- SCU Home Page
- About SCU
- On Campus
- News & Info
- About Us
- Activity Areas
- CSTS News & Media
- CSTS News
Nothing New Under the Sun? A Comparison of Edisonian and Silicon Valley Startups
Thomas P. Hughes
A historian makes history useful by drawing analogies between the past and the present and projecting patterns taken from these analogies to envision future scenarios. If by analogy a historian finds, for instance, a pattern that applies to startup projects or companies in the past and present, then the historian can suggest to his readers that this repetitive pattern may prevail in the future. Finding such a pattern would leave the impression that there is nothing entirely new about startups, even in Silicon Valley.
Searching for such a pattern, I shall focus upon past and present inventor-entrepreneurs who preside over startup projects and companies. I will seek analogies between Edisonian era inventor-entrepreneurs who flourished in the late nineteenth century and informa-tion-era inventor-entrepreneurs who were active in the late twentieth century. Inventor-entrepreneurs of the past and of the present differ from ordinary inventors in that they not only invent, but they develop their inventions until they are ready to be brought into use or onto the market. Another major characteristic of in-ventor-entrepreneurs, past and present, is their freedom to choose the problems that they want to solve.
Entrepreneurs of the Edisonian Era
Thomas Edison is the best-known American inventor-entrepreneur. He nurtured numerous inventions and established a number of startup companies. While Edison is the most famous, a number of other independent inventor-entrepreneurs flourished during the Edisonian era. I call them “Edisonian inventors” because they lived at a time in which he towered above them and because his style of invention and development influenced theirs.
Among the other leading Edisonian inventors were entrepreneurs such as: Elmer Sperry, who invented arc lighting systems as well as automatic airplane and ship pilots; William Stanley, a developer of an alternating current transformer; Nikola Tesla, inventor of polyphase alternating current machines; Frank Sprague, who introduced an electric streetcar system; Lee De Forest, a radio, or wireless, pioneer; and Reginald Fessenden, who introduced voice radio transmission. Their inventions, like Edison’s, clustered in electrical power and electric communications. Edisonian era inventor-entrepreneurs freely chose their problems because they were independent. Neither industry nor government employed them. They fre-quently invented new systems. In contrast, inventors and scientists working in industrial laboratories usually made cumulative improvements in the product lines of their employers.
Entrepreneurs of Recent Decades
Because they have often been associated with universities, many leading inventor-entrepreneurs of recent decades have also enjoyed freedom in choosing problems. The inventions of contemporary inventor-entrepreneurs working in Silicon Valley or in the Bos-ton/Cambridge area cluster in the computer and telecommunications realm.1 They have introduced both new hardware and software systems. While a research engineer at MIT, Jay Forrester presided over the early development of the Whirlwind computer, a groundbreaking command and control device. As a graduate student at MIT, Ivan Sutherland developed the Sketchpad, a graphics system that opened the field for development. After he became a professor at the University of Utah, his graduate students continued the development of graphics. The Massachusetts Institute of Technology environment influenced Joseph Carl Robnett Licklider’s vision of the symbiotic relationship between computers and humans. The ARPAnet, the forerunner of the Internet, took root in university computer research centers. The list of uni-versity-based inventor-entrepreneurs in computing can be greatly extended.
A disproportionate number of leading inven-tor-entrepreneurs have been associated with Stanford University. John Hennessey of Stanford University started MIPS Computer Systems, which developed RISC (reduced instruction set computing) chips upon which Silicon Graphics machines are based.2 Andy Bechtolsheim and Forest Baskett, graduate students at Stanford, created SUNet, which interconnected workstations. Vinod Khoshla, when a student at the Stanford Business School, got wind of this idea and formed Sun Microsystems to manufacture the hard-ware.3 Jim Clark, a Stanford engineering professor, founded Silicon Graphics, and then with Marc Andreessen, a student at the University of Illinois, founded Netscape. Jerry Yang and David Filo, Stanford computer science graduate students, developed the Yahoo search engine for the Internet, and Stanford undergraduate students teamed up after graduation to invent the Excite browser.
