The Idea Factory: Bell Labs and the Great Age of American Innovation
Page 43
29 Oxford English Dictionary, 2nd ed. See “innovate” and “innovation,” vol. 8, pp. 997–98. Usually, for science mandarins such as Frank Jewett, the chairman of the Labs and head of the National Academy of Sciences, the descriptive language for innovation in the World War II era, and immediately thereafter, was “ingenuity,” “invention,” and “development.” Though it may well have been used earlier, the first reference I came across to the word “innovate” in the Bell Labs literature was in a 1958 speech by Jack Morton and in the 1959 Bell Laboratories Record touting the Labs’ involvement in the military’s Distant Early Warning (DEW) line. For further reference, see chapter 9, endnote 4.
30 Ernest Braun and Stuart Macdonald, Revolution in Miniature: The History and Impact of Semiconductor Electronics (Cambridge: Cambridge University Press, 1982), p. 5.
31 Ralph Bown, “The Transistor as an Industrial Research Episode,” Scientific Monthly, January 1955.
32 Jack A. Morton, “From Research to Technology,” International Science and Technology, May 1964.
33 Jack A. Morton, “The Innovation Process,” date unknown. AT&T archives.
34 Eugene I. Gordon, “Morton’s Legacy,” New Jersey Council News, 1991; also author interviews.
35 Michael Wolff, “The R&D ‘Bootleggers’: Inventing Against the Odds,” IEEE Spectrum, July 1975.
36 “Kelly’s favorite room” is from an author interview with Robert Von Mehren, Kelly’s son-in-law; the description of the “stately room” comes from my visit to the actual house in Short Hills.
37 Interview of Walter Brattain by Charles Weiner, AIP.
38 Ralph Bown had already explained the technical reasons for spreading the invention around. An internal Bell Labs memo written a decade after the transistor became a commercial product noted that the large semiconductor industry, “with its center of gravity outside the Bell System,” was the deliberate result of a policy that the Labs’ managers settled on in the months after the invention. By involving engineers around the world in the evolution of the device—making it better, cheaper, more reliable—the hope was that everyone would profit from the advances, especially the Bell System.
39 Ralph Bown, letter to M. J. Kelly, Baur au Lac Hotel, Zurich, Switzerland, August 20, 1948. AT&T archives.
40 Oliver Buckley to M. J. Kelly, Savoy Hotel, London, England, September 15, 1948. AT&T archives.
41 M. J. Kelly, “Remarks Before Bell System Lecturer’s Conference,” October 2, 1951. AT&T archives.
42 Milton Silverman, “Ma Bell’s House of Magic,” Saturday Evening Post, July 1947. The publication of the feature sparked an internal correspondence between Bell Labs executives, all of whom viewed the story critically.
43 William G. Pfann, “Some Remarks on the Discovery of Zone Melting,” May 3, 1965. AT&T archives.
44 Mervin J. Kelly, “Semiconductor Electronics: A New Technology—A New Industry,” 1958. AT&T archives.
CHAPTER SEVEN: THE INFORMATIONIST
1 Claude Shannon, Kyoto Prize acceptance speech, 1985. Shannon Collection, Library of Congress. See also Anthony Liversidge, “Profile of Claude Shannon,” in Claude Elwood Shannon Collected Papers, reprinted (in a slightly different form) from Omni magazine, August 1987. Used by permission of Liversidge.
2 Vannevar Bush, letter to Professor E. B. Wilson, Harvard School of Public Health, December 15, 1938. “As he appeared to have great promise, and moreover appeared to be a decidedly unconventional type of youngster, I made it possible for him to go on with his studies.” Shannon Collection, Library of Congress.
3 Vannevar Bush, recommendation for Claude Shannon for a National Research Fellowship, undated, circa late 1939. Shannon Collection, Library of Congress.
4 Erico Marui Guizzo, “The Essential Message: Claude Shannon and the Making of Information Theory” (master’s thesis, MIT, 2003).
5 Len Kleinrock, a former student of Shannon’s, author interview.
6 Liversidge, “Profile of Claude Shannon.”
7 Claude Shannon, letter to Dr. V. Bush, December 13, 1939. Shannon Collection, Library of Congress.
8 Liversidge, “Profile of Claude Shannon.” Biographical facts relating to Shannon’s father are in a personal letter Shannon wrote, October 20, 1981, to Ms. Shari Bukowski: “[My father] was born in Oxford, New Jersey in 1862, came to Ovid, Michigan when very young and was raised and graduated there. He was a traveling salesman for a period and came to Gaylord shortly after 1900. There he bought a furniture and undertaking business, and, having confidence in Gaylord’s future, built the Shannon Block and Post Office building on Main Street.” Shannon Collection, Library of Congress.
