An essential ingredient of knowledge sharing programs in large organizations is the community of practice. (Why? Because trust levels in large organizations are rarely, if ever, sufficiently high to support open sharing of knowledge,)

   Definitions of community of practice vary somewhat, but are usually taken to mean a group of practitioners who share a common interest or passion in an area of competence and are willing to share the experiences of their practice. It differs from a work team, principally in that it has no specific time-bound work objective, but exists indefinitely for the promotion of the issue or issues around which the community is formed.

   In undertaking knowledge sharing programs, most organizations have found – sooner or later – that the nurturing of knowledge-based communities of practice is a sine qua non to enabling significant knowledge sharing to take place. Such communities are typically based on the affinity created by common interests or experience, where practitioners face a common set of problems in a particular knowledge area, and have an interest in finding, or improving the effectiveness of, solutions to those problems. Various tools can be used to strengthen such communities, including the establishment of specific work objectives for the community, the provision of adequate staff, financial resources, technology and management support to enable it to conduct its activities.

   Communities of practice tend to have different names in different organizations, including:

• communities of practice
• thematic groups (World Bank)
• learning communities or networks (Hewlett Packard)
• best practice teams (Chevron)
• family groups (Xerox)

   Communities of practice are relevant both the connecting and collecting aspects of knowledge sharing. 

• Connecting people who need to know with those who do know requires an element of trust that is often lacking in large organizations, particularly when it comes to sharing knowledge across organizational boundaries. Thus, asking for advice or other opinions can be seen in a low-trust environment as tantamount to an admission of ignorance. Advertising that ignorance across the entire organization is unlikely to occur if there is a risk that it may have personnel sanctions, particularly in organizations that are downsizing, or looking to save costs by laying off personnel.

• Collecting knowledge so that it can be shared through the web or other technology also comes to be dependent on communities, since it is only in communities of practitioners that share common objectives and pre-occupations that it can become apparent as to what knowledge needs to be shared. Efforts to build knowledge collections in the hope that "users will come" almost always encounter a disappointing response, since the builders find it difficult anticipate what knowledge users will want, and even if they succeed in theory, the users will regard the collection as something external and foreign unless they had a hand in designing and constructing it. The experience underlines the difference between a reference tool such as an information system and knowledge collections: the former can be effective if the information is relevant and correct, in the same way that a yellow-pages telephone directory can be useful. The latter depends for its dynamic and living quality on the active participation of those who use it, since knowledge is a much more personal affair than data or information.

    Communities depend on the fundamental finding that communities thrive on passion, and die from lack of it. That is to say, communities depend on the members of the community being interested in, enthusiastic about and committed to the issues around which the community is formed. Communities comprise volunteers, not conscripts, and the community exists only so long as the members are willing to contribute their time and effort to promoting the community and its interests. Communities cannot be commanded into existence, and if this is attempted, they will become something other than a community, except in name. 

   Launching and nurturing communities of practice for knowledge sharing programs can be accomplished in a variety of ways. 

• Endorsing informal communities that already exist: In almost any organization, there are bound to be at least some informal communities that exist, without management support or even awareness. Finding out what they are and where they are and how they can be supported can be an important first step in demonstrating that communities of practice are not something foreign, but something "home-grown".

• Asking practitioners what issues they care about: Often the most effective way of nurturing communities of practice is to consult practitioners as to what issues they care about, and worry about, and would like to find out more about. The results of such consultations which can take place in informal focus groups or in structured groupware sessions can then be used to elicit volunteers to come forward and lead communities built on the issues about which the practitioners care most. In this fashion, the chances of having communities built on issues issues about which members are passionate is enhanced. 

• Instructing leaders to form communities: Having the management instruct leaders to form communities is usually a recipe for disaster, unless the leaders so designated happen to be informal leaders on the relevant issues, or unless they elicit the co-leadership of  others who possess this informal leadership status. Without such support, the communities may exist on paper and may occasionally perform functions for the organization, but the dynamism and energy of genuine communities is likely to be lacking.

• Launching purely virtual communities: Purely virtual communities are known to have come into existence where the members are so strongly passionate about a particular issue that the question of trust is set aside, e.g. communities of individuals who are affected by an obscure disease, and who take to the Web in order to make contact with doctors or other patients who might be able and willing to share relevant experiences related to the disease. Inside large organizations, this kind of passion is rare, and organizations tend to find that purely virtual communities either cannot be effectively launched without face to face meetings, or if launched, cannot be sustained without regular physical meetings (say, once a year). Once the community is formed, however, a community may be very effective in its virtual communications even though its members are scattered around the globe.