Comparing Entrepreneurship of the Edisonian Era with Recent Decades
Having studied over a number of years the history of Edisonian inventor-entrepreneurs, I am confident that I recognize their pattern of invention, development, and innovation as they launched startup companies. On the other hand, my knowledge of Silicon Valley entrepreneurs and their startup projects and companies is limited. To see if the pattern that I found for the Edisonian startups applies for recent Silicon Valley ones, I asked my forty or so students in two Stanford University seminars in 1999 and 2000 to write case histories of Silicon Valley startups to see if the Edisonian pattern prevails.
Drawing upon interviews with founders of startups, information on the web, and written sources, they found that in general the Edisonian pattern continues to prevail today. In this essay, I draw upon and cite several of their term reports. Because the pattern seems to prevail for Edisonian and information era startups presided over by inventor-entrepreneurs, I propose that it may prevail for entrepreneurial startups in the future. Such an exercise, however, should not be taken as an assertion of truth, but simply as a hypothesis that raises questions and provokes discussion.
The following chronologically listed concepts constitute the pattern suggested by the startup activities of the inventor-entrepreneurs, past and present, and one that may prevail in the future.
Simultaneity of Invention
Inventing by Analogy
Invention to Management
In this brief essay, I cannot rehearse the evidence that can be drawn from my own essays and books about Edisonian inventors, nor can I present the abundant evidence in the Stanford student essays, but I can define the pattern concepts by summarizing cases from the Edisonian era and from the recent past in Silicon Valley and Boston/Cambridge.4
Eureka moments sometimes trigger an invention and development process. Having learned from one of his professor’s lectures of the need for an alternating current motor, Nikola Tesla claims that his inventive idea for a polyphase electrical system occurred to him while walking through a park recalling lines from Goethe’s Faust. “In an instant I saw it all,” he recalls, “and I drew with a stick on the sand the diagrams which were illustrated six years later in my fundamental patents of May, 1888.”5
Edison vividly recalls a Eureka moment com-polyphase, alternating current motor. The problem and ing when observing a flawed incandescent lamp sys-response associated with reverse salients continued as tem invented by another inventor prior to Edison’s light and power evolved. successful one. Edison wrote to an associate:
“Have struck a bonanza in electric light…I have the right principle…but time, hard work and some good luck are necessary too. It has been just so in all of my inventions. The first step is an intuition, and comes with a burst, then difficulties arise—this thing gives out and [it is] then that ‘Bugs’—as such little faults and difficulties are called—show themselves and months of intense watching, study and labor are requisite before commercial success or failure is certainly reached.” 6
While flying in 1987 on the private plane of Mitch Kapor, the developer of the Lotus spread sheet, Jerry Kaplan and Kapor had a Eureka moment. When transferring by keyboard his numerous notes scribbled on scraps of paper into a computer, Kapor asked if there might not be an easier way of entering information into a computer directly. Kaplan took a brief nap and then woke up with an idea for a stylus “inputting” directly into a computer rather than using a pencil, paper, and keyboard: “We both sat in stunned silence as this insight sunk in…We had put it all together, combining simple, familiar elements into something radically new.”7 Thus was born the idea for an early, but premature, version of today’s widely used pocket computer.8
A reverse salient is an inadequately functioning component in a complex system consisting of many components. Reverse salients constitute problems that inventors focus upon and often solve. As electric light and power systems evolved after 1880 in the United States, a host of reverse salients emerged. After Edison introduced direct current electricity in 1881 to light the Wall Street District in Manhattan, other inventors quickly recognized a reverse salient. Direct current could not be transmitted economically for more than a mile or so, thus limiting the area that a central generating station could supply.
In the United States, William Stanley and George Westinghouse solved the problem by introducing alternating current systems that could economically transmit electricity at high voltage over many miles. Another reverse salient emerged, however, because alternating current systems did not have a satisfactory motor. Tesla and others responded with a polyphase, alternating current motor. The problem and response associated with reverse salients continued as light and power evolved.
Countless reverse salients stimulate the problem choices of Silicon Valley inventors. Yet, not enough time has passed to give historians a perspective that allows them to discern a pattern of reverse salients over time comparable to that seen in the history of light and power. There are a number of isolated examples, however. The burgeoning increase of information on the World Wide Web in the 1990s, for example, caused a major reverse salient because of the mountain of information through which users had to sift in order to find information that they needed. This need stimulated near simultaneous development of several Internet search engines.
