Visualizing World Wide Web Information Resources

Nahum Gershon, Joel Winstead, Joshua LeVasseur, James Croall, Ari Pernick, and William Ruh

 

The MITRE Corporation
7525 Colshire Drive
McLean, VA 22102

{gershon,winstead,jlevasse,jcroall,apernick,war}@mitre.org

Abstract

The objective of this work is to improve the ability of people from all walks of life and interests to access, search, and use the information distributed on the World Wide Web (WWW). The process of interacting with information resources starts with browsing, continues with digesting and assimilating pieces of information, terminates with generation of new information, and begins anew with analysis of pre-existing and new information. Our approach is user-centric- taking users' needs into account by allowing them to interact with the information contained in large arrays of documents. The visualization process is an integral part of the overall process.

 

We have covered five related categories in this methodology. The first one is browsing through the World Wide Web (WWW) hyperspace without becoming lost, based on a visual representation of the hyperspace hierarchical structure (hyperspace view). The second category is the ability to define a piece of a document and point to it. The third one is allowing one hyperlink to point to more than one document. The fourth category is extending the link concept to include meaning or uncertainty. The fifth one is overcoming the rigidity of the WWW by allowing the user to construct interactively and visually a personal hyperspace of information, linking the documents according to the application or problem domain, or to the user's own perception, experience, culture, or way of thinking.

 

Keywords: World Wide Web, Visualization, User Interface, Human Information Interaction

 

  1. Introduction
  2. Browsing Through Hyperspace without Getting Lost
  3. Pieces of Information
  4. Dealing With Multiple Hyperlinks
  5. Hyperlinks with Multiple Meanings
  6. Applying External Tools to Web Documents
  7. Making the WWW Non-Rigid: Building One's Own Information Hyperspace
  8. Conclusion
  9. Acknowledgments
  10. References

1. Introduction

The World-Wide Web (WWW) has created an exciting universe of information by linking numerous information resources residing on computers dispersed around the world and making them easily accessible. This revolutionary capability has enabled us to explore this universe of distributed information resources from our desktop computers. If set up, managed, and presented correctly, these distributed information resources will enable us to:

 

Ordinarily, users would like to interact with distributed resources to find out:

 

The WWW, or simply the Web, developed at CERN, Switzerland [T. Berners-Lee et al, 1994], and Hyper-G, developed at Graz University of Technology [Andrews and Kappe, 1994], allow the user to roam via menus and embedded links through information spaces of documents or images. Engaging browsers, such as the National Center for Supercomputing Applications (NCSA) Mosaic (for the WWW) [Schatz and Hardin, 1994] and also Harmony (for Hyper-G) have transformed the process of getting information from Internet servers. However, some difficulties still remain.

 

In spite of the captivating nature of the World-Wide Web, we still have a long way to go before the use of this information universe is easy and intuitive. To find required information, users need to browse through multiple pages or sort out a large quantity of search results, and extract the relevant information from the accessed documents. While interacting with information over the WWW, users feel at times lost, confused, and overwhelmed. How can the user know where the sources of the relevant information reside, how to get them, and, once the sources are accessed, how to get the relevant information from them?

 

To a large extent, users are not directly involved in the development of the Internet and its capabilities. If we do not involve the users in designing distributed information resources and their interfaces, we will create useless information systems. As long as there is a human being sitting in the front of the screen, the interface to information resources needs to be user-centric, taking the user's needs into account. Users would like to interact with the information, preferably forgetting that there is a computer separating them from the information. A good human-computer interface (HCI) is a must, but it is not enough.

 

Recent developments in visualization, interactive computer graphics, and mass storage have created new possibilities for information navigation, access, and retrieval in which visualization and user interface could play a central role. The question is how to exploit the advances in visualization and graphics technologies and experience while understanding how the human mind works in order to reduce the frustration, time, and cost of using information dispersed over the Internet.

