Episode 8: The Birth of the Internet

Like other media before it, the internet was possible because of pioneering individuals, government supported research, and several defining events that seemed to unfold independently. Each of these factors were, in fact, curiously related, creating the conditions for the world’s first interactive network during the second half of the 20th century, just as broadcast and cable television expanded.

By the 1990’s it was clear that all media were changing. Radio, the old workhorse of communications, was becoming more marginalized, losing greater audience shares to network television and cable, while searching for a new business model that would prove elusive until the introduction of satellite radio later in the decade.

Broadcast television was also showing its age. The old monopoly of NBC-CBS-ABC was not only losing viewers to cable, but its status as a “news platform” was being challenged by a fourth network with roots in the print world: Fox Television. This new contender employed an aggressive strategy, shedding “objective reporting” for blatant political coverage, then challenging the networks by signing up disenchanted affiliate stations, while allocating vast sums for sports, business, and children’s programming.

Cable television was faring better. In the 90’s subscriptions were nearing 50% of all TV households. System capacity reached 100 channels, taking advantage of breakthroughs in fiber optic technology and an increase in programming services distributed by satellite. Regulatory oversight also was weakening, in part the result of well funded lobbying efforts by the National Cable Television Association, restrictions on municipal policies by the FCC, and federal legislation in 1984, 1986, and 1996, which removed most rate ceilings.

Cable also introduced high speed internet access,called “broadband”, to emphasize large capacity fiber optic “pipes” used to send video, data, and soon, voice transmissions. And yet, even as cable acted as a mid-wife to these changes, neither cable, broadcast television, or radio, was prepared for the “dot.com” revolution that announced the rise of the Internet in the early 90’s. How this new medium developed is a fascinating story in itself.

“Augmented” Computers

A key element in the introduction of the internet was a device that was envisioned in 1956: the personal computer. The person credited with first articulating this vision was Douglas Engelbart.

In December 1968 Engelbart conducted the first public demonstration of his “ONLine System” (NLS), the culmination of a decade’s work at Stanford Research Institute in California to develop “augmented” computers. Speaking before a rapt, overflow crowd in downtown San Francisco, Engelbart sat in a darkened auditorium before a 3 foot computer “work station” of his own design, with a small screen adapted from a radar scanner, and a brick-like pointing device to direct a “cursor”. The computer showed text, black and white graphics, and multiple windows, with hyper-text links that could call up files stored on a larger computer connected by telephone lines to SRI. Behind him, the computer’s output was projected on a 20 foot screen, the result of a video camera operated at the back of the hall and instructions communicated by a telephone headset Engelbart wore connecting him to engineers and research associates in Palo Alto and Menlo Park. It was, as one author described it, “The Mother of All Demos”, unprecedented in scope and potential, and the first public demonstration of interactive computing.

Engelbart’s journey to that triumphant debut had not been smooth. A former Navy radar operator, Engelbart returned from World War II to earn a doctorate in engineering from Berkeley, then worked briefly at the Ames Research Center in Mountain View, and later taught at Stanford before quitting to start his own company, Digital Techniques, in 1956. His goal was to create a modern computer workstation that could “augment” human intelligence by organizing and sharing information that was stored on it. When the venture failed, Engelbart found a home in 1959 at SRI, a research facility supported by Stanford University and the Department of Defense.

SRI allowed Engelbart to organize his own research group: The Augmented Human Intellect Research Center, and gave him virtually free rein to develop his concept. Engelbart’s vision, however, was a radical departure from conventional scientific wisdom that confined computers to card punch input and mathematical calculations. Even at SRI, in an environment that encouraged all forms of personal and intellectual whimsy, Engelbart competed both for respect and research funds with advocates of “Artificial Intelligence”, a theory that envisioned the replacement of human intelligence by large scale computers.

Engelbart’s refinement, in contrast, emphasized supplementing human knowledge and collaboration, which would be possible because of the miniaturization of processing chips and computers. As a result, communities of scientists would have instantaneous access to information, and would be able to collaborative with colleagues in far-flung research centers, all from the comfort of an office workstation. This would replicate the environment Engelbart enjoyed at SRI by creating working forums which would connect scientists, engineers, and entrepreneurs, and encourage them to seek “dynamic breakthroughs”. Engelbart’s vision, which anticipated modern computers and global networks, was first summarized in a report for the Air Force in October 1962 called: “Augmenting Human Intelligence: A Conceptual Framework”.

The augmentation study was the catalyst that sparked the imagination of several key figures in Internet history. The first were J.C.R. Licklider and Robert W. (Bob) Taylor, the directors of a Department of Defense funding unit called the Information Processing Techniques Office (IPTO). Licklider had written a similar paper in 1960, “Man-Compter Symbiosis”, and provided Engelbart’s first research funds. Taylor succeeded Licklider and encouraged Engelbart to contribute to the development of ARPA-Net, a computer network established by the Defense Department to link scientists and military officials around the country. Taylor then provided research grants that led to the proto-types Engelbart debuted at the 1968 demonstration.

