Categories

The parable of the transistor

This weblog could be read as a diary of the disruption of the mobile industry. There is lots of topical analysis and opinion, but sometimes I’ll post on the “theory” which describes what’s going on in a more abstract, and long-term, level. Theory is like a bullion cube: savory, but too concentrated to be enjoyed undiluted. For this reason, the most palatable way of administering theory is through example. So we kick off with an example of what happened to another technology industry as a result of the emergence of a disruptive technology.

The following is a draft article I wrote describing the “electronics industry” for POSTWAR AMERICA: AN ENCYCLOPEDIA OF SOCIAL, POLITICAL, CULTURAL, AND  ECONOMIC HISTORY. It is itself based on a presentation by Clayton Christensen at the Open Source Business Conference in 2004.

—-

The consumer electronics (CE) industry began in 1920 when radio broadcasting commenced in the United States. Radio technology sustained a growing electronics business throughout the 1930’s and 1940’s, however, the electronics industry underwent a major transformation after the war. The critical development came after 1948 when the transistor was first invented. The transistor and the subsequent development of solid-state semiconductors dramatically and profoundly changed the industry. The change the transistor brought was not just evident in measures of technical performance but also in the economics of the industry and the business models required of its participants

“Electronics” has a precise definition: technology based on the controlled flow of electrons. Prior to the transistor, the control of electron flow in a system was through a component called the electron tube (or vacuum tube). The tube in its simplest form is a gas-filled glass cylinder, much like an incandescent light bulb, with two electrodes or terminals inside. When the negative terminal is heated it emits a cloud of electrons which the gas inside the tube carries to the positive terminal. This device allows current to flow in only one direction since the positive terminal is not capable of emitting electrons. It is called a diode and is simply a one-way current or conduction valve. Hence the “semi” in semiconductor. Semiconductors can be designed into complex products that amplify, store and communicate; the core applications of CE devices.

The transistor is significant because it is a “solid-state” semiconductor. Solid-state means that it is composed of solid components, unlike tubes which required a sealed gas to operate. The consequence of this shift from gas to solid components completely changed the industry and created unforeseen growth, wealth and societal change.

NEW ECONOMICS

The invention of solid-state semiconductors allowed new economies of production: components were smaller, cheaper and could be produced with a process closer to printing than to assembly manufacturing. Because of the subsequent advances in assembly, the cost of electronics products decreased even as quality and reliability increased. Power requirements and physical dimensions were also reduced, allowing greater portability.

These implications might have seemed foreseeable to the engineers and businessmen of the 1940s who first looked at a transistor, but what happened is something completely unexpected: almost all the companies dominant in the electronics industry in the 1940s and 1950s who spent huge fortunes to perfect the transistor for their products went out of business or withered away to insignificance. New, giant CE companies emerged in Japan who perfected new product categories rather than challenge the incumbents and their existing products.

The emergence of Japan as a powerhouse of CE has been debated as a motivation for changing US industrial policy toward a more interventionist model. Transistor technology was seen as a strategic national defense asset and the innovations and production associated with it were not to be allowed out of US control. Additionally, American politicians argued the case that it was US taxpayer-funded defense, educational and space efforts which produced these innovations, hence they “belonged” to the United States government, and future innovation will similarly need taxpayer funding. Unless something is done, Japan would dominate American industry.

What happened in fact was far more mundane. The transistor disrupted the industry. Small upstarts were able to take the invention, wrap a new business model around it that motivated the current players to ignore or flee their entry. They thus successfully displaced the entrenched incumbents even though the incumbents were investing heavily in the technology and the entrants weren’t.

THE INDUSTRY THAT EMERGED

The postwar half-century began with the radio as the center of the CE world. It ended with the television dominant, alongside a whole galaxy of mobile devices.  All employed the transistor, which itself underwent dramatic increases in power and equally dramatic decreases in size. Some of the significant breakthroughs along the way are listed in Table below. Note in the table the absence of any companies which dominated the CE industry before the transistor (RCA, GE, Zenith, Westinghouse, Philips, Magnavox, Curtis-Mathes, ITT, Packard-Bell, Philco, Siemens, Sylvania, Western Electric).[3]

