email icon Email this citation


ARPA Does Windows: The Defense Underpinning of the PC Revolution

Glenn R. Fong

Department of International Studies
Thunderbird, the American Graduate School of International Management

International Studies Association
41st Annual Convention
Los Angeles, CA
March 14-18, 2000

The PC industry is leading our nation’s economy into the 21st century . . . There isn’t an industry in America that is more creative, more alive and more competitive. And the amazing thing is all this happened without any government involvement.

— Bill Gates (1998) 1

The personal computer revolution, born out of risk-taking corporate ventures and garage-based innovative individualism, is the epitome of the heights than can be achieved by private sector, free-market entrepreneurialism.

Not quite.

Personal computer (PC) technologies that have revolutionized our everyday lives whether at the office or at home have been deeply rooted in public sector initiatives as well. As communities throughout the country, and countries around the world rush to clone their own Silicon Valleys, the governmental underpinnings of the original Valley’s success should not be overlooked.

This story parallels the widely-recognized government role in spurring a second revolution in information technology: the internet. The current-day internet traces its origins back, of course, to the late 1960s ARPANET project of the Defense Department. 2 However, when it comes our main window on cyberspace — the personal computer — a defense or government link to such a broadsweeping business and consumer appliance is almost inconceivable. Instead, when it comes to the origins of what makes a PC a PC — its graphical user interface, windows, the desktop metaphor and icons, and the mouse pointing device — the genealogy is usually traced back industrially from Apple and Microsoft and then back to the Xerox Palo Alto Research Center or Xerox PARC for short. This accepted history is embodied in the mainstream business literature, general media, and popular culture.

What is less well-known — and serves as the foci of this paper — is that Xerox PARC along with other pioneers of PC technology were associated with a significant government-sponsored thrust in desktop computing. The Air Force, Army, Navy, NASA, National Science Foundation, and most notably, the Defense Department’s Advanced Research Projects Agency (ARPA) 3 aggressively and persistently supported technologies key to the PC revolution.

Uncovering this political-economic link provides an important corrective to the popular lore surrounding the origins of the personal computer. In their emphases on private sector initiative and entrepreneurial risk-taking, conventional PC histories conform to orthodox market-based explanations of technological and economic progress. The role of government in spurring innovation and encouraging risk-taking is downplayed if not outright dismissed. In contradistinction, this paper "brings the state" into the PC realm of apparent market purity.

In making this case, we start mid-story with the Xerox-Apple-Microsoft connection. Reflecting a balance in political-economic analysis, this portion of the paper is business-centered as it is important to briefly establish what would come of earlier R&D efforts. The paper then jumps back to the pre-Xerox, pre-commercialization story where the government role takes center stage. Next we fast forward by briefly looking ahead to the government’s continuing influence on personal computing with the onset of the 21st Century.

The pervasive economic and social impact of the personal computer renders the story of its development as important in and of itself. But in the final section of the paper, we briefly consider the wider implications of this case both for the process of innovation and for the role of the government in that process.


Xerox could have been the IBM of the 90’s . . . could have been the Microsoft of the 90’s.

— Steve Jobs (1996) 4

Microsoft Windows, the Macintosh, the mouse, the desktop metaphor with icons, file directories, and folders — indeed the very notion of computing at the individual, personal level — can all in the first (but not last) instance be traced directly back to the Xerox PARC Alto computer. The first two aforementioned systems were introduced in 1985 and 1984, respectively. 5 The Alto was completed in 1973!

Before tracing this genealogy, it would be appropriate to briefly demarcate what we are tracing — our dependent variable, if you will.

What the layperson calls a ‘personal computer’ is, of course, an integration of a plethora of different technologies. A core subset of these technologies — and the core focus of this paper — is what computer scientists call ‘human-computer interface’ or HCI. HCI is concerned with enhancing the performance of joint tasks by humans and computers. To improve the structure of communication between human and machine, HCI brings together (1) the computer science and engineering fields of computer graphics, operating systems, programming languages, and software development; (2) behavioral science disciplines in communication theory, linguistics, learning theory, and cognitive psychology; and (3) graphic and industrial arts and design as well as ergonomics. Examples of HCI techniques include keyboard commands; pointing devices; touch screens and other display technologies; voice, handwriting and gesture recognition; eye tracking; biological and psychic sensing; computer speech; graphical user interfaces; user navigation and menu selection tools; windows environments; and desktop metaphors. 6

Ultimately, from the user perspective, HCI technologies result in the user-friendliness and "look and feel" — or lack thereof — of our PCs.

HCI technology provided the crucial linkage between two other developments in the 1970s that brought us the PC. 7 In a top-down development, the processing power of mainframe computers was slowly being brought to individual users through computer time-sharing. 8 The computer was still in the basement but scores of users could tap into its resources through remote terminals. While it represented a disservice to the mainframe’s prowess, simple computer games such as Spacewar offered a glimpse of real-time interactive computing. Time-sharing, however, could reach only relatively limited numbers of users.

A second, bottom-up development of the 1970s would bring individualized computers to users but encumbered by primitive features. Here we have the rise of rudimentary computing devices cobbled together by and offered to electronics hobbyists and enthusiasts. While computers such as the Altair 8800 were the size of a bread box, it could be argued that the food container was more functional than the machine. The major output of the Altair was getting the lights on its front panel to blink in a programmed sequence. And the interfaces of these machines were just as limited. To program the Altair (and those lights), users had to flick a series of toggle switches for each program step. Hardly a model of interactivity, these machines had neither displays nor keyboards.

The first major effort to develop a broadly functional individualized computer with HCI-inspired interactivity and user-friendliness took place at the Xerox Palo Alto Research Center. PARC was established in 1970 to provide the technological undergirding for Xerox — the king of paper photocopying — to move into the "paperless" world of office computing. In the process, PARC became the premier draw for the country’s best computer scientists — "like Disneyland for seven-year-olds." 9

PARC’s strategy centered on what it called "distributed interactive computing," and was embodied in the Alto office computer. "Distributed" because the Alto was all about getting the computer up from the basement and on to individual desktops. "Interactive" both in the sense that Altos were to be networked with one another, and in their design for real-time responsiveness and user-friendly approachability for individual users. 10

The Alto was intended for use by one individual with stand-alone processing power and memory (at least on the part of the machine). It was configured much like today’s PC — only this was 1973. It had a high-resolution monitor that could display a full-sized 8 1/2 by 11-inch page, a keyboard, a three-button mouse, a removable hard disk cartridge, and ports for printer and Ethernet connections. What today we would call the computer’s tower holding the guts of the system sitting under a desk was an Alto cabinet about the size of a portable refrigerator that can be found in today’s college dorm rooms (see Figure 1).

Figure 1 11


The Alto had a full slate of applications for word processing, graphics (including animation), printing, email, and playing music. The Alto operating system even allowed for task-switching — the capability to easily and quickly switch between programs.

The Alto’s monitor was a key feature of its user interface. Beyond its full-page dimensions, it had the power to display more than the standard-of-the-day fully formed text characters in a preset font. Instead it could display high-resolution, user-defined fonts and graphics. Using now standard bit mapping technology, the Alto could turn on and off half a million dots across its monitor — essentially turning everything on screen, including text, into pictures. Bit mapping also allowed the computer screen to display exactly what would be output from a printer — a feature that is known as "what you see is what you get" or WYSIWYG.

The Alto’s user friendliness is now almost second nature, but was revolutionary in 1973. 12 Xerox designers began with the assumption that computer users were more interested in getting their work done than being interested in the computer itself. Therefore an important Alto design principle was to make the computer as invisible and as intuitive as possible.

