Content

Or: Content, really

Knowledge is how we make life better. We learn about seeds so we can grow food. We learn about the weather so we can protect ourselves, be more comfortable, and be better at growing food. Within the confines of this collection of web data you can expect to enhance your knowledge about the topics listed below. These topics are covered in summary. See the Resources for more information.

1 The nature of knowledge
2 The evolution of knowledge
3 Enhancing knowledge creation
4 Criticism
5 Summary

As a user of this system you can empower yourself to make changes to selected parts. This provides an opportunity for the system to be improved or corrupted, again demonstrating some of the opportunities and dangers in the types of knowledge tools discussed herein. If you would like to make changes and you have not already read the Howto information, please do so.

1 The Nature of Knowledge

Douglas Foskett, former Director of Central Library Services at the University of London provides an elegant model for understanding data, information, knowledge and wisdom [foskett_fruitful]. In that model there is a path from raw phenomenon-based data, through information and knowledge, to wisdom. The model is represented by a pyramid with data at the bottom and wisdom on top.

In the pyramid data are "the single items of consciousness and thought" at the "first stage of cognition" [foskett_fruitful]. Data are perceptions of phenomena.

If data is processed or organized into some form of representation it becomes information. Information is data which have been:

organised by the relations of one to another into systems of ideas...The distinction is real because the process of organising is real and requires a conscious mental activity. It is possible because we have the power of the imagination: we can convert the isolated percepts into organised concepts, and this is the secondary stage of cognition [foskett_fruitful].

Knowledge is created when information has been synthesized to create a sense of understanding that is held in the mind of an individual. Knowledge is information "which has been assimilated into a human mind and thus given a pattern or structure which is unique" [foskett_fruitful].

Wisdom, for Foskett, is "the distilled essence of knowledge" and "is founded on a wide knowledge and deep understanding of all the facets and events of life in a community" [foskett_fruitful]. Whereas individual knowledge cannot be communicated, its essence can be passed on to a community as wisdom, as an understanding of what is good and what is not good.

Foskett portrays his pyramid as a one-way path, but it is clear that concepts at play along the path can be labeled differently depending on how they are passed. Data may be processed into information and then learned into knowledge but that same information may also become data for some other type of organization.

Context and synthesis are very important to this model. Being able to adequately and appropriately place some data in a context helps to make it understandable information. With synthesis of the information and with additional context we achieve knowledge. That knowledge is our understanding of the universe. It helps us to understand and operate in the world. Knowledge (created from context and synthesis) provides us with additional context and models that allow us to apprehend the unknown in a mechanism like this: "I don't know what this is but I know something like it so I have something to compare it to. I have a handle on it."

[ Top ]

2 The Evolution of Knowledge

Eric Drexler, the chairman of the Foresight Institute, believes that in order for society to adequately deal with the problems and challenges that will come in the near future, humans must be more able to create knowledge. In his model knowledge is created and sustained by an evolutionary process:

Knowledge is valuable and grows by an evolutionary process. To gain valuable knowledge more rapidly, we must help it evolve more rapidly.

Evolution proceeds by the variation and selection of replicators. In the evolution of life, the replicators are genes; they vary through mutation and sexual recombination and are selected through differential reproductive success. In the evolution of knowledge, the replicators are ideas; they vary through human imagination and confusion and are likewise selected through differential reproductive success - that is, success in being adopted by new minds [drexler_hyper].

Foskett and Drexler both see knowledge as a process. Information or ideas enter the world, are compared with other ideas or information resulting in the creation of new ideas or information in an endless cycle going down through history. This is fairly close to the definition of the Hegelian dialectic wherein a thesis (an idea) upon reflection leads to and becomes associated with its antithesis (an opposing idea); but yet more reflection leads to synthesis: a new idea that encapsulates and accounts for the thesis and antithesis. The new idea becomes the launching point for repeating the process: thesis to antithesis to synthesis [philosophy_pages].

To enhance knowledge creation, as Drexler proposes, we must accelerate the process. That is, the following processes must be made more efficient:

• generation of information or theses
• organization or representation of the information (getting it in front of people and placing it in a context)
• generation of additional information which contrasts with or allows comparisons to the original information
• synthesis of information at an individual level

To improve critical discussion and the evaluation of knowledge, we can seek to improve the variation, replication and selection of ideas. To aid variation, we can seek to increase the ease and expressiveness of communication. To aid replication, we can seek to speed distribution, to improve indexing, and to ensure that information, once distributed, endures. To aid selection, we can seek to increase the ease, speed, and effectiveness of evaluation and filtering. The nature of media affects each of these processes, for better or worse.

