Douglas C. Engelbart - Interviewed March 1999 by David Tebbutt

To his eternal chagrin, Douglas C Engelbart's name is likely to go down in history as the man who invented the mouse. Why chagrin? Because the mouse was just one tiny step on his journey to give mankind the means to tackle its growing problems. His original aim was to "augment the human intellect". He reasoned that we would need all the help we could get in a world of accelerating change and increasing complexity. This observation seems extremely pertinent today. What's astonishing is that Engelbart had these insights in 1951.

After 17 years of effort, his ideas made their public debut at the Fall Joint Computer Conference in San Francisco. 2,300 computing professionals witnessed an astonishing demonstration of teleconferencing, on-screen graphics, shared whiteboard, word processing, outline processing, hypertext links, windows, hierarchical file structures containing time-stamped and individually addressable components and more besides. I asked him if he realised at the time that his demonstration was a pivotal moment in the history of computing. In his self-deprecating way, he replied, "I doubt it." Then after a pause he added, "I was actually disappointed. I really hoped the world would just start talking about augmenting things and we'd have lots of company. That isn't what happened."

The reason it didn't happen is because the audience almost certainly missed the point of the demonstration. They were so dazzled by the technology innovations that they forgot that their purpose was to support new ways of working with a shared knowledge repository which would help amplify the efforts of people working towards common goals.

Even now, thirty years later, he still believes that his journey has only just begun. While others have made fortunes directly or indirectly from his technology and software ideas, Engelbart still dreams of a day when people finally understand that his dreams are not just about technology, they're about human systems too. I asked him to compare his progress since 1951 with a journey from San Jose to San Francisco, some 48 miles. He reckons he's still only reached Santa Clara, a distance of four miles.

In 1950, he was working as an electrical engineer at NACA Ames Research Laboratory (the predecessor to NASA) in what is now Silicon Valley. He had just become engaged and was looking forward to a fulfilling life of regular work, marriage and living happily ever after. He reflects, "for some reason I wanted to invest the rest of my heretofore aimless career towards making the most difference in improving the lot of the human race." To this day, he has no idea where that thought came from.

During the next few months, he realised some fundamental truths. Even in 1950 the world was facing ever-accelerating change and ever-increasing complexity and these changes were already outstripping mankind's ability to cope effectively. Into his head popped a vision of networks of knowledge workers sitting in front of large computer screens working on rapidly evolving displays of mixed text and graphics. He had worked on radar systems during the war and he had also read Vannevar Bush's classic paper "As we may think" in Atlantic Monthly in 1945 (plenty of copies on the Internet). Both were contributory factors to this insight.

His mission led him first to Berkeley, where he earned a PhD and an assistant professorship, then on to Stanford Research Institute (SRI) which gradually let him spend more and more time pursuing his dreams. At its peak he was running a laboratory of 47 people. It was while working at SRI that he conducted the famous demonstration. Today Engelbart is 74, an age when most of us would have settled for retirement and decline. But he still travels the world, sharing his insights with anyone who will listen.

Engelbart likens today's organisations as "vehicles that are moving faster and faster, through rougher and rougher, stranger and stranger terrain." He then poses the question, "but what is being done about improving their forward visibility? Headlights don't shine very far." He draws parallels with organisms which thrive by spotting threats and opportunities in the surrounding environment and drawing on experience to react appropriately. The same applies to organisations. The speed at which they spot threats and opportunities, and the smartness of their subsequent plans and reactions, will determine their eventual success.

Engelbart likes to refer to an organisation's ability to perform these tasks as its "collective IQ". But, unlike human IQ, it can be systematically improved. And, although he has created a number of computing 'firsts' along the way, this is the destination towards which he's travelling.

He admits that people still have trouble buying into his ideas. This is partly because he is not one of the world's best communicators. Like many technicians, he is quiet, hesitant and unnervingly truthful. A marketeer and salesman he is not. He called his augmentation system NLS (for oNLine System). Clearly fond of initials, he also coined the terms H-LAM/T and CODIAK. His latest is OHS, of which more later. Engelbart draws his strength from the fundamental justness of his cause. He also appears to be something of an idealist, giving an impression that he cares little for those who commercialise elements of his vision rather than buying into the big picture.

