What do you think?
Rate this book
Computer Lib/Dream Machines
Nelson writes passionately about the need for people to understand computers deeply, more deeply than was generally promoted as computer literacy, which he considers a superficial kind of familiarity with particular hardware and software. His rallying cry "Down with Cybercrud" is against the centralization of computers such as that performed by IBM at the time, as well as against what he sees as the intentional untruths that "computer people" tell to non-computer people to keep them from understanding computers. In Dream Machines, Nelson covers the flexible media potential of the computer, which was shockingly new at the time.
330 pages, Paperback
First published June 1, 1974
About the author
Ted Nelson
54 books25 followersSharlene and Ted Nelson have been co-authoring books for twenty years. The Nelsons live in Washington State.
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 26 of 26 reviews
June 29, 2016
°͜°
There are two books here, very dated but fascinating nonetheless. The first book, Computer Lib, is about computers in general, as they were in 1972-74 or so. Computers were becoming generally known to the public, and more and more widely used in all kinds of business and academia.
Flip the book upside down to the back, and you get Dream Machines - a look at the most clever computer-based and computer-related technologies of the day.
This book was the first "popular book" for the general public about computers and interactive systems.
Wonderful for those who want to understand the early days of computers and the coming of "personal computers"
There are two books here, very dated but fascinating nonetheless. The first book, Computer Lib, is about computers in general, as they were in 1972-74 or so. Computers were becoming generally known to the public, and more and more widely used in all kinds of business and academia.
Flip the book upside down to the back, and you get Dream Machines - a look at the most clever computer-based and computer-related technologies of the day.
This book was the first "popular book" for the general public about computers and interactive systems.
Wonderful for those who want to understand the early days of computers and the coming of "personal computers"
September 24, 2014
After borrowing the title of my latest book, Intertwingled, from one of the many neologisms in this brilliant manifesto by Ted Nelson, I knew I had to own a copy. So I bought a used First Edition. It wasn't cheap. But it's filled with all sorts of fascinating ideas and inspirations. And I love the sprawling magazine layout and two-books-in-one design. Computer Lib/Dream Machines is/are a wonderful, refreshing book/s that could never be contained in a Kindle.
February 23, 2008
Ted Nelson is, in my opinion, the most influential systems thinker of the past 60 years.
August 6, 2020
Until we overthrow the myth that people always have to adapt to computers, rather than the other way around, things will never go right.
Nelson coins a lot of terms in this book, most of which never went anywhere. One that sounds useful is “cybercrud”, or “putting things over on people using computers”.
No man has a right to be proud that he is preserving and manipulating the ignorance of others.
Sadly, he’s a little prone to it himself. He devotes a couple of pages (and they’re big pages with tiny text) to cybercrud from the Club of Rome, something I’ve seen recently while rereading old OMNI magazines from 1979.
Civilization, and the bulk of mankind, have about forty years to live…
This was 1974 or earlier—the book was published in 1974, but appears to have been written in spurts starting well before that—putting the end of the world in about 2014. Like most religions, the end of the world has a fallback position of 2100, far enough into the future that it can’t be used to test the hypothesis. He describes those who disbelieved the Club predictions as taking the “ostrich position” that “science will save us”. And yet that’s exactly what happened. Food production and distribution soared faster than population growth. Energy extraction soared faster than energy use.
Which makes it, reading the book almost half-a-century later, an interesting confirmation of his main point, which is to avoid cybercrud.
The book is rambling and incoherent, by design (he designed it deliberately after the Whole Earth Catalog, among other publications), but if there is another main point, it is that computers should not be the domain of a technological elite.
“Rigid and inhuman” computer systems are the creation of rigid and inhuman people.
He takes direct aim at the excuses, still used today, that “computers can’t do that” and “the computer won’t let me do that”, euphemisms for “I don’t want to program that” and “the software was programmed without regard for who was going to use it”.
Using a computer should always be easier than not using a computer.
Another thread running through the book is what was apparently a cultural identification war between fans of IBM and fans of DEC. Most techies, and many non-techies, are familiar with this today as “format wars”, because it usually manifests itself in identifying too strongly with a particular format, such as Betamax over VHS, or 8-track over cassette, or even, oddly, Laserdisc over DVD. A more on-target comparison might be Mac vs. Windows, except that from what I see of his descriptions in Computer Lib there is far more of a difference between Macintosh and Windows than there ever was between DEC and IBM.
Nelson was a DEC man.
I hope that this book will help people who are inconvenienced by computer systems to understand and pinpoint what they think is wrong with the systems—in their data structure, interactive properties, or other design features—and that they will try to express their discontents intelligently and constructively to those responsible. Including, where appropriate, International Business Machines Corporation, Armonk, NY.
