Hypertext : Encyclopedia of New Media

Hypertext is a form of electronic writing that allows writers to link paragraphs or pages of text using computer-mediated connections that, in essence, enable computer users to draw relationships and make connections between discrete ideas. In the words of hypertext inventor Ted Nelson, however, the definition is much simpler. In his influential 1974 treatise, Computer Lib/Dream Machines, Nelson described hypertext simply as “non-sequential writing.”

However, hypertext does not involve simply scrambling the words in sentences to create inscrutable babble. Hypertext, which Nelson devised and named in the early 1960s, was initially his attempt to align the processes of writing and reading more closely with the processes that he believed are evident in human thinking. Although it was not exactly Nelson's intention that it should be so, the underlying engine of today's World Wide Web is a universal computer language called hypertext transfer protocol (HTTP), and hypertext is the Web's primary driver. The Web uses highlighted hypertext words and phrases, often underlined, that can be clicked on with a computer mouse to allow users to leap from Web page to Web page, skipping passages or pages that are not of interest to explore ideas that are interesting. These hypertext links, or hyperlinks, allow users to engage in informational flights of fancy based entirely on their unique interests, assuming only the presence of hyperlinks on the page.

NELSON DISCOVERS HYPERTEXT

In shaping hypertext, Nelson's fundamental insight was the notion that human thinking is non-linear. Take a fictitious example: A man named Mark is at work. Around noon, he realizes that he is hungry, and the image of a hot dog forms in his mind. From there, his thoughts flip to his pet terrier Jack, who, Mark suddenly remembers, was never let out for his morning toilet run. From there, of course, Mark's thinking moves to the idea that he will probably have a big, disgusting mess to clean up in his apartment that evening. Mark has forgotten all about his appetite; now he is thinking about cleaning solvents.

In strictly logical terms, there is no clear linear connection between the idea of a hot dog and the idea of cleaning solvents (at least none that we know of). But strictly logical links are not necessary for thoughts to emerge. Mark's mind followed his own internal mental links, a trail of thought unique to his own perspective, life situation, and interests. His mind followed that path from one concept to the other fluidly, in the span of seconds. Non-linear links like these are the basis of much of, if not most of, human thought.

However, as Nelson noted in Computer Lib/ Dream Machines, this is not how writing usually works. Writing traditionally has been bound inflexibly to the printed page, in linear structures that can be broken only by flipping through a book or following footnotes away from the main text, and these are not particularly efficient or convenient methods. “Ordinary writing is sequential for two reasons,” Nelson wrote. “First, it grew out of speech and speech making, which have to be sequential; and second, because books are not convenient to read except in a sequence. But the structures of ideas are not sequential.”

This is especially true of the infamously disorganized Ted Nelson. According to Roger Fidler's 1997 book Mediamorphosis: Understanding New Media, Nelson's breakthrough with hypertext was the direct product of his frustration with his own inability to stay organized or to keep track of his written notes on standard 3 x 5 index cards. “For him,” Fidler wrote, “all methods of paper to manage information seemed inadequate and imposed restrictions that masked the true structure of his ideas.”

Nelson has had little formal computer-science training, but nonetheless has achieved prominence in the pantheon of computer scientists because of his discovery, which he first arrived at in 1960, at age 23, while a student at Harvard; the innovation was part of a term project intended to create a “writing system,” which featured the ability to use computers to compare alternate versions of texts side-by-side, backing up sequentially through earlier versions of documents, and revising by outline.

To describe his idea, Nelson coined the terms “hypertext” and “hypermedia” (the latter of which extends the concept to digital sound, pictures, and video). The terms were first used in a paper that Nelson delivered at an Association for Computing Machinery (ACM) conference in 1965, wherein he described a nearly Utopian dream of a global “docuverse” that he called Xanadu.

Xanadu, a conceptual precursor to the World Wide Web, reflected Nelson's dream “for everything to be in the hypertext,” but it never got off the ground. What's more, Nelson was slow in showing how hypertext could function practically. Hypertext was not actually demonstrated until Andries and Nelson van Dam showed the IBM-funded Hypertext Editing System at Brown University in 1967. Nevertheless, there is no serious dispute that hypertext is Nelson's discovery.

“AS WE MAY THINK”

Nelson does, however, acknowledge being inspired by the influential 1945 Atlantic Monthly article “As We May Think” by Vannevar Bush, a key science adviser to President Franklin Delano Roosevelt during World War II. The article, which was written just as the war was ending and just prior to the advent of digital computers, foreshadows with astonishing clarity the underlying hypertext concept.

