to the top archive

The Information City - A Step towards Merging of Hypertext and Virtual Reality

Andreas Dieberger
outdated contact address kept on this page for consistency reasons!
Dept. for Design and Assessment of Technology
Vienna University of Technology
A-1040 Vienna, Moellwaldplatz 5/187
Tel: (+43-1) 504-11-86 Fax: (+43-1) 504-11-88
email: andreas.dieberger@acm.org

Keywords: Hypertext, Spatial User Interfaces, Virtual Reality

Abstract

The Information-City is a spatial user interface for large hypertext collections. It uses the metaphor of a city to represent documents as buildings in a virtual environment the user navigates and makes use of Jay Bolters "writing on the world" concept to visualize as much information about the documents as possible without overloading the user (e.g. age, relevancy, accessability). The city is restructured according to user interest profile and document similarity to create "districts of interest". The city metaphor shall help the user to build a conceptual map of the environment. As autonomous actor the system helps the user to locate information by providing light cues and transportation (taxi and subway metaphors). Significant information is pointed out by open doors for instance. Visualization of hypertext nets as house structures is done using hypertext abstractions, so called building blocks. These constucts define a hierarchical subset of the hypertext web that can be visualized different than the rest of the web - namely as halls and rooms. A second, purely textual version of the information city is constructed in a MUD (Multi User Dungeon) system and will be available on the internet. This system is used to study collaborative navigational behaviour in a multi user environment (development of landmarks and navigation paths). Findings of this version will be used to improve the design of the graphical version of the information city. At the time of writing an authoring tool using building blocks is under construction.

It has often been argued that many problems of today's hypertext systems could be overcome by using a more graphical and more spatial user interface. Proposed solutions are fisheye views which focus on an area of interest and adding more depth to the display of hypertext - leading to three-dimensional user interfaces [STAP93] and to virtual reality hypertext as suggested by Jay Bolter in his ECHT'92 lecture. Transferring hypertext to virtual reality cannot be achieved easily. A strong spatial interface metaphor is needed which supports orientation within and between hypertexts.

More spatial metaphors can help considerably in building and maintaining a cognitive map of the information space (see also [SHUM90]). Examples of spatial metaphors for information systems are manifold [DIEB93] - the city metaphor being an early example - used frequently to illustrate navigational behaviour.

Providing a three-dimensional version of the desktop alone does not help. On most machines users are unable to locate files since the desktops tend to be cluttered and objects are indistinguishable. A richer environment is needed that gives more information about objects. Jay Bolter talked in his keynote lecture about "Writing on the world" as a means of giving more information to the traveller in spatial environments. In real life traffic signs, shop signs or floorplans in large buildings are examples for "writing on the world". Examples from earlier ages are the gable-stones in Amsterdam gracht-houses or wall paintings in egyptian graves.

The enriched city metaphor is a non-immersive virtual reality like the Information Visualizer [RoCM93] and makes use of "writing on the world" to convey as much information as possible to the user. A house with open doors means a document in strong relation to the topic looked for, a half-closed door signifies weak relation. The outside of the building shall convey information about internal complexity and age of the document [HiHo92]. Thus a worn doormat could show a house that is entered very often.

As houses represent single hypertexts, navigation between hypertexts is like navigating in a unknown city. On entering the information city for the first time, it is restructured according to an interest profile provided by the user, so that the topics of interest are grouped into 'districts of interest'. Walking along the street the user can look at related documents - like in a library where all books about a subject are on the same shelf.

When the house is entered for the first time, it should be restructured to provide easy access to topics of most interest (red carpet metaphor). The halls and rooms try to convey as much information about their contents to the user as possible. Wall signs, guides, light, colour and sound cues help navigating inside the information house. The system updates the interest profiles accordingly to the users actions in the city. Since many houses in the information city will be entered after the user has travelled a lot in the city, these houses can be restructured according to the users needs. If the interest profile of the user changes, a second city (or some suburbs) can be created using the same documents or a selection thereof. It is possible to explore different cities according to specific needs, thereby working with clusters of information related to various tasks.

In a real city people seldom really get lost, beause there are always other people they can ask for help and because a real city is an environment filled with various sorts of information that help in navigation.

The information city consists of more than simply hypertext houses. The information city can be used as a front end for Internet services like WWW, Gopher, and others. It can also contain electronic mail, bulletin boards, and meeting rooms. There are information rooms that represent links to other cities (data-bases at remote internet sites), and people can automatically ftp data by requesting a copy of something they saw in the city.

