Skip to Content

The Corporate Logic

Alternatives to silicon-based computing are long shots. Knowing that, why do HP, Lucent and IBM spend time and money pursuing them? Their reasons may surprise you.

An elbow-to-elbow crowd swarmed into the Bell Labs auditorium on the western border of Greenwich Village on June 30, 1948. Onstage before the guests and reporters stood bow-tied research director Ralph Bown-the small sign at his feet telling the story in a nutshell: “The TRANSISTOR.” The Bell Labs folks would soon launch into full-scale demonstrations of the revolutionary device, invented the previous December. But Bown spoke first about how AT&T had engineered its achievement.

“What we have to show you today represents a fine example of teamwork, of brilliant individual contributions and of the value of basic research in an industrial framework,” Bown proclaimed.

It should have been a great moment for Bell Labs. After all, technology revolutions don’t happen every day. But from a bottom-line point of view, AT&T’s transistor breakthrough was less than transformative. That’s because when all was said and done, it’s doubtful Ma Bell netted a dime from its invention. Instead, the real winners were the specialized firms with better business plans and focus-names like Texas Instruments and Fairchild Semiconductor-who won the race to ready the transistor for mass production and distribution.

“What we have to show you today represents a fine example of teamwork, of brilliant individual contributions and of the value of basic research in an industrial framework,” Bown proclaimed.

It should have been a great moment for Bell Labs. After all, technology revolutions don’t happen every day. But from a bottom-line point of view, AT&T’s transistor breakthrough was less than transformative. That’s because when all was said and done, it’s doubtful Ma Bell netted a dime from its invention. Instead, the real winners were the specialized firms with better business plans and focus-names like Texas Instruments and Fairchild Semiconductor-who won the race to ready the transistor for mass production and distribution.

Granted, the transistor might seem to be a special case, since as part of an antitrust consent decree AT&T was forced to sell rights to the invention for a modest fee. But to students of the role of science in industrial research, the outcome is all too typical. In fact, for reasons that include corporations’ inability to embrace radical change and a lack of commercial applications, a number of other profound discoveries have failed to produce big returns on investments. The list includes semiconductor lasers (GE and IBM), cosmic background radiation (Bell Labs), the scanning tunneling microscope (IBM) and even semiconductor and superconductor tunneling phenomena (Sony and GE); the last three won Nobel Prizes.

From the perspective of corporations that sponsor the research the lesson is clear: Breakthroughs are hard to come by, and the financial payoffs have a troubling tendency to go to someone other than the originator.

This provides some food for thought about today’s fervent race to push computing beyond silicon. The field is already littered with failures-think gallium arsenide and optical computers-and current contenders include blue-sky propositions ranging from biological systems to quantum computing. This sort of research fits well in an academic environment, where making money is secondary to the advancement of scientific knowledge (at least in theory), yet much of it takes place in industrial labs. So, since history tells us that the chance of a big payday is remote, why do these companies bother?

The answer is that there are many “hidden” benefits to engaging in basic science-from creating a climate of discovery to staying in touch with the cutting edge. Indeed, the extras are so compelling that the firms bankrolling these studies often don’t expect their researchers to produce much of direct market value. “Why is any curiosity-driven research supported in industrial labs?” former IBM vice president for science and technology John A. Armstrong once asked. “There are several reasons, but they do not include the expectation that out of the company’s own scientific left field,’ so to speak, will come new insights or inventions that will radically alter the nature of the company’s business.”

If Armstrong’s statement appears to run counter to the popular notion that far-sighted corporations invest in basic science to plant the seeds of future growth, it shouldn’t. The two views actually complement each other. For one thing, betting on basic research does sometimes pay off financially: DuPont’s fundamental polymer studies led to the invention of nylon, and Irving Langmuir’s Nobel Prize-winning surface chemistry investigations enabled GE to build a revolutionary light bulb.

Yet by its very nature, most exploratory work fails. What’s more, scientific leadership has never been a prerequisite of marketplace triumphs. Witness Japan’s dominance in steel, autos, consumer electronics and semiconductor memories-or the rise of Dell, Compaq and Gateway in personal computers.

These truths have led many, Intel co-founder Gordon Moore among them, to conclude that wide-ranging basic research simply isn’t worth it. Moore, formulator of the “law” that has long governed semiconductor manufacturing, points to IBM’s Nobel Prize-winning invention of the scanning tunneling microscope (STM)-which does not fit into any of the company’s business lines-as a case in point. The STM “is really a great tool,” he says, “but IBM is not going to get anything out of it.” Moore stresses that society benefits tremendously from basic research-and that Uncle Sam should support it vigorously. But don’t expect Intel to dive into the realm of biological processing or quantum computing anytime soon.

