Guest Editors' Introduction: Scaling to New Heights
JULY/AUGUST 1998 (Vol. 18, No. 4) pp. 16-17
0272-1716/98/$31.00 © 1998 IEEE

Published by the IEEE Computer Society
Guest Editors' Introduction: Scaling to New Heights
Nahum Mitre

Stephen G. Bell Labs/Lucent Technologies
  Article Contents  
  Conclusion  
  References  
Download Citation
   
Download Content
 
PDFs Require Adobe Acrobat
 
"The proof of the pudding is in the eating."
An English proverb
One of the major challenges of information visualization is using it to solve real-world problems in diverse areas such as telecommunications, financial analysis, software engineering, command and control, and information systems. The real world is complicated. Data and information are often complex, massive, time dependent, of diverse types, and not always reliable. To make the situation even more complex, users come with all levels of capabilities, education, and tastes. All of these facts further complicate the visualization process.
Unfortunately, academic interests rather than applicability to real-life situations tend to motivate a large part of the research and development in visualization. This creates a visualization "bag of tricks" according to our colleague Ken Boff.
An alternative, more demanding—but necessary—approach is to study the problem first and then look for appropriate solutions (for example, using information visualization, if appropriate). Those of us in the visualization community need to realize that in many real-world applications, visualization is just one component of a complex system rather than a stand-alone entity. We also need to understand the system and the user needs in order to create effective visualizations.
The section on information visualization in this special issue presents examples of the potential usefulness of information visualization in real-life environments containing large amounts of complex data and information. From the wide range of issues concerning this practical area, we chose to focus mainly on one aspect—scaling. This scaling issue—how to deal with massive amounts of complex data and information practically—is very important.
Methodologies for dealing with large amounts of data and information can be classified as follows:
  • Condensing. Condensing large amounts of information into a given display space. Consider the following:
    • Tile maps—assigning an information unit to each pixel (for software management 1 and long lists 2 ).
    • Hyperbolic representation of large hierarchies 3 (see also Munzner's article in this issue).
    • ConeTrees—representation of large hierarchies. 4
    • Icons (or glyphs)—representing information in multidimensional icons (see Chuah and Eick's article in this issue).
    • 3D—representing the information in a 3D layout on a flat screen (such as 3D rooms 4 and the Perspective Wall 4 ).
    You can also view the condensed information in detail by various techniques, such as using selective magnification. 6
  • Organization. Information can be organized according to its contents or to another natural order. You can arrange the representation in varying levels of detail, such as summaries, and in different formats, including a WebBook, which represents a collection of pages in a book format, 7 tables, a structured display space, 5 rooms, 4 and spreadsheets (see Chi et al. in this issue). In addition, proper interaction techniques (such as simple clicking or brushing) can reveal various details of the information.
  • Segmentation. Displaying the information one segment at a time using side-by-side representations, such as spreadsheets, direct manipulation to display parts of the information contained in the data set, 8 and animation, which presents information segments sequentially with time. 9
The information visualization section in this special issue is composed of two parts. The first contains three articles addressing the problem of how to deal with huge amounts of information—a problem prevalent in current information systems. Munzner's article describes ways to represent large graphs using a 3D hyperbolic space. Chi et al. demonstrate the usefulness of the spreadsheet metaphor to represent visualizations of many data or information sets. Finally, the article by Chuah and Eick represents novel ways of using complex icons (glyphs) to manage massive complex software environments.
Conclusion
The second section comprises three short articles describing various aspects of information visualization applications. Wright highlights the Birds-of-a-Feather session held at the Information Visualization Symposium 97 (InfoVis97) about the potential, road blocks, and trends in using information visualization in the business environment. Andrews depicts the use of information visualization in making structured hypermedia (including large hierarchical structures) more useful to users. Finally, the short article by Gershon discusses the difficulties of visually representing imperfections of data and information, and making accurate presentations. Imperfection of data, information, and presentation can be improved up to a point and, as in real life where nothing is perfect, users must learn how to cope with it. The article suggests a number of ways to handle these complex problems. Enjoy!

References

Nahum Gershon is a principal scientist at the Mitre Corp. His work concerns data and information visualization, Web browsers, image processing, information and data organization, and analysis of medical, environmental, and other multidimensional data. He pursues research in the use of understanding of the perceptual system in improving the visualization process and dealing with information.

Stephen G. Eick is the chief technologies officer of Bell Lab's Visual Insights venture and leads the Interactive Data Visualization Research group at Bell Laboratories. Eick's interests include visualizing databases associated with large software projects and networks, and building high-interaction user interfaces.

John Dill is a professor in the School of Engineering Science at Simon Fraser University, Burnaby, British Columbia. His current research interests include information visualization, engineering visualization, user interfaces, and intelligence in computer-aided design. Dill received a BASc in engineering physics from the University of British Columbia, an MS from North Carolina State University, and a PhD in engineering science from the California Insitute of Technology. He is active in ACM's Siggraph and is a member of the IEEE Computer Graphics and Applications editorial board. Contact him at dill@cs.sfu.ca.