They chose a graphical user interface or GUI for personal computing. A graphically simulated office served as a working metaphor. Images on screen represented the physical objects of an office — documents, folders, file cabinets, in-baskets, out-baskets, waste baskets, mail boxes, printers — all on an electronic rendition of a desktop. These images or icons could be manipulated with a mouse pointer to simulate the physical actions of opening, moving, filing, saving, deleting, etc. The goal was to make everything needed visible on screen and subject to direct manipulation rather than requiring indirect and memory-taxing (not for the computer, but for humans) keystroke combinations. 13

The Xerox GUI is displayed in Figure 2. More than a decade before the Mac and Microsoft GUIs, the Alto had windows to display document contents; multiple windows could be open at the same time, they could be overlapped, and each could be resized; documents could integrate text and graphics; the windows had title bars, mouse-clickable command buttons, and scroll bars; plus you had movable desktop icons, varied fonts, and WYSIWYG rendition of the printed page.

Figure 2 14


Nearly two thousand Altos were built and in use in government, industry and universities. A commercial version of the Alto, renamed the Xerox Star, was introduced in 1981 — a full three and four years ahead of the Mac and Windows, respectively. The Star was marketed as "a new personal computer designed for offices . . . intended for business professionals who create, analyze and distribute information." 15

By current standards the Xerox interface did suffer from certain limitations. Commands such as open, copy and move required a combination of mouse manipulations and special function key operations. Resizing windows and moving icons also required mouse and function key combinations. The selection of text passages required multiple mouse clicks rather than the simpler click-and-drag movement. Menu bars were at the top of each window rather than a single set of menus at the top of the screen as a whole.

At the same time, and more significantly, the Alto suffered from being ahead of its time. While it was marketed as the dream machine for the "knowledge worker," such workers — in any actualized sense — did not exist in 1981 let alone 1973. 16 Plus this pathbreaking user-friendly machine had a very unfriendly price tag of over $16,000. For sixteen grand, you got all of 512 kilobytes of memory and 10 megabytes of disk storage. It was a commercial bust.

The commercial demise of the Alto and of Xerox PARC generally is legend in the business world. A popular recounting of this disaster was titled "Fumbling the Future: How Xerox Invented, Then Ignored, The First Personal Computer." 17

Alto’s Offspring

When Apple sued Microsoft in 1988 for stealing the "look and feel" of its Macintosh graphical display to use in Windows, Bill Gates’ defense was essentially that both companies had stolen it from Xerox.

— Dealers in Lightening (1999) 18

Xerox "fumbling its future" does not mean that its technologies were commercial failures. Indeed, many of the PARC and Alto technologies were spectacularly commercialized — but just not by Xerox. For instance, outside of the HCI area, notable PARC alumni have made market blockbusters out of their Xerox work:

When it comes to HCI, the Xerox legacy and progeny is even greater. In particular, the transfer of technology and, even more importantly, the transfer of people from PARC has been crucial to developments at both Apple and Microsoft. 19

Apple’s Day in the PARC. The flipside of Xerox’s fumbling the PC’s future is the Macintosh story. These two stories are, in fact, opposite sides of the same coin. And it begins when Steve Jobs, Apple’s CEO, takes a tour of Xerox PARC in December 1979.

In 1979, Apple was concerned it would soon lose its first mover advantage in the PC industry. Apple employee Jeff Raskin suggested that Xerox PARC held the keys for Apple’s future. In the early 1970s, Raskin had spent considerable time at PARC while he was a visiting scholar at Stanford’s Artificial Intelligence Laboratory. 20 After Apple arranged for Xerox to purchase $1 million dollars of Apple’s skyrocketing shares, PARC agreed to show Apple the Alto.

The Alto team made not one, but two presentations — and not just to Jobs, but to a dozen of Apple’s leading executives and programmers. Upon seeing the Alto, Apple software designer Bruce Daniels would declare, "That’s it — that’s what we want to build." 21 While no ‘blueprints’ were transferred, Apple came away from these sessions with a vision of the future of personal computing, and eventually key members of the PARC team.

The Xerox visit first inspired the development of the Lisa computer system — the Apple computer that immediately preceded the Macintosh. The Lisa was in development before the Xerox visit, but it was slated to have a non-graphical user interface and a non-bit mapped character-generator display. It also did not have a mouse. All this changed after the Xerox visit. In the words of Apple executive Larry Tesler, the Lisa was "completely redefined . . . only the code name, some of the hardware components, and a few of the staff members stayed the same." The Lisa would directly borrow the desktop metaphor, pop-up menus, overlapping windows, and scroll bars from the Alto. After the 1981 introduction of the Xerox Star, the Lisa team made further changes to their GUI including the incorporation of desktop icons. On Apple’s part, the Lisa would be the first to introduce the menu bar at the top to the screen (instead of menus atop each window), the one-button mouse, pull-down menus (point-and-drag mouse movement), and icons that could be dragged with the mouse and double-clicked to open. 22

Akin to the fate of the Alto, the Lisa was also a commercial failure when it was introduced in January 1983. But its graphical user interface was transferred directly into the Macintosh. Indeed, PARC-savvy Jeff Raskin had begun development of the Mac in Spring 1979. After the Xerox visit, Raskin added the mouse to the Mac. 23 Beginning in January 1982, key members of the Alto-inspired Lisa team were transferred to the Macintosh division. Lisa software programs for word processing and graphics (LisaWrite, LisaDraw) would be converted to the Mac (MacWrite, MacDraw). The two product teams were completely merged in November 1983, and the Mac was introduced January 1984. 24

Besides inspiration, the Xerox influence on Apple took on a second major form: the transfer of key PARC personnel to Apple. PARC alums Alan Kay and Larry Tesler were two of the major coups for Apple.

Alan Kay was PARC’s chief evangelist for personal computing. In his 1969 dissertation, Kay outlined a Dynabook — a computer the size of a notebook with an 8 by 10-inch flat screen, integrated keyboard, all of 2 inches thick, weighing in at two pounds. He had essentially envisioned today’s laptop computer.

For the Alto, which he viewed as an "interim Dynabook," Kay led the development of its overlapping windows capability. The Alto not only allowed users to work in and see more than one window at a time, but it was the first system that allowed windows to be resized and moved — including over one another. This overlapping capability was a major advance over the pre-existing standard of tiled multiple windows there were fixed in place, and virtually expanded the working space of a computer monitor. Kay also inspired the Alto’s pop-up menus — where the click of one of the mouse’s buttons would cause menu options to appear on screen from which a command (e.g., paste) could be selected. 25

In 1980, Kay became chief scientist at Atari where he applied his HCI visions to interactive gaming. In 1984 be became an Apple Fellow, and inspired the company’s successful PowerBook laptop computer line, and the Newton — the industry’s first personal digital assistant (PDA) and forerunner to the Palm Pilot and other handheld computing devices. Since 1996, Kay has been a Disney Fellow and Vice President of Research and Development at the Walt Disney Company.

Alan Kay was preceded by Larry Tesler in moving from Xerox to Apple. Tesler worked in Kay’s section of PARC where he was dedicated to making computing more intelligible to the average user. For the Alto, Tesler designed Gypsy, a powerful word processing program that employed a graphical user interface with extensive icons and menus. In Gypsy, the mouse could point to and select blocks of text whereas previous applications only used the mouse to position the cursor and called for keyboard commands for text selection. As an illustration of its user friendliness, Gypsy was the first program to replace on-screen commands for deleting a block of text and then placing it elsewhere with the simple labels of ‘cut’ and ‘paste.’ 26

In December 1979 Tesler was one of the two major presenters of the Alto to Steve Jobs and company. In July 1980 he would make the move to Apple. Tesler first headed up the Lisa user interface team, helped design the Macintosh including its one-button mouse, and then led the Newton PDA development team. He eventually rose to the position of Vice President and Chief Scientist before leaving Apple in 1998 to found a children’s software startup.

