Abstract
This paper uses empirical evidence drawn from a sample of IT-based innova tion projects to examine theoretical models of the organization of technical knowledge. It suggests that Williamson's transaction cost account is incom plete, in that it fails to take account of the role of social action in communicat ing technical knowledge and organizing transactions. As an alternative model, it describes the 'strategies of social closure' developed by IT experts to achieve effective knowledge trading and a mutually acceptable division of the eco nomic gains of innovation. Empirical data is used to describe three such strat egies : 'blackboxing', 'hostage' and 'prisoner' strategies.