He was a big fan of the PDP series, and it led to an interesting, in retrospect, aspect of some format wars: dismissing anything outside of the dichotomy. Nelson dismissed microprocessors as only really for toasters and cars and other dedicated purposes. Minicomputers were the way to go, and you really needed to get together with some friends and neighbors to pool together and buy one.
The book is interesting both for its generally good advice about what to put up with and what not to put up with from computer experts and software writers, and the glimpse it provides of a computer subculture dedicated to everyone understanding computers—and being comfortable working with them—before it was at all feasible for most people to own or for the most part even access one for general use.
It is only by clarifying distinctions that people are ever going to get anything straight.
This is a flip-book; the front is “Computer Lib” with a stereotypical sixties/seventies clenched fist on the cover. The back is “Dream Machines”, with a superman in torn jeans pressing a glowing screen.
It is, in fact, very much about computer screens, or, only slightly more generally, computer displays. They were the next big thing, because they could change—the teletype, once it printed a response, could not change that response.
The computer display screen is the new frontier of our lives.
A timely criticism of computer display is that it needs electricity. But it saves paper, and, importantly, it bodes to save energy as well.
IF WE SWITCH TO COMPUTER SCREENS FROM PAPER, PEOPLE WON’T HAVE TO TRAVEL AS MUCH. Instead of commuting to offices in the center of town, people can set up their offices in the suburbs, and share the documentary structure of the work situation through the screen.
His screens are also responsive.
Responsive computer display systems can, should and will restructure and light up the mental life of mankind… The computer’s capability for branching among events, controlling exterior devices, controlling outside events, and mediating in all other events, makes possible a new era of media.
Until now, the mechanical properties of external objects determined what they were to us and how we used them. But henceforth this is arbitrary.
Despite the person on the cover touching the screen, he expects, and appears to prefer, light pens for handling this branching. His description of how his hypertext works makes me glad for the computer mouse. And he did know about the mouse—he also has a page on Doug Engelbart and The Mouse.
Dream Machines also mentions briefly one of the then-modern microcomputers, the Altair, in a supplement to the 1975 edition.
His big push on both sides is that computers ought to be easy to use. And by easy to use he specifically does not mean, more human.
If like the author you are bemused by the great difficulty of getting along with human beings, then the creation of extraneous beings of impenetrable character with vaguely human qualities can only alarm you, and the prospect of these additional crypto-entities which must be fended and placated, clawing at us from their niches at every turn, is both distasteful and alarming.
The pages in this book are usually self-contained articles, and the text made small enough to fit on the page, rather than large enough to read. He apologizes in the intro for not fixing this; ironically, the smallest text and the hardest to read is his description of his proposed Xanadu system for hypertext.
There’s a neat reference to David Gerrold’s The Man Who Folded Himself, as the closest analog to Nelson’s vision for editing documents. It is a great book, but I suspect any document editor that used it as a model would end up in as sad a position as Gerrold’s protagonist.
In the movies and photographs sections, he talks a lot about a form of “halftone image synthesis”; his preferred method appears to be the precursor to modern ray-tracing, tracing light rays from their destination back to their source.
Nelson is an odd figure; he criticizes others for keeping their systems closed, but kept Xanadu completely under wraps even though it would have been more likely to succeed had there been content for it from third-party sources. He champions computers for all, but mostly ignores the microprocessor revolution that would bring computers into the home.
As an overview of what people thought the future was going to be, however, this is a great read.
I’d really have hated to miss being in this field, just for the thrilling madness of it all.
September 3, 2019
About: Ted Nelson's dual-book Computer Lib/Dream Machines is a 1974 overview of the field of computing, both as practice (at the time) and as vision (much remains a dream). The book is largely forgotten now, but for decades - and surely before the resurgence of 'the cloud' in the mid-2000s - it was hailed as a masterpiece and must-read of the field. I'm glad I did, even belatedly. You should, too.
I won't spoil the fun by saying this is a book started from a genuine desire to tell everyone about the (then) mysterious emerging artifact called 'the (digital) computer'. Sure, digital computers have been around for a few decades already, but they were accessible to rhe select few and often for classified projects. But, in the early 1970s, PLATO and various Dartmouth/Minnesota projects and the Altair 8800 personal computer were around the corner or making inroads with the general public. Ted Nelson took the risk of postulating 'You can and must understand computers now!' and this book is now (good) history.
