Xanadu

 

Background

Xanadu is Ted Nelson's vision and project

Ted Nelson is often considered the fathers of computer hypertext

Worked at Keio University in Japan

    Now self-employed? (2004)

He ignores the culture and customs of academia

Examples

   Almost all citations he makes are to his own work

   He arguably sounds conceited

   His writing style on technical matters is unusual

But: he is acknowledge as a founding father of hypertext

 

Xanadu and the Web

Let's view the Web as the worlds largest digital library

Xanadu is a claim, vision, and example of something better

"The World Wide Web was...what we were trying to *prevent*"

"The...Web ...trivializes our...hypertext model with one-way ever-breaking links and no management of version or contents" - http://xanadu.com/ (4/23/04)

"The Web displaced our principled model with something...chaotic and short-sighted"

"There is an alternative. Markup must not be embedded. Hierarchies and files must not be part of the mental structure of documents. Links must go both ways. All these fundamental errors of the Web must be repaired. But the geeks have tried to lock the door behind them to make nothing else possible." - http://ted.hyperland.com/buyin.txt

"Project Xanadu has had as its purpose to build a deep-reach electronic literary system for worldwide use"

 

Terms and Concepts

Nelson has quite a few of these!

Xanadu - The name given to the mansion built by Kane in "Citizen Kane"

    Famous 1941 movie, available on video

    Seems to portray publishing giant William Randolph Hearst

    (Worth seeing!)         

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 http://www.xanadu.net/

    Deep interconnection

    Intercomparison

    Reuse/frictionless reuse

http://www.sfc.keio.ac.jp/~ted/

    Transcopyright (cf. Free Software Foundation's copyleft concept)


Include the following HTML transquote-string in the SOURCE of your Web page:

<P><IMG SRC="http://www.sfc.keio.ac.jp/~ted/tednobg.gif" HEIGHT=216 WIDTH=256>
<A HREF="http://www.sfc.keio.ac.jp/~ted/TedPicPermish.html"><IMG SRC="http://www.sfc.keio.ac.jp/~ted/TPUB/tcosymb.gif" HEIGHT=14 WIDTH=36></A></P> .

(Note: I expanded the transcopyright symbol for easier viewing)


Recent History

Nelson obtained his B.S. in 1959

PhD in 2002 from Keio University

 

What are his comments on the Web?

"I DON'T BUY IN"

"The Web isn't hypertext, it's DECORATED DIRECTORIES!"

"What we have ...is the ...victory of typesetters over authors"

"...the most trivial form of hypertext that [can be] imagined"

"The original hypertext project, Xanadu®..."

"Xanadu has... a pure structure of links"

"and facilitated re-use of content"

"Instead, today's nightmarish new world is controlled by "webmasters", tekkies unlikely to understand the niceties of text issues and preoccupied with the Web's exploding alphabet soup of embedded formats.  XML is not an improvement but a hierarchy hamburger.  Everything, everything must be forced into hierarchical templates!"

 

"the "semantic web" means that tekkie committees will decide the world's true concepts for once and for all."

 

"Enforcement is going to be another problem :)"

"It is ...strange ... but...people...think that's how it must be"

"There is an alternative"

"Markup must not be embedded."

"Links must go both ways."

"All these fundamental errors of the Web must be repaired.  But the geeks have tried to lock the door behind them to make nothing else possible."

"We fight on."

"More later."

*Note*: he may be right about technical needs

*But*: his way of thinking seems warped

    (he may be exaggerating a bit, one hopes)

And in another URL...

"The most important thing is to re-introduce the concept of

    deep quotability [in] hypertext"

 

"I have worked out, with my colleagues...

formats that allow quotation in any part."

"Naturally they are completely incompatible with current Web browsers"

"We hope to fix that."

"using these formats, it is ...possible to open quotations in a window"

 

"we hope to have a browser plug-in one of these days that can assemble quoted portions taken from different documents."

"To practice what I preach,"

 

" I will be publishing mostly in text format"

"It may not look as pretty as some stuff that's out there, but I prefer to offer this deeper functionality."

(Referring to a special quote-friendly server)

 

 

 

    ZigZag

        Visualization that is "equivalent to arbitrary graph structures with colored arcs"

        Good for future programming languages, etc.

http://xanadu.com.au/

    Docuverse

        A "universe" of documents and the system that manages them

        Example:

            The Xanadu docuverse

            Perhaps the Web is a docuverse

    udanax.com

        Site for open source Xanadu code

        Why "udanax"?

http://www.sfc.keio.ac.jp/~ted/XUsurvey/xuDation.html

    Most up-to-date survey paper

    Xanalogical structure

    Bidirectional links

    Profuse links

    Content links (survivable deep links)

    Transclusion

    Parallel documents

    Transpointing windows

Transpointing windows: screen shot of the 1998 working demo

 

Fig. 5.  Screen shot of transpointing windows by Ka-Ping Yee, showing his PYXI viewer served from Udanax Green server (xu88 model).  Only transclusions happen to be shown, though PYXI also handles content links.

...early and final drafts of the Declaration of Independence (111), highlighting transclusions and differences with color.

 

Why Have Transpointing Windows?

Supports "parallel commentary"

Supports document comparison 

    (e.g. declaration of independence figure)

Useful whenever documents must be seen in relation to each other

    scholarship, legislation, diplomacy, etc.

Comparison of successive document versions

    Better than what Word does?

To assist in reorganizing during complex rewrite tasks

    Claims reorganization is central to complex rewrites

    My experience: yes, but would transpointing windows help? 

 

 

Pullacross editing

Fig. 6.  Pullacross editing is another use of transpointing windows (simulated graphic).

 

Transclusion Vs. Content Links

    Transclusion is an approach to reuse

    Involves giving credit, payments, permissions, versions...

Content Links are "better" hyperlinks

Transpointing windows can show transclusion or content linking

Example of transclusion:

    An HTML document with frames

        ...Transcludes other documents by dynamically bringing them into the individual frames

        Is that potentially a copyright violation?

        Is there an easy fix to this?

 

 

 

 

Content links

    (The following is modified from Nelson's paper. Since the Web does not support content links, I cannot conveniently give the exact details of what the changes are, however they are relatively minor in my judgment.)

Adam creates document A, version 1. 

It has 300 characters, which have consecutive universal addresses. 

The text contents of document A v.1 are therefore registered as having consecutive addresses from 1 to 300.

 

 

 

Fig. 9.  Screen visualization of Adam's document A,
still an unbroken stretch of content bytes.
(Numbers on the side are for following further examples.)

This document is internally represented as a table of pieces. 

It currently has only has one piece

    So the table currently has only one reference pointer

        ...which points to the consecutive characters 1 to 300.

    Note this pointer *spans* text in contrast to HTML links

Fig. 10.  Internal representation of Adam's document: a list consisting of one reference pointer, since it references a consecutive span of content.

Let's make a CONTENT LINK

Barker publishes document B, version 1. 

The characters in document B v.1 are registered  with permanent addresses 301 to 517

Fig. 11.  Screen visualization of text portion of document B  v.1. 

Barker comments in his document on Adam's

The text of the comment consists of characters 351 to 400 in B

They comment on characters 101 to 200 in A

Barker uses a content link to make this explicit

Figure 12.

The link is from an entire span on the first side to an entire span on the second side

Note that anyone from anywhere can make a content link to any published content

One way this might look on the screen: a stripe between both documents (fig. 12).

 

Fig. 14.  Virtual contents of Barker's document are the text and the link.
 

 See also the next several figs. in the paper.