Ted Nelson

Ted has been working on the XanaduProject for 40 years - his vision of a world-wide, public HyperText publishing system. He expects it to save the world.

• http://xanadu.com.au/ted/

Before After

Wired 3.06 ran a lengthy history of the project in an article by GaryWolf? titled TheCurseOfXanadu. (Ted really doesn't like the article.)(A number of the rest of the former Xanadu dev team don't appear to like it either...even when it's right it says it in the worst possible way)

For a quick summary of Nelson's past and present, see this Economist article [which now costs money to read -- use the WaybackMachine: http://web.archive.org/web/*/http://www.economist.com/science/displayStory.cfm?Story_id=442985 -- the full article is still accessible]: http://www.economist.com/science/displayStory.cfm?Story_id=442985 - it claims he's now trying to build Xanadu on top of the Web.

He's written a book, Literary Machines, in which he explains how XanaduProject should work. There's a reference to it in the Wired article, see chapter 8, http://www.wired.com/wired/3.06/features/xanadu.html#8 . The edition ( ISBN 089347052X , was Microsoft Press(!), I think) is long since out of print, but a newer edition is available on Eastgate System's Serious Hypertext page at http://www.eastgate.com/catalog/LiteraryMachines.html .

• see http://www.xanadu.net
• ZigZag, shareware from xanadu.
• WikiIsNotXanadu
• WikiWikiClone
• also see open source (Smalltalk, C, Python) Xanadu: http://www.udanax.com
• Site for analysis of Xanadu Project at Autodesk Sunless-Sea: http://www.sunless-sea.net
• CritLink

---

One or Two other books:

• Computer Lib / Dream Machines (Two books in one); first edition 1974, then 1975 another printing with corrections, then a second edition in 1987 with significant additions due to intervening events.

Sample: defined CyberCrud.

In Dream Machines, TedNelson discussed some still-rare forms of hypertext/hypermedia, some of which he may have invented; they're not often discussed, so whether someone else has a prior claim in any given case is unclear. The most basic kind of hyperlink was never claimed to have been invented by Nelson (priority usually is given to Memex with followups by e.g. DougEngelbart).

• He defines "HyperText" (and he did coin the word) as "non-sequential writing" and as "forms of writing which branch or perform on request".
  • "speech and speech-making have to be sequential, and books are not convenient to read except in a sequence. But the structures of ideas are not sequential. They tie together every which way."
  • "Presentational sequences are arbitrary; hierarchies are typically spurious; boundaries of fields are arbitrary; compartmentalized and stratified teaching produces compartmentalized and stratified minds.
• Basic/chunk style Hypertext -- "offers choices, either as footnote-markers (like asterisks) or labels at the end of a chunk. Whatever you point at then comes to the screen."
• Collateral Hypertext -- "compound annotations or parallel texts".
  • "Ideally, chunk and continuous and collateral hypertext could all be combined (and in turn collaterally linked; see 'Thinkertoys,' DreamMachines? p50)."
• Specific Hypertext -- written on and for a particular topic/purpose.
• Anthological Hypertext -- "materials brought together from all over, like an anthological book."; "it is reasonable to expect the connective structures to cluster to the same general form [as the general interconnective structure of the field...the very same field of knowledge that others represent as an explorable, formalized whole, I am out to represent as an explorable informalized whole, with anecdotes, jokes, cartoons, 'enrichment materials,' and anything else people might dig."
• Grand Hypertext -- "a hypertext consisting of 'everything' written about a subject, or vaguely relevant to it, tied together by editors (and NOT by programmers, dammit), in which you may read in all the directions you wish to pursue...everything you read, you read from the screen (andn can always get back to right away); everything you write, you write at the screen (and can cross-link to whatever you read; see 'Canons,' Dream Machines p 149)."
• Discrete hypertexts -- similar to Memex links and to Web links, but Nelson clearly has in mind at least a mildly extended vision, e.g. not just page linkages but also a smaller granularity of text chunk size, including smooth treatment of "footnotes on footnotes on footnotes, and pathways of any structure the author wants to create".
• Performing Hypergrams -- "a performing or branching picture", for instance a manipulatable bar graph showing trigonometric relationships, where a user change to one bar causes reactive changes in the others via constraints (Nelson clearly got the basic idea of constraints of this sort from Sutherland's Sketchpad, which he discusses). "Hypergrams may also be programmed to show the consequences of a user's prod -- what follows or accompanies some motion of the picture that he makes with a pointing tool, like [the sequence of events as a heart pumps blood]."
• Stretchtext -- a "stretchable" form of text, which becomes increasingly detailed in length and description in reaction to a user's "throttle", in approximately continuous jumps over a broad span of lengths. So a very terse description could expand to a verbose description and thence to a very detailed multi-paragraph description.
  • Published "in 1967" (no other cite given), and collected in "A Hypertext Editing System for the 360", Steven Carmody, Walter Gross, Theodor H. Nelson, David Rice, Andries van Dam; in Faiman and Nievergelt (eds.), Pertinent Concepts in Computer Graphics (U. Ill. Press 1969), pp 291-330.
• Hypermap -- a continuously navigatable map (similar to the Google Earth stuff) allowing movement in 3d, changes in magnification, but also "the user may request additional display modes or 'overlays,' such as population, climate, and industry."
• Queriable Illustrations: a form of Hypergram -- Such as a line drawing of a camera with terse labels protruding from various parts which, when selected by user, bebcomes sliding descriptive ribbon, giving more description of the thing labelled (e.g. the lens has a label "the lens" protruding, which expands to "the lens projects the picture onto...")
• Dissection on the screen -- use lightpen as virtual scalpel to dissect virtual frog on screen; can be either complex simulation or fairly schematic
• Hypercomics -- educational comics; user selects which character (and thus which personality and style) to explain e.g. how an internal combustion engine works.

