The Idea Factory: Bell Labs and the Great Age of American Innovation (59 page)

35 “Charge to a Study of Foreign Intelligence Gathering,” May 13, 1957. The quote I’ve included reflects some of the inserted text that Baker had noted in the margins of his draft. AT&T archives.

36 Bamford,
The Puzzle Palace
, pp. 427–29.

37
New Scientist,
January 30, 1975, p. 274.

38 Baker’s handwritten notes. Baker Collection, Princeton University.

39 William O. Baker, National Reconnaissance Office Oral History. “I was sitting around
the Cabinet room during the Cuban missile crisis of October 1962,” Baker would recall, and ultimately served as a “legman,” going back and forth between the White House and State Department, relaying information about the Russian fleets steaming toward Cuba with their cargo.

40 Clark Clifford, “Serving the President,”
New Yorker
, May 6, 13, 20, 1991. According to both Baker and Clifford, the two men were eating lunch together in the White House mess when they were informed that Kennedy had been shot in Dallas.

41 Louis Tordella, letter to William O. Baker, January 15, 1959. Baker Collection, Princeton University.

1
Ian Ross, author interview.

2
William O. Baker, “Transistor Making and Mind Stretching,” speech given at the transistor’s twenty-fifth anniversary, Western Electric, Allentown, Pennsylvania, December 5, 1972. AT&T archives.

3
William O. Baker, Murray Hill, New Jersey, June 21, 1966. In his closing remarks at a two-day Bell Labs symposium entitled “The Human Use of Computing Machines,” Baker said, “Thus our nation, and to some degree mankind altogether, have become trustees for one of the heroic capabilities so far evolved from science and engineering—that of the giant computing machine and its associated networks of interactions and man-machine complementarities.”

4
Jack Kilby, oral history with Michael Wolff, December 2, 1975, IEEE Global History Network. For a deeper discussion of how Morton and Bell Labs failed to perceive the integrated circuit, see T. R. Reid’s
The Chip
, and Michael Riordan, “How Bell Labs Missed the Microchip,”
IEEE Spectrum
, December 2006.

5
William O. Baker, letter to Jim Fisk, July 11, 1958. Baker Collection, Princeton University.

6
In Kilby’s device, the different parts of the circuit—transistors and resistors and capacitors—were connected by wires rather than “etched” connections in the silicon, as was the case with Noyce’s design.

7
Arthur L. Schawlow,
Optics and Laser Spectroscopy, Bell Telephone Laboratories, 1951–1961, and Stanford University Since 1961
, an oral history conducted in 1996 by Suzanne B. Riess, Regional Oral History Office, Bancroft Library, University of California, Berkeley, 1998.

8
The first laser patent—which happens to be filed casually in a box buried in the AT&T archives—resulted in a ferocious and long-standing litigation. Gordon Gould, a former colleague of Charles Townes, laid a competing and (ultimately) partially successful claim to the laser.

9
Thanks to Herwig Kogelnik for pointing this out. The Schawlow and Townes March 1960 patent—No. 2,929,922—is titled “Maser and Maser Communications System.” In later years, it would become popular to say that the laser, during its early incarnation, was a solution looking for a problem—a quip apparently made by a colleague of Ted Maiman’s at Hughes Aircraft. In his bestselling book
The Black Swan
(New York: Random House, 2007), for instance, Nassim Nicholas Taleb asserts that the laser had “no real purpose.” The evidence contradicts this.

10 Joan Lisa Bromberg,
The Laser in America, 1950–1970
(Cambridge, MA: MIT Press, 1991), p. 92.

11 The reason that a laser is so much more suited to carrying information than ordinary light is that it is highly ordered (coherent) and monochromatic.

12 “Research Breakthroughs in Optical Masers and Superconductors,”
Bell Laboratories Record
, March 1961. In technical terms, Javan and colleagues “impressed a telephone conversation on a maser signal … by using an electro-optical device—the Kerr Cell.”

13 Herwig Kogelnik, author interview.

14 An undated explanatory flyer about the Bell laser research explained, “When Picturephone service becomes common, when high-speed data communication between computers is more widespread, and when all of today’s communications services have expanded, present message carrying capacities may not be enough.” AT&T archives.