 

The work described in this paper is concerned with improving the way users interact (visually and non-visually) with information embedded in distributed information resources. The process of interaction with the information starts with browsing, continues with digesting and assimilating pieces of information, terminates with generation of new information, and begins anew with analysis of pre-existing and new information. The categories covered in this work include browsing through hyperspace without becoming lost, interacting with pieces of information, extending the definition of hyperlinks to include meaning and uncertainty, and overcoming the rigidity of the WWW. The last capability allows the user to construct interactively and visually a personal hyperspace of information, linking the documents according to the application or problem domain, or to the user's own perception, experience, culture, or way of thinking. Moreover, in this work, the visualization process does not stand alone, but rather it is an integral part of the process of interacting with the information.

 

2. Hyperspace View: Browsing Through Hyperspace without Getting Lost

Using current browsers to "surf" over the Web, users often traverse a multitude of documents via hyperlinks. After opening a number of linked documents on the WWW, users often do not know where they are in the information space and do not remember how they got there. In short, users feel lost or become disoriented. This is a major problem. One solution is to provide users with both a local and a global view of the information space. These views should be represented visually to promote quick perception and understanding of the hierarchical structure of the hyperspace and to help the users quickly locate where they are in hyperspace, i.e., to re-orient themselves.

 

We developed an enhancement of NCSA Mosaic that allows the user to view the hyperspace depicted as a visual "tree" structure (see Figure 1). In addition to viewing, users can "jump" from one document to another by pointing and clicking the mouse without having to go back resource by resource or "page by page." This representation also displays any additional html documents created during the session (see Sections 3,6, and 7). Recently, two additional approaches for visualizing hyperspace structures were independently proposed [Mukherjea and Foley, 1995, and Wood, et al, 1995]. Both interesting approaches focus on how to make complicated hyperspace structure more comprehensible by letting the user view the hierarchy globally and in detail. The latter allows restructuring the view by introducing physical repulsive and attractive forces among the hierarchy elements.

 

 Figure 1. Hyperspace View: A graphical view of the hyperspace emerging from a document depicted as a "tree" structure. The user can "jump" from one document to another by pointing and clicking the mouse without having to go back one by one.

 

While browsing, users often would like to view the names of documents and how they are linked to each other without actually opening and reading each document. Our enhancement allows the user to let the hyperspace view grow automatically up to a specified number of levels. After observing the hyperspace structure and contents visually, the user could decide to open and read none, some, or all the documents represented on the "tree" or to save them in his/her own personal space. In cases where changes to the content of a document over time is an important piece of information, saving Web documents allows the user to compare old and new versions. The user can also print a hierarchy of documents without having to open each one of them in turn.

 

To allow the use of these enhancements in different browser environments, we developed these enhancements in a way that will not require the modification of the Mosaic browser itself using Common Client Interface (CCI).

 

3. Pieces of Information

The smallest unit of text information in the current versions of the WWW is a document. Users often find that a paragraph, a sentence, a word, or even a part of an image is a piece of information relevant to their problem. Our enhancements of NCSA Mosaic allow the user to define his/her own pieces of information and include them in newly defined html documents. These new documents containing fragments of existing documents are linked to their source documents and could be linked to other documents as the user wishes. The newly created html document are displayed in the hyperspace view representation. With this enhancement, the user could create his or her own version of the information space, thus reflecting the current problem or his or her own interests and view of the information.

 

4. Multiple Hyperlinks

An idea expressed in one document could be logically linked to more than one document, or to a specific location in another document. We developed a provision that allows the user to link a phrase to more than one document. By clicking on the multiple hyperlink, the list of links is displayed allowing the user to view one or more of the linked documents.

 

5. Hyperlinks with Multiple Meanings.

The links in this work are more than pointers to where the documents reside. For people using the information stored in documents, a link could carry a meaning (related to the informational content) or strength (e.g., how strong the link between 2 documents in a probabilistic sense). We developed the capability to store and express graphically a quantity or quality in links. In an html document the type of the link is expressed by color or font size while at the hyperspace view ("tree"), it is designated by color or type of line connecting adjacent documents.

 

6. Tools

Our enhancement of NCSA Mosaic allows to call a suite of tools that could operate on the current html document (using CCI). The list of tools, chosen by the user, could be displayed by clicking on a special button appearing on the browser screen. The user can pick up the required tool. The results of applying the tool are displayed as a new html document.

