Information nowadays is available to a large number of users over the Internet. Along with the ease of access at the technical level conies the need to support a wide range of users having different skills and information needs. Users' skills may vary at the level of computer literacy - knowing how to operate a computer and the software to access the information - and at the level of task or domain knowledge which is necessary to find relevant information. In existing information systems, there often is a sharp distinction between user interfaces and functions for beginners and for experienced users, making a smooth transition nearly impossible. Difficulties add up, if multiple heterogeneous information sources have to be accessed to satisfy the user's information needs. On the basis of Visual Formalisms and Query Previews we propose a user interface design for information systems which visualizes semantic dependencies between different information sources and the indexing vocabularies used by them. It exploits human's information processing capabilities and avoids the often incomplete transfer of knowledge involved with metaphors. At the same time it allows for user-controlled adaptation of the information presented on the screen, showing the inexperienced user only the information he can understand while allowing the experienced information seeker to stepwise add information necessary for advanced search strategies. Intelligent components transform search terms between the vocabularies of the information sources, a process that is transparent to the beglnner but can be controlled by the expert at a very detailed level.
展开▼
