Introduction

In order to fulfil the needs for organisational competitiveness and innovation, a significant role is generally assigned to knowledge management, and, more specifically, to the information system as a manager and a repository of knowledge.
This "hard" (technology centred) point of view is bound to the hypothesis that it is possible to use information systems for capturing and storing knowledge residing in the mind of people working in organisations. An additional assumption is that this stored knowledge can be use other people, other projects or other organisations. The technology-centred scenario is today debated by a number of authors which point out the need to focus on people and their interactions (Swan, 1999). Technology should be integrated with a consideration of knowledge transformation problems and with adequate relevance to the role of social interaction in sharing and gathering knowledge (Nonaka and Takeuchi, 1995). Authors like Davenport e Prusack (1998) also highlight the significant role of people in this context: they combine a sense of urgency for starting new initiatives with the indication of avoiding the mistake of taking into account only technology issues.
There is a discrepancy between the apparently good heath of the discipline on one hand and the appearance of concrete failures on the other hand. The problem seems to descend from a fundamental hypothesis: the belief that it is possible to build information system for managing knowledge without centring the methodology on people and their interactions.
The crisis of the methodological apparatus of the information systems discipline has been underlined also in a wider context. Ciborra (1998) pointed out that, notwithstanding the fast growth of the discipline (highlighted by the number of relevant applications and scientific papers), there are significant signs of crisis. The two most significant events in the last few years were not designed nor forecasted. The events are the birth of the Linux operating system and the explosive growth of the Internet: neither of the two was planned. It was not simply a failure of interpreting signals, but rather as a methodological failure of the discipline. The engineering-centred approach shows its limits when there is need to design system which will be used by people and organisations (Lewis, 1994).
Communities of practice (Wenger, 1998) are of prominent interest because of the role of knowledge and representation sharing, beside the importance of the creation of practices and the building of collective identity. If the community does not share a context physically and organisationally delimited, the use of technology becomes relevant. This is the main focus of this project: the role of technology in a virtual community of practices. The reference model for our analysis is the one defined by Mantovani (Mantovani, 1996) that underlines the social context in which the technology act as a mediator of human interaction.