Simultaneity of Invention
Because inventor-entrepreneurs tend to congregate at pressing problem sites, or where there are reverse salients, simultaneity of invention is commonplace. Throughout history, inventors and scientists have become involved in futile priority disputes because they do not realize that their peers also perceive problems ripe for solution. A famous case of simultaneity is Alexander Graham and Elisha Gray appearing at the patent office within hours of one another making claims for similar telephone systems.9 About the same time as Tesla patented his polyphase system in 1888 at least four other inventors patented similar devices. They all perceived the need for an efficient alternating current motor.
Information age inventors also experience simultaneity of invention. About the same time that six former Stanford undergraduate students were developing a search engine for the Internet, David Filo and Jerry Yang, two Stanford computer science graduate students, were developing a text-based searching tool, too. Because the amount of electronically stored texts on the Internet was increasing by leaps and bounds around 1993, others besides the two Stanford teams were simultaneously seeking means to retrieve documents. The need for a search engine was an obvious reverse salient for inventive types keeping track of Internet progress. After graduation, the six Stanford students developed their search and retrieval software in a garage near Stanford University. With an impressive prototype, they successfully found funding and established a company named Excite. The two Stanford graduate students founded Yahoo (Yet Another Hierarchical Officious Oracle) to market their search engine. They went public with stock a few weeks after Excite.10
Invention by Analogy
Analogy sometimes motivates inventors. The possibility of moving into the unknown drawing upon the known emboldens them to forge ahead. Analogies provide for an inventor a bridge from the discovered into the realm of the undiscovered. Edison used analogies extensively. He worked out the quadruplex telegraph, one of the most elegant and complex of his inventions, “almost entirely on the basis of an analogy with a water system including pumps, pipes, valves, and water wheels.”11 Later, thinking anologically, he saw the similarity between existing centralized, illuminating- gas distribution systems, and the illuminating, incandescent-light, distribution system that he intended to invent.12
An analogy deeply influenced Joseph Carl Robnett Licklider’s concept of the computer. In the 1960s numerous young computer scientists and engineers considered Licklider a visionary who was defining the young computer field. He had a varied background that prepared him well to play such a role. Before 1962 he had been an electrical engineering professor at MIT, worked for Bolt, Beranek and Newman, a highly innovative, research -oriented acoustics and computer firm, and served as a group leader at MIT’s Lincoln Laboratory, which pioneered in computer development. In 1962, he became the head of the Information Processing Techniques Office (IPTO) of the Defense Advanced Research Projects Agency (DARPA), a Defense Department agency. As head of IPTO he dispensed funds generously to university computer research centers. Recipients often had projects that related to Licklider’s vision of the future of computers.
Licklider defined a highly influential concept of the computer in a seminal 1960 paper entitled “Man- Computer Symbiosis.” It opened with this preliminary to an analogy:
“The fig tree is pollinated only by the insect Blastophaga grossorum. The larva of the insect lives in the ovary of the fig tree, and there it gets its food. The tree and the insect are thus heavily interdependent: the tree cannot reproduce without the insect; the insect cannot eat without the tree; together, they constitute not only a viable but a productive and thriving partnership. This cooperative ‘living together in intimate association, or even close union, of two dissimilar organisms’ is called symbiosis.”13
He went on to develop the analogy by arguing that the relationship between man and the computer should be symbiotic.
In contrast, many artificial intelligence scientists and engineers at the time were predicting that computers would replace humans, not partner with them. Licklider said that humans should turn over to computers tasks mostly clerical or mechanical. He had found by self-observation that he gave most of his socalled thinking time to clerical or mechanical activities involving searching, calculating, plotting, transforming, and determining the logical consequences of hypotheses, and otherwise preparing the ground for his occasional—but essential—insights and decisions. All but the insights and decisions he wanted to turn over to his computer partner. 14 Time has reinforced his vision, rather than that of artificial intelligence experts.
Once embarked upon a project, inventor-entrepreneurs of the late nineteenth century organized a development team. The size might vary from the four or five crafts-persons who worked with prolific inventor Elmer Sperry when he was developing electrical, chemical, and feedback control systems to the twenty or so who constituted Edison’s team at his well-equipped Menlo Park Laboratory. Stories abound about the informal relations among the Edison group. A memorable image has Edison sitting in his Menlo Park Laboratory. Stories surrounded by a dozen or so of his mechanics, chemists, and model builders. One of the team played a pipe organ as food was brought in for a late evening break.
