formed the basis for the SERCnet network between British academic and research sites, which later became JANET. The initial ITU Standard on X.25 was approved in March 1976.

Stemming from the first specifications of TCP in 1974, TCP/IP emerged in 1978 in nearly its final form, as used for the first decades of the Internet. which is described in IETF publication RFC 791 (September 1981).

The specification of the resulting protocol, the Transmission Control Protocol (TCP), was published as RFC (Request of Comments) 675 by the Network Working Group in December 1974. It contains the first attested use of the term internet, as a shorthand for internetworking.

The term "internet" was reflected in the first RFC published on the TCP protocol (RFC 675: Internet Transmission Control Program, December 1974) as a short form of internetworking, when the two terms were used interchangeably. In general, an internet was a collection of networks linked by a common protocol. In the time period when the ARPANET was connected to the newly formed NSFNET project in the late 1980s, the term was used as the name of the network, Internet, being the large and global TCP/IP network.

After the ARPANET had been up and running for several years, ARPA looked for another agency to hand off the network to; ARPA's primary mission was funding cutting edge research and development, not running a communications utility. Eventually, in July 1975, the network had been turned over to the Defense Communications Agency, also part of the Department of Defense.

In 1976, 12 computers and 75 terminal devices were attached, and more were added until the network was replaced in 1986. NPL, followed by ARPANET, were the first two networks in the world to use packet switching, and were interconnected in the early 1970s. The NPL team also carried out simulation work on packet networks, including datagram networks.

With the role of the network reduced to a core of functionality, it became possible to exchange traffic with other network independently from their detailed characteristics, thereby solving Kahn's initial problem. DARPA agreed to fund development of prototype software, and after several years of work, the first demonstration of a gateway between the Packet Radio network in the SF Bay area and the ARPANET was conducted by the Stanford Research Institute. On November 22, 1977 a three network demonstration was conducted including the ARPANET, the SRI's Packet Radio Van on the Packet Radio Network and the Atlantic Packet Satellite network.

The British Post Office, Western Union International and Tymnet collaborated to create the first international packet switched network, referred to as the International Packet Switched Service (IPSS), in 1978. This network grew from Europe and the US to cover Canada, Hong Kong, and Australia by 1981. By the 1990s it provided a worldwide networking infrastructure.

In 1979, two students at Duke University, Tom Truscott and Jim Ellis, originated the idea of using Bourne shell scripts to transfer news and messages on a serial line UUCP connection with nearby University of North Carolina at Chapel Hill. Following public release of the software in 1980, the mesh of UUCP hosts forwarding on the Usenet news rapidly expanded. UUCPnet, as it would later be named, also created gateways and links between FidoNet and dial-up BBS hosts. UUCP networks spread quickly due to the lower costs involved, ability to use existing leased lines, X.25 links or even ARPANET connections, and the lack of strict use policies compared to later networks like CSNET and Bitnet. All connects were local. By 1981 the number of UUCP hosts had grown to 550, nearly doubling to 940 in 1984.

By 1981, the number of hosts had grown to 213, with a new host being added approximately every twenty days. Building on ideas developed in ALOHAnet, the ARPANET grew rapidly

In 1981 NSF supported the development of the Computer Science Network (CSNET). CSNET connected with ARPANET using TCP/IP, and ran TCP/IP over X.25, but it also supported departments without sophisticated network connections, using automated dial-up mail exchange.

In 1982, one year earlier than the ARPANET, Peter Kirstein replaced the University College London transatlantic satellite links with TCP/IP over IPSS.

South Korea's first Internet system, the System Development Network (SDN) began operation on 15 May 1982. SDN was connected to the rest of the world in August 1983 using UUCP (Unixto-Unix-Copy); connected to CSNET in December 1984; and formally connected to the U.S. Internet in 1990.

ARPANET became the technical core of what would become the Internet, and a primary tool in developing the technologies used. The early ARPANET used the Network Control Program (NCP, sometimes Network Control Protocol) rather than TCP/IP. On January 1, 1983, known as flag day, NCP on the ARPANET was replaced by the more flexible and powerful family of TCP/IP protocols, marking the start of the modern Internet.