  • 1957 Sony TR-63 Transistor Radio. Portability enabled privacy which allowed young people to listen to Rock-n-Roll.
  • 1960 Sony portable TV. Made a second television in the bedroom possible.
  • 1962 Sony CV-2000, first home videotape recorder. The concept of “time-shifting” was born.
  • 1970 Olympus Zuiko Pearlcorder Microcasette recorder. Audio recording became a discrete affair.
  • 1972 Pulsar Quartz Digital Watch. Quartz meant cheap but accurate watches for everybody.
  • 1972 HP-35 Pocket Electronic Calculator. The end of the slide-rule.
  • 1976 Atari Pong Game. A new use for TV: interactive games.
  • 1976 JVC HR-3300 VHS VCR. Mass-market cartridge-based video recording. Video rentals took off.
  • 1977 Atari 2600 Game Console. Game cartridges added a new category of content consumption.
  • 1979 Sony Walkman. Portable high fidelity music for your ears only.
  • 1980 Sony Boom Box. Portable high fidelity music for you and your neighborhood.
  • 1983 Sony CDP-101 CD Player. Loss-less audio reproduction for all.
  • 1984 JVC Camcorder. Home videos were born.
  • 1986 Motorola Bravo Pager. Contact somebody even if they are not near a phone.
  • 1996 Toshiba SD-3000 DVD Player. Loss-less video reproduction for everybody.
  • 1998 Nokia 5100 mass-marketed phone. Easy to use, customizable cheap mobile phone.
  • 1998 Diamond Multimedia Rio. Mobile digital compressed music consumption.
  • 1999 Tivo Series1 Digital Video Recorder. Skip commercials and freeze real-time video.
  • 2001 Apple iPod. All your music with you all the time.

DISRUPTIVE IMPACT

When the transistor was introduced in the late 40s and early 50s, most designers considered it inferior to existing vacuum tube technology as far as performance was concerned; transistors could not handle the power that was required to be used in the main CE products popular at the time, notably, the big tabletop radios and floor standing televisions. The transistor could not amplify sound as well as a tube, but it was small, solid, and used low power. Therefore, RCA, Zenith, and Westinghouse looked at the transistor as a electronics component with great potential, but one needed to be improved to match the tube as an amplifier before it could be used as a substitute.

Every CE company was required to receive a license for the transistor from its creator AT&T, before they could put it in their products. But they took out these licenses in order to experiment with the new technology, hoping they themselves could improve it sufficiently for use in the CE products.  Altogether, these CE companies spent an aggregate $2 billion (in today’s dollars) trying to make solid-state electronics good enough that they could make popular high-end CE products with them.

And while tube companies were trying to fit the transistor into large products, other companies, especially in Japan, were developing the first transistor-based products.  One critical one was a transistor hearing aid, first marketed in 1952. While hearing aids were a tiny market in comparison to radios and televisions, it was a market that valued the miniature size and low-power requirements of the transistor. In 1955, Sony introduced its first pocket radio. The sound quality and reception of those radios was very poor relative to the “console” systems available in living rooms. But Sony chose to sell the pocket radio to teenagers. Teenagers were delighted to have a product that was inexpensive and that they could carry with them, allowing them to listen to the rock and roll music their parents may have disapproved of.

Thus, a booming new market emerged based on a hitherto neglected feature that transistors made possible: portability. Meanwhile, the existing big American CE companies felt no need to change their business plans because the new products, in their opinion, were going to new customers not existing customers. Had Sony tried to sell its pocket radio to the parents, the product would have been judged to be inferior because they had the alternative of a high quality vacuum tube radio in their homes. Then in 1959, Sony introduced its first portable television, and again, the portable television was marketed to a whole new cohort of consumers who could not afford expensive, tube-based consoles. Since the alternative for these consumers was no TV at all, they were willing to accept a product with a somewhat inferior picture and reception.

By the 1960s, solid state electronics had become capable of handling the power required to make larger CE products, making it possible for the big American CE companies to incorporate solid state electronics in the big home products area that they dominated.  And, indeed, by the end of the decade all of the products for the mainstream living room market were using solid-state semiconductors. However, they were largely being made by Japanese companies. The Japanese perfected the semiconductor components through their foothold markets to such an extent that the American offerings were not nearly as good. The venerable vacuum tube companies were doomed.

