ACM Computing Surveys 31(4), December 1999, Copyright © 1999 by the Association for Computing Machinery, Inc. See the permissions statement below.

The Unfinished Revolution and Xanadu

Theodor Holm Nelson
Project Xanadu*     Web:
Keio University    Web:
* "Xanadu" is a registered trademark; "xanalogical structure" is intended as a related generic (non-trademark) for general use.

presented at

Engelbart's Unfinished Revolution
A Symposium at Stanford University

December 9, 1998

Ted's Presentation

I love and hate paper. I have more of it than anyone I know. Stanford University has asked for it. They don't know what they are getting into.

But, since childhood I've been outraged by the imprisonment implicit in paper. Having to write to length because there isn't enough paper in the magazine or book. Having to truncate lines of throught because they have just banged their nose on the right margin of the page. Having to stick to the subject, which of course always means sticking to the subject as someone else sees it. I believe that there are no subjects, actually, in the popular sense, because the word subject in the popular sense suggests sharp lines of demarkation, which is precisely what the mind should not be restricted to.

A little on background. I'll try to be quick, the mandatory story, because I'm not an engineer, not a computer scientist. I was going to be an intellectual/'show biz' guy. When I was eleven, in the June 1949 issue of the Reader's Digest I read about a guy named Buckminster Fuller who'd invented his own geometry, and I said 'Wow! You're allowed to do that?! Hey!"

And I was keenly aware of new media. I grew up as a radio child. I still remember vaguely sitting on my grandfather's knee when we were listening to a symphony, and the Pearl Harbor bombing was announced. I also remember playing with my trains with the radio on and hearing the news of Hiroshima, which I conveyed to my grandparents.

At the age of twelve, about, I sat behind my father, who was a director in the early days of live TV, in one of his greatest moments. It was a show called 'Mama' a comedy show, a family comedy show on 8PM on Friday nights, sponsored by Maxwell House coffee. There was no video tape, there were no retakes, and camera one went out on the opening shot. Flawlessly and seamlessly he talked the whole thing through on the microphone and nobody knew that anything had gone wrong. It was magnificant!

At the age thirteen I read a book called 'The Voyage of the Space Beagle" by A. E. Van Vogt and the narrator said he was a 'nexialist', someone who finds connections, and I typed up little business cards calling myself a nexialist. And about the same time a movie opened with an astonishing story line because it was one story told six times over. I think I saw it three times. The year was about 1950 and the film was called 'Rashomon', pronounced 'Rashomon' {{I can't capture the difference in text}} by most Americans, but essentially about versioning. About the fact that there are many stories in the big city and you'll never know which one is correct so you have to tell them all.

In college I majored in extra-curriculars. By the time I got out... I don't say that I had done these things well, but I had produced an LP, I'd made a movie, I'd done the world's first rock musical, in which van Dam played a spy, I had done newspaper and magazine stuff, and the fighting over creative control seemed to me

the center of the universe. At dress rehersal we had a fight, my stage manager and I, and he said he was going to, we were going to do this his way, and I walked off, and ten minutes later I came back and he capitulated. His name was David Baltimore. I don't know if he is still in show business...

{{they didn't laugh! boy, you can tell these are computer people, not scientists...}}

OK. Outraged by the confines of paper and the problem of version management in ever-changing versions, whether its a script or a seminar paper that's due at 9AM, how do you manage the swirl of ever-changing thought and different ideas? How do you keep track? I was trying to keep..., I was resolved to take notes and understand more than most people did. And how express minority views? There are minority political parties, minority religions, which are satirized in the present. You can never find out what they actually think.

And finally, and most important, the individual's fight for creative control vs. the obtuse, shallow, and conventional majority.

OK. So, unable to decide between intellect and show business, I went to graduate school {{laughter}} and there took a course in computers. Bang! Lightening bolt! Heavens open. It was all obvious because, you know, at the age of eleven I saw people, men, suddenly working at screens in darkened rooms. We're going to change. This is the next thing. For the rest of human life we'll be working at screens. And so the question was how to design tomorrow's, what will replace paper for tomorrow. But it had to be better than paper. It had to do everything paper did and be better. For example, everything has to be annotatable.

