At the 1999 German Database Conference (BTW) Gerhard Weikum gave this invited presentation on quality of service issues in information systems. The paper first reviews the notions of service quality and dependability and classifies the parameters into system-centric, process-centric and information centric notions. Gerhard argues that, despite the modest advances the community has achieved in proving overall system correctness, one should strive to establish certain local service guarantees. He illustrates the derivation of service quality guarantees on a distributed workflow processing application. The paper is concluded with a list of strategic objectives and research avenues that need to be taken in order to advance the state of the art in this area. In particular, the need for long-term research agendas rather than "short-time-to-market" research regimes is demanded.

I particularly liked this paper because of its insightful and critical assessment of current distributed information system's service quality (which is often non-existent as we see on the Internet too often). Furthermore, the paper looks beyond the database area and surveys some relevant work done in tele-communication and distributed program verification. The case study of the workflow system is an encouragement that, by using the tools developed in these areas, one can already obtain useful local results - even though we are a far cry away from a comprehensive "tool box".

a service of  Schloss Dagstuhl - Leibniz Center for Informatics