“Hypermedia” is defined by the Oxford English Dictionary as “a method of structuring information in different media for a single user… whereby related items are connected in the same way as a hypertext.” The term “hypertext” was coined in 1965 by Ted Nelson, who defined it as “non-sequential writing—text that branches and allows choices to the reader.”  Nelson similarly coined the term “hypermedia” to refer to non-textual information, such as images, movies, and sounds, that is connected in this way.
The literal meaning of the word, from the Greek prefix hyper (“over, above, beyond measure” ), suggests a super-text, a beyond-text, or, for “hypermedia,” super-media or beyond-media. A “hypertext” differs from a conventional text in its organization. While a conventional text, such as a book or magazine article, has a sequential structure that is determined by the author, a hypertext has a fluid structure that is determined by its reader. The hypertext-reader navigates as he or she chooses through linked “chunks” of information, which can be viewed in any order.  A hypertext can be experienced in countless different ways, and it allows the reader a level of control over his or her reading experience that isn’t possible with a conventional text.
Many authors use the terms “hypertext” and “hypermedia” interchangeably.  As the OED definition suggests, the distinction between the two words lies in the content of the document (whether it is “text” or another medium), rather than in the hyper-organization. Espen Aarseth distinguishes a “hypertext” from a “cybertext” (a term coined by mathematician Norbert Weiner in 1948 and recently reclaimed by new media theory) in the way that each engages the reader: a hypertext can be read in any way, whereas a cybertext contains rules that must be followed to produce a “successful” reading.  (An example of a cybertext would be a narrative videogame, which can be won or lost depending on the player’s ability to navigate according to the rules.) N. Katherine Hayles distinguishes “hypertext” from “cybertext” through the connotations the words carry. “Hypertext” suggests an emphasis on linkages and a more literary understanding of “text,” whereas “cybertext” suggests a more computational approach to analysis and inclusion of less conventional works (such as videogames) as texts6.
The concept of a “hypertext,” or of hypermedia, originated with engineer Vannevar Bush, who, in the 1945 article “As We May Think,” argued for a better means of organizing and locating information. Bush envisioned a device, the “memex,” which would allow a user to track freely through a variety of information sources. The memex would contain a wide selection of materials on microform and would allow users to freely move from one text to the next, adding notes and making connections between sources as they choose. The machine remembers and archives the user’s links and comments as they are made, so that the user can always return to and continue a particular train of thought. Bush argues that such a machine would facilitate research and thought far better than contemporary indexing systems because it, unlike existing systems, would operate in the same way the human mind does, snapping from one idea “instantly to the next that is suggested by the association of thoughts” .
In his 1981 book Literary Machines, hypertext theorist and pioneer Ted Nelson outlined a similar system, arguing that computers should be structured to facilitate human intellectual inquiry—not vice versa.  Nelson writes that ideas, literatures, always function in terms of “a system of interconnected writings” , both explicitly, in quotations and citations within a document, and conceptually, in the works that inform the document. No work is truly independent: new ideas build on old ideas, old theories are reformulated and revised as new theories, old stories are retold and revisited as new stories.  An ideal organization of information would make these connections between works clear. Instead of forcing a reader to hunt through journals in order to find a referenced work, a good hypermedia system should offer related works simply, via links. More importantly, works should always be ensconced in related documents. New or altered versions of a particular work, or documents that directly address or revise a particular work, should be clearly grouped together.  Nelson’s proposed hypertext /hypermedia system would reify connections between documents and more easily allow readers to follow a progression of ideas.
George P. Landow notes that textual innovations along the lines of hypermedia, meant to make texts more easily navigable, have existed for centuries. Manuscript scribes introduced countless innovations over the ages to make reading easier: individual pages and numbers, discrete paragraphs and sentences, titled chapters, and tables of contents and indexes.  Landow notes that most conventional texts function as hypertexts, situated among other extant texts by notes and references, which readers may choose to read, or not, at their discretion.  Unlike a text that incorporates notes, references, and quotes, though, true hypermedia obliterate the distinction between primary and subordinate texts (what Lev Manovich refers to as a “master-slave” relationship) , and allow texts to co-exist at equivalent levels of authority . In a hypertext system, a reader who peruses an article and finds an interesting citation is able to instantaneously call up the cited text in full, and can read it alongside the original—or drop the original altogether.