So I started designing. This was 1960. The problem is, and most people don't understand yet, all documents are parallel. Consider a well-known example, the Bible. What is the Bible? Well, it begins with the Hebrew Torah, and wait a minute, there's that nifty Hebrew Torah that just came out of Ethiopia that's only, what, the first four books. heh - they separated early. And then there are the different translations and there's the New Testament - thousands of different translations. And consider the subtle differences between the Doue version, the King James version, this version and that version, all of which essentially have important differences from the theological point of view. Now, what is the Bible? It's all of them. Now, some people may only want you to see one version, heh heh,{{laughter}} but the whole point, the whole point is the Bible is all of them and being able to see them side by side. Now, even in the 50's they had the Interpreter's Bible, which printed the Bibles side by side. OK. Shakespeare, Rashomon, its all one. So that was the first epiphany.

Second ephiphany. Interactive software. Software was going to be interactive on screens and what is the computer screen? I was frozen. I saw a picture in Datamation of a map - this was 1960 - of a map on a computer screen. Holy smoke! This was going to replace the printed word. And, once we do that, what does it become? Obviously, interactive software is a - I'll put it how I state it now to my classes - is a branch of movie making. This is *not* a metaphor. This is *not* an analogy. Interactive software *is* a branch of movie making and virtually most of computer science is irrelevant. What is relevant is studying your Welles, studying your Hitchcock, studying your good documentaries. Because right now we're in the stage of software which compares to the movie business before 1904. Between 1892, say, and 1904 movies were made by the cameraman because he understood the equipment. And that is exactly where we are now.

{{clapping, whistles}}.

Thank you. In 1904 they invented the director. What was the director? The director was the guy who didn't have to know how to load a camera, didn't have to know how to sew costumes, didn't have to play a violin, dance, fence, or hang the lights. But, he had to know how to make those effects come together in a unified experience.

Now, a very simple rhetorical question for you. Why are video games so much better designed than office software? {{ laughter}} The answer is preposterously simple: video games are designed by people who love to play video games... {{ clapping}} Office software is designed by people who want to do something else on the weekend. {{laughter}}.

Now, what does show business teach you? It teaches you that design is war. Design is war, its a power struggle between the producers, the directors, the authors, everybody who's going to be involved. And now we see it in the following form. What is the software business about? The software business is about the politics of standardization and almost nothing else. So that all the politics comes down to this one thing. How did we get...For example, Eisenbard Kingdom Brunnell in 1820 said that if we had railways where the two tracks were 20 feet apart we'd be able to drive those trains at 200 miles an hour and they'd be smooth. And he was right. And he also knew that ships could be made of iron and nobody believed him. But, what happened? According to the one version I heard, the opposite party's men came around and got his workmen drunk the night before the queen's or king's inspection party and the result was that we got the present railroad standard where the railroad width between the two tracks is the same as the distance between the wheels of the Roman chariot, for traditional reasons we needn't get into.

OK. So the problem is always how do you establish the standards, and everybody has worked on something that, everybody in the field, the computer field, who has more than four years of experience has worked on something that is better than what is out there now. Right? But the problem is, how do you get your standard done. And the great line, which you have to know from Hollywood, is everybody wants to direct. So, how, who gets to direct, who gets to make those decisions. Right now its committees who design the software and the guy who did some difficult subroutine gets to do the little interface popup for that, and what we have is today's garbage.

Now, I tried to explain to people in 1962 or 3 that we would be reading and writing on screens, it would be hypertext, and the copyright issue would be solved because the original quote would be bought from the original publisher and would come instantly, and - I never talk slowly - and there would be a long pause and people would say "Is it like a tape?" {{laughter}} and only years later did I realize that the answer is "Yeah, its like a tape" {{laughter}}

Now, I visited Doug in the Spring of 1967 because we talked about my coming out to work for him. Two things on that trip. Well several, one was - I loved the guy of course, everybody does, and I saw the mouse. Up til then I thought it would be light pens, well obviously the mouse thing worked better. And also next door there was a place that sold skate boards and I went "Wow, I'm going to ride skate boards". {{laughter}} One of those came true. I skate boarded a little while in my thirties but I quit.

The thing about Doug was his emphasis on collaboration seemed to me completely naive. I've always been very sensitive to conflict {{laughter}} and the notion that there ..., that you can bring agreement to people..., agreement between people is a miracle. OK? It has nothing to do with loving them, right? {{laughs}} Does really loving somebody mean that you get along with them well? OK. {{laughter}}. But, one of the things that moves me greatly is seeing this audience here because it shows me that the emphasis on collaboration and on working together does have a meaning, can have results, and I'm learning that also working with people in Japan.

