Jacquards' Web (33 page)

In evaluating Babbage’s possible influence on Aiken, one must keep in mind the enormous gulf between today’s image of Babbage as a computer pioneer and his relative obscurity before the
1940
s. Today every historian of technology, every computer scientist, and every computer buff is aware of the prescient nature of Babbage’s ideas and of the machines he either proposed or constructed in full or in part … But in the
1930
s Babbage was a rather obscure figure whose ideas were generally unknown to working engineers and applied mathematicians. His writings had been out of print for many decades and were to be found only in large research libraries.

Today, the Harvard Mark
1
is no longer entirely intact, although a large, non-working part of its mechanism is in the Smithsonian Institution in Washington DC. The importance of the Harvard Mark
1
as the world’s first automatic computer can hardly be overstated, but it must also be conceded that in many respects it was a technological dead end much as Babbage’s Analytical Engine was. Aiken’s computer was used in a number of applications for the United States Navy, mainly based around 235

Jacquard’s Web

calculating mathematical tables. Aiken himself had great hopes for the importance of his computer in the creation of mathematical tables. Eventually the Harvard Mark
1
was used to produce twenty-five volumes of these tables. But they had a limited use and a relatively short shelf life.

The reason for this was that even during the period when the Harvard Mark
1
was being built, a new and much better way of building computers was being developed. It involved the use of purely electronic components with no moving parts: a striking contrast to the cumbersome and slow electromagnetic relays.

The development of electronic computers is the last episode in the story of Jacquard’s Web, or at least so far. These electronic computers soon became so powerful and rapid in their operation that compiling mathematical tables ceased to be important because whenever a particular value was needed, it was much easier to instruct the much faster electronic computer to provide the value in a fraction of a second.

Nor did the Harvard Mark
1
ever meet Aiken’s hopes for a machine that could perform all the calculations required by advanced science. The machine was reliable and worked perfectly well, but during the building process there seems to have been an acceptance on the part of its engineers that it was never going to be as fast in its operation as Aiken had hoped. It was true that—as Aiken liked to claim—the machine was more than a hundred times faster than a human ‘computer’ and that it could perform six man-months of computing in a day. However, within a few years electronic computers were making this level of performance look slower than the growth of a cactus.

Furthermore, Aiken unfortunately made a serious error of judgement at the dedication ceremony for his computer. He neglected to acknowledge the enormous importance of IBM’s role in making the machine happen. It appears that he was so carried away with his own sense of historical destiny that he had forgotten about the reality of the struggle to build the machine.

Thomas Watson was incensed by Aiken’s lack of gratitude. A 236

IBM and the Harvard Mark 1

generous man by nature, Watson expected generosity and loyalty from those whom he had assisted, and at this crucial moment in Watson’s life and in the history of technology, Aiken let him down. Watson called it betrayal. He was infuriated, but he was a man whose emotions quickly tended to coagulate into some course of practical action. Aiken’s short-sighted approach may actually have been a disguised blessing for the computer industry, for it led Watson to seek revenge over Aiken by ensuring that IBM soon recaptured the spotlight by building something even better than the Harvard Mark
1
.

The better machines that IBM and other organizations created carried the computer forward into realms unimaginable even by a genius such as Charles Babbage. Aiken, who based his machine on Jacquard’s and Hollerith’s technology and Babbage’s inspiration, was about to be superseded. Machines of almost unimaginable power would soon be built that would change the world for ever. And yet the punched card, so far from being abandoned as part of an obsolete technology, was about to enter its heyday.

237

This page intentionally left blank

●
16
●

Weaving at the speed of light

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

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

I’ll put a girdle round about the earth

In forty minutes.

Puck in
A Midsummer Night’s Dream

The history of a successful invention is like the story of a river as it flows from its source to the ocean.

The invention starts out at the source as an idea born in the mind of a remarkable innovator whom we may be entitled to regard as a genius. Gradually, like tributaries joining the main flow of the stream, other inventors and thinkers become involved. Finally, as the river approaches the ocean, it widens into a delta. There, hundreds or even thousands of expert technicians toil away in well-organized, well-funded, yet comparatively anonymous groups such as the research and development teams of major international corporations. The river’s flow into the ocean represents the idea’s comprehensive acceptance by the mainstream of human culture and society.

Along the way, a number of possible new tributaries are very likely to show promise for a while before drying up. Charles 239

Jacquard’s Web

Babbage’s plan of building a computer using cogwheel technology was brilliant from a conceptual point of view, and posterity has benefited from his insights—enhanced by Ada Lovelace—

into the potential of the computer. Purely mechanical cogwheels, however, proved to be an inadequate technology for enabling a computer to attain the potential which Babbage instinctively knew was possible.

Similarly, the electromagnetic technology of Howard Aiken’s Harvard Mark
1
was also consigned to technological oblivion. In practice, the mechanical element of electric relays—switches that physically move from the ‘off ’ to the ‘on’ position when triggered by an electric current—limited how fast the relays could operate. Like any other mechanical device, they were prone to jamming and overheating. They were also comparatively large, meaning that any machine comprising many thousands of relays had to be enormous. Aiken’s computer filled a very large room, and its processing speed was only modest.

