History of the Internet

The earliest idea of a computer network intended to allow general communication between users of various computers was the ARPANET, the world's first packet switching network, which first went online in 1969.

The Internet's roots lie within the ARPANET, which not only was the intellectual forerunner of the Internet, but was also initially the core network in the collection of networks in the Internet, as well as an important tool in developing the Internet (being used for communication between the groups working on internetworking research).

Motivation for the Internet

The need for an internetwork appeared with ARPA's sponsorship, by Robert E. Kahn, of the development of a number of innovative networking technologies; in particular, the first packet radio networks (inspired by the ALOHA network), and a satellite packet communication program. Later, local area networks (LANs) would also join the mix.

Connecting these disparate networking technologies was not possible with the kind of protocols used on the ARPANET, which depended on the exact nature of the subnetwork. A wholly new kind of networking architecture was needed.

Early Internet work

Kahn recruited Vint Cerf to work with him on the problem, and they soon 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 Herbert Zimmerman and Louis Pouzin (designer of the CYCLADES network) with important influences on this design. Some accounts also credit the early networking work at Xerox PARC with an important technical influence.

With the role of the network reduced to the bare minimum, it became possible to join almost any networks together, no matter what their characteristics, thereby solving Kahn's initial problem. (One popular saying has it that TCP/IP, the eventual product of Cerf and Kahn's work, will run over "two tin cans and a string".) A computer called a gateway (a name later changed to router to avoid confusion with a number of other kinds of devices, also called gateways) is provided with an interface to each network, and forwards packets back and forth between them.

Happily, this new concept was a perfect fit with the newly emerging local area networks, which were revolutionizing communication between computers within a site.

Early growth

In 1983, TCP/IP protocols replaced the earlier NCP protocol as the principal protocol of the ARPANET; in 1984, the US military portion of the ARPANet was broken off as a separate network, the MILNET.

The early Internet, based around the ARPANET, was government-funded and therefore restricted to non-commercial uses such as research; un-related commercial use was strictly forbidden. This initially restricted connections to military sites and universities. During the 1980s, the connections expanded to more educational institutions, and even to a growing number of companies such as Digital Equipment Corporation and Hewlett-Packard, which were participating in research projects, or providing services to those who were.

Another branch of the US government, the National Science Foundation, became heavily involved in the Internet in the mid 1980s. The NSFNet backbone, intended to connect and provide access to a number of supercomputing centers established by the NSF, was established in 1986.

At the end of the 1980s, the US Department of Defense decided the network was developed enough for its initial purposes, and decided to stop further funding of the core Internet backbone. The ARPANET was to be gradually shut down (its last node was turned off in 1989), and NSF took over responsibility for providing long-haul connectivity in the US.

In another NSF initiative, regional TCP/IP-based networks such as NYSERNet (New York State Education and Research Network) and BARRNet (Bay Area Regional Research Network), grew up and started interconnecting with the nascent Internet. This greatly expanded the reach of the rapidly growing network. On April 30, 1995 the NSF privatized access to the network they had created. It was at this point that the growth of the Internet really took off.

Commercialization and privatization

Parallel to the Internet, other networks were growing. Some were educational and centrally-organized like BITNET and CSNET. Others were a grass-roots mix of school, commercial, and hobby like the UUCP network.

During the late 1980s the first Internet Service Provider companies were formed. Companies like PSINet, UUNET, Netcom, and Portal were formed to provide service to the regional research networks and provide alternate network access (like UUCP-based email and Usenet News) to the public.

The interest in commercial use of the Internet became a hotly-debated topic. Although commercial use was forbidden, the exact definition of commercial use could be unclear and subjective. Everyone agreed that one company sending an invoice to another company was clearly commercial use, but anything less was up for debate. The alternate networks, like UUCP, had no such restrictions, so many people were skirting grey areas in the interconnection of the various networks.

Many university users were outraged at the idea of non-educational use of their networks. Ironically it was the commercial Internet service providers who brought prices low enough that junior colleges and other schools could afford to participate in the new arenas of education and research.

By 1994, the NSFNet lost its standing as the backbone of the Internet. Other competing commercial providers created their own backbones and interconnections. Regional NAPs (network access points) became the primary interconnections between the many networks. The NSFNet was dropped as the main backbone, and commercial restrictions were gone.

Early applications

E-mail had existed as a message exchanging service on early time sharing mainframe computers connected to a number of terminals. In about 1971 it developed into the first system of exchanging addressed messages between different, networked computers; in 1972 Ray Tomlinson introduced the "name@computer" notation that is still used today. E-mail turned into the internet "killer application" of the 1980s.

The second most popular application of the early internet was usenet, a system of distributed discussion groups which is still going strong today. Usenet had existed even before the internet, as an application of Unix computers connected by telephone lines via the UUCP protocol.

It wasn't until the early to mid 1980s that the services we now use most on the Internet started appearing. The concept of "domain names" (like "wikipedia.org") requiring "Domain Name Servers" wasn't even introduced until 1984. Before that all the computers were just addressed by their IP addresses (numbers) or used a central "hosts" file maintained by the Network Information Centre (NIC). Most protocols used for email and other services were significantly enhanced after this.