 

7. Making the WWW Non-Rigid: Building One's Own Information Hyperspace

The rigid organization of hypertext documents often makes it difficult to use information distributed over the World-Wide Web. Creators of documents placed on the Web link them to other documents. These links form a rigid structure where no changes are allowed. Quite often, these hyperlinks reflect the document creator's own point of view and current interests, or some arbitrary considerations.

 

Depending on the application, problem, personal way of thinking and perception, experience, or culture, these pieces of information could be related to each other in various ways. For example, an information analyst reads 100 documents and finds out that only 12 are important to the problem he or she is working on. In addition, the analyst could feel that the documents are related to each other in a different way from what the existing hyperlinks specify. For information resources to be effective and to enhance problem solving and analysis, they should allow each user to construct his or her own information space with links and associations (among pieces of information and whole documents and images) that fit the personal problem, application, or ways of thinking and perception.

 

Our enhanced version of NCSA Mosaic enables the user to modify visually and interactively the links among the documents and images using a point-click-and-drag operation on the display of the hyperlink hierarchical structure. This enhancement allows the user to effectively generate new, personalized links and to (visually) view the new and "old" information space globally and locally (see Figure 2). The new hyperlinks are stored at the end of duplicates of the corresponding documents and could be saved for future viewing and sharing with other users. The user can explain his/her thoughts and findings by adding annotations accompanying the new links.

 

 Figure 2. Making Hyperspace Flexible: The MITRE enhancement to NCSA Mosaic enables the user to interactively modify the links among the documents and images using a point-click-and-drag operation on the display of the hyperlink hierarchical structure. The user can generate new, personalized links and to view the new and "old" information space globally and locally. The new hyperlinks are stored at the end of the documents and could be saved for future viewing and sharing with other users.

 

In Conclusion

The use of computers, networks, and the WWW as information resources is still a new concept involving new enabling technologies and permitting new types of interactions between humans and information resources (This is in contrast with traditional human interactions with information contained in books). The situation has similarities with the one prevailing when the movie camera was invented. At that time, people tried to imitate theater via a stationary camera. They later discovered that these two media have different characteristics and were able then to better use the advantages of the new medium. Learning the new medium of computers, networks, and the WWW as information resources and further understanding how humans process and interact with information will enable us to provide new methods for discovery and searching of resources, enhancing creativity and thinking.

 

The developments reported in this paper make use of advances in visualization and interactive computer graphics technologies as well as the understanding of how humans search and process information. In this work, the visualization process does not stand alone, but rather is an integral part of the process of interacting with the information. Advances along these lines could improve the ability of people from all walks of life and interests to access, search, and use the information distributed in Internet resources. This will enable full use of the Internet's information universe from our desktops.

 

Acknowledgments

The authors would like to express their appreciation for the creative advice on design and color given to us by Elaine Mullen.

 

References

Andrews, K. and F. Kappe, Soaring through Hyper Space: A Snapshot of Hyper-G and its Harmony Client, in Proc. of Eurographics Symposium of multimedia/Hypermedia in Open Distributed Environments, Graz, Austria, W. Hezner and F. Kappe, (editors), pages 181-191, Springer Verlag, June 1994.

 

Berners-Lee, T., R. Cailliau, A. Luotonen, H. F. Nielsen, and A. Secret, The World-Wide Web, Communications ACM, 37, No. 8, 76-82, 1994.

 

E.A. Fox (ed.), Source Book on Digital Libraries, Version 1.0, December 6, 1993.

 

Gershon, N.D., and W.A. Ruh, The Information Highway: Putting the User in the Driver's Seat (If we do not balance user needs with technical innovation, we will create useless information systems), IEEE Spectrum, to be published, 1995.

 

Mukherjea, S., and J.D. Foley, Visualizing the World-wide Web with the Navigational View Builder, Computer Networks and ISDN Systems, 27, 1075-1087, 1995

 

Schatz, B.R, and J.B. Hardin, NCSA Mosaic and the World-Wide Web: Global Hypermedia Protocols for the Internet, Science, 265, 895-901, 1994.

 

Wood, A., R. Beale, N. Drew, and R. Hendley, HyperSpace: A World-Wide Web Visualizer and its Implications for Collaborative Browsing and Software Agents, submitted to HCI '95, UK.