Having said that, he seems happy to accept the hospitality of Logitech, a company which has waxed fat on the mouse, an invention which in 1965 took just two minutes of his time to conceive. The company has given him a corner office at its Fremont headquarters from which he runs his Bootstrap Institute (, where you will find mountains of information about his work).

The Institute exists to promulgate his ideas and to play host to the Bootstrap Alliance, which set out to be a consortium of companies which collaborate to find better ways of working. Unfortunately, the Alliance has achieved little success in America where the NIH (Not Invented Here) ethic is strong. And, it has to be said, the organisations which have shown an interest have backed off when they discover that they are expected not only to contribute money but also to share their insights for the common good. Engelbart admits that although things are moving faster than they were a year ago, the Alliance "still hasn't gotten off the ground." He hopes that the recently formed Japanese chapter will be more successful.

Looking from the outside, both the Logitech association and the name 'Bootstrap' appear counterproductive. The first indelibly associates him with a single artefact, the mouse, while the second brands him a 'techie'. In Engelbart's radar world a bootstrap was the name given to an electrical device which feeds its output to its input in order to achieve a smoother signal. I asked Engelbart whether he had ever tried to get marketing people on board. He said he had, three or four times, but reckons they could never properly grasp what he was saying. He recalls, "It wasn't very long before I found they were going West instead of North."

To understand the man, we need to understand his philosophy. In 1962 he published a paper, the first public articulation of his dreams, called "Augmenting Human Intellect: A conceptual framework. In those days, computers were used for processing numbers and very limited amounts of textual information. They certainly weren't used for processing words, let alone ideas. Here's the opening paragraph:

"By 'augmenting human intellect' we mean increasing the capability of a man to approach a complex problem situation, to gain comprehension to suit his particular needs, and to derive solutions to problems. Increased capability in this respect is taken to mean a mixture of the following: more-rapid comprehension, better comprehension, the possibility of gaining a useful degree of comprehension in a situation that previously was too complex, speedier solutions, better solutions, and the possibility of finding solutions to problems that before seemed insoluble. And by 'complex situations' we include the professional problems of diplomats, executives, social scientists, life scientists, physical scientists, attorneys, designers - whether the problem exists for twenty minutes or twenty years. We do not speak of isolated clever tricks that help in particular situations. We refer to a way of life in an integrated domain where hunches, cut-and-try, intangibles, and the human 'feel for a situation' usefully co-exist with powerful concepts, streamlined terminology and notation, sophisticated methods, and high-powered electronic aids."

It's clear from this that Engelbart wasn't interested in producing gizmos or electronic crutches to help us through the minutiae of our problems. He wanted to create the all-singing, all-dancing, electronic support environment within which we could work and collaborate and improve our performance, no matter what the nature of our problems or the size of our community.

Engelbart also makes it clear that results would not be obtained without some adjustment on our part. The references to "streamlined terminology and notation" hinted that humans would need to change their working language. When I challenged him on this, he pointed out that, "languages have evolved through centuries. They contain a lot of redundancy because we speak in a lot of different environments and there has to be enough redundancy to give us all the error correction we need." He added, "we could probably start by making significant simplifications in the language that would improve the efficiency of transfer." He went on, "I could learn to communicate more efficiently and more rapidly than I do now in some significant way. Then I could start reading and listening to something that would be much more efficient. So until we start opening up to things like that, we're just making this implicit assumption that the way things have evolved in the past is the most efficient and effective way to do it." He quite reasonably points out that people in workshops full of specialised tools use many words that outsiders would not understand. So why should a knowledge workshop be any different?

Two of the guiding principles of the personal computer revolution have been WYSIWYG (what you see is what you get) and user-friendliness. Neither is wrong in the lower reaches of computing but at the higher levels Engelbart sees them as disadvantages. WYSISWYG is, by its very nature, limiting the user to a print-centric view of the world. The world of hyperlinked ideas does not lend itself at all well to being printed. The value of a web page drops sharply when it is printed out because you can no longer follow the links.

And ease-of-use, however laudable, is like a high jump bar permanently set at a height that the least athletic among us can clear. How can we move onwards and upwards with these twin shackles on progress? He says, "If our wheeled vehicle transportation evolved in that way, then everybody would be riding tricycles because, of course, they are easy to learn and natural to use. No-one would consider riding a bicycle."

