2.1 HISTORY AND DEVELOPMENT

Team-Fly

In the early 1960s, the Defense Advanced Research Projects Agency (DARPA)^[3] of the U.S. Department of Defense (DoD) sponsored some research and development projects that were asked to develop and come up with network technologies and topologies that provide maximal reliability even in the case of node and circuit failures. Packet switching was considered to be a possible answer to this problem [1–4]. Consequently, some DARPA-sponsored organizations and research laboratories started to think about interconnecting hosts using packet switching in the late 1960s. The resulting network was called the ARPANET, and the ARPANET was the predecessor of the Internet. In fact, the Internet has evolved from the ARPANET and sometimes people still confuse the ARPANET with the Internet. This confusion mainly results because the ARPANET was the first backbone network of the early Internet and remained a part of it until it was finally retired in 1990.

In 1969, the ARPANET interconnected four hosts located at the University of California at Los Angeles (UCLA), the Stanford Research Institute (SRI), the University of California at Santa Barbara (UCSB), and the University of Utah. After this initial phase, more hosts were added to the ARPANET, and work proceeded on specifying and implementing a functionally complete host-to-host protocol and other networking software. In 1970, the first ARPANET host-to-host protocol, called Network Control Protocol (NCP), was specified and deployed. In the following years, the NCP became the protocol of choice for ARPANET researchers and developers (there were not many users at this time).

In 1972, the ARPANET was demonstrated to the general public, and the first application, namely, electronic mail (e-mail), was introduced to be used on the ARPANET. Afterward, the name of the original ARPANET changed, from ARPA Internet to the Federal Research Internet to TCP/IP Internet and finally to its current name of just the Internet. The Internet, in turn, was (and still is) based on the idea that there would be multiple independent networks of rather arbitrary design, beginning with the ARPANET as the pioneering packet switching network, but soon to include packet satellite networks, packet radio networks, and some other networks.

Fairly soon it was realized that the NCP did not sufficiently meet the needs and requirements of the evolving Internet. Consequently, it was decided to develop a new version of the protocol. The development effort was mainly driven by Robert E. Kahn at Bolt Beranek and Newman (BBN), Inc., and Vinton G. Cerf at Stanford University. In the early 1970s, the resulting approach was publicly announced and the corresponding protocol was named Transmission Control Protocol (TCP). TCP was originally intended to support a range of transport services, ranging from totally reliable sequenced data delivery (following the virtual circuit model) to a datagram delivery service in which the application made direct use of the underlying network service, which might imply occasional lost, corrupted, or reordered packets.

However, the initial effort to implement TCP resulted in a version that was considerably simpler and only provided support for the virtual circuit model. This simpler version of TCP worked fine for file transfer and remote terminal access, but some of the early work on advanced network applications, in particular voice transmission over packet switched networks, made clear that in some cases packet losses should not be corrected by TCP, but should be left to the application to deal with. This led to a reorganization of TCP into two protocols:

The Internet Protocol (IP), which provided only for the addressing and forwarding of individual packets;
The new TCP, which was concerned with more sophisticated service features, such as flow control and recovery from packet losses.

For those applications that did not need the services provided by TCP, an alternative transport protocol was added to provide direct access to the basic packet delivery service provided by IP. This protocol was called and further referred to as the User Datagram Protocol (UDP). UDP is primarily used in situations where applications can live with packet losses (e.g., real-time communications), or in situations where the establishment of TCP connections is simply too inefficient or costly (e.g., multicast applications).

After this design effort, DARPA had three contractors at Stanford, BBN, and UCLA implement TCP/IP. The Stanford team produced the detailed specification and within about a year there were three independent and interoperable implementations of TCP. This was the beginning of long-term experimentation and development to evolve and mature the Internet concepts and technology. Since the first three networks (ARPANET, Packet Radio, and Packet Satellite) and their initial research communities began, the experimental environment has grown to incorporate essentially every form of network and a very broad research and development community.