More ambitious machines could only get bigger. By
1944
, when the Harvard Mark
1
was launched to the world, the technology on which it was founded was already starting to seem inadequate to those who understood just how slowly electromagnetic relays operated, and how prone they were to faults and breakdowns.

Was there a way to improve the processing time of computers and reduce the sheer size of the equipment? It seemed at least possible there might be.

In
1881
, a young engineer, William J. Hammer, who was working for the great inventor Thomas Edison at Edison’s laboratory in Menlo Park, New Jersey, made an accidental discovery that turned out to be of great importance.

At the time, Edison was at the forefront of the struggle to produce a reliable and commercial electric light bulb. The essence of his idea was that a filament would glow within a glass sphere from which the air had been excluded. The exclusion of air was 240

Weaving at the speed of light

essential to maximize the life of the filament. The work of Edison’s team focused on finding the best material for the filament, and the most efficient way to exclude the air. Edison’s employee Hammer made the curious discovery that under certain conditions of vacuum and voltage a bluish glow appeared in an evacuated light bulb, suggesting the existence of an unex-plained current between the bulb’s two filaments in a direction
opposite
to that of the main current.

Despite his inability to understand the phenomenon, initially known as ‘Hammer’s phantom shadow’, Edison was never one to miss an opportunity. He immediately took out a patent for the discovery, prudently renaming it the ‘Edison Effect’. But Edison’s time was soon fully absorbed in other projects, and as the Edison Effect seemed to have no immediate commercial applications he chose not to investigate it any further.

Other scientists, however, heard about the Edison Effect and, profoundly intrigued by it, started to try to understand what was going on inside the vacuum tube. In due course, this new research concluded that the Edison Effect was due to the surprising fact that there was a passage of electrons—the particles from which electricity is considered to be made—from the negative terminal (the cathode) of the filament inside the tube to the positive terminal (the anode). A tube in which this process was taking place became known as a ‘thermionic tube’, or more colloquially a ‘vacuum tube’ or ‘valve’. It looked very much like a light bulb, but was designed to maximize the strength of the electron flow rather than actually to produce light.

The discovery of the Edison Effect was to prove one of the crucial discoveries that has made the modern world possible.

The origin of the Effect in the quest for a reliable electric light bulb is yet another example of how one area of technological exploration may ‘accidentally’ create a gateway to another area, and may even create a completely new technology. The link between electric light and the science of electronics is as momentous as the one between weaving and computing.

241

Jacquard’s Web

The ‘Maltese Cross’ experiment.

In time, some scientists wondered whether a thermionic tube could be developed as a new kind of switch. The illustration above shows the ‘Maltese Cross’ experiment, which projects a beam of electrons within a vacuum tube against a screen at the far end of the tube. The Maltese Cross object placed in the path of the cathode rays makes them cast a shadow on the screen. This experiment proved that electrons were particles with a negative charge and a finite mass. These features of electrons meant they could be used in switching devices.

Innovation and refinement of the vacuum tube continued throughout the
1930
s and
1940
s. When World War II broke out, the possibility of using vacuum tubes as rapid switching devices in computers that would handle calculations with unprecedented speed was vigorously explored around the world—especially in the United States and Europe. Experts hoped to employ electronic computers, among other things, for the rapid decryption 242

Weaving at the speed of light

of coded enemy communications traffic and the solution of complex calculations relating to the velocities of projectiles. Not for the first time in history, war was proving a powerful catalyst for technological evolution.

Howard Aiken had stolen a march on his rivals in the new field of computing, but only just. The implications of the vacuum tube as a switching device turned out to be extremely important for the history of the computer. Even while the Harvard Mark
1

was in the final stages of being completed, two brilliant engineers from the University of Pennsylvania—J. Presper Eckert and John W. Mauchly (another two intriguing names)—were engaged in building a machine they christened the Electronic Numerical Integrator and Calculator (ENIAC).

Eckert had first showed an interest in electronics when he started building working radios at the tender age of five. After earning a degree in electrical engineering from the university of Pennsylvania in
1941
, he was offered a graduate fellowship at the Moore School of Electrical Engineering at the same university.

There, he and his professor, Mauchly, made numerous valuable contributions to improving the electronic equipment then available.

The quality of their work became known to the United States Government. They were awarded a contract to construct a digital computer from electronic components. During the construction phase they decided to name this ENIAC. It possessed the enormous technical advantage of not using any electromagnetic relays, but only vacuum tubes. Punched cards were still used to program ENIAC, but its processing speed was much faster than any of its predecessors.

This computer was designed for work analysing ballistics trajectories and producing new tables that allowed a gunner to aim with great accuracy, assuming that the gunner knew a range of key parameters—such as weight of projectile, distance of the target, angle of elevation, and so on. ENIAC is justifiably regarded as the world’s first all-purpose, all-electronic digital 243