Kay and Tesler were not alone in making the move from Palo Alto to Cupertino, where Apple is headquartered. For instance, Dan Ingalls — Kay’s right hand man and co-author of the Alto’s operating system — would follow Kay to Apple. 27 Tom Malloy, who worked on word processing programs for the Alto, would go on to Apple and write the word processor for the Lisa (LisaWrite). 28 Former Xerox PARCers Bruce Horn and Steve Capps would co-write the Macintosh Finder, its graphical file directory. Altogether, some 15 PARC alums would make the move to Apple. 29

Microsoft’s Window on Xerox. While the Xerox-Apple story is better known, Microsoft was also a major beneficiary of PARC’s work. First, Microsoft Windows drew directly from the Alto-inspired Macintosh. Not unlike Jobs’ 1979 visit to Xerox, Microsoft’s Bill Gates visited Apple in 1981. There he saw a Mac prototype, and immediately thereafter began development of Microsoft’s GUI, Windows. In 1982, Mac prototypes were delivered to Microsoft in order for the software company to develop MS Word and Excel for the new machine. At the same time, the prototypes were used to guide the development of Windows. The deliberate intent behind this effort was conveyed by Gates’ disappointment at the first versions of Windows. Gates’ complaint: "That’s not what a Mac does. I want Mac on the PC, I want a Mac on the PC." 30

Akin to the movement of Xeroxers to Apple, members of the Mac team would move on to Microsoft. For instance, Susan Kare did graphic design work for Windows 3.0 after designing the first icons, typefaces, and other graphics for the Macintosh. 31

While the Mac served as a go-between for Xerox’s influence on Microsoft, there were direct Xerox-Microsoft connections as well. To begin with, Gates got his tour of PARC and an Alto demonstration in 1980. Soon thereafter, Microsoft purchased a Xerox Star, the commercial version of the Alto. Microsoft did not intend the machine for operational use. Instead, in the words of one of Microsoft’s leading programmers, "we just wanted everybody in the organization to get used to the desktop and to the mouse . . . we used it for education of the people." 32

That programmer was Charles Simonyi who embodies a yet another type of Xerox influence on Microsoft: PARC alums who moved from Palo Alto to Bellevue and Redmond, Washington, where Microsoft has been headquartered.

Not unlike Larry Tesler’s 1979 presentation to Steve Jobs and subsequent move to Apple, it was Simonyi who demonstrated the Alto to Gates in November 1980 and who, in February 1981, would join Microsoft as Employee #40. At PARC, Simonyi wrote the Alto’s "killer app" — Bravo, its first word processor. Both Simonyi’s Bravo and Tesler’s Gypsy were Alto word processors, but Bravo preceded Gypsy by a year in 1974. And while Bravo did not have Gypsy’s graphical user interface, it was the first program that could insert text in the middle of a document, display fancy typefaces, number pages, format odd margins, and print almost exactly what was on screen. 33

When Simonyi went to Microsoft, he essentially "brought Microsoft Word with him." 34 Gates has said he specifically got Simonyi "on board to help us write applications that would eventually become very graphical." 35 In Simonyi’s words, his mission was to spread the "PARC virus" in Bellevue. 36 As director of advanced product development, Simonyi hired and managed the teams developing the entire suite of Microsoft applications including Excel and PowerPoint as well as Word. When Microsoft’s Research Division was established in 1991, Simonyi became its Chief Architect.

Microsoft’s Research Division is, in fact, the site of Xerox’s continuing influence on the software company. At the Microsoft labs, Simonyi has been joined by four other of PARC’s leading lights: Chuck Thacker, Butler Lampson, Gary Starkweather, and Alvy Ray Smith. 37 Chuck Thacker, the lab’s Director of Advanced Systems, was none other than the chief designer of the Xerox Alto. He championed the Alto’s high-resolution bit-mapped display over the monochrome green monitors of the day, and designed the Star’s first central processor.

Butler Lampson, now a Microsoft Fellow, first conceived of and started work on Alto’s Bravo word processor — work that Simonyi would pick up on. Lampson also designed the second processor for the Star. Gary Starkweather developed the Alto’s laser printer and launched a whole new industry, desktop publishing. Alvy Ray Smith, another Microsoft Fellow, wrote the Alto’s graphics program. Before joining Microsoft, Smith would design for Lucasfilm, co-found Pixar, and win two technical Academy Awards. For Microsoft, these PARC alums have worked on advanced programming and graphics, hand-held and wireless computing devices, and computer security.

Xerox’s legacy extends, of course, well beyond Apple and Microsoft. Its current-day manifestations are innumerable, but let me mention three. Akin to Alan Kay’s move to from Xerox to Atari (before moving on the Apple), HCI advances have been a driving force behind the interactive gaming industry with applications ranging from game consoles and joy sticks to virtual reality environments. The World Wide Web, which began with text-based interfaces like Gopher, exploded in popularity only after user-friendly graphical user interfaces were employed by the Mosaic and Netscape web browsers. And members of the original Macintosh development team are about to give the open-source Linux operating system a major shot in the arm by applying a user-friendly GUI to the up-and-coming challenger to the Windows and Mac OS’s. 38

Even without a more comprehensive assessment Xerox’s legacy (a project worthy of an entire paper on its own), its import should not be in doubt. That import sets the proper perspective for considering the R&D that preceded and led into Xerox’s effort — a task to which we now turn.

The Rest of the Story

Silicon Valley. The World Wide Web. Wherever you look in the information age, Vannevar Bush was there first.

— Wired Magazine (November 1997) 39

The Alto system grew from a vision of the possibilities inherent in computing: that computers can be used as tools to help people think and communicate. This vision began with Licklider’s dream of man-computer symbiosis.

— Butler Lampson, Xerox PARC (1986) 40

Ivan Sutherland’s Sketchpad program is one of the most significant developments in human-computer communication.

— Transforming Computer Technology (1996) 41

While the commercial ramifications of Xerox PARC’s work cannot be over emphasized, Xerox was not the sole source of the HCI revolution. Just as Apple and Microsoft have benefited from Xerox, so too was Xerox the beneficiary of the prior work of others.

At this point, the political-economic balance of this account shifts. While the narrative thus far has been heavily business-oriented, what follows is more of a political dynamic. Most of the innovations and people discussed thus far were in fact influenced by government-sponsored initiatives. Those initiatives began with Vannevar Bush, J.C.R. Licklider, and Ivan Sutherland.

Vannevar Bush. The Online Encyclopedia Britannica entry for ‘graphical user interface’ reads: "There was no one inventor of the GUI; it evolved with the help of a series of innovators, each improving on a predecessor’s work. The first theorist was Vannevar Bush." 42

The source of this attribution was Bush’s vision of a "memex"

in which an individual stores all his books, records, and communications ... It consists of a desk ... On the top are slanting translucent screens, on which material can be projected for convenient reading. There is a keyboard, and sets of buttons and levers ... if the user inserted 5000 pages of material a day it would take him hundreds of years to fill the repository ... If the user wishes to consult a certain book, he taps its code on the keyboard, and the title page of the book promptly appears before him, projected onto one of his viewing positions ... [with] one of the levers to the right he runs through the book before him, each page in turn being projected at a speed which just allows a recognizing glance at each. If he deflects it further to the right, he steps through the book 10 pages at a time; still further at 100 pages at a time. Deflection to the left gives him the same control backwards ... he can leave one item in position while he calls up another.

