|
Quick Links: [Tool Name here] |
||
Hypviewer allows a user to layout and interactively navigate node-link graphs in 3D hyperbolic space. The current node being observed is located in the center of the display, at full size. The parent, children, and sibling nodes extend away from the current node, growing exponentially smaller as they approach the edges of the display. According to its documentation, this program can handle quite large graphs, up to 100,000 edges, quickly and with minimal visual clutter. The hyperbolic view allows the user to see a great deal of the context around the current focus node, as the relation and relative position of a node to others in the graph remain visible. This tool can be used as a standalone viewer, but is intended to be integrated into large UI applications.
This type of tool is of use in any domain that can be represented as a large link-node graph. Many systems can be represented as link-node systems, such as Internet backbone routes, cross-referencing dictionaries, and the US highway system are all good candidates. File-folder displays, which often represent the hierarchical organization as a linear tree, that requires much scrolling to view, could more clearly display information concerning the relative position of a file within the system, and it’s relation to other files using a spanning tree in hyperbolic geometry. This type of display can also be used by web site engineers to better visualize and manage large collections of web pages, files, or documents, related through their embedded hyperlink relations. Web browsers could also use a system like this to offer the user the ability to jump forward, backward, or even laterally, through a connected graph of web sites, without having to tediously page through each one, allowing each to load before moving on. Data arranged in hierarchies, such as biological taxonomies, ontologies, or other classification schemes, can also be viewed and navigated using hyperbolic trees.
New knowledge is added to the system in the forms of nodes in the graph that the system then abstracts into a hyperbolic spanning tree in the graphical display. That node may be just a data point, or it may be a document, web page, or other content (or a pointer or link thereto).
As mentioned, this system is designed to be integrated into a larger and/or more complete UI system. That larger system may or may not limit who has authorization to add, delete, or otherwise manipulate knowledge, but I was unable to find any mention of such limitation in the Hypviewer documentation.
Hypviewer is a KM tool to help organize and display information and knowledge. The system allows the user to view a large amount of data or information, and the contextual links that related it all in a spanning tree, exhibiting hyperbolic geometry. By presenting the knowledge in this way, the system makes it easy for the user to see where and how a new piece of information or knowledge should be inserted to be meaningful within the context of the whole body of knowledge.
The knowledge itself is in the form of a web or graph, that can then be viewed in the form of a hyperbolic spanning tree, starting from any an arbitrary root node, depending on the user’s needs. The display shows a node in the center, or focus, in the hyperbolic viewing space, and it’s surrounding context, in the links and nodes it relates to extending towards the edges, growing smaller with distance from the center, or focus.
The major drawback, that I can see, is that only the name or title of each node is visible in the display. If the names are poorly chosen, cryptic, or ambiguous, then the user may have trouble determining the relative value to them of any given node.
Any node in the system can be expanded in the manner of a hyperlink, to display the data, information, knowledge it contains. To find any given node, the user can traverse the tree from its root, through each link, sequentially. However, because many levels and branches of the tree are displayed at once, in the hyperbolic geometry of the display, the user can also skip up directly to a node several levels above or below the current node, or even move laterally across the branches.
Also, as with many systems, Hypviewer allows the user to search for a node by a keyword search of its name.
As far as I have been able to determine, the Hypviewer system neither collects, nor utilizes meta-data about the usage of knowledge, being primarily a presentation tool. However, there is nothing in its design or implementation that I am aware of that would preclude the use or collection of such data, if, for example, the system were to be integrated into a larger UI.
From what I have read, regarding the Hypviewer system, it could easily be used to display and make sense of the contents of a knowledge repository. Whether that repository is used by one or by many, it would not matter to the knowledge presentation system. However, I do not know of any explicit support for collaborative efforts.
As stated, Hypviewer is primarily a tool for viewing and comprehending knowledge in terms of context and position. I have not been able to find any mention of Hypviewer, or a system incorporating it, in the establishment and management of organizational memory, and I doubt that has any expressly built in. However, Hypviewer could certainly be incorporated into a system designed to capture and maintain such an organizational memory.
The first examples of hyperbolic trees were apparently developed in-house by Xerox’s Parc research labs, inspired by an M.C. Escher woodcut, Circle Limit IV (Heaven and Hell), 1960. Inxight Software, Inc. developed a relatively popular system based on this research, and, from my research, appears to be the dominant leader in this type of visualization software. Xerox apparently does not license this technology, and the software from Inxight is expensive enough to be a barrier for many users.
The Hypviewer system was developed by Tamara Munzer, for her doctoral thesis in computer science at Stanford University, and is license to Silicon Graphics. Unlike those mentioned above, anyone may
copy, modify, use and distribute this software and accompanying
documentation free of charge provided (i) you include the entirety of this
reservation of rights notice in all such copies, (ii) you comply with any
additional or different obligations and/or use restrictions specified by any
third party owner or supplier of the software and accompanying documentation in
other notices that may be included with the software, (iii) you do not charge
any fee for the use or redistribution of the software or accompanying
documentation, or modified versions thereof.
[Excerpted from the
Copywrite agreement accompanying the Hypviewer software]
In other words, Hypviewer is free to use and to share, within broad constraints, for non-commercial purposes. This makes it excellent as teaching tool, for academic projects, or for personal use.