Hardware and software teams today also favor informal and flat, rather than hierarchical, relationships. Several teams to whom I have shown the Edison image of the evening break identify with the informal interactions of the Edison team. An Edisonlike atmosphere seems to have prevailed among a team of engineers working at the firm of Bolt, Beranek and Newman in the late 1960s on the Interface Message Processor (IMP), a major component of the ARPAnet. They had offices side by side so they could engage informally in hallways and over coffee. When one of them achieved a technical breakthrough he would run to the others exclaiming, “Look, I got this running… This is exciting. Something is cycling.” Frank Heart, an engineer who headed the IMP team, recalls the project as “a labor of love;” members of the team called the work “fun.” Heart encouraged the team to reach decisions by consensus. Finding bright people who were interested in the project, he gave them free rein.15
Obtaining funding for a project presents a difficult hurdle for inventor-entrepreneurs. Edisonian types often raised funds by persuading renowned capitalists to invest money in their projects.16 Edison and Tesla appealed successfully to J.P. Morgan who was a sentimental admirer of Edison whom he considered a truly creative person. Tesla, on the other hand, disillusioned Morgan by failing to fulfill grandiose projects and promises. Lee De Forest, a wireless inventor, mastered the art of fund raising through public display. He installed, for example, his wireless telegraph systems on a tugboat from which he reported to newspapers the progress of international yacht races. Operating his system across New York Bay brought “a gratifying amount of public recognition” and the sale of stock of the company holding his patents.17 Edisonian inventors had to describe their projects simply and vividly in order to have newsworthy stories for the press and to engage the enthusiasm and acquisitiveness of financiers devoid of technical knowledge.
University-based inventor-entrepreneurs often need to get the attention of technically and scientifically informed project officers at federal government agencies such as the National Science Foundation. As a result, composing detailed grant applications describing projects that appear to be in the public’s, or taxpayer’s, interest has become an art sometimes farmed out by the inventor-entrepreneurs to specialists. National defense needs often opened Defense Department coffers during the Cold War.
In Silicon Valley in the 1970s, inventor-entrepreneurs turned away from government funding and sought support from venture capitalist firms, many of which are located near Stanford University on Palo Alto’s Sand Hill Road.18 Needy inventor-entrepreneurs traveled from venture capitalist to venture capitalist hyping their software and detailing their business plans, which were often more fiction than fact.
John Doerr of the venture capital firm of Kleiner Perkins Caufield & Byers has become a legendary figure. By 1997, through funding startups, he created what has been called the largest legal accumulation of wealth in the history of the planet. Before the collapse of the “dot com” bubble, Kleiner Perkins-backed companies, many of which Doerr nurtured, may have been worth $125 billion in stock.
Startup companies take on momentum as they mature. A metaphor borrowed from physics, momentum is the inertia of a mass in motion. The mass tends to maintain its velocity and direction when it has high momentum. A startup company with high momentum tends to continue its rate of growth and its direction of development, continuing, for example, to produce the same hardware and software. The investment of capital in a company’s means of production and the particularized skill and experience of its technicians, engineers, and managers create its mass. The level of production contributes to its velocity. A startup with low momentum is relatively easily shaped by forces in its environment, such as the availability of venture capital and the market for its software or hardware. As the startup takes on relatively high momentum, it tends to shape its environment by attracting capital and stimulating a market.
Momentum can work for or against inven-tor-entrepreneurs initiating a startup company. The momentum of the Microsoft Company is well known. Having acquired human and capital resources and overwhelming market share, its momentum can carry it forward over obstacles. On the other hand, the high momentum of large corporations, such as Microsoft, can thwart an inventor-entrepreneur’s young company with a competing product. This policy of Microsoft is a major reason for its having been brought into federal and state courts.
