From Transcontinental Telephone Service to the Internet
Part I of "A Slice of History" (see the fall 1998 issue
of Academic Computing), covers the early landmarks of
the whirlwind evolution of the telecommunication industry.
Part II describes the evolution of the telephone from curio
to commercial success and that of the Internet from
Licklider's academic dream, the "Galactic Network," to its
current pervasive reality.
What hath God wrought?" was the first telegraph message tapped in the form of dots and dashes and sent from Washington, D.C. to Baltimore, Maryland in 1844. This short phrase ushered in the era of instant, although mostly one-way communications. Just over 30 years later, Alexander Graham Bell laid claim to the patent of the telephone and opened the era of instant two-way communications with the request to his assistant: "Mr. Watson come here. I want you!"1 Barely a decade afterwards, lines were being strung from city to city in the United States paving the way for rapid intercity long-distance communications.
The American Telegraph and Telephone Company had been chartered early in 1885 as a subsidiary of American Bell; its sole objective was to build and operate long-distance phone lines. In May of that year, AT&T Vice-President Edward Hall, Jr., recommended the construction of a long-distance line between New York and Philadelphia. Referring to an existing line he said: "We know that a metallic (two wire) circuit can be operated between New York and Boston . . . but we do not know that additional parallel circuits can be successfully worked."1 The New York - Boston circuit was still considered experimental even though it was in commercial service, and the New York - Philadelphia line would test whether or not the line would support added circuits. With the construction of the new line, Hall proposed several innovations. He suggested splitting the toll charge between the local operating company and AT&T and that local subscribers be able to use the line without having to go to special long-distance telephones. The first telephone conversation took place over the line that year in December. Commercial telephone service however, would not begin operation until 1887 over a year later.
It was five years before AT&T opened its next long-distance line. In 1892, the heavy No. 8 gauge copper 920 mile circuit from New York to Chicago opened for commercial service. Alexander Graham Bell himself, long since officially retired from the Bell system, took part of the opening ceremony in the AT&T main office in New York City.
Over the next decade city-to-city lines appeared in
close succession: Chicago, Cincinnati, Nashville, Kansas City, Omaha,
St. Louis, Minneapolis, Norfolk, Richmond and Atlanta joined
the growing network which had originated in the northeast.
Despite this rapid introduction of intercity long-distance service,
technological hurdles would prevent coast to coast service from being
established for many more years.
Local and long-distance telephone service to this point was unaided by amplification devices, requiring very thick (and expensive) copper wire, such as that used in the New York-Chicago circuit, to minimize attenuation (signal degradation). Even so the voice currents were often maddeningly weak; in poor weather conditions they were often nonexistent.
In 1900, as knowledge of electric theory was better understood, Bell system engineers began experimenting with devices know as loading coils. They were tested on two separate 24-mile circuits between the cities of Boston, Jamaica Plain and West Newton, Massachusetts. Loading coils only reduced voice current attenuation under certain conditions and they did not actively amplify signals.
A Bell System engineer, H.R. Shreeve, developed the first amplification device in 1904. A Shreeve repeater amplified the signal by passing the circuit through the working parts of a telephone receiver, creating in a sense, a device that talked to itself. Shreeve repeaters were first placed into service near Pittsburgh, Pennsylvania on the New York - Chicago circuit.
The long-distance service established between New York City and Denver, Colorado in 1911 represented the practical limits of phone service given the technology of the day. But the forward looking Theodore Vail who had returned to AT&T as president, anticipating further advances in amplification technology, promised transcontinental telephone service by the opening of the 1914 Panama-Pacific Exposition in San Francisco.
The advance that would make Vail's vision real was the vacuum tube triode. The vacuum tube triode, invented in 1906 by Lee DeForest, would successfully be adapted to function as a true amplifier in 1912.
Advances in vacuum tube electronics also made possible an improved ability to increase telephone circuit capacity, which had first been made possible by the 'Phantom' telephone circuits, adapted into use by Bell Engineer John Carty in 1886. Further progress in vacuum tube technology by Bell System engineers also allowed the development of practical active amplification devices, thus transcontinental telephone service became possible.
On June 17, 1914, near the Nevada-Utah border, the last pole of the transcontinental telephone line was set. The first test of the New York - San Francisco circuit of this line, which used both loading coils and vacuum tube repeaters, took place a month later, but commercial service would not commence until January of the following year.
On January 25, 1915 at 4 PM EST commercial transcontinental phone service was officially inaugurated with ceremonies at AT&T headquarters in New York City and at the Panama-Pacific Exposition in San Francisco. In New York city, sixty-eight year-old Alexander Graham Bell connected a replica of his original "gallows-type" telephone to the line, and repeated the first words ever spoken and heard by telephone. "Mr. Watson, come here. I want you" he told his former assistant in San Francisco. Thomas Watson replied that he would love to, but that now it would take him a week to get there.1
The achievement of transcontinental telephone service
brought a new challenge: transatlantic telephone service.