Bush went on to consider the memex’s applications:

The lawyer has at his touch the associated opinions and decisions of his whole experience, and of the experience of friends and authorities. The patent attorney has on call the millions of issued patents, with familiar trails to every point of his client’s interest. The physician, puzzled by its patient’s reactions, strikes the trail established in studying an earlier similar case, and runs rapidly through analogous case histories, with side references to the classics for the pertinent anatomy and histology. The chemist, struggling with the synthesis of an organic compound, has all the chemical literature before him in his laboratory, with trails following the analogies of compounds, and side trails to their physical and chemical behavior.

This vision of the memex is widely recognized in government, industry, and academic circles as the first major articulation of the modern personal computer including hypertext and internet links. Xerox-Apple alum Alan Kay observes that "Bush’s vision of a hyperlinked 10,000 volume library in a desk had a great impact on the development of personal computing." 43 Tim Berners-Lee, inventor of the World Wide Web, notes that "to a large part we have MEMEXes on our desks today." 44

The memex was not the product of a science fiction writer conjuring up visions of the future; nor an entrepreneur toiling away on a garage work bench; nor an industrial researcher supported by a well-equipped corporate laboratory. Instead, Vannevar Bush was a government official. And not just any government official, but the Director of Office of Scientific Research and Development — the chief science advisor to the President. When Bush envisioned the memex, the President was Harry Truman; the date July 1945. 45

Between 1941 and 1947, Vannevar Bush served as science advisor to both Franklin Roosevelt and Truman. His greatest contribution in office is highly debatable in both the best and worst of senses. First, he organized the 6000-strong scientific enterprise to help prosecute the U.S. war effort. While he was not physically in the sands of New Mexico, Bush oversaw the Manhattan Project to create the first atomic bomb. Second, he established the structure of the country’s postwar science and technology effort — including the prominent roles played by military R&D, the National Science Foundation, and university-based research.

Then there is the memex. Bush’s vision inspired R&D efforts throughout government, industry, and academia. The lead player in this effort was the Advanced Research Projects Agency (ARPA).

JCR Licklider. Since its inception in 1958, ARPA has supported both the development of military-specific weapons technologies as well as more generic technologies with the potential for military application. The former have included ballistic missile defense and tactical anti-tank weapons technologies, and even the M-16 rifle. The latter include R&D in new materials, novel energy sources, and biomedical technologies as well as computer science.

ARPA began its computer science work in 1962 when it established its Information Processing Techniques Office (IPTO) as one of a half dozen technology-specific offices within the agency. 46 Starting off with a $7 million annual budget, IPTO’s funding was larger than the computer research budgets of the rest of the government combined. Over the next eight years, the IPTO budget would more than quadruple.

Most of IPTO’s funding went to university research. It is hard to imagine now, but before 1962 no formal university computer science programs existed. Not until ARPA came along. IPTO grants were crucial in establishing the country’s first graduate programs in computer science including those at MIT, Stanford, Berkeley, Utah, and Carnegie Mellon.

More will be said of these and other ARPA-funded programs, but first we turn attention to the ARPA official that served as the guiding light behind this effort. That official was J.C.R. Licklider. As quoted above by Xerox PARCer Butler Lampson, the Alto would grow out of Licklider’s vision.

J.C.R. Licklider was IPTO’s inaugural director from 1962 to 1964. Earlier, as an MIT professor, Licklider "got fired up about the idea Vannevar Bush had mentioned in 1945, the concept of a new kind of library to fit the world’s new knowledge system." Licklider’s 1959 book, Libraries of the Future, was not only dedicated to Bush but expanded upon the memex concept. When he moved on to ARPA, he brought with him his "religious conversion" to interactive computing. 47

From ARPA, Licklider galvanized the computing research community around two pathbreaking concepts. Given the first one — "the intergalactic network" — it is almost understandable to overlook the second. The intergalactic network was "the first concrete proposal for establishing a geographically distributed network of computers." 48 As initiated by Licklider, the network would first take the form of computer time-sharing links and later transform into the ARPANET/internet.

As consequential as this first concept has been, the second — "man-computer symbiosis" — is arguably just as profound. Licklider came to computing not as a computer scientist, but as an academic psychologist. His interest was in how computers could contribute to, rather than replace, human cognitive processes. He was concerned that the rudimentary user interfaces of computers of the 1950s hindered the technology’s true potential. To realize that potential, he called for computing advances in real-time processing and interactivity. He called for advances in the computer’s outward face to its user — its display — and in how users input instructions into the computer including via graphical input and automatic speech recognition. In calling for a "much tighter coupling between man and machine," Licklider sought to realize "interaction with a computer in the same way that you think with a colleague whose competence supplements you own." 49

These are all matters of human-computer interface, and Licklider defined the HCI agenda for decades to come. ARPA-supported research centers not only took part in building Licklider’s "intergalactic network," but they launched major HCI initiatives as well.

Ivan Sutherland. When Licklider prepared to leave ARPA in 1964, he selected Ivan Sutherland to replace him as IPTO director. Sutherland was one of the first researchers to take up Licklider’s HCI challenge. His 1962 Ph.D. project at MIT, called Sketchpad, was the first-ever computer graphics program where the user could make drawings on screen interactively.

Sketchpad is widely recognized as the seminal program that started off the entire field of computer graphics. 50 But Sutherland’s immediate motivation was to advance human-computer interactivity. Indeed, the subtitle of his project was "A Man-Machine Graphical Communication System." 51 Three features made Sketchpad, as quoted above, "one of the most significant developments in human-computer communication."

First, Sketchpad was one of the first computers with a monitor, and a user’s work would immediately be represented on screen. This form of interactivity is now easy to take for granted, but before Sketchpad users would have to wait for a print out in order to see their work. 52

Second, Sketchpad was one of the first computers to use a pointing device. A hand-held "light pen" was employed to make drawings. The pen would make physical contact with the screen and its "light" would be picked up by the computer. Moving the pen would draw lines on screen in real-time. The pen could also be used to grab-and-drag images as well as rotate, expand or contract an image. A major user interface breakthrough, before Sketchpad users had to express object geometry by typing coordinates on a keyboard. The light pen would later lead to today’s mouse.

Third, Sketchpad was the first system with a rudimentary windowing system. The Sketchpad screen could be split to produce two work areas or windows.

The Sketchpad project was sponsored by the Army, Navy, and Air Force. This funding is a reminder that government agencies other than ARPA have also supported HCI technology. In this particular case, the three military services provided support to Sutherland before IPTO was even established.

Licklider hired Sutherland to explicitly carry on IPTO’s HCI work. As IPTO director, Sutherland would fund major university computer graphics programs. Besides fueling the burgeoning field of computer generated images, this research would provide the foundation for computers with "graphical" user interfaces, "picture" icons, and high-resolution bit-mapped displays. Such displays, interfaces, and icons along with Sketchpad-derived windows and pointing devices would be incorporated into the Xerox Alto.

Xerox’s ARPA Brats

Xerox PARC was set up near the Stanford campus. For the next ten years the ARPA dream took up residence at PARC.

— West of Eden (1989) 53

A veritable "ARPA Army" — a term coined at PARC — would fill the ranks of computer scientists at the Xerox. This influx into Xerox was led not by a researcher from an ARPA-supported university, but by an official direct from ARPA itself: Robert Taylor.