Engelbart's solution is to provide levels of user interface. Each may be harder to learn, but each amplifies the knowledge worker's power over the underlying information. Users themselves elect when they need to move to the next level. Once there they operate on the underpinning information in ways that people lower down the skill hierarchy would barely understand.

Even as far back as 1962, Engelbart was proposing "giving the human the minute-by-minute services of a digital computer equipped with computer-driven cathode-ray-tube display, and developing the new methods of thinking and working that allow the human to capitalise upon the computer's help." These remarks clearly identified the needs which were addressed by the personal computer many years later. However, the original personal computers were a step backwards from his vision because they weren't networked. Without communal effort and a shared knowledge repository, his dreams could not be realised. It wasn't until groupware and computer supported cooperative working (CSCW) came into vogue that Engelbart's pioneering work started to be recognised. He won his first award in 1987, almost 20 years after he had first tried to explain his vision. Now, 24 awards and honours later, his efforts are still being recognised, the latest award being by the IEEE (June 1999).

In his 1962 paper, Engelbart laid out four ways in which a human's abilities can be extended: "1. Artifacts - physical objects designed to provide for human comfort, for the manipulation of things or materials, and for the manipulation of symbols. 2. Language - the way in which the individual parcels out the picture of his world into the concepts that his mind uses to model that world, and the symbols he attaches to those concepts and uses in consciously manipulating the concepts ('thinking'). 3. Methodology - the methods, procedures, strategies etc., with which an individual organises his goal-centred (problem-solving) activity. 4. Training - the conditioning needed by the human being to bring his skills in using means 1, 2 and 3 to the point where they are operationally effective." (This was the H-LAM/T system mentioned earlier.) Engelbart saw quite clearly how information technology could be used to support and amplify the abilities of these target individuals. Never mind that the price would have been astronomical in 1962, he knew that costs would fall over time.

Here's an extract from what he had to say at the time on the subject of a 'writing machine', "This writing machine would permit you to use a new process of composing text. For instance, trial drafts could rapidly be composed from re-arranged excerpts of old drafts, together with new words or passages you stop to type in. Your first draft could represent a free outpouring of thoughts in any order, with the inspection of foregoing thoughts continuously stimulating new considerations and ideas to be entered. If the tangle of thoughts represented by the draft became too complex, you would compile a reordered draft quickly." This sounds like a fair description of a word processor with some decent outline processing and hypertext thrown in. He continues, "You can integrate your new ideas more easily, and thus harness your creativity more continuously, if you can quickly and flexibly change your working record." Not only is this impressive for the age in which it was written, it gives a clear example of how an improvement at one level - being able to shuffle text - amplifies capabilities, thinking and speed of action, at higher levels. At the time he referred to this suggestion of text manipulation as a 'minor' advance.

Engelbart regards language as a key element in successful augmentation. He believes that our thinking power is directly related to the symbols that we use. To test this hypothesis, he decided to de-augment someone by the simple expedient of tying a brick to their pencil. Needless to say, they couldn't match a typewriter for speed and legibility. He uses this to pose a question: "How would our civilization have matured had this been the only manual means for us to use in graphical manipulation of symbols?" Going to the opposite extreme, he asks what would have happened if we had developed "a high-speed, semi-automatic table-lookup device, cheap enough for almost everyone to afford and light enough to be carried on the person?" He describes a sample application: "Publishers would produce cartridges containing unabridged dictionaries - a one-paragraph description could be located and displayed on the face of the device by the average practised individual in less than three seconds." He then poses questions about organisational change if everyone could get their hands on relevant information and on the educational impact if students and teachers could reach out in this way. Within a quarter of a century, the PC industry had more than exceeded his expectations in this respect.

Engelbart's proposition to SRI was that he should examine the human language, artefacts, materials, and training systems with the aim of identifying all points which could benefit from some kind of augmentation. His view of our systems is hierarchical. All of our capabilities can be decomposed into component capabilities until the decomposition becomes pointless, providing a perfect framework within which he could consider ways of improving each element. As time went by, new developments in the outside world would support improvements in our capabilities at specific points in the hierarchy. The benefits of these improvements would flow up and down the hierarchy. And the resulting amplification of our abilities at the highest levels - organisational or personal - would raise both our personal and our organisational intellectual effectiveness.