The software was redesigned as a modular protocol stack, using full-duplex channels. Originally named IP/TCP it was installed in the ARPANET for production use in January 1983.

In 1983, the U.S. military portion of the ARPANET was broken off as a separate network, the MILNET. MILNET subsequently became the unclassified but military-only NIPRNET, in parallel with the SECRET-level SIPRNET and JWICS for TOP SECRET and above. NIPRNET does have controlled security gateways to the public Internet.

Between 1984 and 1988 CERN began installation and operation of TCP/IP to interconnect its major internal computer systems, workstations, PCs and an accelerator control system. CERN continued to operate a limited self-developed system (CERNET) internally and several incompatible (typically proprietary) network protocols externally. There was considerable resistance in Europe towards more widespread use of TCP/IP, and the CERN TCP/IP intranets remained isolated from the Internet until 1989.

In 1986, the NSF created NSFNET, a 56 kbit/s backbone to support the NSF-sponsored supercomputing centers. The NSFNET also provided support for the creation of regional research and education networks in the United States, and for the connection of university and college campus networks to the regional networks.

In 1988, the first international connection to NSFNET was established by Piet Beertema at the Centrum Wiskunde & Informatica (CWI) in the Netherlands.

The use of NSFNET and the regional networks was not limited to supercomputer users and the 56 kbit/s network quickly became overloaded. NSFNET was upgraded to 1.5 Mbit/s in 1988 under a cooperative agreement with the Merit Network in partnership with IBM, MCI, and the State of Michigan. The existence of NSFNET and the creation of Federal Internet Exchanges (FIXes) allowed the ARPANET to be decommissioned in 1990.

Sublink Network, operating since 1987 and officially founded in Italy in 1989, based its interconnectivity upon UUCP to redistribute mail and news groups messages throughout its Italian nodes (about 100 at the time) owned both by private individuals and small companies. Sublink Network represented possibly one of the first examples of the Internet technology becoming progress through popular diffusion.

Since the 1990s, the Internet's governance and organization has been of global importance to governments, commerce, civil society, and individuals. The organizations which held control of certain technical aspects of the Internet were the successors of the old ARPANET oversight and the current decision-makers in the day-to-day technical aspects of the network. While recognized as the administrators of certain aspects of the Internet, their roles and their decision-making authority are limited and subject to increasing international scrutiny and increasing objections. These objections have led to the ICANN removing themselves from relationships with first the University of Southern California in 2000, and in September 2009, gaining autonomy from the US government by the ending of its longstanding agreements, although some contractual obligations with the U.S. Department of Commerce continued.

By 1990, ARPANET's goals had been fulfilled and new networking technologies exceeded the original scope and the project came to a close. New network service providers including PSINet, Alternet, CERFNet, ANS CO+RE, and many others were offering network access to commercial customers. NSFNET was no longer the de facto backbone and exchange point of the Internet. The Commercial Internet eXchange (CIX), Metropolitan Area Exchanges (MAEs), and later Network Access Points (NAPs) were becoming the primary interconnections between many networks.

In 1991, the People's Republic of China saw its first TCP/IP college network, Tsinghua University's TUNET. The PRC went on to make its first global Internet connection in 1994, between the Beijing Electro-Spectrometer Collaboration and Stanford University's Linear Accelerator Center. However, China went on to implement its own digital divide by implementing a country-wide content filter.

NSFNET was expanded and upgraded to 45 Mbit/s in 1991, and was decommissioned in 1995 when it was replaced by backbones operated by several commercial Internet service providers.

In 1992, the U.S. Congress passed the Scientific and Advanced-Technology Act, 42 U.S.C. (g)1862, which allowed NSF to support access by the research and education communities to computer networks which were not used exclusively for research and education purposes, thus permitting NSFNET to interconnect with commercial networks.