[ Top ]

3 Enhancing Knowledge Creation

3.1 Emanuel Goldberg
3.2 Paul Otlet
3.3 Vannevar Bush
3.4 Ted Nelson
3.5 Doug Engelbart
3.6 Tim Berners-Lee
3.7 A Knowledge Enhancement System

The first known references to technological systems that contain aspects of what is now known as hypertext start to show up in the 1930s and 1940s. At that time microfilm was showing that it was possible to maintain great volumes of information in a very small space. People speculated that if it were possible to produce and access these films quickly then research would be greatly accelerated. Three names stand out: Emanuel Goldberg, Paul Otlet and Vannevar Bush.

3.1 Emanuel Goldberg

Emanuel Goldberg was a Russian native who emigrated to Germany in the early 1900s. He was responsible for developing very high resolution microfilm. At the 1931 International Congress of Photography he presented a paper, "The retrieval problem in photography" that is, "perhaps, the first paper on electronic document retrieval and describes what seems to have been the first functioning document retrieval system using electronics" [buckland_goldbush].

3.2 Paul Otlet

Paul Otlet, responsible for starting the International Federation for Information and Documentation, wrote a book in 1934, Traité de Documentation, in which he envisioned a machine that would act as a scholarly workstation. The workstation would have access to remote resources that could be presented on the workstation. In an article summarizing Otlet's (extensive) contributions to information science Boyd Rayward says:

Similarly, he thought it should be possible to add long distance, as it were, to existing texts held remotely and to do this in such a way that the original texts were not disturbed [rayward_otlet]

We should have a complex of associated machines which would achieve the following operations simultaneously or sequentially: 1. Conversion of sound into text; 2. Copying that text as many times as is useful; 3. Setting up documents in such a way that each datum has its own identity and its relationships with all the others in the group and to which it can be re-united as needed; 4. Assignment of a classification code to each datum; [division of the document into parts, one for each datum, and] rearrangement of the parts of the document to correspond with the classification codes; 5. Automatic classification and storage of these documents; 6. Automatic retrieval of these documents for consultation and for delivery either for inspection or to a machine for making additional notes; 7. Mechanized manipulation at will of all the recorded data in order to derive new combinations of facts, new relationships between ideas, new operations using symbols [buckland_goldbush].

3.3 Vannevar Bush

Vannevar Bush was the Director of the Office of Scientific Research and Development, a World War II organization that coordinated the research efforts of approximately 6000 scientists. After the war he wrote an article, which, despite Goldberg and Otlet's contributions, is heralded as the predictor of hypertext. The article, "As We May Think" [bush_think], discusses his version of a scholarly workstation called the memex.

Bush believed the memex was important because it would allow researchers to do their work (of generating knowledge) more effectively:

There may be millions of fine thoughts, and the account of the experience on which they are based, all encased within stone walls of acceptable architectural form; but if the scholar can get at only one a week by diligent search, his syntheses are not likely to keep up with the current scene. [bush_think]

The important part of the memex is its ability to create trails. The memex

affords an immediate step, however, to associative indexing, the basic idea of which is a provision whereby any item may be caused at will to select immediately and automatically another. This is the essential feature of the memex. The process of tying two items together is the important thing [bush_think].

That--the notion of trails--is hypertext: associative linkings between things. Any things.

3.4 Ted Nelson

The term hypertext was introduced by Ted Nelson in 1963 "to mean a body of written or pictorial material interconnected in such a complex way that it could not conveniently be presented or represented on paper" [wordnet, hypertext]. Nelson's "principal insight was in place by the end of December 1960. It was this: if text and other media are maintained as referential structures, the resulting structure will have numerous powerful advantages over merely moving the contents around" [xanadu_tech].

