Spatial User Interface Metaphors in Hypermedia Systems

Workshop at the European Conference on Hypermedia Technology, September 1994, Edinburgh, Scotland

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outdated contact address kept on this page for consistency reasons!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

Keith Andrews
Institute for Information Processing and Computer Supported New Media
Graz University of Technology
A-8010 Graz, Schiessstattgasse 4a
email: kandrews@iicm.tu-graz.ac.at

The position papers are availabe as .zip archive

The workshop on spatial user interface metaphors in hypermedia systems was held on September 18, 1994 in Edinburgh in conjunction with ECHT'94. Participants had to provide position papers which were distributed to all participants before the workshop. To enable discussions also before the workshop an e-mail discussion list was installed. There were 11 participants at the workshop.

Spatial user interface metaphors essentially try to make use of humans abilities to use space to organize objects or human skills in navigating structures like houses or cities. The use of space in a particular application depends on the goal of that application - be it to make structure explicit, to support navigation or to support organization. Well-known examples of spatial user interfaces metapors are the desktop metaphor, the Xerox's Information Visualizer, and most virtual reality systems.

The position papers and the first round in the discussions brought forth that the main interests of many participants was in easing information access and navigation using spatial user interface metaphors.

As a starting points of the workshop Keith Andrews from the Graz University of Technology gave a presentation of the Harmony browser for Hyper-G. One of the navigational tools supported by Harmony is navigation using an information landscape. (For more information contact Keith Andrews: Institute for Information Processing and Computer Supported New Media, Graz University of Technology, Schiessstattgasse 4a, A-8010 Graz, kandrews@iicm.tu-graz.ac.at)

Several topics were isolated as being relevant to the field of spatial user interface metaphors. Due to time constraints not all of these issues could be discussed during the workshop:

How to layout spatial structures? Should hand-crafted or automatically generated layouts be used? This issue is related to the goal of a system. Systems where users modify the spatial layout sometimes use spatial relationships to communicate with the machine or other users. An example for such a system is VIKI which was presented in an ECHT'94 paper session. Automatically crafted spatial layouts can be used to show structure inherent to the information in the system.
How to support large, dynamic, multi-user environments? An advantages of spatial metaphors it the ease in which people are able to communicate about spatial relationships. This ease in communication allows users to communicate about spatial relationships. This would allow people to collaborate in navigation in such systems. Communication about the environment and navigation relies partly on static features (orienteering points) in the system. Therefore systems should not be too dynamic.
Searching, hierarchical structures, contents of the space. The spatial layout expresses a (spatial) relationship. What is the meaning of this relationship? How can it be used to support searching?
Evaluating spatial metaphors. Although many spatial metaphors have been proposed or even realized it seems that few systems have been evaluated on the usability of the spatial metaphor - at least in the field of hypermedia. This lack of evaluation makes the field of spatial user interface metaphors seem very immature.
Especially in large systems "conventions" on the meaning of spatial layout seem necessary. Conventions also should clarify the meaning of spatial icons or widgets (user interface elements). Spatial user interfaces and their elements should show clear affordances (that is: they should visually communicate their use).
User interface metaphors are not an identical mapping from a source domain to a target domain (for instance objects cannot fall off a virtual desktop). Features which extend the metaphor beyond the source domain (sometimes called "magic features") often make a metaphor really useful. In the context of spatial metaphors such magic features are especially interesting in navigation. The use and realization of such features -- especially for supporting navigation -- has not been discussed much in literature till now.
Perhaps the most important issue concerning spatial metaphors is to decide when to use them and when not to use them. As user interfaces today tend to be very realistic a spatial conception seems logical. However spatial user interfaces are no universal solution to navigation problems.

The results of the discussions clearly point at the need of detailled evaluation of spatial user interface metaphors. They might be useful in a smaller range of applications than is believed today. Like in all other user interface issues these evaluations will probably show that there is no universally valid spatial metaphor. This is especially true when social and cultural issues are taken into account. These issues are very complicated ones. Most spatial metaphors today (and also most metaphors discussed at the workshop) were quite detached from social beings.

Spatial metaphors have to bring real advantages for the system: it is hard to sell a metaphor alone. Albeit the advances in virtual reality systems spatial metaphors are very much at the start. One of the conclusions of the workshop therefor was that we should stop talking, build something and evaluate it.

last modified on 10/26/96

Andreas Dieberger
andreas.dieberger@acm.org