The format is difficult for the starting reader, especially for readers used to the secure, uniform, standard formats of rhe 2010s bookselling industry. It's two books, where the pages of one are displayed on the back of the other's, and sometimes made to match. It's magazine-like formatting, but a creative magazine at that, with diverse column-breaking layouts, zany hand-drawn graphics (and cartoons, and comics), and font sometimes digital and sometimes hand-drawn as well. Ironically and unfortunately, the format cannot be reproduced in modern digital readers - turns out the dream of Ted Nelson of making a good display of complex information (see among others the pages on Xanadu, DM56-7) available and affordable for the general public is still open. (So are Vannevar Bush's Memex, and Doug Engelbart's System for the Augmentation of Intellect.)
The Computer Lib side explains, humorously and with many side-notes, but with a clear direction and excellent presentation, the technology and business of computing. It's:
+ the basics (bits, bytes, electronics as much as a computer scientist needs, the full computer, running a program);
+ the software / early languages (BASIC, TRAC, APL, FORTRAN including Pi - "not quite 3, not quite 4", ALGOL, PL/I, COBOL, LISP, JCL);
+ the hardware / early systems details (IBM, DEC and its PDP and LINC lines, CDC and its 6600, Univac and its 1106/08s, Burroughs 5500 and later);
+ the mainframe, minicomputer, microprocessor (and the sales pitfall related to the microcomputer);
+ the advanced programs, operating systems, batch processing, multi-programming, and time-sharing (from McCarthy and Licklider's early success to TENEX, MULTICS, and IBM's promises on OS/VS2-2);
+ advanced software for a variety of domains.
+ there's so much more! The jokes, the insider info, the credited rumor (Datamation articles, often), the constant IBM bashing, etc.
+ there's even an auto-biography! (p. 70) On the opposite page, in Dream Machines, there's even an explanation about the path taken by the book itself!! (p. 126 in the 1975 edition) Plus, lots of exclamation underlines and weird page layouts!!! (makes for fantastic reading, try it out, then check out what Stewart Brand's Whole Earth Catalog was).
+ the end: "By Computer Lib I mean simply: making people freer through computers." (p. 70)
The Dream Machines part is too big to summarize here. You will find:
+ the early networks and distributed computing projects (PLATO, Dartmouth, ARPA and military projects);
+ hypermedia and hypertext (before the World Wide Web, there was this notion of hypertext invented by a young academic called... Ted Nelson), + AI:
+ IR;
+ computer-assisted instruction (CAI in the book, but one of the few terms where the acronym did not gain ubiquitous acceptance);
+ some weird concepts and ideas that are still not possible today in computing (including the seemingly easy to achive idea of a document reader that would allow opening the document at different points, with different views and annotations for each - can you Acrobat Reader or PDF viewer do this? Mine can't!)
+ There's Xanadu and Thinkertoy, both nice ideas that were not ultimately successful.
I found it fascinating how many things he gets right, but also how many sound by now obsolete, quaint, of slideware (promises made on PowerPoint slides; of course, PowerPoint always computes.) I won't spoil the fun, because reading this is rewarding also for the comedic effect.
Also, you may want to know you'll find here new cyberspace terms, worthy of a Vernor Vinge, William Gibson, or any other excellent cyberpunk writer, among which intertwingling and intertwingled to express the making and existence of interdependencies between hypertextual objects.
I won't spoil the fun by saying this is a book started from a genuine desire to tell everyone about the (then) mysterious emerging artifact called 'the (digital) computer'. Sure, digital computers have been around for a few decades already, but they were accessible to rhe select few and often for classified projects. But, in the early 1970s, PLATO and various Dartmouth/Minnesota projects and the Altair 8800 personal computer were around the corner or making inroads with the general public. Ted Nelson took the risk of postulating 'You can and must understand computers now!' and this book is now (good) history.
The format is difficult for the starting reader, especially for readers used to the secure, uniform, standard formats of rhe 2010s bookselling industry. It's two books, where the pages of one are displayed on the back of the other's, and sometimes made to match. It's magazine-like formatting, but a creative magazine at that, with diverse column-breaking layouts, zany hand-drawn graphics (and cartoons, and comics), and font sometimes digital and sometimes hand-drawn as well. Ironically and unfortunately, the format cannot be reproduced in modern digital readers - turns out the dream of Ted Nelson of making a good display of complex information (see among others the pages on Xanadu, DM56-7) available and affordable for the general public is still open. (So are Vannevar Bush's Memex, and Doug Engelbart's System for the Augmentation of Intellect.)