This notion deserves some explaining but more important is what it allows (a more detailed explanation of Xanadu's referential structures will be added if time allows). Nelson took his insight and began a 40-year crusade to develop a full_hypertext publishing system called Xanadu. He is the hypertext pioneer, the nexus of all the possibilities, but unfortunately his pursuits have resulted in little. Much of his work has been overshadowed and corrupted by more recent developments. The existing world wide web is, as they put it on the Xanadu website, "a diabolical dumbdown of our original hypertext idea" [xanadu_web]. Nelson's confrontational style has probably not helped his crusade, but that's how crusaders are.

3.5 Doug Engelbart

Later in the 1960s Douglas Engelbart was also having visions of how technology, especially computer-based technology, could be used to enhance knowledge. Engelbart is primarily remembered for his development of the mouse but that was simply a side effect of a far greater vision to "increase the capability of people to deal collectively with urgent complex problems" [softky_engelbart]. In the 1960s he pursued his research at the Stanford Research Institute where he and his team "developed hypertext linking;...real-time text editing; integration of text and graphics; on-line journals; shared-screen teleconferencing; and technology that allowed people to collaborate on problems from different remote locations" [softky_engelbart]. Today he pursues his vision at the Bootstrap_Institute.

3.6 Tim Berners-Lee

In 1990, while working at CERN, Tim Berners-Lee convinced the management to devote some resources to the development of a network- transparent hypertext data access system he initially called "Mesh" [berners_lee_info]. While working on the code he changed the name to World Wide Web, the system you are using today. The World Wide Web was primarily designed as an information management tool and thus falls shorts of the knowledge enhancement tools envisioned by Otlet, Bush, Nelson and Engelbart and described by Drexler. The most noticeable difference is that while the World Wide Web does have the linking behavior of hypertext the links are uni-directional. In other words the hypertext in the World Wide Web is semi_hypertext and not the full_hypertext required to meet the requirements outlined by Drexler. (There are many other things wrong [pam_where].)

3.7 A Knowledge Enhancement System

So what, then, comprises a knowledge enhancement system that would satisfy Otlet, Bush, Nelson, Engelbart and Drexler? Drexler lists several requirements for enhancing expression, transmission and evaluation. The system must:

• provide access to a hypertext publishing systems that has the following attributes:
  • full_hypertext: links which can be followed in both directions and which can be attached to documents without the author being involved (but the author can know)
  • fine_grained_hypertext: short works can be effectively published and links can be made to and from any object in a document, not just chunks defined by the author
  • public_hypertext: the system is open to an indefinitely large community and geographically and organizationally dispersed and diverse
  • filtered_hypertext: the system has the ability to filter links based on arbitrary, definable, rules
• support effective criticism
• serve as a free press, including anonymous reading
• handle machine-use charges so the system is funded
• handle per-use royalties so there is incentive to contribute
• support filtering so that low-quality material does not choke the system [drexler_hyper]

• alter publications, except by attaching commentary
• hide commentary
• skirt royalty payments
• monitor who is reading what
• steal identities [drexler_hyper]

The system enhances expression by lowering the barriers to publishing. Anyone can publish for a very low cost in a very short amount of time. If someone has a response they can quickly attach a comment, again at very low cost.

The system enhances transmission by existing as a ubiquitous network entity. Documents are available anywhere the network can reach.

The system enhances evaluation by allowing ideas to be exposed to wide criticism and placed in contrast with other ideas that are supported by context that is easy to access and assess.

Thus such a system, if well implemented, would accelerate knowledge evolution. Why isn't it here? There are many reasons but most prominent is the presence of the existing World Wide Web. Many people are not interested in discussing hypertext because we already have it and it must be wonderful because we are in the midst of a revolution. That's simply not the case: the revolution has yet to begin.

[ Top ]

4 Criticism

All ideas fall on their face from some perspective. The idealism rolling around here is all well and good but how do things look from either the real world or the squishy world of critical theory?

4.1 Authorship and Authority
4.2 Power and Knowledge
4.3 Sound and Fury
4.4 Capitalism Rears its Ugly Head
4.5 Digital Divide
4.6 Technical Divide

4.1 Authorship and Authority

Much of the criticism aimed at systems that are deeply hypertextual tend to be associated with the way in which the notion of "who wrote this" can be lost. As you navigate amongst a collection of links there is no immediately apparent way to determine if the author of the current text is the same as the author from the previous side of the link. This is especially true if the system provides fine-grained links into the middle of documents. You may be able to make assumptions but you can't necessarily be sure.

