Chapter 1: A History of Information Highways and Byways

Development of Modern Computers
pp. 27-35

Howard Rheingold observed that "every desktop computer connected to the Internet is a printing press, broadcasting station, place of assembly, with world wide reach."75 This suggests that the Internet and the Web, far from being solely the product of engineers and computer scientists, are deeply rooted in the history of publishing. One could argue that the Web in particular is a direct descendent of the printing press. However, in order to complete this historical sketch of the intellectual and cultural sources for the Internet, we must look at the development of computers and the way these machines shaped the Web.

Jay David Bolter explains that "the computer rewrites the history of writing by sending us back to reconsider nearly every aspect of earlier technologies."76 Modern computers started out as machines created to solve complex mathematical problems. While the sixteenth century witnessed the revolution in printing, other developments were taking place that would advance the progress of technology in this period. In the first half of the fourteenth century, large mechanical clocks were placed in the towers of several large European cities. Determining time accurately is not only the antecedent to modern computers, it is also fundamental how society operates.77 Early European clocks, like the one installed in the clock tower of Strasbourg Cathedral in the mid-fourteenth century, could be heard for miles and contributed to productivity and work.78 However, the European clocks did more than "remember" time; they tracked it and measured it in new ways that had a dramatic impact on people's lives. The clocks did work that humans had done. This mechanization of timekeeping, coupled with the concept of time, created a revolution in concepts of space and the self. From this point forward, as machines were developed to do the work previously done by people, they often outstripped the simple expectations of their inventors.

These machines also spawned metaphors that described existing human activities. Thus, a clock came to be seen as a metaphor for the cycle of the day, as well as for the action of the heart or the span of a human life. However, machines also defined new kinds of human activities. The typewriter, invented in the nineteenth century, is a prime example of this. Mark Twain was the first American author to use a typewriter. In 1905, he recalled his initial experience working with the new machine: "Dictating autobiography to a typewriter is a new experience for me, but it goes very well, and is going to save time. . . . At the beginning . . . a type-machine was a curiosity. The person who owned one was a curiosity, too. But now it is the other way about: the person who doesn't own one is a curiosity."79 What is interesting here is not just the novelty and then the assimilation of the typewriter, but also Twain's description of the typewriter not as a machine at which he is working, but rather as one to which he is "dictating." In the nineteenth century, "typewriter" referred to both the machine and the operator. (As we will see, "computer" also referred to a human calculator before it referred to the machines that took over this work.)

Richard Polt, scholar and avid collector of antique typewriters, argues that "there is a certain truth in this usage: when you type, your body becomes a typewriter. More precisely, you reinforce certain bodily habits that act symbiotically with the machine in order to carry out the activity of typing. A new activity thus opens up new, unpredictable possibilities for the human body."80 This argument has also been put forth by David Stork, who works in the field of artificial intelligence, in an interesting parallel to computers. Stork argues that "before you have tools, the only device you have for getting things done is your own body. But with tools, you can go beyond that. Still, once you've built a tool, you're stuck with that particular tool. The idea of a universal computer is that you make a universal tool--a general purpose object--that you can program to do absolutely anything."81

In the hands of the people they were invented for, clocks and typewriters, along with many other mechanical inventions, became not only agents of social change, but also critical metaphors describing the workings of daily life. This assigning of machine metaphors to human life and interactions is described succinctly by Stork:

The problems addressed by [Artificial Intelligence] are some of the most profound in all of science: How do we know the world? . . . How do we remember the past? How do we create new ideas? For centuries, mankind had noticed hearts in slaughtered animals; nevertheless, the heart's true function was a mystery until one could liken it to an artifact and conclude: a heart is like a pump. (Similarly, an eye is like a camera obscura, a nerve is like an electric wire. . . .) In the same way, we have known for centuries the brain that is responsible for thoughts and feelings, but we'll only truly understand the brain when our psychological and neurological knowledge is complemented by an artifact--a computer that behaves like a brain.82