Bush's “Memex” (memory extender), as he called it in his article, was a mechanical machine, not a digital one, and it was based on technologies that were either in existence or just emerging in 1945. Rather than discussing the highlighted point-and-click techniques that came to characterize hyperlinks, Bush's Memex involved a special desk affixed with view screens capable of selecting from vast stores of miniaturized microfiche documents and instantly zooming in on the one that was wanted. The Memex would use a system of “dry photography,” allowing its owner to input new records at will. It employed a typewriter-styled keyboard, through which special codes (recorded in an easily used index) could be inputted, telling the machine which document to display. A lever on the desk would then be pushed, instantly drawing up the desired document. Through a kind of mechanical memory, Memex users could traverse a unique path of information according to their interests of the moment, and the lever could be used as the equivalent of the “Back” and “Forward” buttons on the modern Web browser, allowing users to retrace their idea paths. What's more, Bush proposed, that document trail could itself be stored and shared.

The similarities of this system to Nelson's hypertext concept are obvious, despite its mechanical foundation and its comparative limitations. A researcher could not cut and paste material from a microfiche into a written document as easily as today's Web surfer can when drafting a paper using Microsoft Word, for example.

Bush had high hopes for the Memex. He wrote that it could allow users to store ideas confidently, and to know exactly where and how they could be retrieved if they proved useful later. Meanwhile, the mind's faculties could be freed up from the task of trying to remember where old documentation was stored and how to get at it. This, in turn, would free the human intellect to focus on what it does best—generating new ideas.

THE VISION OF DOUGLAS ENGELBART

Computer-science engineer Douglas Engelbart, head of the Augmentation Research Center in San Francisco, was fascinated by the ways that computing technology could be used to augment and increase human intelligence, and as a result his work brought the concept of hypertext and hypermedia, indeed virtually of all personal computing and community networking, into sharp focus. Engelbart staged a 1968 demonstration at the Fall Joint Computer Conference in San Francisco that Paul Saffo, director of the Institute for the Future, would later say was no less remarkable than if Engelbart had landed a UFO on the White House lawn.

At that conference, Engelbart demonstrated his “oNLine System” (NLS) in a 90-minute multimedia computer presentation. The demo introduced the tool that is synonymous with the now ubiquitous point- and-click hypertext-selection method, the computer mouse. The NLS contained a “shared journal,” storing more than 100,000 papers, reports, memos, and cross-references, all of which were accessible through hypertext links. His presentation also included a live videoconference with staff members who were in a lab 30 miles away.

The demonstration in many ways represented the first time that computer technology was shown to be interactive. However, Engelbart's ambitions were far removed from simply creating an interactive tool for human productivity and play. His insights about how the human mind works—relying on paths, branches, and webs of thought rather than operating on a strictly linear plane—played a key role, and were very much in keeping with Nelson's ideas. If anything, Engelbart's ambition was even greater than Nelson's. Using networked computing in a hypertext environment, he theorized, would make people more intellectually effective by fostering a collaborative method of sharing their knowledge. The title of his 1962 white paper, in which he laid the groundwork for his NLS, defines Engelbart's vision fully: It was called “Augmenting Human Intellect: A Conceptual Framework.” But Engelbart's ideas were perhaps too far ahead of their time, and even today his contributions are not popularly known.

MAKING HYPERTEXT A REALITY

The use of hypertext grew rather slowly over the next decade and a half. The ZOG Project, founded in 1972 by researchers at the Computer Science Department of Carnegie Mellon University, was intended to create a front-end application for use in a summer workshop on artificial-intelligence research, and was one of the first pure hypertext systems. ZOG (the initials apparently don't stand for anything) was first made available as a commercial application in 1983 as the Knowledge Management System (KSM), a publishing-software package that allowed the creation of well-formatted paper documents from hypertext elements.

In 1978, Andy Lippman of the MIT Architecture Machine Group (later known as the MIT Media Lab), helped push the hypertext concept into multimedia by creating the Aspen Movie Map, a virtual tour of Aspen, Colorado. Regarded as the first hyper-media videodisk, it contained video images from four cameras, pointing in different directions, that had been mounted on a truck as it was driven through the streets of Aspen. The resulting pictures were compiled onto videodisk and linked to allow the user to start anywhere in the city and move forward, backward, left, or right.

Beginning in 1985, with the introduction of the Intermedia program, hypertext authoring systems capable of producing interactive documents began appearing in computers. Janet Walker's Symbolics Document Examiner, a hypertext manual for Symbolics computers that was distributed as an alternative to the 8,000-page printed manual, became the first widely used hypertext-based application. Other hypertext editing systems followed, including Guide, Linkway, Writing Space, and HyperCard, the last of which was distributed free on Macintosh computers and arguably contributed the most to hypertext's popularity. The first major hypertext convention, Hypertext '87, was organized by the Association for Computing Machinery in 1987.