Another important contributor was a young graduate student named Alan Kay. Kay was a brilliant computer scientist and visionary from the University of Utah, who also attended Engelbart’s 1968 San Francisco presentation. Kay quickly realized that Engelbart’s proto-types were similar to his own ideas for “FLEX”, a desktop computer Kay was developing for children. Like Engelbart, Kay was hired by a research facility near Stanford, in this case Xerox’s Palo Alto Research Center (PARC) in 1970, to work on his design and an innovative programming language called “Smalltalk”. His supervisor was Bob Taylor, who left government work in 1969 and was hired by Xerox in 1970 to organize PARC’s Computer Science Laboratory. In 1974, after visiting PARC and hearing Taylor and Kay speak, two young visitors, Steven Jobs and Steve Wosniak, adapted some of Kay’s concepts into their own design for a “home computer”, and founded Apple in 1976. Their second visit to Xerox in 1979 resulted in the creation of the Lisa and Macintosh computers in 1980.

The “Intergalactic Computer Network”

While Jobs, Wosniak, Kay, and Engelbart refined their designs for personal computers, Licklider and Taylor organized teams of computer scientists and engineers to create an efficient means to connect them, what Licklider called an “Intergalactic Computer Network”.

Their collaboration with Engelbart was especially important. The initial interest in his work was aligned with a plan to seek an alternative to punch card driven mainframes that were connected in the same facility to “drone” workstations. Engelbart’s demonstration that smaller, more powerful computers could be connected by telephone lines to different research centers provided support for Taylor’s argument that future scientists should be able to collaborate in real time from equally powerful machines. Engelbart’s influence was also evident in an essay written by Licklider and Taylor in 1968, “The Computer as a Communication Device”, which predicted that “augmented” computers and communications networks would “…change the nature and value of communication even more profoundly than …the printing press and the picture tube, …” The problem now was how to make it work.

When Taylor succeeded Licklider as IPTO director in 1966 he envisioned ARPA-Net as an extension of Licklider’s early research in “time-sharing”. The essence of time-sharing was the necessity of parsing out computer time on mainframes among different users. Each would have to “batch load” endless numbers of punch cards by hand, wait for processing, and return hours later when the cycle was complete. If the software program represented on the cards was flawed, the time was wasted, and the operator was forced to start the cycle all over again. Taylor saw this as highly inefficient and an obstacle to interaction between the operator and the computer. Moreover, time-sharing encouraged the isolation of scientists and engineers, effectively preventing them from brainstorming and problem-solving.

Taylor’s idea was to fund the development of separate computers for each group of users, equipped with powerful processors, storage capacity and visual displays, then connect them through telephone lines into a national computer network. The process Taylor adopted was to organize periodic meetings of project leaders in computer science, engineering, programming, graphics, video, communications, and other ARPA-Net funded projects to work out the details. For most, it was the first time they had ever met. As Taylor recalled during an interview:

“I constructed the meetings so they all had to get to know one another and argue with one another technically in my presence. I would ask questions that would force people to take sides on technical issues. Lasting friendships were built from the give and take…. Then, after they went back to their laboratories and campuses, their communications increased in both quality and quantity, because they knew each other.”

These meetings resulted in specifications for modifying computers, protocols for communicating between different time-sharing systems, equipment for wiring campuses so that individual researchers could collaborate, software to make collaboration easier, and packaging and compressing messages, called “packet switching”, for efficient transmission over telephone lines.

The meetings also determined where the first “nodes” of the new network were to be located: one at System Development Corporation in Santa Monica, another at UCLA, a third at Berkeley, and a fourth node at MIT. When Taylor needed a network application to initiate the network, he used Engelbart’s software from his ON-Line System. Appropriately, Taylor added a fifth node at Engelbart’s Augmented research facility at SRI and transmitted the first internet message between SRI and UCLA in October 1969.

By the time Taylor ended his stewardship of ARPA-Net later that year, the result of his disenchantment with Defense Department policies in Vietnam, the basic structure of the network had been established. The final piece, a standard protocol for connecting different networks called “Transmission Control Program/Internet Protocol”, or simply “TCP/IP”, was formulated in 1974 by another Defense Department funded group led by Vinton Cert, a computer scientist from Stanford University. Cert’s paper, “Specification of Internet Transmission Control Program” is credited with the first public recognition of the “Internet”.

Following a demonstration of the network in the San Francisco Bay Area in 1977, and the standardization of all host computers to the protocol in 1983, the Internet in its present form became operational to government users in 1984. By 1989 the network was estimated to connect 20,000 scientists around the world. Commercial use of the Internet followed the development of a graphical interface by Timothy John Berners-Lee called "the world wide web" in 1990, through the use of “dial-up” service providers using local telephone exchanges, and the provision of the first fiber-optic “broadband” connections in 1994.

Despite the apparent smooth transition, commercial use of the Internet was slowed by the high cost of long distance telephone transmission through the 1980’s, and the slow convergence of voice, video and data into digital until the late 1990’s. When these obstacles were overcome, American media, now reliant on high speed “broadband”, was under the influence of a new set of players.

Next Time: Telephone vs Cable: The Battle for the Internet Begins

Copyright © 2008 R.E. Xavier