There are many other examples of a high momentum company thwarting competing ventures. Paul Baron, while a RAND Corporation researcher, encountered the frustrating momentum of a large corporation. In the early 1960s, he proposed a digital packetswitching communications network of the kind later adopted by the creators of the ARPAnet. Engineers and managers at AT&T, which then dominated the long distance telephone system, talked politely to Baron, but he found them resolutely committed to their analog system. Presiding over a totally integrated system, AT&T insisted that new technology had to fit into its system. One AT&T engineer after an exasperating session with Baron told him, “First it [a digital system] can’t possibly work, and if it did, damned if we are going to allow creation of a competitor to ourselves.” 19
Invention to Management
Inventor-entrepreneurs of the late nineteenth century presided over a project from invention to innovation, or entry into the market. Typically inventor- entrepreneurs, as their companies grew large, turned over the company management to managerentrepreneurs. Sperry, for instance, always left management to others. Edison did not enter the management structure of the General Electric Company, which grew out of his inventions.
Similarly, the founders of start-up companies today give way to managers often at the insistence of the venture capital firms backing them. The young founders of Excite put their egos on the back burner and brought in a seasoned manager to head their company. The Wall Street Journal praised their decision to bring in “adult supervision” as a critical reason for Excite’s successful transition from a start-up to a public company.20
In summary, the stages of development of invention and development projects of the golden era of independent inventors and of the frenetic era of later twentieth century start-ups resemble one another. The usable history approach suggests that those who embark upon such projects in the future will experience similar stages of development.
1 It would be interesting to see if biotechnology
startups appearing so rapidly in Silicon Valley and
in the Cambridge/Boston area develop in stages similar
to the ones I suggest prevail among Edisonian
and computer startups.
2 Hennessey was the director of the VLSI (Very Large
Systems Integration) project at Stanford.
3 I am indebted to Professor Tim Lenoir of Stanford’s
history department for this information.
4 My conclusions about Edisonian inventors can be
found in chapters 1 and 2 of American Genesis: A
Century of Invention and Technological Enthusiasm,
1870-1970 (New York: Penguin Books, 1990). I
have the Stanford student papers on file.
5 For more on Tesla and his invention, see Thomas
P. Hughes, Networks of Power: Electrification in Western Society, 1880-1930 (Baltimore: The Johns Hopkins University Press, 1983), 112-117.
6 Edison to Puskas, 13 November 1978. Edison Papers, West Orange, New Jersey.
7 Jerry S. Kaplan, Startup (New York: Houghton Mifflin, 1995), 14.
8 I am indebted to Randall Eason, a student in my Stanford course, History 274 B (2000), for calling my attention to this anecdote in his term paper “Jerry Kaplan: The Phoenix of Silicon Valley.”
9 David Hounshell, “Bell and Grey: Contrasts in Style, Politics, and Etiquette,” Proceedings of the IEEE 64. September (1976), 1305-1314.
10 I am indebted to Matt Goldberg and Madhu Tadikonda, students in my Stanford course, History 274 B (2000), for calling my attention to this case history in their term paper: “Case History: Excite@Home.”
11 Theodore M. Edison, “Diversity Unlimited: The Creative Work of Thomas A. Edison,” a condensation of a paper given before the MIT Club of Northern New Jersey, January 24, l969, 2.
12 Robert Friedel and Paul Israel with Bernard S. Finn, Edison’s Electric Light: Biography of an Invention (New Brunswick: Rutgers University Press, 1986), 63-64.
13 Joseph Licklider, “Man-Computer Symbiosis.”
IRE Translations on Human Factors in Electronics,
HFE-1 (March 1960), 4-11.
14 For more on the early computer pioneers and the Internet, see Thomas P. Hughes, Rescuing Prometheus (New York: Pantheon, 1998), 255-300.
15 Interview with Frank Heart conducted by Judy O’Neill, 13 March 1990 and with David Walden, February 6, 1990. Archives (Charles Babbage Institute).
16 For more on Sperry’s early inventive activity, see Thomas P. Hughes, Elmer Sperry: Inventor and Engineer (Baltimore: The Johns Hopkins Press, 1971 and 1993), 16-19.
17 Thomas P. Hughes, American Genesis: A Century of Invention and Technological Enthusiasm, 18701970 (New York: Penguin Books, 1990), 89-90.
18 David A. Kaplan, The Silicon Boys and Their Valley of Dreams (New York: William Morrow and Company, 1999), 156 ff.
19 Thomas P. Hughes, Rescuing Prometheus (New York: Pantheon, 1998), 273-274.
20 Hal Lancaster, “Web Pioneer Succeeds by Knowing His Limits,” The Wall Street Journal Interactive Edition (October 13, 1999).