Jacquards' Web (30 page)

Finally, there was the
methodological
stage, in which there is actually a danger of the findings and pronouncements being accepted without question, with original thought being stifled by the sheer need to master a vast body of knowledge before any original thinking could take place. A case could at least tentatively be made that computer science is close to reaching that stage today. That said, constant problems with bugs (i.e. errors in software) bedevil even mass-manufactured, highly popular programs. This problem, and other problems such as the challenge posed by viruses and the often tortuous difficulty of integrating 213

Jacquard’s Web

disparate computer systems without causing failures in communication and conflicts between programs, indicate that in practice computing is a rather less precise science than might be expected.

In
1923
Aiken was awarded a Bachelor of Science degree in electrical engineering. He continued his studies at Wisconsin and later at the University of Chicago, working full-time as usual.

This explains why he did not matriculate as a graduate student in Chicago until
1932
. In practice his paid work complemented his academic career because his employers recognized his talents and gave him extensive responsibilities that provided practical training in the very subjects he was studying. For example, during Aiken’s period as a graduate student, his jobs included that of electrical engineer responsible for the design and reconstruction of an electric generating station and also chief engineer of another company. At the time when Aiken was studying electrical engineering, the link between the new science and useful and major practical applications of it were particularly strong: Aiken’s work helped him to be familiar with these applications. It was typical of the strength and resilience of his character that he turned an apparent disadvantage, such as the need to work full-time, into an advantage in this way.

Aiken continued his studies by enrolling at Harvard in the autumn of
1933
. He did so well that he was invited to teach there for the academic year
1935
–
6
. For several years he continued at Harvard, teaching and researching in electrical engineering.

From early in his career Aiken was known for his skill at visualizing abstract mathematical or physical situations in terms of physical processes. He was always deeply interested in any piece of mathematical equipment that allowed a mathematical process to be carried out by physical means. In particular, he was intrigued by the punched-card tabulation machines available in the
1930
s and in mechanical analogue devices such as tide predictors. These were basically machines that used a complex 214

Howard Aiken dreams of a computer

combination of gear-wheels and gear-shafts to provide a reasonably accurate representation of the constantly varying action of the tides. The more advanced of these machines were used to handle complex differential equations. They became known as
differential analysers
. Differential analysers were wonderfully subtle pieces of machinery, but the fact of their being machinery placed inherent limitations on their speed, reliability, and accuracy. Eventually, in the age of electronics, the functions formerly carried out by tide predictors and differential analysers were performed by special types of computers known as
analogue
computers
because they deal with information as a continuously variable quantity.

The vast majority of computers in the world today, however, are known as
digital computers
because they deal with information in the form of digits or similar discrete elements. Similarly, the Jacquard loom handled the elements of a complex woven picture in an essentially digital fashion.

Toward the end of his life Aiken traced his passionate interest in digital computers to his time at Harvard. He said, ‘I was obliged to conclude that the area of electronics in which I was interested could never be explored properly with the calculating machines that were available at the time. The reason was that it would simply be too laborious and time-consuming to undertake the very numerous, lengthy, and extremely repetitive calculations needed to complete this exploration.’ He believed that some way had to be found to ensure that the labour of calculating

‘could be mechanized and programmed’, and that an individual needed to have access to a very special type of device in order to do this properly.

More than a century had elapsed since Babbage had started work on his Analytical Engine. In
1936
, the word ‘computer’ still meant in popular parlance a human clerk who actually carried out the mathematical calculations with pen and paper.

215

Jacquard’s Web

At the time Aiken was working on his doctoral thesis, most computers were women. The job was badly paid; men usually considered it beneath them. In practice, anyone undertaking a complex mathematical calculation in
1936
was just as dependent on mathematical tables as Babbage and his colleagues had been.

The tables—laboriously put together by people working with calculating machines—were more reliable than those available to Charles Babbage and John Herschel, but they were still by no means error-free.

There were at the time only a few facilities in the United States capable of large-scale calculation. One of the best-known was the Computation Center at Columbia University, later known as Columbia’s Watson Scientific Computing Laboratory, named after Thomas Watson himself, who sponsored it. By the mid-
1930
s the Computation Center was making extensive use of the latest generation of IBM punched-card machines that had been modified for calculation. The Center was engaged in important work involving the computation of lunar orbits; again, one is reminded of Babbage and Herschel’s work for the Royal Astronomical Society back in the
1820
s. But even with the latest punched-card machines, the work was hampered by precisely the same problem that had confronted Babbage and Herschel, and which Aiken also needed to solve in order to make progress in the field of electronics. The problem was that computers did not yet exist.

It was in
1936
, while working on his thesis, that Aiken started to put together informal plans for a completely new kind of calculation machine. One of his mathematician friends recalls a discussion around this time, in which Aiken discussed the possibility of directing the activities of scores of computing units, initially talking in terms of racks of interrelated machines, each roughly equivalent to the mechanical calculators of the day. By late in
1936
, the friend recalls, Aiken had started to give careful thought 216

Howard Aiken dreams of a computer

to how such a machine might have instructions inputted into it.

He was determined to use paper-tape or cards but initially he had not decided whether the holes in the tape or cards should be sensed mechanically or electrically.

Finally, after careful thought, Aiken decided that he preferred the electrical way of sensing the holes in the tape or cards.