Robert Taylor. J.C.R. Licklider not only selected Ivan Sutherland to replace him as director of IPTO, but chose Robert Taylor to be associate director. When Sutherland finished his stint as director in 1966, Taylor took his place.

Robert Taylor "heartily subscribed" to Licklider’s vision of computing even before joining ARPA. 54 In his first year in office, he advanced the latter’s "intergalactic network" transforming it from a computing time-sharing paradigm to a decentralized packet-switching network, the ARPANET. While ARPANET’s construction would begin under Taylor’s successor at IPTO, Lawrence Roberts, the network’s design would be completed under Taylor.

Taylor was also a true believer in Licklider’s theme of "man-computer symbiosis." Taylor held a NASA research post in HCI just prior to joining ARPA. Distributed interactive computing became a "sacred cause" of Taylor’s as director of IPTO. 55 As described in a 1968 paper, co-authored with Licklider, he envisioned a computer for each individual user; each with a large graphic display, a keyboard, and "electronic pointer controllers called ‘mice’ [that could] control the movements of a tracking pointer on the TV screen." 56 This vision grows directly out of the memex of Vannevar Bush. It also presages Xerox PARC’s Alto.

When Xerox started forming its PARC facility in 1970, one of the first people they tapped was Robert Taylor. As quoted above, "for the next ten years the ARPA dream took up residence at PARC."

Taylor has been called "the impresario of computer science at Xerox PARC." 57 Forbes has described Taylor as playing "the Robert Oppenheimer role" at PARC — making a parallel to the infamous director of the Manhattan project. 58 Taylor exercised this influence as head of the Computer Science Laboratory (CSL) — the largest of PARC’s four internal labs. It was CSL that would become the mecca for 50 of the country’s top computer scientists.

In Spring 1971, Taylor set CSL’s agenda by proposing that it build the machine he had written about in 1968. Two years later, the Alto realized his vision. While his researchers would undertake the Alto’s design and development, the general concept and the "Alto" name came from Taylor. 59

Besides setting the lab’s agenda, Taylor hired its staff. He did so not by merely reading their resumes. Instead, he chose his people from the ARPA-funded research centers. Indeed, he choose researchers whom he along with Licklider and Sutherland had directly and personally supported through IPTO.

ARPA’s Army. Stanford, Berkeley, Utah, and SRI were the major programs that Taylor drew from. Most of these researchers — and their exploits at Xerox, Apple, and/or Microsoft — have already been noted in the first half of this paper. Here we reveal their university and ARPA pedigrees. To help keep the names and affiliations straight, Figure 3 graphically displays some of these people and places.

Figure 3: From ARPA to Windows

Stanford’s Artificial Intelligence Laboratory was established in 1962 with ARPA funding. Indeed into the 1970s, most if not all of the computing research conducted at Stanford would be supported by ARPA — as would be the case at Berkeley, Carnegie Mellon, Illinois, MIT, UCLA, and Utah. 60 Out of Stanford, Taylor hired Larry Tesler and Charles Simonyi, who would later go on to Apple and Microsoft fame, respectively.

In 1963, IPTO began supporting Project Genie at Berkeley, a small-scale computer time-sharing project. Charles Thacker and Butler Lampson as well as Simonyi from Stanford would first come together to work on this project. 61 While burdened by the mainframe paradigm, Project Genie sparked their pursuit of interactive computing. The three were considered among the country’s top programmers, and Taylor hired them as a group to join PARC in 1970. Taylor would hire others from Berkeley including Peter Deutsch, Ed Fiala, Jim Mitchell, Dick Shoup, Alvy Ray Smith, and Gary Starkweather. Thacker, Lampson, Starkweather, and Smith along with Simonyi would all end up at Microsoft.

One of the Berkeley faculty members that directed Project Genie, David Evans, would not go to Xerox. Instead he remained in academia training students, many of whom would make the trek to PARC. This is the Utah connection, where Evans became head of the computer science department in 1966. As IPTO director, Taylor would make a $5 million award to Evans to transform Utah into a center of excellence for computer graphics. 62 Ivan Sutherland, Taylor’s predecessor and creator of the Sketchpad program, would be on the Utah faculty from 1968 to 1973. Taylor himself would spend a year at Utah between his ARPA and PARC tenures.

Taylor would bring to CSL many Utah students including Jim Curry, Bob Flegal, Martin Newell, and John Warnock. But the key hire for the Alto and HCI at Xerox was Alan Kay in 1972. Kay came to Utah in 1966 as one of Evans’ first graduate students. At their very first meeting, Evans assigned the new student Sutherland’s Sketchpad dissertation. In a reaction any professor would die for, Kay has described his reading of Sketchpad as "seeing a glimpse of heaven." 63 Kay would try to capture a bit of that heaven first in his own dissertation, then at PARC, and later at Apple.

One major ARPA-supported research center that has yet to be mentioned, and that has made major contributions to the PC industry is the think tank Stanford Research Institute (SRI). SRI was the home of computer scientist Douglas Engelbart from 1957 to 1975. Engelbart was inspired by Licklider’s notion of augmenting (rather than replacing) human intellect via "man-computer symbiosis." Indeed Engelbart’s lab at SRI was called the Augmentation Research Center. 64

In designing a system to augment human intelligence, Engelbart used Vannevar Bush’s memex concept as an ideal. 65 Over a two decade period, Engelbart would develop a computerized personal information storage and retrieval system to replace paper and hardcopy filing systems. Called NLS (for oN Line System), the system was not a personal computer, but rather a networked workstation. It had a large video monitor and input devices to manipulate information on screen, but it was all cabled into a remote mainframe computer.

Still, NLS introduced two major innovations that advanced the "man-computer symbiosis" and HCI. First, it introduced a dynamic windowing environment. While the earlier Sketchpad system of Ivan Sutherland could display two different work areas in split-screen mode, NLS could "cut" out "boxes" on screen to display document contents. 66 And multiple document "boxes" could be displayed at the same time.

Second, NLS introduced a new pointing device to move a cursor within and between document windows. Engelbart conducted a series of studies comparing various pointing devices including Sketchpad’s light pen, track balls, joysticks, even a knee-switch under the desktop. What he decided upon was a device that "stays put when your hand leaves it do something else (type or move a paper) and reaccessing proves quick and free from fumbling. . . and it doesn’t require a special and hard-to-move work surface." 67 This device is, of course, the mouse (see Figure 4).

Figure 4


The size of a brick and carved out of a block of wood, the underside of Engelbart’s mouse housed two wheels positioned at right angles to one another that could digitally track and convey its position to the computer. While the wheels would be replaced with a ball, the computer mouse was not invented by Xerox in 1973 let alone Apple in 1984. It was created by Engelbart in 1964.

The system described earlier in Robert Taylor’s 1968 paper — a large video screen, keyboard, and a mouse — was Engelbart’s NLS. Not only did Taylor properly cite Engelbart in that paper, but Engelbart had three major connections to Taylor and ARPA.

To begin with, Taylor — while at NASA — provided initial funding for Engelbart’s project. The Air Force did as well. Both NASA and the Air Force were interested in how operators in their command centers could best interface with their computers. 68 As in the case of Sutherland’s Sketchpad project, Engelbart received support from these other organizations before IPTO was even established.

Then, with IPTO’s establishment in 1962, "Douglas Engelbart was one of the first persons to apply for funding." 69 Not only did he gain IPTO funding, the support would significantly rise during Taylor’s tenure. ARPA funding would continue until 1975, and Engelbart’s research team would expand from 2 to nearly 50. In 1968, ARPA and NASA co-sponsored a major presentation of the NLS to the public that amazed the wider computing research community.