The Computer Lib side explains, humorously and with many side-notes, but with a clear direction and excellent presentation, the technology and business of computing. It's:
+ the basics (bits, bytes, electronics as much as a computer scientist needs, the full computer, running a program);
+ the software / early languages (BASIC, TRAC, APL, FORTRAN including Pi - "not quite 3, not quite 4", ALGOL, PL/I, COBOL, LISP, JCL);
+ the hardware / early systems details (IBM, DEC and its PDP and LINC lines, CDC and its 6600, Univac and its 1106/08s, Burroughs 5500 and later);
+ the mainframe, minicomputer, microprocessor (and the sales pitfall related to the microcomputer);
+ the advanced programs, operating systems, batch processing, multi-programming, and time-sharing (from McCarthy and Licklider's early success to TENEX, MULTICS, and IBM's promises on OS/VS2-2);
+ advanced software for a variety of domains.
+ there's so much more! The jokes, the insider info, the credited rumor (Datamation articles, often), the constant IBM bashing, etc.
+ there's even an auto-biography! (p. 70) On the opposite page, in Dream Machines, there's even an explanation about the path taken by the book itself!! (p. 126 in the 1975 edition) Plus, lots of exclamation underlines and weird page layouts!!! (makes for fantastic reading, try it out, then check out what Stewart Brand's Whole Earth Catalog was).
+ the end: "By Computer Lib I mean simply: making people freer through computers." (p. 70)
The Dream Machines part is too big to summarize here. You will find:
+ the early networks and distributed computing projects (PLATO, Dartmouth, ARPA and military projects);
+ hypermedia and hypertext (before the World Wide Web, there was this notion of hypertext invented by a young academic called... Ted Nelson), + AI:
+ IR;
+ computer-assisted instruction (CAI in the book, but one of the few terms where the acronym did not gain ubiquitous acceptance);
+ some weird concepts and ideas that are still not possible today in computing (including the seemingly easy to achive idea of a document reader that would allow opening the document at different points, with different views and annotations for each - can you Acrobat Reader or PDF viewer do this? Mine can't!)
+ There's Xanadu and Thinkertoy, both nice ideas that were not ultimately successful.
I found it fascinating how many things he gets right, but also how many sound by now obsolete, quaint, of slideware (promises made on PowerPoint slides; of course, PowerPoint always computes.) I won't spoil the fun, because reading this is rewarding also for the comedic effect.
Also, you may want to know you'll find here new cyberspace terms, worthy of a Vernor Vinge, William Gibson, or any other excellent cyberpunk writer, among which intertwingling and intertwingled to express the making and existence of interdependencies between hypertextual objects.
February 2, 2011
Fascinating glimpse to the future in a highly readable and thoroughly entertaining book. As well as original text from 1974 there's extra text written for the 1987 version, updating on what has changed since 1974. Wish I'd read this years ago. I can easily see why it became a cult classic among the hackers (programmers) when first published.
July 16, 2021
The copy I got my hands on, through Inter-Library Loan, was the first edition, published in 1974.
The book was self-published. It looks like the author put some fanfold paper into an IBM Selectric typewriter, typed up his articles about 6" wide on the paper, then used a copy machine to reduce them down to 3" wide galley strips and cut and pasted them onto layout pages. He'd make a reference to another article in the book, with a page number, but he'd leave the space for the page number blank. Then, after pasting his galley strips onto larger pages (with copious amounts of hand-sketched stuff added), and he determined where various articles "landed," he went through and hand-wrote the page numbers. At least a couple of those spots were still blank. After all that, the layout pages were imaged for printing.
It has to be seen to be believed.
I mention the Selectric because I learned to type on one of those and the main typeface looks very familiar. Additionally, the Selectric had a replaceable type "ball," not much larger than a golf ball. You turned off the typewriter (which left the platen in place), lifted a tab on the ball, lifted the ball out, set another on the platen, aligned it, locked the tab down, turned the typewriter on and ... voila! you've changed typefaces, in the middle of a word if you wish. Tedious, yes, but most of the galley strips making up this book are one typeface, with occasional excursions into another. He has a section on the APL programming language (is that redudant? APL stands for A Programming Language ...). Selectric-based teletypes, with a special ball, would allow you to type APL programs. I suspect he had such a ball when he wrote that section. Back in the day, you could rent a collection of balls instead of needing to buy them all. Finding APL type balls on eBay is rather difficult, today. The Computer History Museum seems pretty chuffed that they have one.
The book is written as two books. Open it from one side and you have Computer Lib, which is mostly background information about computers in his era, including in-depth discussions of the architectures of the major computers of the era (including DEC PDP models) and microprocessors (some of which are now obscure). When you get to about the mid-point of the book, you close it, flip it over (no, I'm not joking) and start reading from the other end of it. The second book is Dream Machines, hence the odd-looking title of the book. The first part is about history and "where we are now," the second part is "where we could be going."