The other side of it, of course, is still how to empower the dissenter and how to empower the person who wants to package the material differently. Because in the old world of media, every information package gets the lid nailed on shut, you can't reuse it, you can sort of point to it vaguely from all sides. Now, i'll skip that.

Now, to talk a little about today's horrible computer world. To quote Eric Raymond: "Microsoft is not the problem, Microsoft is the symptom." So the problem is - how do we alert people's intelligence to what can be, rather than taking it down to the lowest possible common denominator, and this, of course, is the center of Doug's work.

Paper. We've got to have media which are better than paper. Let's just begin... You can flip through paper. Opening a file on a computer screen is like opening a packing crate. The hierarchical assumption of structures in todays software... we've got hierarchical directories which we accidentally invented like in 1947 - 'where are we going to put all this stuff?' 'well, lets make a file' - they named it files, right. So, the hierarchical assumption has passed on to us which assumes that there is no overlap between things we do. You work on one thing, then you finish that, put it away neatly and then you work on something else. {{laughter}}. right. There is no overlap, there is no interpenetration, projects are never redefined, we don't have to change our terminology once we've started...

In other words, today's software was designed by and for clerks and engineers, but not for people who think, who are paid to think. {{laughter}}. Pardon me. Let me restate that. {{laughter}} Not for people who are incessently plagued by problems of re-thinking.

Now the Macintosh and the PC. To me, this is Tweedle Dum and Tweedle Dee. OK. It masks ...

this so-called GUI - I would rather call it the PUI (PARC User Interface) {{laughter}} because there are so many millions of graphical user interfaces possible and yet we are stuck with one in which we have a single fixed little area called the desktop - I don't know why, I've never seen a vertical desktop - where the icons are a fixed size and then they open to flat windows which don't have any perspective and don't connect.

To me calling a hierarchical directory a folder is like calling a prison guard a counsellor. {{laughter}} As Zimbardo {{Dr. Philip Zimbardo}} demonstrated a few hundred yards from here on the Stanford campus, anyone will behave like a prison guard under circumstances that put you in control of people like a prison guard and hierarchical directories create a truncated and cut-up thoughts.

Word processing is a geek's notion of what writers need in the same way that MIDI is a geeks notion of music. In both cases, misled by the notation -'oh, those are characters on a page' 'well we'll just have those characters on a page' and 'those are just notes on paper' - 'well we'll just play those notes'. So now we get MIDI and we get word processing. Whereas the problem in writing is the duality of text between the items, the items you wish to include and whose position you wish to track from version to version, and you want to make sure you get in, and you want to make sure you reused, and the surface structure, which is the knitted sentences resulting from those items projecting upward.

So we have forever this duality. Doug understood that from the beginning, merging outline processing and text but that was just the beginning.

Now, this prophet without honor business is all completely true, but I was misled by the old slogan 'In the country of the blind, the one-eyed man is king". The real slogan ought to be "In the country of the one-eyed, the two-eyed man had better watch himself". {{laughter}} because people don't like your perspective. The lack of appreciation of Doug - its like you stand next to the Empire State building and you don't know how tall it is. You just know that its taller than you and a lot of people resent that. So just to keep standing, for somebody like Doug to keep standing, with all the disappointments and all the crud is just a remarkable, wonderful achievement and I'm very very happy he's here.


The software industry and the understanding of software is not merely about the blind men and the elephant, its the blind venture capitalist trying to take parts of the elephant public. {{laughter}}. 'Like this trunk-thing. Can you give us a three-year exit strategy' {{laughter}}

Integrated software. Well, Doug knew from the beginning that software had to be integrated, but to most people its like trying to explain the idea of cuisine to someone's who's only eaten at McDonalds. 'Let's see, cuisine, is that you put the fries and the burger and the shake next to each other on a tray?' Or, you could be really radical and dip the fries and the burger in the shake - that's Windows. {{laughter}} Or you could put the burger on your head, put the french fries up your nose, and pour the shake over your head. That was called the Lotus Symphony. {{laughter}}. I finally found out - I dearly love Mitch Kapor, but I found out how Lotus Symphony was actually designed. Its true. I had dinner with Ray Ozzie, who created Lotus Symphony, apparently inspired by a conversation we had that I don't even remember, and he said - he was working for Mitch at Lotus - and he said 'I'd like you to back my idea" and Mitch said "OK, I'll make you a deal. Here's the Lotus wish list, the things we've had requested by customers. Will you implement it exactly as stated?" "OK" So Ray programmed this, very carefully, exactly according to the list, and got backing to develop Lotus Notes, the list was released as Lotus Symphony. That's software design for you.

