From Agents to Electronic Order

Trust, reputation, norms and organisations are all relevant to the effective operation of open and dynamic multiagent systems. Inspired by human systems, yet not constrained by them, these concepts provide a means to establish a sense of order in computational environments (and mixed human-machine ones). In this talk I will review previous work across a range of areas in support of the need to develop theories and systems that provide the computational analogue of common social coordination mechanisms used by humans, in addition to those that might only find favour in computational systems. I will focus on particular examples that illustrate different approaches, including through the use of norms and contracts, and suggest some key challenges that need to be addressed to drive the field forward.