In addition to making amplification devices practical for use, Lee DeForest's vacuum tube became the foundation for the development of the high frequency oscillator and modulator circuits, which made radiotelephony possible. In 1901 Guglielmo Marconi transmitted the first transatlantic radio signal, the letter S, in Morse Code, between Cornwall, England and Newfoundland, Canada. That same year Vail had written to one of AT&T's London bankers that "the difficulties of wireless telegraph are nothing compared with the difficulties in the way of the wireless telephone."1 One of the main problems was that a submarine transatlantic telephone cable and amplifiers would be unrealistically expensive to maintain and therefore unprofitable. Thus AT&T focused its efforts on radiotelephone research. The first radiotelephone communication would take place in 1915, the same year commercial transcontinental long-distance telephone service began. It was a successful wireless transmission of signals over the 1,000 miles between Long Island, New York and St. Solomon's Island, Georgia. This was followed later that same year by a successful transatlantic transmission from Arlington, Virginia, to the Eiffel Tower in Paris, France, where Bell Engineer H. R. Shreeve was able to hear a few, weak words. 'Practical' transatlantic telephone service however, would not exist for more than another decade. This commercial transatlantic radiotelephone service would be often unreliable, plagued by weak signals, fading, and static.
In the late forties, researchers at Bell Labs resumed the work they had started before the war: researching solid state electronics in order to develop a more efficient switching system for telephone communication. In addition to making feasible a durable submarine transatlantic cable, the transistor, humbly described by Bell Labs as "a device...which has several applications in radio where a vacuum tube is ordinarily employed" 1 would lead to the development of integrated circuits and microchips thus laying the groundwork for the modern computer and the Internet.
What we have come to know today as the Internet was originally conceived in the early 1960s in the era of large time-sharing computers. It has continued to evolve in the present era of desktop and network computers and local area networks (LANs). It was designed in a time before LANs were even imagined, and has accommodated the computing revolution with new network technology. It was first expected to support a range of functions such as file sharing and remote login. It would lead to electronic mail and the World Wide Web. As with the telephone and telegraph, it started as the creation of a small band of dedicated researchers and has grown to be a world-wide commercial success.
In 1962 MIT professor J.C.R. Licklider, who became the first head of the computer research program at the Defense Advanced Research Projects Agency (DARPA, later ARPA), put forward his "Galactic Network" concept. In the "Galactic Network," much like the modern Internet, users would be able to quickly access data and programs from any site.
In 1964, while researching secure communication via packet switching with colleagues W. D. Davies and Paul Baran, Leonard Kleinrock, also of MIT, published the first book on the subject, "On Communication Networks" based on a 1961 paper about packet switching theory. Their research confirmed the theoretical feasibility of communication using information packets rather than circuits, an important step towards computer networking.
A key step was to make computers talk. In 1965 researchers connected the TX-2 computer in Massachusetts to the Q-32 in California with a low speed dial-up telephone line creating the first wide-area computer network. Although the experiment was considered a success, the communication itself was slow, expensive, and inefficient. It did confirm the theory that computers could work well together in both running programs and retrieving data. Importantly, it also proved as Kleinrock had predicted, that packet switching would be the most promising solution to computer networking. The circuit-switched telephone system was totally inadequate for the job for reasons of speed and scalability.
1966 marked the birth of the ARPANET, a government-funded project aimed at establishing a network which, among other objectives, would test the packet-switching theory. The first host-to-host message sent over the experimental network was exchanged between UCLA and the Stanford Research Institute in 1969. Two more nodes, added at UC Santa Barbara and University of Utah, incorporated application visualization projects: display of mathematical functions at UCSB and network 3-D representations at Utah. Systems were added to the ARPANET for the next few years at the rate of one per month and by the end of 1971, nineteen nodes were connected to the ARPANET.4
In 1970 ARPA's Network Working Group (NWG) completed the preliminary ARPANET Host-to-Host protocol, Network Control Protocol (NCP), which systematized the way computers talked to one another and meant that network users could finally begin to develop network applications. A year later NWG finished the telnet protocol and began work on the file transfer protocol, both of which are still in use today.
In March of 1972 Ray Tomlinson of BBN, a company integrally involved with the evolution of ARPANET, was motivated by the need of the ARPANET developers for an easy collaboration mechanism, to develop the basic e-mail send-and-read software. ARPA's Laurence Roberts expanded the capabilities of e-mail by writing the first e-mail utility program to list, selectively read, file, forward, and respond to messages. From there e-mail took off as the largest network application for over a decade.