If we decide what purpose an author serves, that can help us to understand how this purpose can be filled in hypertext. If we think about how hypertext has filled these purposes thus far we can have a better understanding of how the hypertext pioneers think about authors.

For most people identifying the author of a given work allows them to make some judgments about the quality and appropriateness of the work. We can look at this as an aspect of Foucault's author function as discussed in "What is an Author?" [foucault_author]. In much modern criticism there is an effort to downplay the notion of the author and the author's intent in interpretation of a discourse. It is replaced by the primacy of the text (or sometimes the reader, but it is the reader responding to the text, so in a sense the text is still primary) and the multiple layers of interpretation that can be found in the text by unfolding the different aspects of significance. This replacement, however, does not remove the author but recenters the author away from a person who wrote a text to an ephemeral function existing in a cultural context. Or as Foucault puts it "to a transcendental anonymity" [foucault_author, 141].

Foucault wonders what's left in the space where the author used to be. Readers use the notion of author, so if it is not the engine of meaning behind a discourse, what is it? An author is not the proper name that has signed a work, because some works have authors and others do not (books versus contracts, for example). Foucault settles on the notion of an author function:

In this sense the function of an author is to characterize the existence, circulation, and operation of certain discources within a society [foucault_author, 142].

Is it possible, then, that the notion of author and authorship as a conceptualization of a proper name in a hypertext system is not as important as the author function as described by Foucault? Consider this web site. I, the writer of this text, am essentially unknown. To attach the name Chris Dent to this text does little (amongst those who don't know me) to enhance the value of the information. The value of the information is present from the way in which it may be internally consistent, coherent and supported by its external sources. As it is read it may gain a reputation that enhances its apparent value. Instead of my name reflecting value onto the text, the text may reflect value onto me thus creating an author function that is somewhat associated with the person perceived to be the author of this text. Chris Dent is not the author. The author is some thing named Chris Dent associated with a work that gained a reputation for being valuable. We'll see if that happens.

Furthermore, the text that is down inside the warp engine--the text of the definitions--is openly acknowledged to be fully anonymous. It may or may not be written by the author function with the label Chris Dent. There is, though, a community that is associated with the text: people who have been granted access to make changes. As above, if the text in the definitions is perceived to have value then that community becomes an author function and is granted a positive reputation. Identification of the author then becomes a function of identifying the community of people for whom access has been granted or identifying the guidelines by which access is granted.

So, if an author serves a function that grants some dynamic measure of authority, a hypertext system does not necessarily need to provide proper name identification of the writer of some text. It does, though, need to provide some way of identifying how the text came to be where it is. That is, the cultural context under which the text was generated needs to be apparent.

The current world wide web hypertext system doesn't make that context very clear. Xanadu attempts to address some of these issues in its design, especially with the notion of a transclusion.

The hypertext pioneers care more about the author function than about authors as proper names. In fact, it can be said that the hypertext pioneers agree with text having primacy over author. What is produced and the value of that product is important. Or to put it another way, what is said and whether what is said is valuable is of far more importance than who said it because the goal of the system is to inform the reader, not satisfy the needs of the author or publisher.

[ Top ]

4.2 Power and Knowledge

- those who hold the power must control expression
- if content is more important than author, power is lessened for the
  author
- people used to being authors want to stay authors
- people are used to having domains of control
  - existing network based publishing systems preserve the notion of
    a "place" where information lives and that place having an owner
- bidirectional linking destroys author control
- transclusions and transcopyright in Xanadu attempt to address this

4.3 Sound and Fury

- if there is easy access to publish, a great deal of garbage will be
  published
- but the costs of storing garbage are low so if only a small % of the
  garbage is worthwhile that is still a lot of good stuff
- drexler's notions on reputation and filtering [drexler_hyper]
- current web situation not indicative of how content should be
  managed. lack of bidirectional linking means that individual pages
  are difficult to judge.

The author function concept (described above) supports much of the
positive energy associated with full_hypertext system. In those
systems there is hope that author-reputation associations will allow
people to effectively filter content. There is also a desire to allow
anonymous and pseudonymous contributions. Both can work well if the
notion of an author function, disassociated from a "real" person, is
accepted. 

A great deal of the issues associated with reputation can be seen
in action at web sites that allow commentary. Slashdot is a good
example.