9 Robert McEliece, Claude Shannon, Father of the Information Age, directed and written by Doug Ramsey, produced by Ramsey and Mike Weber; http://www.youtube.com/watch?v=z2Whj_nL-x8. In a profile of Shannon in his Collected Papers (New York: IEEE Press/John Wiley & Sons, 1993), the book’s editors, N. J. A. Sloane and Aaron D. Wyner, refer to H. H. Goldstine’s book The Computer from Pascal to Von Neumann, which described the thesis as “one of the most important master’s theses ever written … a landmark in that it helped to change digital circuit design from an art to a science.”
10 Claude Shannon, letter to Dr. V. Bush, December 13, 1939. Shannon papers, Library of Congress.
11 Norma (Levor) Barzman, author interview.
12 Claude Shannon, letter to Dr. V. Bush, March 8, 1940. Shannon Collection, Library of Congress.
13 Claude Shannon, letter to Dr. V. Bush, February 16, 1939. “Off and on I have been working on an analysis of some of the fundamental properties of general systems for the transmission of intellegence [sic], including telephony, radio, television, telegraphy, etc.” Shannon, Collected Papers, p. 455.
14 Claude Shannon, oral history conducted in July 1982 by Robert Price, IEEE History Center, New Brunswick, NJ.
15 Claude Shannon, interview with Robert Price, December 20, 1983. Shannon said: “… a lot of information theory … I had worked out before, during the five years between 1940 and 1945. Much of that work I did at home.” Shannon’s penciled manuscripts: author’s review, Shannon Collection, Library of Congress.
16 Betty Shannon, author interview.
17 T. C. Fry, oral history, unpublished. Interviewed by H. O. Pollak and D. M. La Porte. Author’s copy, p. 4.
18 Ibid., p. 55.
19 Ibid., p. 33.
20 Henry Pollak, author interview.
21 M. D. Fagen, ed., A History of Engineering and Science in the Bell System: National Service in War and Peace (1925–1975) (Bell Telephone Laboratories, 1975), p. 172.
22 Claude Shannon, oral history with Robert Price, IEEE.
23 David Hagelbarger, author interview.
24 David Kahn, a historian of cryptography, has noted, “Shannon’s insight, his great contribution to cryptology, lay in pointing out that redundancy furnishes the ground for cryptanalysis.” Essentially, Shannon’s analysis had “given an explanation for the constancy of letter frequency … he has thus made possible, for the first time, a fundamental understanding of the process of cryptogram solution.” David Kahn, The Codebreakers: The Comprehensive History of Secret Communications from Ancient Times to the Internet (New York: Scribner, 1996), p. 748.
25 Like many innovations in the Bell System, the precise history of PCM includes competing claims. AT&T has maintained, somewhat unpersuasively, that it was an early, albeit unrecognized invention of P. M. Rainey of Western Electric in 1926. But most accounts credit the invention to A. H. Reeves of International Telephone and Telegraph in 1938. Shannon’s primary contribution to the technology is in B. M. Oliver, J. R. Pierce, and C. E. Shannon, “The Philosophy of PCM,” in Collected Papers.
26 An internal history of Bell Laboratories described two major reasons for PCM in slightly more technical terms: First, its traffic capacity could be high. The pulses and codes of a PCM system “can be very short in duration,” allowing for the “multiplexing” of other signals (audio as
well as video) in the intervals between. Also, it would be easier to preserve the quality of a digital signal, since “the uniform pulse coded samples may be exactly reconstructed at regenerators along a line so that there need be no accumulation of impairments with distance, as there is in all analog systems.”
27 Claude Shannon, oral history with Robert Price, IEEE.
28 “We had an internal distribution system,” recalls a Bell Labs mathematician, Brock McMillan, who worked in the office next to Shannon’s at Murray Hill at the time. “So people in the math department could read things before publication. Shannon was not particularly talkative. He talked to us about his theories maybe a little bit. But that 1948 paper caught me almost stone cold.” Most of the Bell researchers, with the exception of Shannon’s friend Barney Oliver, experienced a similar sense of shock. John Pierce likened the surprise he felt upon encountering his friend’s ideas to the dropping of a powerful explosive.