In 1980, TCP/IP was adopted as a DoD standard, and on January 1, 1983, the ARPANET host protocol was officially changed from NCP to TCP/IP. This transition was carefully planned within the Internet community for several years before it actually took place and went surprisingly smoothly (in fact, it resulted in a distribution of buttons saying "I survived the TCP/IP transition"). The transition of ARPANET from NCP to TCP/IP also permitted it to be split into a MILNET supporting operational requirements for the military and an ARPANET supporting research needs. Thus, by 1985, the Internet was already well established as a technology supporting a broad community of researchers and developers, and was beginning to be used by other communities for daily computer communications. E-mail was being used broadly across several communities, often with different systems, but interconnection among different e-mail systems was demonstrating the utility of broad-based electronic communications among people.

Much of the early popularity of the TCP/IP protocols was due to their implementation in version 4.2 of the BSD UNIX operating system. The BSD UNIX was developed at the University of California at Berkeley (UCB) by the Computer Systems Research Group (CSRG). Again, the development was partly funded by DARPA. Because of the source of its funding, BSD UNIX version 4.2 was made publicly available at the cost of its distribution and so its use spread quickly. Co-incidentally, BSD UNIX version 4.2 became available at the same time as some inexpensive microprocessors, such as the Motorola 680x0 and the Intel 80x86 chip series. Both startup and established companies took advantage of the combination to build computer systems, mainly workstations, incorporating both the newly available microprocessors and the BSD UNIX operating system. Notably the most prominent startup company of this kind is Sun Microsystems. Since then, the development of the UNIX operating system and the TCP/IP protocols became intimately intertwined. Nevertheless, it is important to say that the Internet is not a UNIX network and that the TCP/IP protocols have been implemented on most other operating systems, including, for example, Microsoft Windows NT and Windows 2000.

At the same time that the Internet technology was being experimentally validated and widely used among computer science researchers, other networks and networking technologies were being pursued. The usefulness of computer networking demonstrated by DARPA and the DoD contractors on the ARPANET was not lost on other communities and disciplines, so that by the mid-1970s, computer networks began to spring up wherever funding could be found for the purpose. For example, the U.S. Department of Energy (DoE) established MFENet for its researchers in Magnetic Fusion Energy, whereupon DoE's High Energy Physicists responded by building HEPNet. The National Aeronautics and Space Agency (NASA) followed with SPAN, and some researchers established CSNET for the (academic and industrial) computer science community with an initial grant from the U.S. National Science Foundation (NSF). AT&T's free-wheeling dissemination of the UNIX operating system spawned USENET, based on UNIX's built-in UUCP communication protocols, and in 1981 BITNET was initiated to interconnect academic mainframe computers.

With the exception of BITNET and USENET, these early networks (including the ARPANET) were purpose-built, meaning that they were intended for, and largely restricted to, closed user communities. Consequently, there was little pressure for the individual networks to be compatible and, indeed, they largely were not. In addition, alternate technologies were being pursued in the commercial sector, including, for example, XNS from Xerox, DECNet from DEC, and SNA from IBM. It remained for the British JANET and the U.S. NSFNET programs to explicitly announce their intent to serve the entire higher education community, regardless of discipline.

During the 1980s and 1990s, the roles of DARPA and NSF fundamentally changed, and NSF's privatization policy culminated in 1995 with the defunding of the NSFNET Backbone. Today, the Internet is mainly funded by private companies, collectively referred to as Internet Service Providers (ISPs). ISPs can operate in the backbone of the Internet or at the edges. Big telecommunications companies typically operate in the backbone, whereas most smaller ISPs only operate at the edges and provide local Internet connectivity to their subscribers.

Today, the Internet is growing exponentially and diversifying rapidly, hence the exact form of its future is unpredictable. Beginning about 1988, the size of the Internet has been more than doubling every year and promises to continue doing so. The current and future status of the Internet is overviewed next.

^[3]The Advanced Research Projects Agency (ARPA) changed its name to Defense Advanced Research Projects Agency (DARPA) in 1971, then back to ARPA in 1993, and back to DARPA in 1996. In this book, we use the term DARPA to refer to either of the two acronyms.

Team-Fly