The open agent society and its enemies: a position statement and research programme

Virtual enterprises and connected communities are common visions of future commercial and social structures. Underpinning these visions is the idea of the open agent society: a flexible network of heterogeneous software processes, each individually aware of the opportunities available to them, capable of autonomous decision making to take advantage of them, and co-operating to meet transient needs and conditions. Furthermore, the society should be regulated by the kind of relations (contractual and normative) found in human business and social interactions. This paper reviews experience with developing, deploying and evaluating multi-agent systems, and distills from this some of the drivers for the open agent society. We then identify three crucial innovations which will help realise this idea. However, the development of the open agent society is exposed to certain risks, and we consider a number of `enemiesʹ which threaten its development. We conclude that an awareness of the risks entails new paradigms for engineering multi-agent systems, in which dynamic social relationships are as important as interface definitions in providing the interoperability required for the open agent society.