Standards and control

The Internet has developed a significant subculture dedicated to the idea that the Internet is not owned or controlled by any one person, company, group, or organization. Nevertheless, some standardization and control is necessary for anything to function.

Many people wanted to put their ideas into the standards for communication between the computers that made up this network, so a system was devised for putting forward ideas. One would write one's ideas in a paper called a "Request for Comments" (RFC for short), and let everyone else read it. People commented on and improved those ideas in new RFCs. (With its basis as an educational research project, much of the documentation was written by students or others who played significant roles in developing the network but did not have official responsibility for defining standards. This is the reason for the very low-key name of "Request for Comments" rather than something like "Declaration of Official Standards".) The first RFC (RFC1) was written on April 7th, 1969. There are now over 3500 RFCs, describing every aspect of how the internet functions.

The Internet standards process has been as innovative as the Internet itself. Prior to the Internet, standardization was a slow process run by committees with arguing vendor-driven factions and lengthy delays. In networking in particular, the results were monstrous patchworks of bloated specifications.

The fundamental requirement for a networking protocol to become an Internet standard is the existence of at least two working implementations that inter-operate with each other. This makes sense looking back, but it was a new concept at the time. Other efforts built huge specifications with many optional parts and then expected people to go off and implement them, and only later did people find that they did not inter-operate, or worse, the standard was completely impossible to implement.

In the 1980s, the International Organization for Standardization (ISO) documented a new effort in networking called Open Systems Interconnect or OSI. Prior to OSI, networking was completely vendor-developed and proprietary. OSI was a new industry effort, attempting to get everyone to agree to common network standards to provide multi-vendor interoperability. The OSI model was the most important advance in teaching network concepts. However, the OSI protocols or "stack" that were specified as part of the project were a bloated mess. Standards like X.400 for e-mail took up several large books, while Internet e-mail took only a few dozen pages at most in RFC-821 and 822. Most protocols and specifications in the OSI stack, such as token-bus media, CLNP packet delivery, FTAM file transfer, and X.400 e-mail, are long-gone today. Only one, X.500 directory service, still survives with significant usage, mainly because the original unwieldy protocol has been stripped away and effectively replaced with LDAP.

Some formal organization is necessary to make things operate. The first central authority was the NIC (Network Information Centre) at SRI (Stanford Research Institute in Menlo Park, California).

World Wide Web

(see also Origins of the World Wide Web)

One of the parts of the Internet many people are most familiar with is the World Wide Web.

As the Internet grew through the 1980s and early 1990s, many people realized the growing need to be able to find and organize files and related information. Projects such as Gopher, WAIS, and the Anonymous FTP Archive Site list attempted to create schemes to organize distributed data and present it to people in an easy-to-use form. Unfortunately, these projects fell short in being able to accommodate all the various existing file and data types, and in being able to grow without centralized bottlenecks.

One of the most promising ideas was hypertext, inspired by Vannevar Bush's "memex", Ted Nelson's Project Xanadu and the Note Code Project. Small self-contained hypertext systems had been created before, such as Apple Computer's HyperCard, but before the Internet, nobody had worked out how to scale it up so that it could to refer to another document anywhere in the world.

The solution was invented by Tim Berners-Lee in 1989. He was a physicist working at CERN, the European Particle Physics Laboratory, and wanted a way for physicists to share information about their research. His documentation project was the source of the two key inventions that made the World Wide Web possible.

The two key inventions were the uniform resource locator (URL) and hypertext markup language (HTML). The URL was a simple way to specify the location of a document anywhere on the Internet in one simple name that specified a computer name, a file identification on that machine, and a protocol to use. HTML was an easy way to embed codes into a text file that could define the structure of a document and also include links pointing to other documents. An additional network protocol, a hypertext transfer protocol (HTTP), was also invented for reduced overhead in transfers, but the true genius of the new system was that a new protocol was useful but not necessary; the URL and HTML system was backwards compatible with existing protocols like FTP and Gopher.

Later around 1992 people realized that the simple markup capabilities of HTML could allow graphics to be included in text documents. The first graphical web browsers were developed: Viola and Mosaic. Mosaic was developed by a team at the National Center for Supercomputing Applications at the University of Illinois at Urbana-Champaign (NCSA-UIUC), led by Marc Andreesen. Andreesen left NCSA-UIUC and joined Jim Clark, one of the founders of SGI (Silicon Graphics, Inc.). They started Mosaic Communications which became Netscape Communications Corporation , making Netscape Navigator the first commercially successful browser. Microsoft acquired technology from Spyglass, Inc., (who got their technology from NCSA) to develop Internet Explorer.

The ease of creating new Web documents and linking to existing ones caused exponential growth. As the Web grew, search engines were created to track pages on the web and allow people to find things. The first search engine, Lycos, was created in 1993 as a university project. In 1993, the first web magazine, The Virtual Journal, was published by a University of Maine student. At the end of 1993, Lycos indexed a total of 800,000 web pages.

By end of 2000 the phenomen of Dot-com bubble bursted and many overpriced companies went bankrupt overnight.

By August 2001, the Google search engine tracked over 1.3 billion web pages and the growth continues. In early 2004, Google's index exceeded 4 billion pages.

External links

The Internet Society History Page
A Brief History of NSF and the Internet
Hobbes' Internet Timeline v7.0
The Birth of the Internet, with a picture of IMP #1.