Still, not every company shares Intel’s philosophy. IBM, Hewlett-Packard, AT&T, Lucent-Bell Labs, NEC and Hitachi are among those supporting world-class investigations into quantum systems, carbon nanotubes, biological processing, molecular computing or other alternative means of data crunching.

This work is so important to IBM that it went gangbusters to nab quantum hotshot Isaac Chuang two years ago, beating out a pack of university and corporate rivals with the lure of a generous salary and state-of-the-art equipment.

Similarly, when HP decided to spin off its measurement and equipment business-now Agilent Technologies-management originally leaned toward placing chemist R. Stanley Williams with the new company. But Williams, whose recent advances in molecular computing received international attention (see Q&A, TR September/October 1999), apparently proved such a hot commodity that he was kept in the HP fold.

All of which underscores the fact that there is more to corporate science than just science. The more subtle payoffs include: Covering the corporate backside. While it is relatively easy for research managers to focus R&D on areas likely to affect their firm’s interests, it is much harder to be sure nothing has been overlooked. Small but well-considered basic research projects can keep a company’s hand in the bigger game in case that something else turns up. As HP chief executive Carly Fiorina says about the necessity of pursuing alternatives to silicon: “You’ve got to start now or risk being left behind or missing out altogether.” (see Q&A, this issue) Building ties to university science so that companies are able to understand and exploit what comes out of academic labs. Retired NEC research executive Michiyuki “Mickey” Uenohara, who led his company’s vast expansion into basic science in the late 1980s, says it’s true that universities should be the center of basic studies. “However,” he notes, “it does not excuse industry from performing basic research. We have to have excellent basic research, otherwise we cannot fully utilize university’s basic research.”Creating a “culture of research,” to use the words of Bell Labs vice president of research Bill Brinkman, that will attract top scientists. Hiring the scientific elite raises the cachet and standards of the operation-and in turn brings in more recruits. For example, it was the Bell Labs culture that enticed highly recruited physical chemist Lisa Dhar, who joined the enterprise five years ago after finishing her doctorate at MIT. “Having that mix of long-term and applied research is a very compelling aspect of Bell Labs,” she notes. “And that drew me in.”Getting a fundamental perspective on commercial problems. Locating defects on an integrated circuit holding 200 million transistors, for example, is an immense problem. IBM physicists Jeffrey Kash and James Tsang were studying some exotic aspects of the optical spectroscopy of semiconductors when they realized that the infrared light transistors emit as they switch could overcome this obstacle. Their Picosecond Imaging Circuit Analyzer (PICA) tool now tracks these emissions over intervals of one-trillionth of a second-a far better solution than the Band-Aid approaches often forced on manufacturing engineers. “You can see every transistor light up as it switches,” says Tom Theis, director of physical sciences at IBM’s Thomas J. Watson Laboratory in Yorktown Heights, N.Y. “So if one is slow because of a defect, you’ll find exactly that device.” Last November, IBM licensed PICA to semiconductor test and measurement provider Schlumberger.Public relations. AT&T may not have made money off the transistor. But the PR impact of its six Nobel Prizes (11 prize winners) and litany of important patents is priceless. Chairman and chief executive Rich McGinn recognized this when Lucent spun off from AT&T in 1996. He placed his headquarters inside Bell Labs and brought the famous research facility into the corporate logo: “Lucent Technologies. Bell Labs Innovations.”

Beyond all these factors is one critical point: Although places like Bell Labs, IBM and GE became famous for their basic research, science alone did not make them great. Instead, it was their ability to bring together a wealth of talents and viewpoints-scientists with engineers, chemists with mathematicians, deep thinkers with the practical-minded. And from that volatile combination-rather than from basic research itself-leaps the spark of discovery.

Bell Labs’ Lisa Dhar experienced the power of such combinations firsthand a few years ago, when she began studying optical holography. This field, which seeks to use light to store data, has long promised unrivaled storage capacity, but it has lacked a good recording medium. Dhar was part of a team of engineers, mathematicians, optical experts, chemists and engineers that not only fashioned a new polymer storage material, but also created a working prototype of a high-density recording system. “There was this incredible feedback that would go on that really accelerated the progress,” she recalls. Late last year, Lucent signed an agreement with 3M spin-off Imation to try and develop a product with 25 to 100 times the capacity of today’s DVDs-and may even launch a startup of its own to sell the technology.

Viewed in the light of these experiences, it often makes perfect sense for a firm to participate in far-out ventures like quantum or molecular computing that may never provide their own revenue streams. Not only does it provide a lot of buzz, the work helps attract good people, and researchers probably learn some math, chemistry or atomic physics that can be applied to more practical problems.

Top companies know this and often insist on the full package in research, including some blue-sky studies. These efforts never represent a very large fraction of the company’s overall R&D budget-and they may never yield a Nobel Prize. But even without a scientific breakthrough, the payoffs can be incalculable.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.