Bell Labs 的故事

台積電的8000人研發中心，據稱是TSMC的Bell Labs。
我想這有點比喻失真。

這是第n本關於 Bell Labs 的史詩故事。
又可參考我前年的書 系統與變異: 淵博知識與理想設計法 u.3
也也

Inventing the Future

‘The Idea Factory,’ by Jon Gertner

Reprinted with permission of Alcatel-Lucent USA Inc. and courtesy of the AT&T Archives and History Center

Bell Labs technicians prepare the Telstar 1 communications satellite before its launching in 1962.

By WALTER ISAACSON

Published: April 6, 2012

In 1909, top executives at AT&T decided to commit themselves to a challenge: building a transcontinental phone line that could connect a call between New York and San Francisco. The problem was one that required not just engineering skill but advances in pure science. They needed, among other things, to create a repeater or amplifier for the electric signals so that they would not attenuate after a few miles. Thus was the seed planted for a new collaborative industrial organization — teaming up theoreticians, experimentalists, material scientists, metallurgists, engineers and even telephone pole climbers — that eventually became Bell Labs. Jointly owned by AT&T and its affiliated equipment maker, Western Electric, Bell Labs went on to invent the transistor and make major contributions to the field of lasers and cellular telephony.

THE IDEA FACTORY

Bell Labs and the Great Age of American Innovation

By Jon Gertner

Illustrated. 422 pp. The Penguin Press. $29.95.

Enlarge This Image

Courtesy of AT&T Archives and History Center

The Picturephone from Bell Labs was introduced at the 1964 World's Fair.

Jon Gertner, an editor at Fast Company magazine, has produced a well-­researched history of Bell Labs, filled with colorful characters and inspiring lessons. But more important, “The Idea Factory” explores one of the most critical issues of our time: What causes innovation? Why does it happen, and how might we nurture it? The lesson of Bell Labs is that most feats of sustained innovation cannot and do not occur in an iconic garage or the workshop of an ingenious inventor. They occur when people of diverse talents and mind-sets and expertise are brought together, preferably in close physical proximity where they can have frequent meetings and serendipitous encounters.

Bell Labs was created on Jan. 1, 1925, and was originally based on West Street in Lower Manhattan. During the 1930s it was the site for many lectures, scientific visits (Einstein dropped by) and informal study groups. Among those groups was one that discussed the latest advances in solid-state physics, an emerging field that analyzed the conductivity and other properties of different materials in relation to their atomic structures. It included William Shockley, who was an intense theoretician, and Walter Brattain, who was a verbose experimentalist.

Shockley was inspired by a talk given by the Bell Labs research director, Mervin Kelly, who wanted to find a way to invent electronic switches and amplifiers that could replace the relay switches and vacuum tubes that underlay the telephone exchanges. So Shockley began to think about using a type of material known as the semiconductor, which partly conducted and partly resisted electric current; it also sometimes acted as a rectifier that would allow current to move in only one direction. His idea was that semiconducting materials might be used to make small devices that would serve as electronic amplifiers. He urged Brattain to try to fabricate one using copper oxide.

It didn’t work, and for a while the endeavor was put aside as Bell Labs concentrated on helping the military during World War II. But in the middle of the war, Bell Labs began moving to a new campus in Murray Hill, N.J., and Kelly began to create interdisciplinary teams that threw theorists and engineers together into the same work spaces. “By intention, everyone would be in one another’s way,” Gertner writes. Among the teams was one doing solid-state research. It included Shockley and Brattain. Kelly recruited John Bardeen, a very quiet theorist, to join the group, but there was no vacant office, so Bardeen decided to share space with Brattain, the experimentalist. This was a smart idea. Gertner describes how innovations came not just from new theories but from linking them to advances made by the lab’s experimental chemists and metallurgists who were creating a revolution in materials. “Indeed, without new materials,” Gertner writes, “Shockley would have spent his career trapped in a prison of elegant theory.”

Having failed with copper oxide, the team tried two other semiconducting materials, silicon and germanium. By December 1947, they had rigged up thin slices of those materials with a wire tipped by a gold-foil point and were able to show that the contraption could act as an amplifier. It also proved able to serve as an electronic switch and do everything a vacuum tube could do at a fraction of the size and electricity use. After polling 31 members of the Bell Labs staff, they decided to name the new device a “transistor.” Shockley, Bardeen and Brattain would share the 1956 Nobel Prize in Physics for the discovery.