His approach was to identify five types of structure and to then examine each for augmentation opportunities. The five were: mental, concept, symbol, process and physical. Mental structuring is the internal organisation of conscious and unconscious images, associations or concepts, which provides us with the basis for judgement, intuition, inference and meaningful action. Concepts are tools which can be confidently manipulated by the brain in symbolic form. They can be composed of other concepts and we are happy to work with them providing they are given some kind of symbolic 'handle'. (To take a recent example, I suspect that 'road rage' only became an observable phenomenon once it had been named.) Apart from amplifying our thinking power, such concepts give us a high level of exchange of ideas with others. Both speed and comprehension rise if the listener shares the same concepts as the speaker.

By symbols, Engelbart means the way in which concepts are represented - words, phrases, sentences, paragraphs, charts, lists, diagrams, tables etc. The question here is one of appropriateness. Certain symbols express our concepts more clearly than others. He offers Arabic and Roman numerals as a perfect contrast. (Imagine trying to do complex arithmetic using Roman numerals.) But for symbols to do their job properly, they have to be readily transformed according to the needs of the person working with them. Engelbart again, "we are very much concerned with the speed and flexibility with which one form can be transformed into another, and with which new material can be located and portrayed." Engelbart concluded that "computer-controlled display systems" would be "a very significant capability to build upon." Importantly, he saw that the way in which information is stored would have no direct bearing on the way it would be displayed. The computer could store complex concept structures in ways that humans would find impossible using paper. In fact, the computer would be able to store n-dimensional constructs, even though they would have to be presented through two-dimensional views on the screen.

Process structures are the procedures we adopt to organise, execute, supervise, evaluate and coordinate other process structures in order to reach our goals. Physical structure refers to the physical construction of the artifacts we use.

Engelbart warned against taking a piecemeal approach to augmentation, yet this is pretty much exactly what happened over the ensuing years. Companies want returns on their investments within a reasonable timescale but Engelbart's grand vision did not lend itself to rapid returns. The result is that we now work with a collection of tools, some of which work together, many of which don't. Here is Engelbart's prescient warning from 37 years ago: "To most people who initially consider the possibilities for computer-like devices augmenting the human intellect, it is only the one-pass improvement that comes to mind, which presents a picture that is relatively barren compared to that which emerges when one considers this regenerative interaction." He is referring here to the flow of improvements up and down the capability chain from our ability to learn how to execute new processes at the top to improvements in our artefacts at the bottom.

To give you an idea of the kind of computer world in which Engelbart was living when he laid out his grand vision, here are some prices: To store a single symbol (a letter, a number or a special symbol) in random access memory would cost between 60 cents and $1.60. A magnetic tape drive would cost $30,000 with tapes at $50 a pop. Capacity? About 5MB. A five inch cathode ray tube at the time would have cost between $20,000 and $60,000. Teletypes were $3,000 to $4,000 each. Five years later, when he conducted his public demonstration, he was backed up by 64k memory, a 96MB hard disk drive, a $5,500 cathode ray tube and wireless links to SRI 30 miles away.

For the 1968 demonstration a young, dark-haired Engelbart wore a headset and sat in front of a keyboard flanked by a mouse to the right and an odd looking device like five piano keys to his left. (See photo.) His CRT display was projected onto a massive 27 foot by 18 foot display screen.

The chord keyboard was a classic bit of Engelbart 'augmentation'. It allowed him to key in up to 31 characters with one hand while keeping his other hand on the mouse. He could key commands with his left hand while the mouse was being moved to the text on which the commands should operate. He thought that the operational efficiency more than made up for the time needed to learn the chord combinations. A common utterance during the demonstration was 'oops' as his chording failed to produce the right result. In the main, though, it worked well. Both the Microwriter and the Agenda organiser in the UK tried a similar arrangement, but neither took the world by storm. Engelbart still uses his.

At the demonstration he showed the following revolutionary features of the NLS system: copy, paste, delete, backspace, expand, collapse, replace, automatic line numbering, moving entries up and down a list, altering the line spacing, jumping on links, jumping on keywords, drawing, different views of the same information, split screen working - the forerunner to windows, authorship history for every entry, search by single and multiple strings, collaborative on-screen working and telepresence with someone at SRI. Remember this was 1968.