Ted Nelson is known as the person who coined the term "hypertext." He's convinced that hypertext will save the educational world. His final "destination" in the Dream Machines section is talking about Project Xanadu. His thought is to build a hypertext system which will contain the sum total of the world's knowledge and make it easy, and cheap, for people to explore it. Additionally, he wants as much of it as possible to be written be regular people, not experts in various fields (they tend to write over the heads of most of their potential audience and he wants this to be more approachable). He wants it to be devoid of hierarchy, so much as possible. Hypertext should allow you to link from one article to another, without needing to navigate up from here and down to there through some hierarchy or folder structure. Engelbart's oN-Line System (NLS) had hypertext but it had a rigidly-enforced hierarchy, which made it difficult to find stuff if you didn't already know where it was.
Is it just me, or does this sound an awful lot like Wikipedia?
He points out, in various articles within the book, that most of what we consider "formal education" is counter-productive. First, we have to dumb down the content to various grade levels and separate it into different "subjects," then we have to dedicate a certain amount of time to this subject, then drop what we're doing and devote time to this other subject, ad nauseum. This is not how people learn (not very well, anyway). And it tends to create a very disjointed mind.
The ideal way to teach is to have everything interconnected (because that's how reality works; people recognize this, subconsciously, which is why the artificially-separated "subjects" just feel so wrong) and provide a way to explore it. If you want to explore math today, explore math. If you want to explore literature today, explore literature. If an article in one section leads you to an article in another section, fine. If you mind wants to go there, you can go there. Hypertext makes it easy to interconnect things from different subjects, so you get away from discrete subjects.
I, personally, find Wikipedia to be a rabbit hole; I can get started reading one article, because it's something I need to know something about, and end up, hours later, on something that seems hideously unrelated. It only seems unrelated; the fact that I got from point A to point Z means there had to be stuff connecting them in points B through Y. But that's how reality is. That's how the mind works. That's how his ideal educational system would work.
He also points out how, at one point, he had a small minicomputer at home, with BASIC on it. He let a couple younger relatives (adolescents) play with it. He showed them a few BASIC commands and ... four hours later they were still at it. They weren't reading some book and doing the exercises; they were exploring. It was far more engaging and they were far more motivated. Much of my programming skills (and, to be fair, I do it for a living) were developed in that fashion; I might spend some time reading about theory but I didn't really grok it until I played with it. Once I get going on something, I can stay "head down" in it for hours at a time, suddenly coming out of it at 11 pm or so and ... geez, I need to eat some dinner, feed the kitties and get some sleep. I can empathize with his thoughts about exploring instead of formal education.
There are some very interesting little "asides" in this book.
He talks about a system which could store multiple streams of video onto one, wide piece of film with the ability to create an "interactive movie," where you could change which stream you were watching at various points. Take a moment to stop and ponder that. A movie ... with multiple video streams ... which you can switch between ... on film ... in the 1970s. We didn't really see that realized until the "Dragon's Lair" video-disk-based game hit the arcades, as well as a few others of that type afterwards.
He also talks about someone who put together some very simple biofeedback equipment and attached it to an HP minicomputer, then wrote a program which would show you letters in sequence and you could use biofeedback to "select" letters as they scrolled past. He showed it to a neurosurgeon, who responded "I can type faster than that" and dissed it, so the developer gave up on it.
This would've allowed someone who was a quadriplegic to type.
Before the publish date in 1974.
Uhh ... can we build something like that, maybe a little faster, so I can input text without needing to use a keyboard? I'm thinking of a few devices (smartphone, for example) where that could be very useful, even though I'm NOT quadriplegic.
There are plenty of references to the Nixon Administration and various events of that era. It's really impossible to take this book out of the time in which it was written, so I can't say as it will ever be timeless.
So many of the things he talks about have come to pass. People get far more of their information, these days, from computer screens than they do from printed material (we just call them "smartphones" instead of "computers"). Hypertext is a common thing. There are huge networks out there which allow you to access all manner of content, all for an hourly or monthly fee (remember, minicomputers of that era cost more than new cars so most people accessed a computer via time sharing, usually billed by the hour; since a computer in your home would be limited in its storage capacity, but a network of time sharing systems would be largely unlimited, he figured that model would continue).
His somewhat-breathless discussion of the computer mouse (most computers back then, if they weren't teletypes, used a light pen to select stuff) is downright amusing, today. It was hardcore, cutting-edge stuff back then.