In effect, Aiken had decided what Charles Babbage may well have decided had he been born in
1891
rather than a century earlier.

By April
1937
Aiken had made sufficient progress in his thinking about the general design of the machine, and the tasks it could perform, to start seeking support from industry for actually building the machine. Never one to waste time, on
22
April
1937

he presented his plans to the Monroe Calculating Company, America’s largest manufacturer of calculators. He was given a warm reception by George C. Chase, Monroe’s director of research. Chase subsequently published an account of the visit, during which Aiken explained that:

certain branches of science had reached a barrier that could not be passed until means could be found to solve mathematical problems too large to be undertaken with the then-known computational equipment.

The plan that Aiken discussed with Chase amounted to an enormously important step on the way to the modern computer. As Chase recalled, it consisted of a proposal for a machine that provided automatic computation in the four rules of arithmetic; facilitated storage and memory of installed or computed values; established sequence control that could automatically respond to computed results or symbols, and provided a printed record of all that transpires within the machine; and a recording of all the computed results.

Chase was convinced that the machine specified in the proposal represented one of the most remarkable innovations in the history of technology. He did everything in his power to convince 217

Jacquard’s Web

his colleagues at Monroe that the organization should build it.

Chase accepted that the machine would be enormously expensive, but he tried to persuade the corporation that the machine would play a crucial role in every aspect of the company’s development in the future. Chase evidently based his case on emphasizing how the machine would be an immensely important promotional tool for demonstrating to the world the sophistication of Monroe’s technology. In essence he saw the whole thing as a brilliant public relations exercise.

The problem, however, was that the Monroe Calculating Company made calculators, but Chase was asking it to fund a completely speculative project for another purpose. The Board congratulated Aiken on the inventiveness of his ideas and the ambition of his plans, but declined to be involved. Its reasons were the inevitable huge cost of the machine, and a belief that Monroe was not the organization to build it. Chase’s embarrass-ment at the rejection prompted a decision that would change the history of technology. He introduced Aiken to the one other corporation in the United States with the technical expertise, vision, experience, and—above all, money—to fund the project. IBM.

218

●
15
●

IBM and the Harvard Mark 1

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ❚ 1 1 1 1 1 1 1 1

2 ❚ 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

3 3 3 ❚ 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3

4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 ❚ 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4

5 5 5 5 5 5 5 5 5 5 5 5 5 ❚ 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5

6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 ❚ 6 6 6 6 6 6 6 6 6 6 6

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 ❚ 7 7 7

8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 ❚ 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8

9 9 9 9 9 9 9 ❚ 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ❚ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

If, unwarned by my example, any man shall undertake and shall succeed in really constructing an Engine embodying in itself the whole of the executive department of mathematical analysis upon different principles or by simpler mechanical means, I have no fear of leaving my reputation in his charge, for he alone will be fully able to appreciate the nature of my efforts and the value of their results.

Charles Babbage,

Passages from the Life of a Philosopher
,
1864

Thomas Watson was a businessman rather than a technologist, but his intuition told him that his cherished organization could only flourish if it remained at what would nowadays be called the leading-edge of technological development. As things turned out, Watson’s belief in this principle resulted in IBM being the perfect incubator for Howard Aiken’s ambitious plans.

Charles Babbage, dreaming of mechanical computers in the early nineteenth century, had had little alternative but to try to build his machines in his own workshop. Howard Aiken 219

Jacquard’s Web

was fortunate in being alive at a time when a successful organization with considerable experience in information technology existed and had money available to fund speculative research. As Aiken, with typical generosity of spirit, was subsequently to remark:

If Babbage had lived seventy-five years later, I would have been out of a job.

Chase gave Aiken an introduction to a Professor Theodore (Ted) Brown who taught at Harvard Business School. It was a happy choice. An applied mathematician and a trained astronomer with a PhD from Yale in celestial mechanics, Brown was already a consultant for IBM and a friend of Thomas Watson. One of Watson’s many talents was a gift for winning over people who were smarter than he was and giving them key positions at IBM.

For example, Brown ran special training courses at IBM and often lectured staff himself. At the time he met Aiken, Brown was a member of the Advisory Board of a computation laboratory that Watson had set up at Columbia University. This meant he was directly and intimately involved with the problems of scientific calculation by machine, and well aware of the need to make progress in this area. Aiken met Watson through Brown.

Watson quickly took to Aiken. At the outset everything went very well between the two men. Watson was always intensely respectful of the kind of Ivy League academia that Aiken exem-plified. Besides, there was a strongly philanthropic side to Watson’s personality. He was committed to using some of IBM’s resources for educational purposes and for the advancement of science generally, and Aiken’s project was exactly the kind of venture that appealed to him. Watson arranged for Aiken to be introduced to IBM’s senior engineer, a man named James Wares Bryce. Following this introduction, Aiken prepared a formal proposal for the new machine.

The proposal was entitled, with the directness that appealed as much to Watson as it did to Aiken,
Proposed Automatic Calculating
220

IBM and the Harvard Mark 1

Machine
. Occupying twenty-three double-spaced pages, it was, in effect, the first blueprint of the computer age.

The proposal opens by discussing ‘the desire to economize time and mental effort in arithmetical computations; and to eliminate human liability to error’. This desire, Aiken says, is