[ Top ]

4.4 Capitalism Rears its Ugly Head

- xanadu transcopyright
- other notions of "pay per view"
- is the micropayment notion inevitable?
- must there be a market for information?
- evolution of knowledge implies a marketplace of ideas, but does that
  mean there must be payment?

4.5 Digital Divide

There is an implicit elitism in the concept of knowledge evolution. The concept suggests that tools should be built to allow people to share information so they can create more knowledge that they will use to improve their lives or create yet more knowledge. Being able to do that requires access to an initial tool and knowledge set that many people do not have. Those with access have an advantage that increases with use, widening the gap between the knowledge haves and the knowledge have-nots. Is it not possible, though, that creation of such tools, if done with care, can help to narrow the gap? Just as public print libraries provide access to vast collections of knowledge so too can public hypertext systems provide access to potentially much larger collections of knowledge. In addition to providing access, such systems can also facilitate expression for those who previously did not have the opportunity. We see much of this already, on a limited scale, with the world wide web.

[ Top ]

4.6 Technical Divide

Some readers may be saying, "Hey, we've got hypertext. We've come a long way baby." There's some truth to that but only some. Current widely available hypertext systems are a far cry from the systems hoped for by Otlet, Bush, Nelson, Engelbart, Drexler or other idealists.

The World Wide Web (WWW) has done a great deal to enhance the dissemination and accessibility of information. Taken by itself it is a great leap forward, with the benefits yet to be fully recognized. Compared to full_hypertext systems it is a letdown. The semi_hypertext present in the HTML and HTTP combination travels only a short distance towards enhancing knowledge evolution as described by Drexler. This is because the links are only unidirectional. Why is that?

Plenty of theories about control, power, and fear can be thrown forward to explain why semi_hypertext came to the fore. Many of these are probably quite accurate but there is an additional reason that comes from a misapprehension of the technical situation.

XML came along in response to the difficulties caused by HTML's intermingling of markup for presentation, structure and linking. Unfortunately XML continues to contain one of HTML's flaws: the notion of hypertext and linking is maintained within the document. This means that making links from the document to other resources is within the domain of the document and thus under the control of either its publisher or author.

A full_hypertext system must separate the notion of the document, of content, from the linking system within which it lives. This separation implies that hypertextual linking exists at a layer above documents. It's a conceptual difference that is more fully aligned with Bush's notion of a trail created by his memex. This separation has many implications:

• Documents may be stored anywhere on the network and be referenced from anywhere on the net.
• Collections of references are themselves documents so they can be shared, referenced and annotated.
• Bi-directional links are easy because they exist outside of documents so can reach where they like.
• Including sections of other documents is easy because document presentation is a function of the display tool; therefore, parts can be gathered from various locations.

A good example of a tool that creates links outside of documents, in both directions, is the CritSuite from the Foresight Institute. CritSuite pulls in documents by their URL and overlays bidirectional links and commentary.

Xanadu was designed from the outset to deal with bidirectional linking. In that system documents are maintained as referential tree structures wherein the very small parts of the document may be accessed by reference, even if the document is changed. The tree structures exist in a globally accessible address space. Links between documents exist as documents within that same address space. A link has an address and consists of two addresses pointing not just into other documents but to specific places in those documents. This architecture is elegant and conceptually simple, but Nelson has experienced great difficulty being understood over the years [xanadu_tech].

[ Top ]

5 Summary

Enhancing knowledge evolution is a revolutionary concept. It is explicitly designed to allow more people to learn more and share more in an environment where effective evaluation and criticism of ideas are the primary goals. When ideas are more important than authors and publishers those who have become accustomed to those roles are threatened. So be it: content, information and knowledge are more important to the evolution of society than control, authority and pride.

Knowledge enhancement can help to bring about an environment where understanding is more important than memorization; where students and others are encouraged to think, learn and discuss instead of memorize, organize and regurgitate; where synthesis of disparate ideas leads to new thought and new ideas.

Hypertext publishing systems will not bring about universal knowledge enhancement, but they can be tools that help to reach that goal. There is nothing new to these hopes. They have been around for thousands of years. The ideas, as stated here, are the result of a natural evolution of ideas, tuned and refined through expression, transmission and evaluation.

Previous is Rationale
This is Content
Next is Architecture