Finally, routing technologies were developed for the Internet to remove the remaining centralized routing aspects. The Exterior Gateway Protocol (EGP) was replaced by a new protocol, the Border Gateway Protocol (BGP). This provided a meshed topology for the Internet and reduced the centric architecture which ARPANET had emphasized. In 1994, Classless Inter-Domain Routing (CIDR) was introduced to support better conservation of address space which allowed use of route aggregation to decrease the size of routing tables.

The final restrictions on carrying commercial traffic ended on April 30, 1995 when the National Science Foundation ended its sponsorship of the NSFNET Backbone Service and the service ended.

In August 1995, InfoMail Uganda, Ltd., a privately held firm in Kampala now known as InfoCom, and NSN Network Services of Avon, Colorado, sold in 1997 and now known as Clear Channel Satellite, established Africa's first native TCP/IP high-speed satellite Internet services. The data connection was originally carried by a C-Band RSCC Russian satellite which connected InfoMail's Kampala offices directly to NSN's MAE-West point of presence using a private network from NSN's leased ground station in New Jersey. InfoCom's first satellite connection was just 64 kbit/s, serving a Sun host computer and twelve US Robotics dial-up modems.

In 1996, a USAID funded project, the Leland Initiative, started work on developing full Internet connectivity for the continent. Guinea, Mozambique, Madagascar and Rwanda gained satellite earth stations in 1997, followed by Ivory Coast and Benin in 1998.

In November 2005, the World Summit on the Information Society, held in Tunis, called for an Internet Governance Forum (IGF) to be convened by United Nations Secretary General. The IGF opened an ongoing, non-binding conversation among stakeholders representing governments, the private sector, civil society, and the technical and academic communities about the future of Internet governance. The first IGF meeting was held in October/November 2006 with follow up meetings annually thereafter. Since WSIS, the term "Internet governance" has been broadened beyond narrow technical concerns to include a wider range of Internet-related policy issues.

Tim Berners-Lee, inventor of the Internet, was becoming concerned about threats to the web's future and in November 2009 at the IGF in Washington DC launched the World Wide Web Foundation (WWWF) to campaign to make the web a safe and empowering tool for the good of humanity with access to all.

The first Internet link into low earth orbit was established on January 22, 2010 when astronaut T. J. Creamer posted the first unassisted update to his Twitter account from the International Space Station, marking the extension of the Internet into space.

On April 23, 2014, the Federal Communications Commission (FCC) was reported to be considering a new rule that would permit Internet service providers to offer content providers a faster track to send content, thus reversing their earlier net neutrality position.

On May 15, 2014, the FCC decided to consider two options regarding Internet services: first, permit fast and slow broadband lanes, thereby compromising net neutrality; and second, reclassify broadband as a telecommunication service, thereby preserving net neutrality.

On November 10, 2014, President Obama recommended the FCC reclassify broadband Internet service as a telecommunications service in order to preserve net neutrality.

On January 16, 2015, Republicans presented legislation, in the form of a U.S. Congress HR discussion draft bill, that makes concessions to net neutrality but prohibits the FCC from accomplishing the goal or enacting any further regulation affecting Internet service providers (ISPs).

On January 31, 2015, AP News reported that the FCC will present the notion of applying ("with some caveats") Title II (common carrier) of the Communications Act of 1934 to the internet in a vote expected on February 26, 2015.

On April 13, 2015, the FCC published the final rule on its new "Net Neutrality" regulations.

Finally, on October 1, 2016 ICANN ended its contract with the United States Department of Commerce National Telecommunications and Information Administration (NTIA), allowing oversight to pass to the global Internet community.

On December 14, 2017, the F.C.C Repealed their March 12, 2015 decision by a 3–2 vote regarding net neutrality rules.

In November 2019 at the IGF in Berlin, Berners-Lee and the WWWF went on to launch the Contract for the Web, a campaign initiative to persuade governments, companies and citizens to commit to nine principles to stop "misuse" with the warning "If we don't act now - and act together - to prevent the web being misused by those who want to exploit, divide and undermine, we are at risk of squandering" (its potential for good).