Landow and others  have linked hypermediation with modern critical theory, particularly the work of Roland Barthes. Barthes describes his ideal text as containing “networks [that] are many and interact, without any one of them being able to surpass the rest… [the text] has no beginning; it is reversible; we gain access to it by several entrances” . Barthes’ ideal text is “writerly,” rendering its reader “a producer rather than a consumer of the text” . Rather than having an order and meaning clearly imposed upon it by its author, the ideal text invites its reader’s maximum participation, allowing him or her to develop his or her own connections and find his or her own meaning within the words.
Like Barthes’ open, “writerly” text, hypermedia allow for the maximal involvement of their readers. Not only is the reader or user allowed to navigate through interconnected chunks of information as he or she pleases; he or she can tailor the experience to match his or her preferences. In the current interface for the dominant Hypermedia system, the Inernet, for example, a reader/user/viewer is free to adjust a browser’s text size or window size; he or she can open multiple pages at the same time and view them simultaneously; can view pages with or without music playing in the background, can allow a website to customize itself to him or her by logging in to it, or by activating digital “cookies.” As Lev Manovich writes, “a number of different interfaces can be created from the same data” . Katherine Hayles reinforces the fact that different trappings do in fact alter the meaning of the text: “The physical form of the literary artifact always affects what the words (and other semiotic components) mean” . The content of a hypermedia work remains fixed, but the experience of this content varies according to the user and his or her preferences. A hypertext author can never be sure that a reader will see all of the content within a document, or that he or she will progress through the text in a particular way.
In the realm of art, hypertext/hypermedia allow for the creation of new kinds of artistic and literary experiences. In Michael Joyce’s hypertext tale “afternoon, a story,” for example, the meaning of the narrative changes radically depending on whether the reader follows the default arrangement of information or explores the text more carefully . Videogames and computer games, which Espen Aarseth has argued are forms of “ergodic literature” , frequently allow the player to influence the game’s narrative, creating multiple paths through the same game-world. In the world of print, recent novels such as Don DeLillo’s Underworld and David Foster Wallace’s Infinite Jest act as hypertexts through their copious footnotes and in-text “links,” which allow readers to recall and interrelate portions of the text, or delve deeper into the narrative, at their discretion.
Lev Manovich sees hypertextuality as characteristic of postindustrial society. In today’s world, citizens are no longer bound by stringent, uniform morals and expectations. Instead, “every citizen can construct her own custom lifestyle and ‘select’ her ideology from a large (but not infinite) number of choices” . As the number of choices available in the social milieu increases, so, Manovich argues, does the desirability of choice in other aspects of life, such as the reading of texts. Rather than wanting to experience a text, a film, or a website in the same way everyone else does, individuals in postindustrial society prefer experiences that are custom-tailored to themselves. By Manovich’s rubric, hyper-organization will be increasingly demanded by readers in the presentation of information.
The driving concepts underlying hypermedia follow Marshall McLuhan’s idea of media as extensions of human beings . Bush named his machine the “memex” because he saw it as an extension and aid to human memory—tracking a user’s train of thought and permanently “remembering” his or her responses to various texts . Later researcher Douglas Engelbart, who invented the first hypertext system and helped to pioneer graphical user interfaces for computers, similarly viewed hyperlinked information as a means of extending thinking and problem-solving abilities (“augmenting intelligence”) . More recently, Ted Nelson has argued that hypertext systems can and should take their cues from the human mind, physicalizing and improving the thinking process by reifying connections between texts and making the development of ideas easier to follow .
Hypertext and hypermedia offer unprecedented choice and fluidity in approaching information, allowing connections to be made more quickly and in a more personalized way than ever before. A good hypertext system takes intellectual inquiry as its model: as Ted Nelson writes, “There is no Final Word. There is always a new view, a new idea, a reinterpretation” . Information stays the same, but the way it is viewed, interpreted, and interacted with changes constantly. Given this shifting nature, hyper-organization is an ideal means of organizing information: it offers access that is as fluid and flexible, as in-depth or as shallow, as specific or as broad, as its user wants it to be.