But now, you see, we're in this era of paper simulation. WYSIWYG. Doug already mentioned WYSIWYG. You see, it's one of those misleading terms. What you see is what you get. What do you mean, what are you talking about? What I see is what I see. And I'm getting it, right? No,

not quite. Because, you see, what it really means is what you get when you print it out. In other words, we are tieing your activity to the destiny of paper, we are using the computer as a paper simulator. Which is like tearing the wings off a bus, off a 747 and driving it as a bus on the highway. Now, the usual story is that this evolved at Xerox PARC because they wanted to make computers simple for the man in the street. Actually, the truth is that they did it because they wanted to make computers clear, the idea of computers clear for Xerox management. {{laughter}} Xerox was a paper walloping country, company, and they understood paper. So the whole point was to use the image they have. So like the Roman chariot creating the width of the tracks, the Xerox upper management mind set has created today's computer system. Just think of all the imaginary forests we've had to cut down to fit the simulated paper on the screen. What's worse, is that we are stuck in that prison of paper, those four walls.

So, let's go to the slide, please.

This is a slide which I published in 1972, in the proceedings of a conference. This is the mockup of how computer windows should look. 1972, which I believe was a couple of years before the design of today's prevailing windows at Xerox PARC. So, we should be able to point out the connections of the contents of one window from another window, so you can write your commentaries, you can show your links explicitly, point to point. We've got to have that. People would say "Oh Ted, you don't understand how computers work". Wrong context of discourse. This is how computers *must* work, and if its not convenient for you, buddy, its time...

{{some business with Paul}}

A word about education. People ask me 'Is your stuff useful for education' and the question is, which do you mean by education? Do you mean the process by which people become educated? Great! Do you mean the educational establishment? Thank you, no. Because actually, you buy the farm - oh, that's the wrong expression here - {{laughter}}

The moment you establish a curriculum, you've lost it. Do you know what curriculum means, in Latin? It means, and this is not well publicized, little racetrack. {{laughter}}. So, as soon as you have a curriculum, you have winners and losers, right? You have a schedule, so that the curriculum creates the educational system in exactly the same way that the time slot creates broadcasting as we know it. {{laughter}}

So, here on the screen - this is a closeup of my proposed windowing system. We will now go - control C, double click - we will now go. This is a mock up done in 1960. How do we get out of this guys - {{laughter}}. OK. Thank you. {{some more business with Paul}} So this was programmed by Dr. Ian Heath at the University of Southampton, according to the specs for my 1972 system. So this is how windows look today. Yup. There they are, okay. WinDOze. {{laughter}}.

This is how they ought to look. Now these linked connections should be able to follow the material as you scroll, so that we can actually start to write connections on material that can follow the real interconnection of the material and not be confined to the four walls of windows as we now have them.


OK. So, two cheers for the Web. {{laughter}} The Web is intrinsically broken. It doesn't have version management, it doesn't have rights management, and the reason ... all too late I read a paper that explains why the Web beat Xanadu. Its an article by Richard Gabriel and written in 1987 and called 'Good News, Bad News, and How to Win Big, and Why the Thing that Does the First 50% Flashilly Wins' Trying to add version management and rights management to HTML is like trying to graft arms and legs onto a hamburger.

OK. So, I will leave the last word to Doug, in this little video clip I made six years ago.

{{music. Then Doug speaks}}

"There's a lot of potential out there for humanity and once you get a picture of potential being very much more than what you see, and you think that there's a way to get there, then its very hard to set that aside to do something that's more practical you might say. I really think that the world and humanity could climb a pretty effective ladder, with a lot more capability, human potential, and I think we could go at it with the right strategy. So, that's what I get committed to somehow.


Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Publications Dept, ACM Inc., fax +1 (212) 869-0481, or