Some of the things he talks about have yet to come to pass. For Xanadu, he suggested a reading system with two "throttles," one of which would let you scroll forward and backward in the content and the other would let you scroll the complexity. If you're having a hard time following the terminology being used in this article, use the other throttle to push it toward "simpler," such that terms were replaced (or annotated, in-line) by their definitions and the grammar was simplified. In this fashion, one article could be written to satisfy people who read at varying skill levels. Imagine being able to replace your current Wikipedia article with one written for an elementary school reading level, or a college PhD level, all with a simple control. There are days where I can read at a college level and there are days when middle school might be more my speed. There are subjects which I can read at a college level and others where middle school might be more like it. He was suggesting this in 1974. We've still not accomplished it.
I, originally, requested this book because it has some interesting articles in it about PLATO. Yes, Dream Machines has multiple pages devoted to that. Most of what was in there, I already knew about from other sources. Some of his hand-sketched diagrams, though, were instructive. Sometimes, a picture (even one rough-sketched with a pencil) is worth 1,000 words.
Because of the really small type, I had a difficult time reading it. I tried the "geek magnifying glass" approach of using the camera on my tablet to image the text and blow it up so I could read it on the screen. Scrolling a tablet above a page full of tiny type, in multiple columns, for hours a time ... tedious. I finally went with using my smartphone to "cast" whatever was on the screen to a flat-screen TV with a Chromecast and putting the phone on the video recording app, without actually recording anything. If I put the camera on "image" mode, it was slow to adjust the focus; the "video" mode was pretty aggressive about keeping it in focus. The smartphone was much lighter weight than the tablet so holding it for hours at a time (more than 50 pages in each of the two sections) wasn't as tedious. We don't use that TV very much so my wife found it amusing, rather than annoying.
The book was self-published. It looks like the author put some fanfold paper into an IBM Selectric typewriter, typed up his articles about 6" wide on the paper, then used a copy machine to reduce them down to 3" wide galley strips and cut and pasted them onto layout pages. He'd make a reference to another article in the book, with a page number, but he'd leave the space for the page number blank. Then, after pasting his galley strips onto larger pages (with copious amounts of hand-sketched stuff added), and he determined where various articles "landed," he went through and hand-wrote the page numbers. At least a couple of those spots were still blank. After all that, the layout pages were imaged for printing.
It has to be seen to be believed.
I mention the Selectric because I learned to type on one of those and the main typeface looks very familiar. Additionally, the Selectric had a replaceable type "ball," not much larger than a golf ball. You turned off the typewriter (which left the platen in place), lifted a tab on the ball, lifted the ball out, set another on the platen, aligned it, locked the tab down, turned the typewriter on and ... voila! you've changed typefaces, in the middle of a word if you wish. Tedious, yes, but most of the galley strips making up this book are one typeface, with occasional excursions into another. He has a section on the APL programming language (is that redudant? APL stands for A Programming Language ...). Selectric-based teletypes, with a special ball, would allow you to type APL programs. I suspect he had such a ball when he wrote that section. Back in the day, you could rent a collection of balls instead of needing to buy them all. Finding APL type balls on eBay is rather difficult, today. The Computer History Museum seems pretty chuffed that they have one.
The book is written as two books. Open it from one side and you have Computer Lib, which is mostly background information about computers in his era, including in-depth discussions of the architectures of the major computers of the era (including DEC PDP models) and microprocessors (some of which are now obscure). When you get to about the mid-point of the book, you close it, flip it over (no, I'm not joking) and start reading from the other end of it. The second book is Dream Machines, hence the odd-looking title of the book. The first part is about history and "where we are now," the second part is "where we could be going."
Ted Nelson is known as the person who coined the term "hypertext." He's convinced that hypertext will save the educational world. His final "destination" in the Dream Machines section is talking about Project Xanadu. His thought is to build a hypertext system which will contain the sum total of the world's knowledge and make it easy, and cheap, for people to explore it. Additionally, he wants as much of it as possible to be written be regular people, not experts in various fields (they tend to write over the heads of most of their potential audience and he wants this to be more approachable). He wants it to be devoid of hierarchy, so much as possible. Hypertext should allow you to link from one article to another, without needing to navigate up from here and down to there through some hierarchy or folder structure. Engelbart's oN-Line System (NLS) had hypertext but it had a rigidly-enforced hierarchy, which made it difficult to find stuff if you didn't already know where it was.
Is it just me, or does this sound an awful lot like Wikipedia?
He points out, in various articles within the book, that most of what we consider "formal education" is counter-productive. First, we have to dumb down the content to various grade levels and separate it into different "subjects," then we have to dedicate a certain amount of time to this subject, then drop what we're doing and devote time to this other subject, ad nauseum. This is not how people learn (not very well, anyway). And it tends to create a very disjointed mind.