29 Shannon, Kyoto Prize speech.
30 Robert Lucky, Silicon Dreams: Information, Man, and Machine (New York: St. Martin’s Press, 1991).
31 “Information Theory,” from Shannon’s article on the topic in Encyclopedia Britannica, 14th ed., 1968.
32 John Robinson Pierce, quoted in M. Mitchell Waldrop, “Claude Shannon: Reluctant Father of the Digital Age,” Technology Review, January 7, 2002. For my explication of information theory, I owe a debt to a number of people who have written on Shannon’s theory: John Pierce, Warren Weaver, David Kahn, Robert Gallager, Bob Lucky, Neil Sloane, Aaron Wyner, John Horgan, Wiliam Poundstone, and Erico Marui Guizzo. Before me, Pierce, Poundstone, and Guizzo also made the insightful point juxtaposing the reduction of redundancy in cryptanalysis with the addition of redundancy in coding. Finally, Shannon was especially eloquent on the genesis and meaning of his own work in his speeches throughout the 1950s, which were rich with metaphors to explain his ideas; in his interviews with Robert Price and Anthony Liversidge; and (later still) in his Kyoto prize speech in 1985.
33 Richard Hamming and David Slepian are most often credited for their work on these codes.
34 Waldrop, “Claude Shannon: Reluctant Father of the Digital Age.” As Bob Gallager at MIT points out, communications engineers (Fano included) had long understood that errors could be reduced by increasing the power of a transmission and/or increasing the channel capacity—that is, the “bandwidth” through which transmissions moved. Shannon’s great breakthrough was in showing that they could reduce errors without doing either.
35 Shannon, Collected Papers; Lucky, Silicon Dreams, p. 37.
36 Claude Shannon, letter to Francis Bello at Fortune. Shannon Collection, Library of Congress.
37 Hamming complained about this in an unpublished oral history.
38 The supervisor was John Pierce, whose work is detailed later in this book. The anecdote comes from Henry Landau, a former Bell Labs mathematician. Author interview.
39 William Shockley, interviewed by Harriet Zuckerman, 1964, as part of her larger study of the research careers of Nobel laureates in the sciences, Columbia Oral History Research Office. See Harriet Zuckerman, Scientific Elite: Nobel Laureates in the United States (New York: Free Press, 1977), and the enlarged edition (New Brunswick, NJ: Transaction Press, 1996). Quoted by permission of Dr. Zuckerman.
40 John R. Pierce, “Mervin Joe Kelly: 1894–1971,” National Academy of Sciences, Biographical Memoirs, 1975.
41 C. Chapin Cutler, oral history, IEEE. This point was also made to me in an interview with William Keefauver, a former patent attorney at the Labs.
42 Liversidge, “Profile of Claude Shannon.” When asked, “Can Bell Labs take credit to some extent for your achievement?” Shannon answered, “I think so. If I had been in another company, more aimed at a particular goal, I wouldn’t have had the freedom to work that way.”
43 Walter Brattain conceded this in a 1963 oral history interview with Harriet Zuckerman, Columbia University Oral History Research Office.
44 Toby Berger, Claude Shannon, Father of the Information Age, directed and written by Doug Ramsey, produced by Ramsey and Mike Weber; http://www.youtube.com/watch?v=z2Whj_nL-x8.
CHAPTER EIGHT: MAN AND MACHINE
1 Shannon Collection, Library of Congress.
2 Claude E. Shannon, “Programming a Computer for Playing Chess,” March 9, 1949, Collected Papers, edited by N. J. A. Sloane and Aaron D. Wyner (New York: IEEE Press/John Wiley & Sons, 1993).
3 Claude Shannon, Kyoto Prize acceptance speech, 1985. Shannon Collection, Library of Congress.
4 It was actually built of Meccano parts, rather than Erector Set parts, purchased at the Hudson Dobson store in New York City.
5 Brock McMillan, author interview.
6 William Keefauver, former Bell Labs patent attorney, author interview.
7 “Mouse with a Memory,” Time, May 19, 1952.
8 Claude Shannon, letter to Miss Irene Angus, August 8, 1952. Shannon Collection, Library of Congress.
9 Mervin Kelly to Ralph Bown, March 24, 1950. Shannon Collection, Library of Congress.
10 Walter B. Smith to Dr. M. J. Kelly, May 4, 1951. Shannon papers, Library of Congress.
11 Francis Bello, Fortune, December 1953.
12 Claude E. Shannon, “Game Playing Machines,” 1955, in Collected Papers.
13 THROBAC stands for “THrifty ROman numeral BAckward-looking Computer.” Shannon later admitted that the name (and to some extent the calculator itself) was inspired by the era’s computer acronyms—ENIAC, UNIVAC, and so on. Unlike some of Shannon’s other machines, which are now owned by the MIT Museum, his original “ultimate machine,” with the mechanical hand, has been lost, according to his wife, Betty.