The new ARPANET network technology was introduced to the public in 1972 at the International Computer Communication Conference (ICCC). From there the Internet rapidly evolved out of the idea that multiple independent networks of arbitrary design would join the ARPANET, the first packet switched network. It could then accommodate other networks, such as packet satellite, ground-based packet radio and other networks.
Open-architecture networking, the fundamental idea which underlies the structure of the modern Internet was first proposed by Bob Kahn, at ARPA. Kahn defined "Internetting" as it was called at the time, by four critical ground rules:
Other key issues included: developing algorithms to prevent lost packets from permanently disabling communications; providing for host to host "pipelining" so that multiple packets could be en route from source to destination as allowed by participating hosts and intermediate networks; and a network-to-network protocol. Kahn and Vint Cerf (a Stanford researcher who headed the first International Networking Working Group), presented their first paper on the new internetworking protocol, TCP, in 1973. Three years later they demonstrated 'Internetting' in public demonstrations where they interconnected a Packet Radio network, SATNET, with the ARPANET.4
By 1984 over 1000 hosts were connected to the nascent Internet. To make it easy for people to use the network, each host machine was assigned a name so that it would not be necessary to have to remember each host's numeric addresses. When the network consisted of only a few hosts, each host maintained a single table of all the hosts and their associated names and addresses. With an increasing number of independently managed networks (e.g., LANs) a single table of hosts was no longer feasible. To maintain network scalability without introducing the administrative nightmare of managing such a large hosts table, Paul Mockapetris of University of Southern California Information Sciences Institute (USC/ISI) developed the Domain Name System (DNS). The DNS permitted a scalable distributed mechanism for resolving hierarchical host names into an Internet address. This simplified administration, since by 1987 the Internet had grown to ten thousand sites. By 1992, this number would grow to over a million.4
Early networks were built for and largely restricted to closed communities of scholars. Starting in the early 1980s and continuing the present, the Internet grew beyond its primarily research roots to include both a broad user community and increased commercial activity. Originally, commercial efforts mainly comprised vendors providing the basic networking products, and service providers offering the connectivity and basic Internet services. The Internet has now become a "commodity" service, and much of the latest attention has been focused on the use of this global information infrastructure for support of other commercial services. This has been tremendously accelerated by the widespread and rapid integration of World Wide Web and browser technology, allowing users easy access to information linked throughout the globe. Products which facilitate the distribution of that information, and many of the latest developments in technology, have been aimed at providing increasingly sophisticated information services on top of the basic Internet data communications.
The Internet has its roots in the electric telegraph; these widespread means of communications mirror each other at different stages of their development in significant ways. In 1871 Samuel F. B. Morse, inventor of the telegraph, tapped out his final farewell to a standing ovation from the global community of telegraphers. By this time, this "instantaneous highway of thought" (as it was described at the time) 650,000 miles of telegraph wire and 30,000 miles of undersea cable linked over 20,000 cities and villages throughout the world. A distinct telegraphic subculture had emerged with its own customs and vocabulary, and a hierarchy based on the speed at which operators could send and receive Morse code. A little over a decade later, the incipient American Telephone and Telegraph Company under the guidance of Theodore Vail, was reaching out to link cities and towns with the new telephone network. The telephone's capacity for two-way instant communications brought people into even closer contact.
The first Internet, a natural outgrowth of this mode of personal communication, was built on the many advances made by the telephone industry (i.e. the transistor, microwave radio relay, cable video transmission, teletypewriter networks). Conceived to allow remote logins and data retrieval, today's Internet is evolving to support even higher bandwidth services, such as streaming audio and video; video telephones and teleconferencing. Modern networking combined with powerful yet inexpensive laptop computers, two-way pagers, PDAs (Personal Digital Assistants), and cellular phones is making possible a new paradigm of J. C. R. Licklider's Galactic Network vision with high-speed, portable, and mobile communications. The development of the Internet2 is proof that the Internet's evolution is not yet complete.
Jeff Nucciarone, Center for Academic Computing
Sources
1 Brooks, John. Telephone, the First Hundred Years. New York: Harper & Row, 1976.
2 Coll, Steve. The Deal of the Century: The Breakup of AT&T. New York: Atheneum, 1986.
3 Davie, Bruce S. and Larry L. Peterson. Computer Networks: A Systems Approach. San Francisco: Morgan Kaufmann Publishers, 1996.
4 "A History of the Internet: 1962-1992" (1997) The Computer Museum, with support from the Association for Computing Machinery/Institute of Electrical and Electronics Engineers, Inc.