In recent years, interest in intelligent agents has ballooned. With the WWW as a source of seemingly limitless opportunity and robotics platforms becoming cheaper and more easily maintained, the applications are myriad, challenging and within reach of even small groups.

Software agents have come a long way in the last ten or so years. This paper looks at the practical challenges many software agent developers face including: interfacing to legacy systems; developing distributed systems; agent competence; and trust and privacy issues. Because it covers such a broad area, it will focus on how the problems and issues have changed over the years, and highlight some of the perennial problems.

The First International Conference on Autonomous Agents brought together researchers concerned with implementing systems that perceive and act in dynamic, unpredictable environments, that coordinate interoperation among complementary component capabilities, and that perform significant jobs with a high degree of autonomy.

Autonomous robots are the intelligent agents par excellence. We frequently define a robot as a machine that senses, thinks and acts, i.e., an agent. They are distinguished from software agents in that robots are embodied agents, situated in the real world. As such, they are subject both to the joys and sorrows of the world. They can be touched and seen and heard (sometimes even smelled!), they have physical dimensions, and they can exert force on other objects. These objects can be like a ball in the RoboCup or Mirosot robot soccer games, they can be parts to be assembled, airplanes to be washed, carpets to be vacuumed, terrain to be traversed or cameras to be aimed. On the other hand, since robots are agents in the world they are also subject to its physical laws, they have mass and inertia, their moving parts encounter friction and hence heat, no two parts are precisely alike, measurements are corrupted by noise, and alas, parts break. Of course, robots also contain computers, and hence they are also subject to the slings and arrows of computer misfortunes, both in hardware and software. Finally, the world into which we place these robots keeps changing, it is non-stationary and unstructured, so that we cannot predict its features accurately in advance.

Software agents help automate a variety of tasks including those involved in buying and selling products over the Internet. This paper surveys several of these agent-mediated electronic commerce systems by describing their roles in the context of a Consumer Buying Behavior (CBB) model. The CBB model we present augments traditional marketing models with concepts from Software Agents research to accommodate electronic markets. We then discuss the variety of Artificial Intelligence techniques that support agent mediation and conclude with future directions of agent-mediated electronic commerce research.

This paper presents an Agent-based Process Management System (APMS) architecture that combines recent developments in autonomous agent technology and distributed computing platforms. It argues that a service-oriented view is required in order to meet the requirements of open distributed enterprises, now and in the future. This approach embraces and fully supports the decentralisation of responsibility and local autonomy, within and across enterprises. The paper begins by summarising existing approaches to workflow. It describes the APMS approach which extends workflow with the abilities to anticipate process requirements, to resource processes dynamically, and to adapt to exceptions. An important feature of APMS is the absence of a centralised business process representation. Finally, the paper describes an implementation of the APMS architecture, and discusses the engineering issues related to realising such a system within a large organisation.

This report summarises the papers presented in the expert assistants area at the First International Conference on Autonomous Agents. It describes the relationship between the expert assistants area and the other areas of the conference, and identifies the major research themes within this area.

This workshop brought together researchers concerned with fundamental issues of modelling action, those developing automated planning techniques and those attempting to implement autonomous agents acting in the world. In the past, these research communities have been separate from each other to a surprising extent. Researchers interested in theories of action have busied themselves with finding solutions to the frame and ramification problems, for very expressive theories of action. On the other hand, researchers interested in developing planning systems have typically concentrated on efficiency over expressiveness and assumed away the frame and ramification problems by means of the “STRIPS assumption”. Finally, researchers interested in implementing autonomous agents have found their attention occupied by issues of execution monitoring and sensing.

The first International and Interdisciplinary Conference on Modeling and Using Context (CONTEXT-97) was held at Rio de Janeiro, Brazil on February 4–6 1997. This article provides a summary of the presentations and discussions during the three days with a focus on context in applications. The notion of context is far from defined, and is dependent in its interpretation on a cognitive science versus an engineering (or system building) point of view. However, the conference makes it possible to identify new trends in the formalization of context at a theoretical level, as well as in the use of context in real-world applications. Results presented at the conference are ascribed in the realm of the works on context over the past few years at specific workshops and symposia. The diversity of the attendees' origins (artificial intelligence, linguistics, philosophy, psychology, etc.) demonstrates that there are different types of context, not a unique one. For instance, logicians model context at the level of the knowledge representation and the reasoning mechanisms, while cognitive scientists consider context at the level of the interaction between two agents (i.e. two humans or a human and a machine). In the latter case, there are now strong arguments proving that one can speak of context only in reference to its use (e.g. context of an item or of a problem solving exercise). Moreover, there are different types of context that are interdependent. This makes it possible to understand why, despite the consensus on some context aspects, agreement on the notion of context is not yet achieved.

This report is a summary of the Third International Workshop on Rules in Database Systems (RIDS'97) (Geppert & Berndtsson, 1997) which was organized by the database groups at University of Zurich (Switzerland) and University of Skövde (Sweden), and was held at the University of Skövde, June 26–28 1997. Following the success of RIDS '93 (Edinburgh) (Paton & Williams, 1993) (1st International Workshop on Rules in Database Systems), and RIDS '95 (Athens) (Sellis, 1995), the successor RIDS '97 brought together researchers working on both theoretical and practical aspects of rules in database systems.