15 Stewart Miller, “Communication by Laser,”
Scientific American,
January 1966.

16 C. G. B. Garrett, “The Optical Maser,”
Electrical Engineering
(April 1961): 248–51.

17 The waveguide pipes, as Jeff Hecht points out in
City of Light
(New York: Oxford University Press, 1999), his history of optical communications, were also sensitive to temperature changes and vibrations, not to mention seismic activity.

18 One possible solution to keeping a light beam focused within the pipe seemed to be the insertion of gas “lenses” inside the tubes that would continually refocus the light wave as it moved along. Still, waveguide transmissions might still be ruined by fluctuations in air temperature, ground vibrations, and tremors (see above). The problem appeared to be vexing.

19 Hecht,
City of Light
, p. 111.

20 Tingye Li, author interview. In Li’s view, Kao deserves enormous credit for three separate aspects that hastened the introduction of fiber optics. The first was proposing fused silica as the material for the glass strands. “The second thing was that [Kao] went ahead and measured a very low loss to the material. And the third was, he went around the world to get people to work on this. He came to Bell Labs, and he went to Corning.”

21 Ira Jacobs, author interview.

22 Memo, “Work Forecast,” S. E. Miller to R. Kompfner and J. R. Pierce, July 18, 1969. As Stew Miller put it to his bosses, “Our current mainstream effort is centered around laser transmission systems. Primary effort has been on the transmission media, the following being of prime interest: 1) Glass-lens (beam) waveguides …; 2) Gas-lens waveguides; 3) Glass fiber waveguides.” The memo goes on to detail the Labs’ efforts at all three and acknowledges its work on fiber was lagging behind the work at Corning, which had been described to the Bell Labs’ team through “a private communication.”

23 William O. Baker, “Testimony of William O. Baker,” May 31, 1966. F.C.C. Docket No. 16258.

24 Memo, R. Kompfner to J. K. Galt, November 30, 1970. AT&T archives.

25 Unix was created primarily by Ken Thompson, Dennis Ritchie, Brian Kernighan, Douglas McIlroy, and Joe Ossanna.

26 Willard Boyle and George Smith were credited as inventors of the CCD device and were awarded the 2009 Nobel Prize—to be shared with Charles Kao, the early champion of fiber optics. Immediately after Boyle and Smith won the award, Eugene Gordon, another Bell Labs veteran, challenged the credit they received and claimed that some of the ideas for the device originated with him. And as for making the device usable in a camera? That credit would probably go to Michael Tompsett, another Bell Labs engineer, who worked on developing the actual invention with Carlo Sequin and Ed Zimany. A series of articles in
Spectrum
, the magazine of the IEEE (the Institute of Electrical and Electronics Engineers), ably summarized the dispute:
http://spectrum.ieee.org/tech-talk/semiconductors/devices/nobel-controversy-former-bell-labs-employee-says-he-invented-the-ccd-imager
. Without question, the unsettled controversy highlights the inadequacy of the Nobel awards in singling out a small number of individuals for industrial advances where authorship and credit could be more broadly dispersed. The transistor award that went to Shockley, Bardeen, and Brattain—but not to any of the material scientists who fabricated the silicon and germanium—is arguably another case in point.

27 William O. Baker, National Reconnaissance Office Oral History, conducted by R. Cargill Hall, May 7, 1996. There is some confusion as to the date that Baker brought the CCD to Washington. He claims in this interview, probably mistakenly, that it was 1963. Almost certainly it was several years later.

28 In missing the invention of the integrated circuit, however, Bell Labs and AT&T were not at all precluded from using the idea. Patent agreements guaranteed that the Labs would have access to the technology—though not the glory.

29 By several estimates, the Picturephone cost nearly half a billion dollars to develop.

30 J. P. Molnar, memo, “PICTUREPHONE Program,” August 19, 1966. Baker Collection, Princeton University.

31 “Exploratory Study of the Market Potential for Picturephone Service,” American Telephone & Telegraph Company, Business Research Division, December 1967. AT&T archives.

32 James B. Fisk, speech given at the Midwest Research Institute, Kansas City, Missouri, May 22, 1968. Whether or not the technological hunch behind the Picturephone was correct, Bell Labs executives could rightly point to the device as a significant feat of engineering, one that indirectly yielded not only the CCD device but a variety of new applications in integrated circuits and (during the manufacturing process) ruby lasers.