Then there is the Xerox connection. In the words of Butler Lampson, the NLS "made a profound impression on many of the people who later developed the Alto." 70 Both the mouse and windows were directly incorporated from NLS into the Xerox computer.

Moreover, Taylor hired key members of the NLS team to come to PARC. Akin to David Evans remaining at Utah, Engelbart would not himself make the move to Xerox. But Taylor did hire Engelbart’s right hand man, Bill English. English was NLS’s hardware expert and had done the detailed design on the mouse. Taylor offered English that chance to "reproduce NLS, or something like it, at PARC." 71 That "something else" turned out to be the Alto.

Another member of the NLS team, Roger Bates, would develop the Alto’s high-resolution bit-mapped display. NLS alum Charles Irby would help design the user interface for the Xerox Star. Altogether, a dozen of Engelbart’s team would make the move to PARC. 72

Given these hires from SRI and the universities, ARPA-supported research would leave an "indelible stamp on almost every major innovation to emerge from PARC." 73

In surveying ARPA’s influence on Xerox, it is difficult not to mention other major computer scientists that have been supported by IPTO — including Wesley Clark, Lynn Conway, Michael Dertouzos, Edward Feigenbaum, John Hennessy, Daniel Hillis, John McCarthy, Carver Mead, Marvin Minsky, David Patterson, and Raj Reddy. Then there are those that have left their mark in the commercial world. We have already mentioned Bob Metcalfe of 3COM, John Warnock of Adobe Systems, and Edwin Catmull of Lucasfilm and Pixar — all of whom came out of PARC. We can now note their earlier ARPA-backing at Harvard and Utah (last three). To this list we can add Jim Clark (Utah) of Silicon Graphics and Netscape, and Bill Joy (Berkeley) of Sun Microsystems. 74

But our focus here has been on HCI-specific ARPA-supported researchers who made their way to Xerox PARC and then contributed to or influenced developments at Apple or Microsoft. Even with these restrictors, the ARPA reach is substantial. "ARPA does Windows" is more than a catch phrase.

Windows on the Future

The analysis has now come full circle. Vannevar Bush’s extraordinary vision is followed up by ARPA’s Licklider, Sutherland, and Taylor. They sponsor the Stanfords, Berkeleys, Utahs and SRIs. Xerox draws upon this research and the researchers (plus Taylor). Then Apple and Microsoft commercialize Xerox’s work. The rest, as they say, is history.

But the PC revolution does not stop with Windows (thank goodness). And ARPA’s hand in matters HCI is not confined to just the 1960s and 1970s. Indeed, ARPA’s direct support for further advances in personal computing extend into the 1990s. Before coming to our conclusion, we briefly survey ARPA’s latest initiatives.

[Note: under construction]


Government funding of advanced human-computer interaction technologies built the intellectual capital and trained the research teams for pioneer systems that, over a period of 25 years, revolutionized how people interact with computers.

— Stuart Card, Xerox PARC (1996) 75

This paper has not considered (nor is this the place to do so) the continuing economic and social impact of the personal computer. That pervasive impact renders the story of the PC’s development as important in and of itself. However in this final section, we briefly consider the wider implications of this case both for the process of innovation and for the role of the government in that process.

[Note: under construction]

In contrast to the thrust of this argument are sentiments such as that quoted at the top of this paper. Bill Gates is not alone in holding this view. His is the mainstream perspective on the development of the PC industry; indeed of the development of virtually the entire "new economy."

Case in point is Tim Draper, founder and managing partner of Draper Fisher Jurvetson. DFJ is "perhaps the top venture firm for Silicon Valley startups." 76 Mr. Draper personally provided start-up capital for Hotmail (world’s largest email provider), Four11 (internet white pages directory), and Upside (most widely read business technology magazine).

In 1997, Draper penned a Wall Street Journal editorial that articulated much of Silicon Valley’s attitude towards the government — an attitude legitimated by the premier U.S. business daily. Draper starts by telling us he "earned an MBA from Harvard and an electrical engineering degree from Stanford. I worked at Hewlett-Packard and Alex. Brown before starting a venture capital firm. My favorite periodicals are Upside and the Red Herring, not the Washington Post or the Weekly Standard. In my free time I surf the Net; I don’t watch Capital Gang or C-SPAN." Writing under the title, "Silicon Valley to Washington — Ignore us, Please," Draper then make his case:

We in the high tech business have reason to feel good . . . Our industry now accounts for 11 percent of gross domestic product and a quarter of U.S. manufacturing output. We employ more than 4.2 million people, who earn almost double the average salary of manufacturing workers. Our industry is the biggest reason the U.S. has the world’s most competitive economy . . .

We ought to count our blessings that most of our industry is 2,500 miles from Washington and that most bureaucrats either fear, don’t care about or don’t understand technology. And we’ve done just fine without their help . . . Washington doesn’t understand my business, [and] I’d like it to stay that way. The fact is that politicians and government bureaucrats can’t help us; they can only get in the way . . . If the U.S. wants more good jobs, better lives, and a stronger economy, the best thing lobbyists, bureaucrats and politicians can do is leave us alone. 77

Even putting aside the current-day prescription, "we’ve done just fine without their help"? "They can only get in the way"?

These views permeate coverage in Fortune and Business Week and the general media. Even the highly-regarded six-hour PBS documentary on the history of the PC industry, Triumph of the Nerds, overlooks the government connection. 78

In contrast we have observations like that by Stuart Card at the top of this section. Card might be in the position to know. He has been with Xerox PARC for twenty-five years, and currently heads its User Interface Research Group. The quote is from a fifty-page technical paper he compiled on the historical development of HCI.

Card too is not alone. Dr. Brad Myers, Senior Research Scientist at Carnegie Mellon’s Human Computer Interaction Institute, warns against "the mistaken impression that much of the important work in Human-Computer Interaction occurred in industry." 79 Instead, as computer historians Martin Campbell-Kelly and William Aspray have written, "almost all the ideas in the modern computer interface emanated from laboratories funded by ARPA’s Information Processing Techniques Office. 80

Uncovering this political-economic link provides an important corrective to the popular lore surrounding the origins of the personal computer. This paper "brings the state" back into the PC realm of apparent market purity.

Government support for the development of the PC should take its place on a list that includes the internet, the computer chip, and the PC’s bigger brother, the mainframe. 81

The federal government’s role in supporting the development of the internet is now widely acknowledged. The ARPANET of 1969 was followed by the NSFNET of 1985 before being privatized in 1995. This support extends to the government’s on-going Next Generation Internet project.

The government’s support of the chip industry goes back to military R&D funding in the 1940s and procurements into the 1960s by the Air Force and NASA of 100% of the industry’s production. Government support of the chip industry would continue into the 1980s and 1990s with the Very High Speed Integrated Circuit Program and Sematech consortium.

And, of course, the Defense and Energy Department’s support of the mainframe and supercomputer industry stretches from the ENIAC of 1945, IBM’s 1953 Stretch computer, the SAGE computer in 1954, Cray’s first supercomputer in 1976, the 1996 Intel teraflop machine, and even IBM’s 1997 chess champion Deep Blue. This kind of support continues today with government programs such as the High Performance Computing and Communication Initiative and the Accelerated Strategic Computing Initiative.

The internet, the computer chip, the mainframe, and the PC. Together these four innovations define the information technology revolution that has fueled the new economy of 21st Century. No doubt university researchers and private entrepreneurs have made this revolution possible. But popular mythology, corporate P.R., and political ideology aside, thanks also go to the government.


Note 1: Microsoft News Release, "Remarks by Bill Gates," May 18, 1998 — the day the Justice Department launched its anti-trust suit against the company. Back.