Like Bell Labs, the transistor stood at the intersection of theoretical science and applied engineering. It could be described as both a discovery and an invention. It was also an example of the “linear argument” in the history of science that was expounded by Vannevar Bush, James Conant and other academics who were involved in World War II’s scientific endeavors and wanted to encourage continued government funding of pure research: The theoretical discoveries of pure science would lead to applied science breakthroughs and new technological inventions. Gertner explains how this process could result in sustained innovation:

“If an idea begat a discovery, and if a discovery begat an invention, then an innovation defined the lengthy and wholesale transformation of an idea into a technological product (or process) meant for widespread practical use. Almost by definition, a single person, or even a single group, could not alone create an innovation. The task was too variegated and involved.”

Because Bell Labs was part of the AT&T monopoly, its executives felt a moral, political and legal imperative to share the discovery with other researchers and license it to other companies. “AT&T maintained its monopoly at the government’s pleasure, and with the understanding that its scientific work was in the public interest,” Gertner writes. The transistor would end up doing a lot more than make telephone circuits function better. It would start a digital revolution in computing and information technology.

Bell Labs had just the right person to help its people imagine the transistor’s larger implications. Claude Elwood Shannon was a very eccentric theoretician who amused himself by juggling while riding a unicycle up and down the long Bell Labs corridors. He had the insight that the best way to understand complex circuits that contained many on-off switches was through Boolean algebra, which assigned each operation a value of either 0 or 1. Shannon went on to develop an information theory in which all communications and sequences of information could be measured in the number of binary digits (either a 0 or a 1), known as bits, they required. This was one of the great intellectual achievements of the 20th century, and it correlated neatly with the advent of transistors that could permit circuits to have huge numbers of on-off switches.

The application of these theories and discoveries helped to spawn the computer revolution, but Bell Labs did not lead the way. In part this was because, with the threat of antitrust lawsuits looming, the company decided not to go into the computer or consumer electronics business. In fact, it did not even take its invention of the transistor to the next step, which was figuring out how to etch a circuit of multiple transistors onto a single chip, an idea that became known as the integrated circuit, or microchip. That breakthrough came in 1958 at two small companies that had licensed semiconductor patents from Bell Labs — Texas Instruments, led by Jack Kilby, and Fairchild Semiconductor, led by Robert Noyce and Gordon Moore, who had worked for the increasingly erratic Shockley after he quit Bell Labs but soon rebelled against him.

Bell Labs had other great successes that it failed to capitalize on fully. It did pioneering work in the invention of the laser, a super-focused beam of visible light based on the stimulated emission of a stream of photons. Laser beams could carry information, like voice and data, and Bell Labs began a cumbersome and expensive endeavor to create hollow pipes that would serve as waveguides to send the beam to the proper destination. Another concept, developed elsewhere, was to direct the beams by creating incredibly pure strands of glass fiber. The Bell Labs brain trust concluded this would be unfeasible, and besides, they already had a lot invested in the waveguide pipe infrastructure. Thus Bell Labs lost out to Corning Glass and others on the chance to be a leader in the development of fiber optics.

Gertner draws smart insights from the successes at Bell Labs, but he does not do quite as well drawing lessons from its lapses and failures. Perhaps these were inevitable; in innovation as in hitting home runs in baseball, you have to be willing to strike out a lot to be successful. Whatever the case, it was not the failures that doomed Bell Labs. That was mainly the decision of the Justice Department, in 1974, to file a sweeping antitrust suit against AT&T that led, 10 years later, to the breakup of the company. Bell Labs gradually withered.

Steve Jobs once said that the most difficult and important thing to create was not an innovative product but a great organization that could continually create innovative products. That required joining creative people with product designers and great engineers so that imagination and technology could be connected. For much of the 20th century, Bell Labs played that role. It showed the value of having theoreticians, researchers, developers and engineers all huddled together. “People had to be near one another,” Gertner writes. “Phone calls alone wouldn’t do.” Mervin Kelly even created branches of Bell Labs at the phone company’s factories so that the theoreticians and scientists could be closely involved with the manufacturing workers.

The ability to combine theory, creativity and engineering was a great achievement of postwar America. For 50 years, economic growth and job creation were propelled by transistors, lasers and other discoveries that came from the willingness to nurture theoretical research in conjunction with applied science and manufacturing skills. But these days, manufacturing is being outsourced, and funding for pure science is being curtailed. With Bell Labs and other such idea factories disappearing, and with government research money endangered, what will propel innovation and job creation for the next 50 years?

Walter Isaacson, the chief executive of the Aspen Institute and former editor of Time, is the author of biographies of Henry Kissinger, Benjamin Franklin, Albert Einstein and Steve Jobs.