The ideal way to teach is to have everything interconnected (because that's how reality works; people recognize this, subconsciously, which is why the artificially-separated "subjects" just feel so wrong) and provide a way to explore it. If you want to explore math today, explore math. If you want to explore literature today, explore literature. If an article in one section leads you to an article in another section, fine. If you mind wants to go there, you can go there. Hypertext makes it easy to interconnect things from different subjects, so you get away from discrete subjects.
I, personally, find Wikipedia to be a rabbit hole; I can get started reading one article, because it's something I need to know something about, and end up, hours later, on something that seems hideously unrelated. It only seems unrelated; the fact that I got from point A to point Z means there had to be stuff connecting them in points B through Y. But that's how reality is. That's how the mind works. That's how his ideal educational system would work.
He also points out how, at one point, he had a small minicomputer at home, with BASIC on it. He let a couple younger relatives (adolescents) play with it. He showed them a few BASIC commands and ... four hours later they were still at it. They weren't reading some book and doing the exercises; they were exploring. It was far more engaging and they were far more motivated. Much of my programming skills (and, to be fair, I do it for a living) were developed in that fashion; I might spend some time reading about theory but I didn't really grok it until I played with it. Once I get going on something, I can stay "head down" in it for hours at a time, suddenly coming out of it at 11 pm or so and ... geez, I need to eat some dinner, feed the kitties and get some sleep. I can empathize with his thoughts about exploring instead of formal education.
There are some very interesting little "asides" in this book.
He talks about a system which could store multiple streams of video onto one, wide piece of film with the ability to create an "interactive movie," where you could change which stream you were watching at various points. Take a moment to stop and ponder that. A movie ... with multiple video streams ... which you can switch between ... on film ... in the 1970s. We didn't really see that realized until the "Dragon's Lair" video-disk-based game hit the arcades, as well as a few others of that type afterwards.
He also talks about someone who put together some very simple biofeedback equipment and attached it to an HP minicomputer, then wrote a program which would show you letters in sequence and you could use biofeedback to "select" letters as they scrolled past. He showed it to a neurosurgeon, who responded "I can type faster than that" and dissed it, so the developer gave up on it.
This would've allowed someone who was a quadriplegic to type.
Before the publish date in 1974.
Uhh ... can we build something like that, maybe a little faster, so I can input text without needing to use a keyboard? I'm thinking of a few devices (smartphone, for example) where that could be very useful, even though I'm NOT quadriplegic.
There are plenty of references to the Nixon Administration and various events of that era. It's really impossible to take this book out of the time in which it was written, so I can't say as it will ever be timeless.
So many of the things he talks about have come to pass. People get far more of their information, these days, from computer screens than they do from printed material (we just call them "smartphones" instead of "computers"). Hypertext is a common thing. There are huge networks out there which allow you to access all manner of content, all for an hourly or monthly fee (remember, minicomputers of that era cost more than new cars so most people accessed a computer via time sharing, usually billed by the hour; since a computer in your home would be limited in its storage capacity, but a network of time sharing systems would be largely unlimited, he figured that model would continue).
His somewhat-breathless discussion of the computer mouse (most computers back then, if they weren't teletypes, used a light pen to select stuff) is downright amusing, today. It was hardcore, cutting-edge stuff back then.
Some of the things he talks about have yet to come to pass. For Xanadu, he suggested a reading system with two "throttles," one of which would let you scroll forward and backward in the content and the other would let you scroll the complexity. If you're having a hard time following the terminology being used in this article, use the other throttle to push it toward "simpler," such that terms were replaced (or annotated, in-line) by their definitions and the grammar was simplified. In this fashion, one article could be written to satisfy people who read at varying skill levels. Imagine being able to replace your current Wikipedia article with one written for an elementary school reading level, or a college PhD level, all with a simple control. There are days where I can read at a college level and there are days when middle school might be more my speed. There are subjects which I can read at a college level and others where middle school might be more like it. He was suggesting this in 1974. We've still not accomplished it.
I, originally, requested this book because it has some interesting articles in it about PLATO. Yes, Dream Machines has multiple pages devoted to that. Most of what was in there, I already knew about from other sources. Some of his hand-sketched diagrams, though, were instructive. Sometimes, a picture (even one rough-sketched with a pencil) is worth 1,000 words.