Note 2: The PC and networking stories do in fact intersect at important junctures in terms of personalities, private initiatives, and public policy. While certain references will be make to computer networking, this account focuses on advances in stand-alone computing. For the ARPANET story, see Sidney G. Reed, et al., DARPA Technical Accomplishments, Volume 1: An Historical Review of Selected DARPA Projects (Alexandria, VA: Institute for Defense Analysis, February 1990), chapter 20; Arthur L. Norberg and Judy E. O’Neill, Transforming Computer Technology: Information Processing for the Pentagon, 1962-1986 (Baltimore, MD: Johns Hopkins University Press, 1996), chapter 4; Vinton Cerf, "How the Internet Came to Be" in The Online User’s Encyclopedia, Bernard Aboba (Addison-Wesley, November 1993); Barry M. Leiner, et al., "A Brief History of the Internet" (1997), website: <>; and Jeffrey Hart, et al., "The Building of the Internet: Implications for the Future of Broadband Networks," Telecommunications Policy, v.16, n.8 (November 1992); and Katie Hafner and Matthew Lyon, Where Wizards Stay Up Late: The Origins of the Internet (NY: Simon & Schuster, 1996) Back.

Note 3: The agency was founded in 1958 as ARPA, changed to DARPA ("Defense" added) in 1972, reverted back to ARPA in 1993, and then back to DARPA in 1995. The ARPA acronym used in this paper reflects the earlier time period largely covered in this paper. Back.

Note 4: "Triumph of the Nerds" (PBS video, 1996), volume 3.


Note 5: Windows 1.0 was introduced in 1985, but would not qualify as a fully functional graphical user interface. While version 1.0 and even version 2.0 had windows containing document contents, and while different programs could be open at the same time, the windows could not be overlapped (only tiled) and neither utilized graphical icons. Only with Windows 3.0 in 1990 would Microsoft offer a functional GUI. "Bill Gates Interview," National Museum of American History, Smithsonian Institution, website: <>. Back.

Note 6: Association for Computing Machinery, Special Interest Group on Computer-Human Interaction, Curricula for Human-Computer Interaction (New York, NY: ACM, 1992). Website: <>. Back.

Note 7: These two other developments are covered by Paul E. Ceruzzi, A Modern History of Computing: 1945-1995 (Cambridge: MIT Press, 1998), ch. 7. Back.

Note 8: Computer time-sharing, like the development of internet and HCI technologies, was initiated by government program, specifically by ARPA. Back.

Note 9: Michael Hiltzik, Dealers of Lightening: Xerox PARC and the Dawn of the Computer Age (New York: Harper Business, 1999), p. 153 Back.

Note 10: Consistent with the HCI focus of this paper, it does not elaborate on the networking aspects of the Alto. Back.

Note 11: Source: Lawrence H. Miller and Jeff Johnson, "The Xerox Star: An Influential User Interface Design" in Human-Computer Interface Design: Success Stories, Emerging Methods, and Real-World Context, Marianne Rudisill, et al, eds. (San Francisco: Morgan Kaufmann Publishers, 1996), p. 75. The Alto can be seen in action in the PBS video "Triumph of the Nerds" (1996), volume 3. Back.

Note 12: David Canfield Smith, et al., "Designing the Star User Interface," Byte, v.7, n.4 (1982), reprinted in Ronald M. Baecker and William A.S. Buxton, Readings in Human-Computer Interaction: A Multidisciplinary Approach (Los Altos, CA: Morgan Kaufmann, 1987), p 653-661; Jeff Johnson, et al., "The Xerox Star: A Retrospective," IEEE Computer, v.22, n.9 (September 1989), reprinted in Readings in Human-Computer Interaction: Toward the Year 2000, Second Edition, Ronald M. Baecker, et al. (San Francisco: Morgan Kaufmann Publishers, Inc., 1995), pp. 53-70; and Miller and Johnson, "The Xerox Star, " pp. 70-100. Back.

Note 13: Interestingly, analysts have pointed out that Xerox pushed the physical desktop metaphor too far — requiring cumbersome mouse manipulations where simple keyboard commands would have been sufficient (i.e., requiring that a document icon be moved over a printer icon instead of a simple key command for printing). Miller and Johnson, "The Xerox Star," p. 93. Back.

Note 14: Figure 2 actually displays a 1985 Xerox interface called ViewPoint and not the original 1973 Alto nor 1981 Star GUIs. In a temporary regression, multiple windows could only be tiled by the Star while the original Alto could overlap them. Overlapping would return with ViewPoint. Source: Jeff Johnson, et al., "The Xerox Star: A Retrospective," IEEE Computer, v.22, n9 (September 1989), reprinted in Readings in Human-Computer Interaction: Toward the Year 2000, Second Edition, Ronald M. Baecker, et al. (San Francisco: Morgan Kaufmann Publishers, Inc., 1995), p55 Back.

Note 15: Smith, et al., "Designing the Star User Interface," p. 653. Back.

Note 16: Ronald M. Baecker and William A.S. Buxton, "The Star, the Lisa, and the Macintosh" in Readings in Human-Computer Interaction: A Multidisciplinary Approach, Ronald M. Baecker and William A.S. Buxton (Los Altos, CA: Morgan Kaufmann, 1987), pp. 649-652. Even the Xerox salesforce had difficulty "getting it"— upon the conclusion of an Alto demonstration one brave soul asked, "Where’s the click?" Hiltzik, Dealers of Lightening, p. 393. Back.

Note 17: Douglas K. Smith and Robert C. Alexander, Fumbling the Future: How Xerox Invented, Then Ignored, The First Personal Computer (New York : W. Morrow, 1988). For PARC’s commercial fate, see also Hiltzik, Dealers of Lightening. Back.

Note 18: Hiltzik, Dealers of Lightening, p. xxv. Bill Gates has remarked: "Hey, Steve, just because you broke into Xerox’s house before I did and took the TV doesn’t mean I can’t go in later and take the stereo." MacWeek, March 14, 1989, p. 1. Back.

Note 19: Other Alto-inspired GUI efforts not covered in this paper include those by Digital Research, IBM, and VisiCalc — efforts that did not match the success of the Mac or Windows. Back.

Note 20: Jeff Raskin correspondence, website:; Owen W. Linzmayer, Apple Confidential: The Real Story of Apple Computer (San Francisco, CA: No Starch Press, 1999), p. 52. Back.

Note 21: "The Birth of the Lisa," Personal Computing, February 1983, p. 89. Back.

Note 22: Linzmayer, Apple Confidential, pp. 54-56; Miller and Johnson, "The Xerox Star," p. 94; Larry Tesler, "The Legacy of the Lisa," Macworld (September 1985); "The Birth of the Lisa," pp. 88-94. Views of the Lisa and early Mac interfaces can be seen at *. Back.

Note 23: Ceruzzi, A Modern History of Computing, p. 273. Back.

Note 24: Tesler, "The Legacy of the Lisa;" Linzmayer, Apple Confidential, pp. 57-75. Back.

Note 25: Hiltzik, Dealers of Lightening, pp. 224-28. Back.

Note 26: Hiltzik, Dealers of Lightening, pp. 201-03, 207-10. Back.

Note 27: Hiltzik, Dealers of Lightening, pp. 214-15, 217-18. 316-17. Back.

Note 28: Miller and Johnson, "The Xerox Star," p. 76. Back.

Note 29: Linzmayer, Apple Confidential, p. 54. Back.

