MeatballWiki | RecentChanges | Random Page | Indices | Categories

Any medium that consists of nodes representing content and arcs representing links between content. The arcs provide no direct content, but indirect content through the wider context which they present. Taken another way, the arcs are non-linear jumps between linear (or at least internally consistent and coherent) content blocks. Over all, the entire structure is typically a directed graph, though it is possible to use a tree structure like the AnimalGame.

(See also WhatIsaMedium, WhatIsContent, WhatIsContext, AboutLinks?)

Coined by TedNelson of the XanaduProject, though the original idea was by Vannevar Bush's (1945) memex device [1].

Hypermedia, generalizing HyperText, extends HyperText Links to any set of multimedia objects, including sound, motion video, and VirtualReality. It allows a higher level of user/network interactivity than the interactivity already implicit in hypertext. Some newer hypermedia systems are even user-programmable by appropriate scripting languages or better.

An excellent practical example is the WorldWideWeb which demonstrates linear pages for the nodes, and hyperlinks as the arcs. Another excellent example is a dictionary. The nodes are the definitions, the arcs are the words used in the definitions (they refer to other dictionary entries). Another example is a "Choose your own adventure" book where the links are page references at the bottom of each mini-chapter. In this latter case, note how the links are out-of-line (not directly embedded in the content) versus in-line (directly embedded in the content). See AboutLinks? for a taxonomy of links.

Hypermedia are useful for providing structure to a set of content. Navigational hints, or an organization. This is similar to common media, as in book/part/chapter/section/subsection/paragraph or television/channel/program/segment/topic. Slightly more interesting is the added context the arcs provide. This is much like the dictionary where the arcs provide more information about the words used in the definition.

A hypermedium is an amazing way of structuring all humanity's information (like the CycProject attempts to do) because they structure information much like our minds appear to. A CollaborativeHypermedium is built on this premise. It attempts to store the information taught to it by its contributors in the most efficient way. This is anthropomorphising, but sometimes they really do feel like they are a CollectiveIntelligence.

Contributors: SunirShah, FridemarPache


Bush, V. (1945) As me way think. The Atlantic Monthly 176(1), pp. 101-108. Available from http://www.theatlantic.com/unbound/flashbks/computer/bushf.htm

In early versions of Netscape (and perhaps even today) there was an undocumented <tt><hype></tt> tag. You click on it and it says "What is worldwide hypermedia". Of course, a lot of it was nothing but hype in 1995, while today, there's some substance and less vapor.


But what makes good hypermedia? Or what is the better future of hypermedia? This was my collection of criteria (a new version of the whole text at http://www.kaapeli.fi/~smaatta/hypertextandcriticaldiscussion.html ) :

1. The public hypertext. The hypertext created must be public. Anyone must have the possibility of commenting or annotating the pieces of work within the sphere of public hypertext. This is most important for the future of critical discussion in the Web, and for the openness of the Web. Creating public documents and annotations bring up the question of author's rights to his/her work and the question of payments on the basis of copyright (or similar). Furthermore, there is the question of VersionHistory. A sound hypertext tool does keep track on different versions of a document. (Should the hypertext software be public, too? The free and the open source software are highly recommended!)

2. The fine-grained, bidirectional links. The hypertext must allow fine-grained, bidirectional linking. Links must be visible and usable to all. Links can contain information on types of links, types of transactions etc. This information should not be saved in linked documents itself but in a third location "between" them; thus the need for mediators.

3. The sensors, filters and evaluation of ideas. The apparatus of sensors and filters let one choose and find the relevant infomation and also to follow the changes being made to the documents. The evaluations part of the apparatus gives a chance to rate or vote on documents or on ideas. The resulting "social software" could be most important in solving problems of today's world or of small communities.

4. Intellectual nodes. In various fields of hypertext intellectual nodes arise for a reader to see and comment on. A node could be a landmark writing or a site that is directed to something very exciting. A node could be a remarkably well elaborated summary of some discussion or of several discussion threads. This is a new kind of touch to collective wisdom and information in hypertext. With finding these nodes one can connect him-/herself to the most relevant content in digital space, become part of something truly important.

5. Archiving. It is our duty to have all the valuable hypertext archived and saved for the future generations. Authors of documents should have the right to determine which documents or which part of material (and for which period of time) to be archived. Public libraries might be involved in this (as in everything else about the hypertext), other solutions are possible depending on the country etc.

6. Open for extension by external helper programs like e.g. LinkItAll -- fp

-- SamiMaatta (updated 21st of April 2003), FridemarPache

Another view to a HyperMedium is to view the nodes as states and the arcs as state transitions of a finite automaton. Smoother state-changes (i.e. more discrete intermediary steps, e.g. morphing) can make hypermedia more attractive for the humans.

The states represent content in such a way, that the user (human or non-human) gets link (suggestions) to activate preferably such state transitions, that are intended by the contentprovider of the current node.

Beyond that hypermedia (browser) may have addressboxes that allow arbitrary jumps in the nodespace.

-- FridemarPache



MeatballWiki | RecentChanges | Random Page | Indices | Categories
Edit text of this page | View other revisions