This is not a comprehensive list.

---

Does anybody have a URL for a project associated with Ted Nelson or Xanadu which allows a web page reader to insert a marker in the web page so future readers could click on the marker and read the marker-author's comment on the text at that point? Colored markers, as I recall. The system could require registration or allow anybody to do it (reminiscent of Wiki) as on their demo page. I thought it should be mentioned somewhere on WikiWiki but I can't find the URL. I could see a combo of the best features of both systems working well.

It sounds like you are talking about now defunct ThirdVoice, or perhaps CritDotOrg.

---

The son of moviestar Celeste Holm (http://us.imdb.com/Name?Holm,+Celeste). Quite possibly the only major computer scientist to have a famous parent (unless you count BillGates' parents as famous; I just think they're rich).

Except that TedNelson isn't a computer scientist. At least, not if you base his credentials on XanaduProject. Rather, Teddy is a cheerleader for TransClusions. A vapid, bottle-blonde, airhead of a cheerleader that masquerades as a programmer and computer scientist.

It's true that TedNelson isn't a computer scientist, nor even a programmer. He is a computer enthusiast and a writer who has popularized computers ("Computer Lib") and hypertext from the early 1960s, and who has advanced a number of vaporware projects such as Xanadu, which, although never completed, many people have found inspirational, so he is often considered a Visionary.

I find TN to be pathetically lacking in vision. Way, way back in the 60s, Xanadu was visionary. By the time the 80s and 90s rolled around, it became pathetic. Now it's 2003 and XanaduProject is simply contemptible. I eagerly await the day when it will be beneath contempt, when TedNelson will be relegated to the trash bin of history.

Harsh.

There is no harsher judgement than from one's peers. If in 10 years' time I have failed utterly in my own projects to transform all media, all computing experience, and even software production, then I expect to be judged harshly. But for now I expected some fucking support (no longer). Rationality dictates this be so. Of course, human beings aren't exactly what I'd call rational.

Consider the difference between the harsh way I am treated and the comparatively royal treatment given to TN by programmers. The key difference is that I might actually achieve what I've set out to do whereas TN is a proven loser and repeat failure. As a consequence, I am a threat to be attacked rather than a harmless icon who can be treated benevolently. People attack me on the basis that my projects might be vaporware. Nobody attacks TN on the basis that his projects have been vaporware. I don't think people care that a project is vaporware. I think they worry that a project might not be. (For example, AnonymousOnPurpose deleted SyndicateOfInitiative on the basis that "we don't need another description of vapourware". Would he ever delete XanaduProject? No.) -- RichardKulisz

I see. It certainly seems consistent to say that you should be judged by the same standards with which you judge others. As for your projects, is there an overview?

As for Nelson, actually he has been attacked quite a lot over the years, although with less media attention than his adulation has gotten.

My understanding is that he had been attacked earlier on, when Xanadu might still have been successful, when it was still to be feared. Nowadays, he is only attacked by ignorant nobodies like the writers at Wired.

I started to describe my projects on this wiki (SyndicateOfInitiative) until I realized the timeframe of their implementation wasn't 10-20 years, but 2-5 years. And I firmly resolved to say nothing more about them when I acquired a partner with a very implementation-oriented mindset.

What I can say is that the ultimate goal of my projects are to implement BlueAbyss. I'm willing to describe one of my other projects, which will run in parallel to them.

I plan to start a business on a SeparatedCooperative? model. Instead of haranguing others to buy into a model of software funding that nobody has ever used, as RichardStallman does, I will implement an ethical model of software funding for my own software development business.

This model includes; a cooperative for every autonomous team of developers, an autonomous non-profit cooperative for every region of users of software, and credit unions to manage and transfer funds from the latter to the former. The funds will be transferred to research or implement specific features, not to pay for permission to use something that already exists.

---

CategoryPerson CategoryAuthor