Because of the really small type, I had a difficult time reading it. I tried the "geek magnifying glass" approach of using the camera on my tablet to image the text and blow it up so I could read it on the screen. Scrolling a tablet above a page full of tiny type, in multiple columns, for hours a time ... tedious. I finally went with using my smartphone to "cast" whatever was on the screen to a flat-screen TV with a Chromecast and putting the phone on the video recording app, without actually recording anything. If I put the camera on "image" mode, it was slow to adjust the focus; the "video" mode was pretty aggressive about keeping it in focus. The smartphone was much lighter weight than the tablet so holding it for hours at a time (more than 50 pages in each of the two sections) wasn't as tedious. We don't use that TV very much so my wife found it amusing, rather than annoying.
December 26, 2014
Computer Lib/Dream Machines contains a treasure trove of information about the pre-PC world of computing which is both fascinatingly alien and eerily familiar. Nelson exclaims "You can and must understand computers NOW!", warning its readers that computers are a tool that can work for you if you understand them, and against you if you don't. It's a must read for any computer nerd.
January 13, 2011
Ted Nelson invented the term 'hypertext', and we still do not have a system as good as Xanadu.
Probably we never will.
Probably we never will.
August 15, 2011
For it's time, this was absolutely incredible. I imagine the first edition is now worth a tidy some of cash if you're lucky enough to have one.
January 29, 2012
Presages the advent of the World Wide Web.
January 8, 2013
Started me on the path.
Humanist vision of computers and their potential. Bridged the gap from Vannevar Bush to modern computing. Taught me to love things I hadn't met yet.
Humanist vision of computers and their potential. Bridged the gap from Vannevar Bush to modern computing. Taught me to love things I hadn't met yet.
June 1, 2013
Batsh*t insane. But in a good way.
July 23, 2013
This is the book that really started the microcomputer revolution; the ideas Ted Nelson presents can be considered as precursors of the internet as well.
July 12, 2014
I stumbled across Computer Lib when I would hunt for alternative zines. I must have read it 5 times because I immediately realized what a treasure this book is. This is one of my treasures.
July 28, 2022
A historical curiosity at this point. If you are looking for details on the origins of hypermedia and/or Xanadu then I suggest you look at his other writings.
June 25, 2023
Chock this up to another book I don't understand the hype for.
The writing is borderline schizophrenic with concepts only loosely related and varying lucidity. The text is small and haphazardly arranged making this book difficult to consume.
"But the ideas are great!" you say! Are they?
Look I'm sure this was a novel book that inspired many people back in the day but it aged into irrelevance and I'm not sure it has that much historical value.
I read the 1987 edition which had been heavily edited by the author who was now much more full of themselves and willing to opine about their own significance.
It's a cool piece of history for a niche audience but it's more of a collector's item than a book I'd recommend you read.
The writing is borderline schizophrenic with concepts only loosely related and varying lucidity. The text is small and haphazardly arranged making this book difficult to consume.
"But the ideas are great!" you say! Are they?
Look I'm sure this was a novel book that inspired many people back in the day but it aged into irrelevance and I'm not sure it has that much historical value.
I read the 1987 edition which had been heavily edited by the author who was now much more full of themselves and willing to opine about their own significance.
It's a cool piece of history for a niche audience but it's more of a collector's item than a book I'd recommend you read.
Shelved as 'to-keep-reference'
December 4, 2019Citado en Hiperculturalidad. Nelson propone una interpretación de la sociedad basada en el hipertexto. En ese sentido la hiperculturalidad sería la interrelación entre distintas culturas contrario al enfrentamiento entre dos culturas opuestas.
August 10, 2022
A bit of computer legend and history. Fabulous at the time. I'll have to read it again. I bet most of it is still visionary!
Read
December 30, 2015This is not a book, it is an experience. This book and its author were and still are visionary. I thoroughly enjoyed the format of pictures, notes, and text in a dual book front and back. Reading the original format of the book(s) is to step into a time machine that simultaneously goes backward and forward whirling through a rush of sometimes crazy, often times brilliant enormous thoughts siphoned straight from the mind of Nelson. The ideas and concepts in this pre-PC book influenced and shaped computing forever. I honestly cannot say enough good things about this book!
January 1, 2014
Tough to rate these days. Only my second read-through since I grabbed the '87 revision when it was new on the shelf. Mostly of historical interest, but still some really cool stuff to think about. There's a PDF scan of the original 1974 edition out there, but I haven't found the later one and the cheapest copy is $80 on Amazon!
Read
January 16, 2014hey book
November 7, 2015
Mostly skimmed, quite dated. Must have been an interesting read in the decade it was published.
Read
June 27, 2016Don't feel it's relevant.
August 18, 2015
Completely worth reading, especially decades after it was written.
Displaying 1 - 26 of 26 reviews