IPSI - Integrated Publication and Information Systems Institute
GMD - German National Research Center for Information Technology
Dolivostr. 15, D - 64293 Darmstadt, Germany
Phone: + 49-6151-869 919, e-mail: {streitz, boecker}@darmstadt.gmd.de
This organization overview describes two research divisions of GMD-IPSI in Darmstadt, Germany: User Interfaces for Information Systems and Cooperative Hypermedia Systems. They are in particular addressing HCI work within the overall framework and goals of IPSI.
organization overview, user-interfaces, 3D visualization, information retrieval, electronic publishing, hypermedia, CSCW, desktop-based collaboration, electronic meeting rooms, shared work spaces, pen-based interaction
© Copyright on this material is held by the authors
The institute (director: Erich Neuhold, deputy director: Norbert Streitz) has a staff of 90 people and is organized in four divisions: User Interfaces for Information Systems (Heinz-Dieter Böcker), Cooperative Hypermedia Systems (Norbert Streitz), Distributed Multimedia Information Management (Wolfgang Klas), and Multimedia Information and Publication Prototypes (Klaus Mätzel). Focussing here on HCI, we restrict this overview to the first two divisions. Members of IPSI are also teaching at the Department of Computer Science at the Technical University of Darmstadt based on a cooperation and a joint appointment of Prof. Neuhold. The work at IPSI is based on a deliberately interdisciplinary approach. This is manifested in the disciplines constituting the background of its staff: computer science, mathematics, linguistics, psychology, sociology, and information sciences. At IPSI, work on HCI is very much application driven but at the same time new problem domains and contexts require an in-depth treatment of rather fundamental issues. Beyond requirements from real world applications the design space is constrained through models from cognitive science. Furthermore, we use rapid prototyping cycles and involve end-users as much and as early as possible.
For content-oriented retrieval from multimedia databases our approach combines methods from logic-based IR and object-oriented database theory [5]. Because of the importance of interaction, we use a comprehensive conversational dialogue model [1,6] to develop a dialogue manager for multimodal interaction. This is complemented by visualization techniques for information seeking that exploit the human visual system to deal with large amounts of information.
We have built sample environments to explore communication processes between information providers and information seekers. Our prototypes allow the communication of goals and intentions, they present problems, suggestions and results to the user, and they are able to follow a continuous problem solving dialogue between human and computer, and keep that dialogue going. Our practical work centers around the design of generic methods and tools to access, visualize, reorganize, and restructure large amounts of multimedia information.
Using high-end Silicon Graphics machines we have also built systems that provide intuitive visualizations and direct-manipulative interaction mechanisms to access and explore information stored in very large multimedia knowledge- or databases. We design and construct 3D-visual information spaces [3] that retain more of the descriptive dimensions of real worlds than traditional information systems based on the desktop metaphor. Explorations in these environments can take place as spatial, 3D navigation on visualizations of information networks or as interactions augmented by retrieval mechanisms provided by the computer system. Appropriate 3D-visual metaphors are derived from real world behaviour of users and from the structures of the explored information content as provided by the data/knowledge base.
From the broad spectrum of hypermedia applications, we selected two areas to explore their innovative potential:
1) Experimental and innovative electronic publications transcend the limits of print media and ask for new tools for the planning, design and production of large hypermedia publications. Examples are tailorable publications (e.g., individual electronic newspaper), large reference works (e.g., Dictionary of Art), and technical documentation. The tools include an editor's workbench [4] utilizing an object-oriented knowledge base and an automatic visualization environment for the dynamic on-demand generation of hypermedia documents including automatic layout.
2) Hypermedia-based collaboration environments are our approach to CSCW. The supported activities range from group problem solving (e.g. brainstorming, discussion, and decision making) in face-to-face and "virtual" meetings to joint viewing and editing of documents in distributed desktop-based collaboration using also audio/video communication links. A prominent example is a project of developing support for government agencies distributed between Bonn and Berlin. The tools we built include the SEPIA cooperative hypermedia authoring environment [7] providing different degrees of shared workspaces and the DOLPHIN electronic meeting room support system [8]. Integrating large, interactive electronic whiteboards led to innovative pen-based user-interfaces using gesture recognition techniques and to supporting a wide range of structures from informal (e.g., scribbles) to formal (e.g., typed node-link structures) and their visualization. These cooperative work scenarios require also support for versioning of hyperdocuments [2].
Finally, we will report about the directions to be pursued in the future. Tools and methods developed to support information seeking activities will be adapted and applied to networked information. Building upon our expertise in constructing 3D visualizations (built for TV productions employing virtual TV studios), we will incorporate virtual reality techniques into CSCW scenarios. This will be applied to our ATM-based coupling of two or more electronic meeting rooms and several desktop-based remote partners in different locations, creating "Ubiquitous Meeting Environments" for Virtual Organisations. We will also continue empirical evaluation studies informing our design.
1. Belkin, N.J., Cool, C., Stein, A., Thiel, U. Cases, scripts and information seeking strategies: On the design of interactive information retrieval systems. Expert Systems with Applications, 1995, 9 (3), 379-395.
2. Haake, A., Haake, J. Take CoVer: Exploiting version support in cooperative systems. Proceedings of InterCHI'93 (Amsterdam, April 24-29, 1993), 406-413.
3. Hemmje, M., Kunkel, C, Willett, A. LyberWorld - A visualization user interface supporting fulltext retrieval. Proceedings of SIGIR '94 (Dublin), 249-259.
4. Hüser, Ch., Reichenberger, K., Rostek, L., Streitz, N. Knowledge-based editing and visualization for hypermedia encyclopedias. Communications of the ACM (April 1995), 38(4), 49-51.
5. Müller, A., Thiel, U. Query expansion in an abductive information retrieval system. Proceedings of RIAO'94, CID-CASIS (Paris), 461-480.
6. Stein, A., Maier, E. Structuring Collaborative information-seeking dialogues. Knowledge-Based Systems. Special Issue on Human-Computer Collaboration, 1995,8 (2-3), 82-93.
7. Streitz, N. Designing hypermedia: A collaborative activity. Communications of ACM (August 1995), 38(8), 70-71.
8. Streitz, N., Geißler, J. Haake, J., Hol, J. DOLPHIN: Integrated meeting support across LiveBoards, local and remote desktop environments. Proceedings of CSCW '94 (Chapel Hill, N.C., Oct. 22-26, 1994), 345-358.
9. Thüring, M., Hannemann, J., Haake, J. Hypermedia and cognition: Designing for comprehension. Communications of the ACM (August 1995), 38 (8), 57-66.