• Launching communities among the "incorrigibles":  In any organization, one will find staff who are naturally and enthusiastically attracted to participate in communities of practice. One will also find staff who are naturally antipathetic to communities for a variety of reasons, including habits of hierarchical behavior, distrust of fellow practitioners, or organizational game playing. Provided that there are many active communities in core areas of the business, the knowledge management program might choose to ignore these "incorrigibles" either totally, or until such time as the measures to encourage compliance with organizational priorities (budgets, formal incentives, measurement) eventually "kick in", so that even incorrigibles find it difficult to go on working successfully in the organization, unless they at least go through the motions of participating in communities.

• A knowledge sharing vision that makes sense for the business: Since communities comprise volunteers, not conscripts, it is vital that the management has articulated a compelling vision of knowledge sharing that makes sense to front-line practitioners.  Without this, communities of practitioners will never form or survive.

• Budgets: Financial resources or budgets for communities can be an unmistakable sign that communities have an official status in the organization, even if they are not part of the formal organization chart. Resources may be needed for instance for: 
• building knowledge collections that are of relevance to community members and that help staff do their jobs better.
• outreach to the community members, such as holding informal meetings and clinics, issuing newsletters and the like; and 
• funding of ad hoc advice for those experts who will be expected to provide unprogrammable advice from time to time.

Care has to be taken that financial resources are made available in such a way that both the priorities of the organization and the independence and the dynamism of the community is preserved. 

• Independence: Some knowledge management practitioners question whether the provision of financial resources will not inevitably and ultimately co-opt the community for the organizational agenda so that it ceases to be a genuine community. Time will tell whether this concern turns out to be generally applicable. Many organizations find however that without the provision of financial resources, it is impossible to get the communities launched in the first instance.

• Incentives for communities: Participation in a community is a volunteer activity, based on personal interest and passion, and cannot be "bought" with financial incentives. Nevertheless, the reflection of knowledge sharing in the personnel system of the organization can send a signal, reinforcing the budget allocations, that participation in communities is not something extraneous or  "extra-curricular" but rather part of the core business activity of the organization.

•  discipline-based communities
•  cross-cutting communities
•  time-bound communities which are similar to organizational initiatives.
•  umbrella type communities which embrace a number of different sub-communities.

   Nevetheless, it is important to recall, in the effort to launch communities, that communities also have a downside: .

• Communities can be exclusionary - professional communites and scientific communities vis-à-vis other disciplines. 
• Communities can stifle innovation -as did the medieval guilds. It is important therefore not to give any one community a monopoly - actual or de facto - over any area of knowledge.
• Communities can be chaotic - boundaries are untidy - anything organic is untidy
• Communities cannot manage activities very well. Management needs a manager. Communites should not be used as a crutch by incompetent managers.

    The effectiveness of a community is most easily measured by surveying the members of the commnity and asking them whether and to what extent the community has proved helpful to them and why. Such surveys can however give false positives, in the case of communities which have become self-serving or exclusionary in their mode of operation. More valid, albeit more expensive, measures can be obtained by tracking client or customer satisfaction in areas covered by the community.

Inspiring communities of practice

You can't direct or order communities of practice to come into existence. Attempts by command-and-control managers to do this will inevitably backfire. But this doesn't mean that leaders can't do anything to bring communities into existence. Instead, they have to inspire people to form communities. How? That's the subject of my new book, The Secret Language of Leadership.

An indispensable reference on communities of practice is: :
   Etienne Wenger, Communities of Practice, Communities of Practice: Learning, Meaning, and Identity (Cambridge University Press, 1999).
   Harvard Business Review: Etienne Wenger and William Snyder: Communities of practice: The organizational frontier.   Volume 78 No.1; Jan/Feb 2000.
   Harvard Business School Case Studies: "Knowledge Management at the World Bank: The case of the urban services group" Forthcoming, October 2000.

  Stephen Denning, The Springboard: How Storytelling Ignites Action in Knowledge-Era Organizations. Boston, London, Butterworth Heinemann, October 2000.

  Stephen Denning: The Leader's Guide to Storytelling (Jossey-Bass, 2005) chapter 8.

  Stephen Denning: The Secret Language of Leadership: How Leaders Inspire Action Through Narrative (Jossey-Bass, October 2007)