33 To some extent, there was an expectation that in preparing the Bell System for Picturephones, Bell Labs was also laying the groundwork for high-speed network services for businesses. “The switching and transmission capacity provided for Picturephone service will also enable us to move into flexible high-speed message data services which customers could use as easily as they place an ordinary call today,” Ray Ralston, the Picturephone’s engineering chief, remarked (
195 Magazine
, November–December 1967). Indeed, the Picturephone was able to connect with a computer and thereby provide rudimentary data to its user.

34 Julius Molnar, “Technical Program of Bell Laboratories: Talk to Department Heads,” March 2, 1972. AT&T archives.

35 One of the more curious things about Picturephones emerged from research in John Pierce’s communications science department. Tests showed that it was easier to lie to someone during a Picturephone conversation than in an ordinary telephone conversation. The researchers concluded that when not distracted by a visual image, a caller attends more closely to the content and veracity of the exchange.

36 I’m indebted to Bob Lucky for this comparison. More precisely, Metcalfe’s law suggests that the value of a network grows in proportion to the square of its number of users.

37
John R. Pierce, “Rudolf Kompfner: 1909–1977,” National Academy of Sciences, Biographical Memoir, 1983.

1
The album of black-and-white photographs is part of the Shockley Collection, housed at the Stanford University archives. In addition to Kelly, Fisk, Shockley, Bown, Baker, Bardeen, Brattain, Molnar, and Pierce, other men at the party included Cal Fuller and Gerald Pearson, who would collaborate nine years later on the solar silicon cell. On the dais with Fisk was Harald Friis, the Danish-born antenna wizard who ran the Holmdel lab and served as a mentor to John Pierce and his Caltech friend Chuck Elmendorf. Friis had been instrumental in work leading to the construction of the microwave long-distance network.

2
Joe Baker, author interview.

3
Peter Temin with Louis Galambos,
The Fall of the Bell System: A Study in Prices and Politics
(New York: Cambridge University Press, 1987), p. 10.

4
Steve Coll,
The Deal of the Century: The Breakup of AT&T
(New York: Atheneum, 1986), p. 59.

5
The post-1956 legal tangles between AT&T, the U.S. Department of Justice, and the FCC are compelling but highly involved; moreover, because those legal tangles were often unrelated (or indirectly related) to the science and engineering of Bell Labs, they are not fully detailed here. Some of the competition to the Bell System dates back to
the 1950s and 1960s, to cases involving two independent companies and their devices, Hush-a-Phone and Carterfone. Hush-a-Phone was an attachment to a handset’s mouthpiece that allowed customers to talk or whisper into the phone with more confidentiality; Carterfone was a device that allowed mobile radios to be tied into the landline network. AT&T contended that both could conceivably harm the network, an assertion that was ultimately proven unfounded. Eventually, based on court appeals and FCC decisions, both devices were allowed. A number of books offer a useful presentation of the issues, players, and implications. See
The Deal of the Century
, above, for instance, or
The Fall of the Bell System
, also above. A concise and valuable summary is also given in the introduction to the three-volume set
Decision to Divest: Major Documents in U.S. v. AT&T, 1974–1984
, edited by Christopher H. Sterling, Jill F. Kasle, and Katherine T. Glakas (Washington, DC: Communications Press, 1986).

6
MCI was started by the entrepreneur Jack Goeken. McGowan joined in the late 1960s and soon became its singular, driving force. Goeken was forced out in the early 1970s.

7
The 1971 FCC decision was known as
Specialized Common Carrier
. It followed on the heels of several earlier decisions that opened the door for MCI. These were
Above 890
(1959), which allowed for private companies to build “point-to-point” microwave systems, such as those that might connect one corporate branch to another;
Carterfone
(1968), which allowed non-AT&T equipment to be connected to the network; and
Microwave Communications Inc.
(1969), which allowed MCI to build a point-to-point microwave service between Chicago and St. Louis.

8
Complaint, November 20, 1974,
United States of America v. American Telephone and Telegraph Company; Western Electric Company, Inc.; and Bell Telephone Laboratories, Inc
., pp. 11–14. As published in
Decision to Divest: Major Documents in U.S. v. AT&T, 1974–1984.
MCI also initiated its own private antitrust suit against AT&T.