THE WORLD WIDE WEB

Unbeknownst to anyone outside of CERN (Centre Européen pour la Recherche Nucléaire), the Swiss nuclear-physics lab, around this time a young computer programmer was independently reinventing hypertext. Tim Berners-Lee, an Englishman who eventually created the World Wide Web, has written in his book Weaving the Web that he was at best dimly aware of Nelson's work, and had not been exposed to Bush or Engelbart when he set to work in 1980 on a program that he called Enquire (short for Enquire Within Upon Everything).

Enquire's mission was to solve CERN's massive data-retrieval problems. The lab had immense amounts of research stored on a variety of computer operating systems, in a number of data formats. Enquire was Berners-Lee's way of solving this problem. It worked somewhat in the style of HyperCard; pages in the system resembled index cards, and the only way to create new pages was to link to old ones. Enquire had two kinds of hyperlinks: “internal” links connected data from one page in a single file to another page, while “external” links could connect unrelated files. This is not unlike Berners-Lee's later design for the Web.

Berners-Lee left CERN in the early 1980s, taking jobs as a networking consultant before returning to CERN in 1989 to follow up on the problem of incompatible computer systems and data at the research lab. In March 1989, he submitted to his supervisors a simple paper titled “Information Management: A Proposal,” which laid the groundwork for the World Wide Web. By this time, Berners-Lee was aware of Nelson's work, and made reference to it in his paper, referring to what later became known as hyperlinks as “hot spots,” and acknowledging Nelson's belief that these hot spots could work not only to connect widely dispersed text, but also to trigger multimedia applications in computers. While he had to approach his project from the limited vantage point of CERN's immediate needs, it is clear from the paper that Berners-Lee was keenly aware of the global possibilities of his proposed networked hypermedia system.

At the time that Berners-Lee started his work, hypertext was used mostly within programs and databases, allowing readers to access, for instance, one part of a database by connecting to it from information stored in another part. Berners-Lee's innovation, gleaned from his experience as a computer-networking administrator, was to use hypertext to connect computers in the outside world. This was important at CERN, where computers in different buildings and in some remote locations had to communicate. The important trick was not to make those incompatible machines conform to a single standard, but to make the data itself conform. Berners-Lee accomplished this by creating HTTP, the computer language that made it possible for hypertext connections to work over wide networks like the Internet.

Through HTTP, a computer could be linked not only to information stored within its own internal database, but also to information stored on a computer far away. HTTP is the language that established the Web address format (e.g., http:/ / www.cnn.com), and it made it possible for pages to be requested and retrieved through the Web. Of course, to make the Web truly work, Berners-Lee also had to create the first Web browser, write the program for the first Web-server computer, and create the hypertext markup language (HTML) that makes it possible for simple documents to be displayed in readable form on a Web browser once a successful document request has been fulfilled.

When the World Wide Web went public in 1992, Nelson's concept of linking everything to everything through hypertext began to be realized—although Nelson has long complained that the Web is something of a “stupid” version of Xanadu, which he imagined would make all the world's literature available online permanently in a single unified system. After the Mosaic browser was released to the public in early 1993, hypertext became popularized as the most revolutionary new form of communication technology since the advent of television—and arguably the most significant since the invention of Gutenberg's printing press, which democratized both reading and writing. The Web, obviously, has only continued to grow since its inception. Meanwhile, hypertext has continued to flourish in other interactive media, such as CD-ROMs, e-books, video games, and interactive television.

THE IMPORTANCE OF HYPERTEXT

Like Nelson and Engelbart, a number of critics and researchers have suggested that profound changes in human comprehension may result from the use of hypertext. Professors George Landow and Paul Delaney, in their 1991 essay “Hypertext, Hypermedia and Literary Studies: The State of the Art,” predict vast changes in the relationships between readers and writers, audiences and creators. Whereas the writer traditionally has been completely in charge, telling the reader exactly what to read and all but dictating the exact order in which it should be read, hypertext places more control in the hands of the reader, who no longer has to plow through pages of uninteresting detail to get to the meat of a work. If there is a link to a concluding paragraph on page one, the reader can jump directly to it, skipping everything else, or start in the middle and fan out in either direction.

These new possibilities have been explored by a number of experimental authors, notably Michael Joyce, whose afternoon, a story, is an early and famous experiment in hypertext fiction. In it, Joyce presents the blocks of text that make up a story, but allows readers to select hyperlinks to connect the prose in any order that they choose. The end result is that the reader is nearly as responsible for the outcome of the story as is the writer; the writer creates the story's framework, and the reader is set loose to explore its many possibilities.

Hypertext “changes our sense of authorship, authorial property and creativity … by moving it away from the constrictions of page-bound technology,” Landow and Delaney write. “In so doing, it promises to have an effect on cultural and intellectual disciplines as important as those produced by earlier shifts in the technology of cultural memory that followed the invention of writing and printing.”

—Kevin Featherly

Entry Citation:

Featherly, Kevin. "Hypertext." Encyclopedia of New Media. 2002. SAGE Publications. 4 Apr. 2010. .