The demonstration was a technological tour de force. But he didn't have time to properly communicate the profound thinking behind it. The upshot was that people came away amazed and in some cases inspired, but mainly by the gizmos - the on-screen graphics, the windows, the teleconferencing, the word processing and, it has to be said, the mouse. Engelbart's impression was that he'd demonstrated how NLS was used within the laboratory for planning, documenting, source-code development, business management and document retrieval.

I watched the video ($125 from Bootstrap, in NTSC format) and found it totally humbling. In 1981, I had invented an 'ideas processor' called BrainStorm. Many of the issues that I'd struggled with had already been tackled by Engelbart almost twenty years earlier. My problem, and that of many others, was that we had no idea that Engelbart had already produced a kind of computing 'Bible' in his 1962 paper and demonstrated its feasibility in 1968.

In the mid-1970's, life changed dramatically for Engelbart. The world decided that Office Automation was the next big thing and that the real users would be secretaries. Not only that, but his ideas of getting people to learn new ways of working would be unnecessary since artificial intelligence would enable computers to make up for the inadequacies of their users. He argued that the real users of tomorrow would be knowledge workers, but his pleas went unheeded. He'd run through about $13m of research funding and although his team had built the NLS system, he still felt he had a long way to go. The world was changing, SRI wanted him to consider using minicomputers, but he felt this would cripple his efforts, not least because of the lack of storage.

1976 was a dreadful time. His research funding started to dry up, an AI specialist took over his job and his house was burned to the ground. He assured me, "there was no evidence of sinister causes." In 1977 people started leaving the team in droves, many to go to Xerox PARC. Finally, in July, SRI terminated him and set about selling NLS to Tymshare. Because his vision was so far outside the boxed-in thinking of his peers, he had been side-lined. He says, "A lot of bad vibes about what I'd been up to started circulating. People were saying that I was a charlatan playing up to popular prejudices, taking the wrong direction but too stubborn to see it."

When Tymshare announced that anyone from Engelbart's team could come and help make the system work, twenty people migrated but the last person they expected to show up was a discredited Engelbart. He told PC Pro, "So much work had gone into the system that I wanted to stay with it." He was taken on as a senior scientist but he accuses them of "fixing it so that I wouldn't talk to any customers." Not only that, they more or less ignored his pivotal role in creating NLS (now renamed AUGMENT) turning instead to one of his erstwhile underlings whenever they needed technical advice.

Seven frustrating years later his hopes were raised again when Tymshare was taken over by McDonnell Douglas. However his new CIO told Engelbart, "If IBM and DEC and HP are not doing any of this crazy link stuff, I don't see why we should invest in it." Ironically, the World Wide Web came along a few years later and showed exactly what could be done, even with relatively crude hypertext links. Although the Web falls short of his ideals Engelbart says, "That's all right. It is still a huge movement and it has taught the world a lot. It is a great place to move on from."

So where does he move from here? He sees a number of opportunities but particularly in the Open Source movement and in XML. Both, he believes can be used to push ahead towards his goals. At the moment, the only software manifestation of his work is in a Sun workstation running a DEC 20 emulator which, in turn, runs AUGMENT. He chuckles when he points out that despite the layers of emulation, AUGMENT runs about twice as fast on the Sun system than it did before. He says, "we had a whole room full of machines to do what a pizza box on the desk now does."

A key element of AUGMENT is the knowledge repository in which all the information and its links are stored as hyperdocuments. These are more like information containers than printable documents because they are hierarchically structured and may contain any kind of information, including video as well as links and authorship history for the individual elements. Different applications will be expected to work with this common knowledge base. He doesn't expect AUGMENT itself to be used to support collaborative working but he does think that it can be used as an operational example for future planners and developers.

Engelbart believes that Java and XML between them offer a strong possibility for his hyperdocuments to find their way into general usage. By defining an Open Hyperdocument System (OHS) template ( he hopes that he can encourage the open source development community to take up his ideas, bring them to a wider audience and improve them in the light of new discoveries. And, needless to say, Engelbart believes that the developers would benefit hugely from using OHS for the source code itself just as his team did all those years earlier at SRI.

He sees OHS as the perfect support environment for collective circles that exist to address certain types of problem. He cites as an example the community which is trying to cure cancer. These pre-existing communities could benefit from a better way of sharing their findings. He points out that the old days of papers being published periodically and, maybe once a year, a text book being written is simply not fast enough. It needs to be much more dynamic. Exactly what OHS promises.