Note 30: Martin Campbell-Kelly and William Aspray, Computer: A History of the Information Machine (NY: Basic Books, 1996); and Linzmayer, Apple Confidential, p. 136. For views of the early Windows interface, see Phil Lemmons, "Microsoft Windows: A mouse with modest requirements," Byte, December 1983, website: <>. Compare to the Xerox (Figure 2) and Apple interfaces (footnote 22). Back.

Note 31: "The New Face of Open Source OS?" MacWeek, February 21, 2000; Linzmayer, Apple Confidential, p. 73. Back.

Note 32: "Intentional Programming: A Talk With Charles Simonyi," Edge Foundation, website: <> Back.

Note 33: Hiltzik, Dealers of Lightening, pp. 198-200, 358-60. Back.

Note 34: Hiltzik, Dealers of Lightening, p. 395; also Miller and Johnson, "The Xerox Star," p. 76. Back.

Note 35: Bill Gates interview with Smithsonian Institution (1993); website: <>. Back.

Note 36: "Intentional Programming." Back.

Note 37: Hiltzik, Dealers of Lightening, p. 397-398. Back.

Note 38: "Apple, AOL veterans making Linux easy," CNET, February 16, 2000; "Old Apple Macintosh Team Aims to Put Linux on the Desktop," New York Times, February 21, 2000; "The New Face of Open Source OS?" MacWeek, February 21, 2000; "A Less Complex Linux," San Francisco Chronicle, February 21, 2000. Back.

Note 39: G. Pascal Zachary, "The Godfather," Wired, November 1997, p. 152. Back.

Note 40: Butler W. Lampson, "Distributed Computing: The Alto and Ethernet Software," in A History of Personal Workstations, Adele Goldberg, ed. (Addison-Wesley, 1988). Back.

Note 41: Norberg and O’Neill, Transforming Computer Technology, p. 36. Back.

Note 42: Encyclopedia Britannica, website: To help redeem myself from the use of such a source, the encyclopedia entry refers to only two of the half dozen individuals that are the focus of this section of the paper. [Even in grade school, I never used the encyclopedia to write papers.] Back.

Note 43: Alan Kay, "Simex: The Neglected Part of Bush’s Vision" presented at "As We May Think" — A Celebration of Vannevar Bush’s 1945 Vision, MIT Department of Electrical Engineering & Computer Science, October 12-13, 1995, website: <>. Back.

Note 44: Tim Berners-Lee in "Hypertext and Our Collective Destiny" presented at "As We May Think" — A Celebration of Vannevar Bush’s 1945 Vision, MIT Department of Electrical Engineering & Computer Science, October 12-13, 1995, website: <>. Back.

Note 45: Vannevar Bush, "As We May Think," Atlantic Monthly; July, 1945. Back.

Note 46: IPTO has undergone a number of name changes over the past 40 years, and is currently named Informationa Technology Office (ITO).


Note 47: Howard Rheingold, Tools For Thought: The People and Ideas of the Next Computer Revolution (NY: Simon & Schuster, 1985), Chapter 7, website: <>. Back.

Note 48: Campbell-Kelly and Aspray, Computer, p. 288 Back.

Note 49: J.C.R. Licklider, "Man-Computer Symbiosis," IRE Transactions on Human Factors in Electronics, v.1 (March 1960), pp. 4-11. Back.

Note 50: Roaslee Wolfe, Seminal Graphics: Pioneering Efforts that Shaped the Field ( Association for Computing Machinery, 1998). Back.

Note 51: Ivan E. Sutherland, "Sketchpad: A Man-Machine Graphical Communication System," Proceedings of the AFIPS Spring Joint Computer Conference (Washington, DC: Spartan Books, 1963), pp. 329-346. See also Norberg and O’Neill, Transforming Computer Technology, pp. 125-128. Back.

Note 52: Wolfe, Seminal Graphics. Back.

Note 53: Frank Rose, West of Eden: The End of Innocence at Apple Computer (New York: Viking Penguin, 1989), p. 45. Back.

Note 54: Arthur L. Norberg and Judy E. O’Neill, Transforming Computer Technology: Information Processing for the Pentagon, 1962-1986 (Baltimore, MD: Johns Hopkins University Press, 1996), p. 29. Back.

Note 55: Hiltzik, Dealers of Lightening, p. 19. Back.

Note 56: J.C.R. Licklider and Robert Taylor, "The Computer as a Communication Device," Science and Technology, April 1968. Back.

Note 57: Hiltzik, Dealers of Lightening, p. 3. Back.

Note 58: Forbes ASAP, October 7, 1996. Back.

Note 59: Smith and Alexander, Fumbling the Future; Hiltzik, Dealers of Lightening, p. 170. Back.

Note 60: Norberg and O’Neill, Transforming Computer Technology, p. 290.


Note 61: Norberg and O’Neill, Transforming Computer Technology, pp. 102-103; Hiltzik, Dealers of Lightening, pp. 18-19, 68-78. Back.

Note 62: Norberg and O’Neill, Transforming Computer Technology, pp. 137-143. Back.

Note 63: Hiltzik, Dealers of Lightening, p. 91. Back.

Note 64: Bootstrap Institute, "Biographical Sketch: Douglas Carl Engelbart," website: <>; Hiltzik, Dealers of Lightening, p. 63. Back.

Note 65: Howard Rheingold, Tools For Though, chapter 9; Ceruzzi, A Modern History of Computing, p.260. Back.

Note 66: Norberg and O’Neill, Transforming Computer Technology, p. 131. Back.

Note 67: Steven Levy, Insanely Great: The Life and Times of Macintosh (New York: Penguin, 1994), p. 41. Back.

Note 68: Norberg and O’Neill, Transforming Computer Technology, p. 131. Back.

Note 69: Ceruzzi, A Modern History of Computing, p.260. Back.

Note 70: Lampson, "Distributed Computing." Back.

Note 71: Hiltzik, Dealers of Lightening, p. 67; Ceruzzi, A Modern History of Computing, p.260. Back.

Note 72: Hiltzik, Dealers of Lightening, p. 173; "The Mouse," San Jose Mercury News, website: <>. Back.

Note 73: Hiltzik, Dealers of Lightening, p. 67. Back.

Note 74: See Computing Research Association, Computing Research: A National Investment for Leadership in the 21st Century (Washington, DC: CRA, 1997); National Research Council, Evolving the High Performance Computing and Communications Initiative to Support the Nation’s Information Infrastructure (Washington, DC: National Academy Press, 1995), chapter 1; and Norberg and O’Neill, Transforming Computer Technology. Back.

Note 75: Stuart K. Card, "Pioneers and Settlers: Methods Used in Successful User Interface Design" in Human-Computer Interface Design: Success Stories, Emerging Methods, and Real-World Context, Marianne Rudisill, et al, eds. (San Francisco: Morgan Kaufmann Publishers, 1996), p. 162. Back.

Note 76: The Argua, July 3, 1999. Back.

Note 77: Tim Draper, "Silicon Valley to Washington—Ignore us, Please," Wall Street Journal, March 4, 1997, emphasis in original. Back.

Note 78: Interestingly, the producer’s sequel on the origins of the internet, Nerds 2.0.1 (PBS, 1998) starts with a major profile on ARPA’s contribution. Back.

Note 79: Brad A. Myers, "A Brief History of Human Computer Interaction Technology," ACM Interactions, v.5, n.2 (March 1998). Back.

Note 80: Campbell-Kelly and Aspray, Computer, p. 266. Back.

Note 81: Regarding these other areas of government support, see Glenn R. Fong, "Breaking New Ground or Breaking the Rules: Strategic Reorientation in U.S. Industrial Policy," International Security, forthcoming Fall 2000. Back.