14 Shannon, “Game Playing Machines,” 1955, Collected Papers.
15 It was unclear, fifty years after the fact, which game Slepian was singling out. It may have been Shannon’s version of the complex game Hex.
16 Robert Lucky, Silicon Dreams: Information, Man, and Machine (New York: St. Martin’s Press, 1989), p. 51.
17 Claude Shannon to H. W. Bode, October 3, 1956. Shannon Collection, Library of Congress.
18 Claude Shannon, oral history conducted in July 1982 by Robert Price, IEEE History Center, New Brunswick, NJ..
19 Shannon Collection, Library of Congress.
20 “I hear via my daughter and your son that you’ve bought a Volkswagen microbus and are going to California.” Edward Moore, letter to Claude Shannon, February 26, 1957. Shannon Collection, Library of Congress.
21 Shannon’s work on cryptography with the NSA probably did not continue beyond the 1960s. Even in the 1980s, though, he was forbidden to speak of it—a hallmark of the secrecy obligations that involve any type of work with the NSA. In response to a Freedom of Information Act request by this book’s author, the NSA responded that “Dr. Shannon was never an affiliate of NSA. He was, however, involved as a consultant to NSA as a member of various committees.” Documents demonstrate that while Shannon was still at Bell Labs he was at least involved with the Mathematics Panel of the National Security Agency Scientific Advisory Board (NSASAB). When asked by Bob Price during an oral history many years later about his National Security Agency contributions, Shannon replied, “I was a consultant. I probably should … I don’t know … I think, I don’t know that I have any … even though this was a long time ago, I’d better not talk about …” Claude Shannon, oral history with Robert Price, IEEE.
22 Claude Shannon, letter to James R. Newman, April 9, 1958. Shannon Collection, Library of Congress.
23 Bob Lucky, author interview.
CHAPTER NINE: FORMULA
1 “Dr. Kelly Visits Technical Groups in Europe,” Bell Laboratories Record, June 1950.
2 Mervin Kelly, letter to Ralph Bown, March 24, 1950. Shannon Collection, Library of Congress.
3 Mervin J. Kelly, “The Bell Telephone Laboratories—An Example of an Institute of Creative Technology,” March 23, 1950. AT&T archives.
4 It is by all m
eans possible that a Bell Labs scientist used the word “innovation” before 1958. However, in a review of thousands of Bell Labs documents and hundreds of speeches from the 1950s, the first instance I came across was in a speech by Jack Morton, “Some Thoughts About the Future,” at the tenth anniversary of the transistor, June 17, 1958. “In the past,” Morton remarked, “American industry has led the world because we have used existing knowledge with ingenuity, energy and innovation.” Perhaps simultaneously, John R. Pierce began using the word, too. “Innovation in Technology,” an essay by Pierce, appeared in the September 1958 issue of Scientific American. Moreover, a draft of a paper of Pierce’s later titled “Myths of Creation,” dated September 10, 1958, was originally titled “On Innovation in Science.”
5 Ernest Braun and Stuart Macdonald, Revolution in Miniature: The History and Impact of Semiconductor Electronics (Cambridge: Cambridge University Press, 1982), p. 33.
6 John Rowell, author interview.
7 Some of the most acclaimed Bell Labs scientists would often stand amazed by the talents of their technical assistants. For Morris Tanenbaum, for instance, it was Ernie Buehler, who was adept at growing crystals.
8 Phil Anderson, author interview.
9 Sara Silver, “With Uncertain Future, Bell Labs Turns to Commerce,” Wall Street Journal, August 21, 2006.
10 Mervin J. Kelly, Bell Telephone Magazine, Summer 1953.
11 William T. Golden, “Conversations with Drs. Oppenheimer, Robert Bacher, and Charles Lauritsen,” Memorandum for the file, December 21, 1950. “We commented on the volume of his [Kelly’s] work and they said that he seems to be working himself to death although he does not look as badly they said as he did during the war.” Golden’s search for a presidential science advisor involved numerous interviews with Kelly and his peers—Oppenheimer, DuBridge, and so forth. These interviews are chronicled in his contemporaneous memoranda, available at http://archives.aaas.org/golden.