The computer has always been a thinking man's machine, or put another way, a machine in the service of human thought. The first computers, in fact, were simple devices that served as aids in solving mathematical problems. The earliest computer was probably the slide rule. Counting devices such as the abacus could be cited as early computers, but the slide rule, developed in 1621 by William Oughtred, was the first analog computer (an analog computer produces an approximate or "analogous result.")83 In 1642, Blaise Pascal (1623-1662) built a computer to help his father, a tax collector, with his computations. Pascal's machine, called a "pascaline," was essentially an adding machine.84 In 1673, Baron Gottfried Wilhelm von Leibniz created the "Leibniz stepped wheel," the first mechanical multiplier.85 Although mathematicians invented mechanical aids to speed up addition and multiplication, the direct ancestor of the computer program came out of the carpet industry. The first "program" was the Jacquard "punch card," invented in 1804 to control the warp threads on a carpet loom. The "punch card" had holes in it and could be mounted on a carpet loom, which would "read" the simple program to produce a particular weave.

These inventions were separate and as-yet-unrelated developments in the history of computers. This changed by the mid-nineteenth century--a period that marked the "true beginning of modern technology, when men embraced the machine as the heart of economic expansion."86 By 1820, science and mathematics were quite complex and relied on long tables of calculations, produced by hand by people called "computers."87

The real beginnings of modern computers lie with the inventions of the British mathematician Charles Babbage and his partner, Ada Lovelace. Babbage was intrigued by the problem of finding a mechanical replacement for human "computers." He was frustrated by the many errors these "computers" produced and declared: "I wish to God these calculations had been performed by steam!"88 Babbage proposed building a "Difference Engine" to mechanically figure out solutions to mathematical problems. His idea was for a machine that not only figured out the solutions to mathematical problems, but also printed out the results, thus eliminating most human error.89

He explained that his machine would work almost as fast as a manual computer, but would be more accurate and have far more endurance than any human. There was no limit to the number of differences the machine could handle; the bigger the task, the more it would outstrip the human computers. The significance of Babbage's proposed Difference Machine was that "it substituted a machine for the human brain in performing an intellectual process, which [was] one of the most revolutionary schemes ever to be devised by any human being."90

Babbage had an unlikely intellectual partner in Ada Lovelace, the daughter of the English poet Lord Byron. Lovelace met Babbage in 1833 and became a critical partner in the development of the "Analytical Engine," his next invention.91 The Analytical Engine in design profoundly foreshadowed the modern computer--there was memory, a central processing unit (CPU) and punch cards for transmitting data back and forth. Lovelace, a gifted mathematician, was fascinated by the potential of Babbage's machine, and in 1843 she published her notes explaining the Analytical Engine. She understood the potential of Babbage's invention, declaring: "No one knows what power lies yet undevelopped in that wiry system of mine."92 Her understanding of the machine made it possible for her to create instruction routines--or "programs"--that could be fed into the Analytical Engine. So significant was her work that Babbage in one letter called her the "Enchantress of Numbers." She is remembered as the first computer programmer, and in the 1980s, the U.S. Defense Department named a programming language ADA in her honor.93

The limitations on Babbage's design had more to do with the lack of technology to produce the machine than whether the machine would work. What Babbage designed outstripped both the materials and machining possible in the early nineteenth century. However, by the late nineteenth century, modern materials and new machining processes set the stage for the first working computers, which were developed as data processors.

By 1870, calculating the figures for the U.S. census had become a cumbersome task and the need for a solution to the problem of tabulating large columns of figures had become a critical one. In 1884, Herman Hollerith, working with Charles W. Seaton, the chief clerk of the census, developed the punch card tabulator by reinventing Jacquard's punch card as a way to simplify the census process.94 By 1899, Hollerith's invention was used to process the U.S. census. This was the first time "a largely statistical problem was handled by machines."95 In 1896, Hollerith formed the Tabulating Machine Company, the world's first computer company.96 After a brief period as the Computing-Tabulating-Recording Company (CTR), in 1924 Hollerith's company became International Business Machines (IBM).97

Hollerith's machine and others like it became mainstays in companies where data was tabulated, and for a time it looked as if the function of computers would always be limited to calculating numbers. Until World War II, the only significant advance in computing was a serious attempt in 1930, by Vannevar Bush, to design a computer called a "differential analyzer." The differential analyzer would provide the foundation for wartime developments in computing.