I pointed out that such a pooling of information could lead to a massive billowing of ideas and asked how people would be able to keep up. Engelbart suggested that a specialised support role could exist, through which team members rotated. Their job while working there would be to sift, analyse and link the material in the growing repository. Lining his sights on software companies that expect everything to be done using their tools, he says "it would be ridiculous if they were forced to work with commercial tools and/or documents produced by them." He draws a parallel with the physical world, "it's totally ludicrous to say that I'm going to buy my whole workshop from company X. Every so often it will replace the whole workshop. I have standard materials moving through my workshop and standard things I want to do with them. I want to be able to buy a drill, a press or a lathe which will operate in standard ways on standard materials."

Despite the years of frustration and misery, Engelbart is undeterred. He justifies his continuing optimism with the remark, "If you come out to the marketplace with a product before the market is ready, then you can't really blame the market place for ignoring it." Yet the man has made an indelible mark on our industry. Up and down Silicon Valley, companies owe their existence to the technological and software by-products of his essentially human-centric thinking. I asked him how much influence he thinks he's had on developments in the Valley. He replied, "It could ruin my health mulling all that over."


Starting in 1966, David trod the traditional path from programmer to IT manager. He was editor of Personal Computer World. He wrote BrainStorm, the first published ideas processor. He currently splits his time between training, writing and software design.


Possible box outs:



1999 Benjamin Franklin Medal in Computer and Cognitive Science.

1999 The IEEE John Von Neumann Medal

1998 Inductee, National Inventors Hall of Fame

1997 ACM A.M. Turing Award

1997 Lemelson-MIT Prize

1997 Inductee into the Discovery Online Hackers' Hall of Fame.

1996 American Society for Information Science Special Award

1996 Distinguished Engineering Alumnus, University of California, Berkeley

1996 Certificate of Merit, The Franklin Institute, Committee on Science and the Arts

1995 SoftQuad Web Award

1995 "Editors' Choice Award", MacUser

1994 "Engelbart Award" established by the International Conference on Hypertext and Hypermedia

1994 "Certificate of Appreciation from Smart Valley" Inc.

1994 "Price Waterhouse Lifetime Achievement Award", Computerworld Smithsonian

1993 "IEEE Computer Pioneer Award"

1992 "Pioneer of the Electronic Frontier", Electronic Frontier Foundation

1991 "Lifetime Achievement Award", Dominican College of San Rafael

1991 "Distinguished Alumnus Award", UC Berkeley Computer Science and Engineering Department

1991 "American Ingenuity Award", National Association of Manufacturers' Congress of American Industry

1990 "ACM Software System Award"

1990 "Lifetime Achievement Award for Vision, Inspiration, and Contribution", Electronic Networking Association

1989 "Citation for Distinguished Service and Outstanding Contributions in His Field", Sigma Phi Epsilon Fraternity

1987 "Distinguished Alumni of the Year Award", Oregon State University

1987 "Lifetime Achievement Award for Technical Excellence," PC Magazine


And this:? (Taken from Engelbart's OHS framework) - Could make a nice table.

Open Hyperdocument Systems versus prevailing information systems

ie Tool-centric vs Document-centric

Function-oriented tool system vs Integrated end-to-end knowledge management environment

Authoring and publishing vs Developing, integrating and applying knowledge on-line

Isolated passive libraries and archives vs Active "living" libraries and seamlessly integrated within the organisation's work processes

One user class: "easy to use" vs Many classes of user: pedestrian to high performance

File level addressability vs Object-level addressability

"Load and scroll browsing" in WYSIWYG or outline view modes vs "Precision browsing" Jumping directly to any object in any file with on-the-fly custom views

Windows tied to a file and to the application used vs Windows are portals into a file repository, a variety of applications may use any window

Each application has unique file design vs Applications use common file design

"Interoperability" means font styles retained vs Means my structured hyperknowledge interoperates in your environment, I can send you a link into my domain

Click to mail David Tebbutt
Address: Canada House, 272 Field End Road, Eastcote, Middlesex HA4 9NA, UK

Tel: +44 (0) 1895 677845 (UK waking hours please)