In this form the information city is a metaphor for reading hypertext only. Authoring for the information city poses some problems, since houses normally are planned before they are built. Hypertexts with unconstrained linking structures cannot be visualized well as a house. So for authoring it is necessary to use buildings blocks which can be visualized. Those building blocks are predefined structures that allow the building of linear structures, hierarchies or other constructs very easily. Each structure defines a set of predefined links ("implicit links") which are typical for the structure type (for instance next / previous-links in a linear structure). Authors don't have to provide those links themselves [MaSc92] - they simply create building-blocks as concept nodes with predefined behaviour. Building-blocks act as higher-order abstractions for hypertext like higher-order constructs do in a programming language. Use of these building blocks alone would result in a strictly hierarchical hypertext. But the author can provide as much additional links as she wishes. Such links are different only in that they are visualized differently - for instance a linear structure would be visualized as a corridor - additional links can be shown as magic mirrors, windows or marked text on an information wall. It is important to note that the construction of the information city using those building blocks does not restrict authors - instead this method just designates parts of the overall linking structure as skeleton around which the entire hypertext is built. The house metaphor is only an example of possible visualizations in the information city. Other examples which would fit neatly into the information city are information streams, gardens etc.

The information city as described above is still in the conceptual stage. First we are building a text-only version of the information city using a MUD. MUDs are interactive multi-user computer games played over a network, frequently refered to as text-based virtual realities [CURT92]. MUDs can be seen as purely textual CSCW environments [MaOs93] and several MUDs on the internet today act primarily as "meeting places" for scientists. MUDs are made up of objects which can be rooms, users, things the users can carry and/or manipulate and robots controlled by the MUD library. Rooms are linked in a network structure by means of exits. This network structure has many similarities to hypertext networks. A big difference between adventure games (or game MUDs) and hypertext is that hypertexts try to help the users to find data looked for, whereas adventure games often hide the solution in order to make the game more interesting [NIEL90]. We intend to use the Information City MUD to study navigational behaviour and spatial perception in spatial information environments. After completion of the graphical version of the Information City, the MUD version can act as a secondary user interface to the city.

During work for the MUD version it turned out that many navigational issues in virtual environments are not understood well enough to provide adequate linking facilities. We performed a pilote study concerning some issues in this area by placing two questionnaires in a MUD - MUDs tend to be very navigation intensive and typical MUD users often have yearlong experience navigating a virtual city-like environment. The results from those questionnaires were not contained yet in the poster. Especially important results are: collaboration is a main ingredient of navigation in cities, local disturbances of euclidean space don't really matter and users perceive different kinds of navigational methods very differently (for instance walking, flying and teleporting have very different semantics). While those results are not scientifically bomb-proof they still show that great care has to be applied when designing or applying transportation metaphors for virtual information environments and virtual realities - something that seems to be forgotten in todays virtual reality euphoria.

For further information on the Information City project see [DiTr93a] and [DiTr93b].

References

[CURT92] Curtis P.: "Mudding: Social Phenomena in Text-Based Virtual Realities", FTP-able from parcftp.xerox.com in /pub/MOO/papers/DIAC92.*

[DIEB93] Dieberger A.: "Summary Spatial Metaphors" posted to newsgroup alt.hypertext in May 1993

[DiTr93a] Dieberger A., Tromp J.G.: "The Information City - A Metaphor for Navigating Hypertexts", recent research paper at the HCI93, Loughborough, Sept 1993, posted to sci.virtual-worlds in Oct. 1993

[DiTr93b] Dieberger A., Tromp J.: "The Information City project - a virtual reality user interface for navigation in information spaces", Proc. of the Symposium VR Vienna, Dec. 1993, forthcoming

[HiHo92] Hill W.C., Hollan J.D.: "Edit Wear and Read Wear", Proc. CHI'92, pp. 3-9

[MaSc92] Maurer H., Scherbakov N.: "Hypermedia Systems without Links", IIG Report Nr. 343, Techn. University Graz Sept. 1992

[MaOs93] Masinter L., Ostrom E.: "Collaborative Information retrieval: Gopher from MOO", Proc. of the INET93

[NIEL90] Nielsen J.: "Hypertext and Hypermedia", Academic Press 1990

[RoCM93] Robertson G.G., Card S.K., Mackinlay J.D.: "Information Visualization Using 3D Interactive Animation", Comm. of the ACM, Vol.36, No.4, April 1993, pp. 57-71

[SHUM90] Shum S.: "Real and Virtual Spaces: Mapping from Spatial Cognition to Hypertext", Hypermedia Vol.2, No.2, 1990, pp. 133-158

[STAP93] Staples L.: "Representation in Virtual Space: Visual Convention in the Graphical User Interface", Proc. InterCHI'93, pp. 348-354

last modified on 10/26/96

Andreas Dieberger
andreas.dieberger@acm.org