The development of the modern computer was greatly spurred by World War II. The concentration of the research community and the huge resources of capital and manpower during the war formed the foundation of the postwar effort, which continued to receive governmental support. With this support, programming languages were developed, as was fundamental software that paved the way for the transformation of the computer from an electronic difference engine and giant calculator to a general-purpose analytical tool. However, the first digital computers built during the war were barely more than electric difference engines. Their purpose was primarily to perform huge calculations rapidly and accurately. The invention of the transistor in 1947 and the creation of the integrated circuit caused a major reorganization and redesign of the nascent industry. With transistors, the new computer became smaller, faster and more reliable and had much simpler requirements for power.

Babbage's design for the Analytical Engine did not incorporate the fundamental idea employed by modern computers--that programs, data and the internal workings of the machines should be carried out by the same form of data. Thus, for example, Babbage's machine was designed to work with programs that were stored on punch cards, while the actual calculations were "performed" by the mechanical wheels and cogs of the machine. Fundamental to Babbage's design was the "the rigid separation of instructions and data. . . . A hundred years later, . . . no-one had advanced on Babbage's principle. Builders of large calculators might put the program on a roll of punched paper rather than cards, but the idea was the same: machinery to do arithmetic, and instructions coded in some other form."98

To understand just how different this is from the way modern computers work, we need only understand that all software, our e-mail messages and Web pages are to the computer fundamentally no different from any other kind of data. Each is carried on a "data-stream" made up of a sequence of electronic on-or-off states. This is the fundamental idea behind the computer byte. A byte (as in "megabyte" or "2 gigabyte hardrive") is the smallest unit of computer information, made up of a sequence of eight "bits" represented by either a zero or a one. This eight-bit sequence is fundamental to most modern computer programs and functions. To get a sense of proportions involved here: a megabyte--as in the storage space on a floppy disk--is equal to literally 1,024,000 of these bytes. This paragraph as a chunk of computer data saved in plain text (that is, with none of the extra coding that my word-processing package inserts) is made up of 950 bytes.

The scientists and engineers who built big electromechanical calculators in the 1930s and 1940s did not think of anything like this. Even when they turned to electronics, they still thought of programs as something quite different from numbers. In the United States, scientists built the first working differential analyzer that solved complex mathematical equations and provided critical solutions for ballistic firing tables. The ENIAC (Electronic Numerical Integrator and Computer), the world's first general-purpose electronic computer, was built at the Moore School at the University of Pennsylvania in 1943. It was an enormous machine that cost $486,804.22 to build and "took up 1800 square feet, was 100 feet long, 10 feet high, and 3 feet deep, and weighed 30 tons."99 The ENIAC prompted one observer to declare that: "computers in the future may . . . weigh only 1 1/2 tons."100

Designed to calculate the firing tables used to aim long-range guns, the ENIAC sped up the process that had previously been done by hand. Kay McNulty, one of a number of "computers" who did this work prior to the ENIAC, later recalled:

We did have desk calculators at that time . . . that could do simple arithmetic. You'd do a multiplication and when the answer appeared, you had to write it down to reenter it into the machine to do the next calculation. We were preparing a firing table for each gun, with maybe 1,800 simple trajectories. To hand-compute just one of these trajectories took 30 or 40 hours of sitting at a desk with paper and a calculator. . . . My title working for the ballistics project was "computer." The idea was that I not only did arithmetic but also made the decision on what to do next.101

Many of the women who had done the figures by hand went to work "programming" the figures into the ENIAC.

Meanwhile, at Bletchley Park in Britain the development of modern computers took a different turn. While the ENIAC in the United States was designed to serve as a giant calculator, the Colossus, its British counterpart, was designed to break German secret codes. What spurred the development of Colossus in 1943 was the need (like that in the United States) for a machine that could work faster than its human operators. Alan Turing, a British mathematician who worked at Bletchley Park on the development of the code-breaking machines, became fascinated by the speed and reliability of machines in solving complex problems.