THE CYCLE REPEATS

The table ends with four company names that are distinctly non-Japanese. The phenomenal process of innovation-based disruption is under-way again. This time, the incumbents are the same Japanese companies that used the transistor to disrupt the previous incumbents while the new entrants are small companies focused on microprocessor-enabled digital media or communications. In a cruel irony, the incumbents framed the innovation of digital media products as not good enough for the living room while the entrants saw it as good enough for a new, non-consumptive mobile or time-shifted usage. The existing customers saw the digital products as inferior while the new customers saw portability as a fair trade-off for quality.

A future article will discuss how the telecom industry saw voice mobility as sustaining but data mobility as disruptive.
—-

REFERENCES

  1. Electronics (2nd Edition), Allan R. Hamley, Prentice Hall, August 3, 1999.
  2. The Innovator’s Solution, Clayton M. Christensen, Michael E. Raynor, Harvard Business School Press, September 2003.
  3. “The Top 100 Gadgets of All Time”, Chris Null, Mobile PC, March 2005.
  4. Sony, John Nathan, Mariner Books, April 2001.
  5. The Road Ahead, Bill Gates, Nathan Myhrvold, Peter Rinearson, Viking Press, November 1995.
  • Roman

    Wow, this may have been overlooked when you had a smaller audience. It's a great piece – I wonder if "Best Of" section could be put in Categories up top, with gems like this one.

    • http://www.asymco.com asymco

      Thanks for the request. Been meaning to do that.

      • http://nmuppala.wordpress.com Nalini Kumar Muppala

        Still meaning to do that?

  • Niilo

    "The transistor could not amplify sound as well as a tube"

    The debate still rages in certain quarters Some things just refuse to be disrupted!

    http://www.thedailyguitar.com/the-debate-solid-st

  • Mark Hernandez

    When I was an 11 year old kid in the early sixties learning electronics from my dad's collection of Popular Mechanics magazines, I still have a very vivid memory of reading a particular article called "Will the Transistor Replace the Tube?"

    I tried to find that issue in an archive somewhere here in the future, but it's lost. I also had this amazing collection of odd and strange vacuum tubes, but also as a kid there was this one time I didn't pay my storage locker bill and they liquidated all my stuff. (sigh)

    But this time around I have a couple pieces of memorabilia that I keep nearby at all times — my '71 high school yearbook, and my August 1981 issue of Byte magazine that reported on another disruptive moment in our industry – the SmallTalk object-oriented computer language and the Xerox Parc computers it ran on with their strange graphical user interfaces.

    As most readers know, Apple's Lisa computer came out a couple years later, then the Macintosh in 1984. Here I am now, an iOS developer using Objective-C which is based on SmallTalk and C (which came along in the early seventies.)

    On page 33 of that issue is an ad for another disruptive moment, the Osborne portable computer with its 4" monitor, two 100K byte floppy drives and 64K of built-in memory, all for $1795. Take it with you ANYWHERE. :-)

  • Pingback: asymco | The parable of the telegraph()

  • John

    The paragraph describing the term "electronics" needs work. Vacuum tubes, as the name suggests, contain a vacuum, they do not contain gas. Although diodes did only have two terminals much more common were triodes which had three connections. The cathode and anode were terminals that sourced and collected electrons. The thirs terminal was called the grid. Small signals applied to the grid resulted in large changes in the current flowing through the tube. Sometimes these were also called valves due to the analogy with the way fluids could be controlled.

    The term "semiconductor" is not related to the one directional flow of current in a diode. It has to do with the physal properties of the materials used. In a conductor there are electrons which are free to move about the material as soon as an electric field is applied. A nonconductor does not have this feature. A semiconductor behaves like both of these. Normally there are no electrons available so they are nonconductors. However, it is possible to boost some electrons into a different part of the electronic structure of these materials so that they are free to move about and the material becomes a conductor. Scientists and engineers have been avoid to work with this to develop a dizzying array of products.

    Incidently, there are new disruptions on the way as scientists have learned how to process information using the spin state of electrons rather than the motion of electric charge.

    A book called The Attacker's Advantage discussed the difficulty established companies have absorbing new technology. They see it coming but they have a vested interest in protecting the status quo till it is too late.

    Great article.