Turing, who is considered the father of modern computers, came to Bletchley Park with a foundation of ideas that made his interest in the code-breaking machines even more compelling. In the early 1930s, Turing began working on the theoretical problem of whether there was one principle by which all mathematical problems could be solved. He argued that in fact such a principle existed. The work at Bletchley Park supplied the machines on which this theory could be tested. For Turing, this theory could also be applied to the workings of the human brain. Turing argued that it was possible, if one assumed that the brain was capable "of finite number of possible states of mind," that a machine could be built to embody these states of mind. In fact, he spoke of "building a brain."102

In 1936, Turing had theorized that it was possible to build a machine that would work something like a typewriter, but would have the additional ability of reading symbols and being able to erase them as well. He theorized that a long tape could be fed into this machine that the machine would then "read." The tape would be divided into squares, with each square carrying a single symbol. The machine would "read" the tape one square at a time and perform a function based on what the tape "sent" in the way of instructions. (In effect, Turing had a vision of a computer program and the computer byte.)103

In 1945, Turing argued that computer programs should be stored in the same way as the data was. Turing had seen a proliferation of specialized machines doing different tasks. He believed that a universal machine could be designed that could switch from program to program and task to task--regardless of whether the task was solving a mathematical problem, playing a game or processing data. In the United States, by contrast, the ENIAC engineers also arrived at the idea of stored programs, but still only thought that computers were limited to performing massive mathematical calculations.

An American mathematician helped to pave the way for the development of the first programming languages. Grace Hopper--or "Amazing Grace," as she was often called--joined the Naval Reserve in 1943 and was assigned to work on the Navy's Mark I computer, designed like the ENIAC to do gunnery calculations. Hopper relished working on the Mark I: "I always loved a good gadget. When I met Mark I, it was the biggest fanciest gadget I'd ever seen. . . . It was 51 feet long, eight feet high, eight feet deep, and could perform three additions per second. . . . I had to find out how it worked."104 After the war, she continued to work with computers. In the late 1940s, she went to work on the UNIVAC project, which led to the development of the UNIVAC I (Universal Auto Computer), the first commercial computer in the United States. By this time, Hopper had begun to believe that the major obstacle to computers in nonscientific and business applications was the scarcity of programmers for these far-from-user-friendly new machines. The key to opening up new worlds to computing, she knew, was the development and refinement of programming languages that could be understood and used by people who were neither mathematicians nor computer experts.

It took several years for her to demonstrate that this idea was feasible. Pursuing her belief that computer programs could be written in English, Hopper moved forward with the development a compiler for the UNIVAC computer that was built in 1950.105 Using this compiler, Hopper and her staff were able to make the computer "understand" twenty statements in English and then recognize keywords in French and German. However, when she recommended that an entire programming language be developed using English words, she "was told very quickly that [she] couldn't do this because computers didn't understand English." It was three years before her idea was finally accepted.106

Turing's idea for a universal machine and Hopper's development of a programming language helped pave the way for the next significant development in computers, time-shared computers, which in turn led to the development of the Internet. These developments are covered in the next chapter. It is important to point out that in the period from roughly 1945 to the early 1960s computers were located in isolated computer centers and worked on by small groups of experts. Computers were so unusual and removed from mainstream culture that George Orwell in 1984 (1948) was able to write an effective futuristic scenario in which computers acted as watchers in the service of a totalitarian state. (The image of machinery controlling individuals predates computers and was used effectively by Fritz Lang in Metropolis [1926], with its dominant theme of people enslaved by machinery. In postwar science fiction films, computers played leading roles as dehumanizing machines in films from Jean-Luc Godard's Alphaville (1965) to Terry Gilliam's Brazil (1985), as well as in numerous low-budget B-grade films.) There was little sense in the postwar period that the evolutionary path of computers would deviate from the development of large, expensive and specialized mainframe computers. However, one man at the war's end envisioned a device that would link computers together as a means to disseminating information. In 1945, an essay titled "As We May Think" was published in the Atlantic Monthly. Written by Vannevar Bush, Harry S Truman's science advisor, this essay served as a conceptual blueprint for a small group of men who became the architects of the Internet. Their story and the computer network they built are the subject of the next chapter.

------------