Let's look at circumstance 1, where no employees are using online technologies in any organized way to share knowledge with each other. Email is available to employees; there may or may not be an intranet or portal with online discussion space. A team or work group has discovered (and agrees) that a more effective environment and process for exchanging knowledge would help them and the company excel.

In such a circumstance, we recommend minimizing the time required to propose and sell the idea and maximizing the use of available technologies and existing social agreements to begin the online knowledge-sharing process. Learn by doing. Demonstrate incremental success. Follow the recommendations made later in this chapter under Spontaneous Conversational Communities, and once the network has proven its value, point to that success in a more ambitious proposal. By that time, the group will understand the technical improvements it actually needs for more productive mutual learning and for putting its shared knowledge to work.

Your organization might be open to the idea of skipping these learning steps and diving right in to redefining itself as a knowledge-sharing company. This is a challenge, and we know of at least one company (Buckman Labs, cited in Chapter 5, "Fostering a Knowledge-Sharing Culture") that has pulled it off successfully and become a good example, but full acceptance and adoption of its new system and practices took 3 years. Two conditions must be satisfied for the strategy of skipping the incremental grass-roots learning phase to be a good bet.

The cost of the interface does not have to be a major obstacle in getting approval or starting the network conversation. In fact, simplicity sometimes works in favor of the social interaction rather than against it. In our experience, we've witnessed higher cohesion and interactivity in online communities where the interface was not ideal and presented users with some challenges. In groups such as the early WELL, the early incarnation of Women.com, and in many email discussion groups we've managed and participated in, lack of features and sophistication in the discussion software and online environment forced people to try harder to communicate. They had to solve problems as groups to make the interface work. They were forced to collaborate to invent solutions and raise the quality of the communication.

In taking the cheap and simple route, there are trade-offs, of course; some people won't participate if the interface lacks certain features. Some of the creative energy of the group is used up in figuring out its online conversational processes. But there is something Darwinian at work when the interface selects for the most committed and innovative users. The filtering effect of what amounts to an interface boot camp is actually a good way to select participants for the startup stages for an online community when what you want are some willing pioneers to settle the new environment and civilize it for the later arrivals.

In sum, where the organization has not yet invested in leveraging online conversation, make the best possible use of tools and internal resources that are already available. Build a pilot network using email or available message boards and take the interaction as far as possible using technology that doesn't cost the company anything. Set reasonable goals, provide good leadership, document achievements, and demonstrate improvements over previous methods of collaboration in the performance of network members or of the group as a whole. The best way to sell an unfamiliar concept is to demonstrate its potential to change the organization in alignment with its strategy.

Once an online conversational community has demonstrated its effectiveness within an organization, circumstance 1 is instantly transformed into circumstance 2, with experiential evidence to support the proposal. A track record and a functioning prototype will increase the chances of the proposal being approved.

The leap in effectiveness from conversations using email to conversations using a well-designed message board with content management can be significant depending on the group and the nature of its knowledge exchange. The proposal must outline how the current technology is holding the community back from achieving its potential. The specific gains that can be realized through better supported conversation must be detailed.

As described in Chapter 7, "Choosing and Using Technology," there is a wide range of software and platforms appropriate for knowledge networks. The scale of the conversational community must be large enough to justify the most expensive platforms by spreading out the costs over more users. There may be substantial costs for integrating the platform into the existing technology. There will be costs in training and refining the design once the platform is in use.

A joint proposal representing the needs of other groups within the organization stands a better chance of approval. There may be many other groups expressing the same needs to engage online through better software and wanting more training or more time to spend in online collaboration. The more groups that are allied behind the proposal, the more leverage there is for getting necessary funding. Be willing to compromise and accept software solutions that are less than perfect, at least as an interim concession. Different groups will have different visions of the ideal platform, but if the major features can be agreed on, all groups may benefit to a greater extent than any one of them could on its own.

Remember above all that it's the motivation and focus of the group, not the interface it uses, that is most responsible for the effectiveness of its knowledge exchange. It can be a blessing in disguise to be forced to bootstrap the community conversation from the ground up. If motivation and focus can carry the load in the beginning, the community will eventually earn itself access to better tools, and selling the idea will be relatively easy.

In summary, a group, community, team, or line of business that recognizes a need for more powerful online knowledge sharing should approach the appropriate decision maker with the following:

Knowledge exchange won't happen online or off if there are cultural disincentives to telling others what one knows. If people feel that they will somehow lose opportunities, status, safety, power, prestige, or value in the company by sharing their expertise, their part in a knowledge-sharing conversation will be pretty quiet. A company may take pains to remove disincentives for contributing knowledge, but rather than rest at a neutral attitude toward sharing knowledge, there should be clear positive incentives for workers to join the online conversation.

Time constraints are a disincentive for many busy workers. One of the main reasons people refuse to adopt any new practice in an organization is the complaint of too little time. Their schedules are full, and asking them to learn and then participate in an activity that may cause them to fall behind in their other work is only going to anger them. Eventually, participation in the knowledge conversation will prove to be a time-saving activity, but there will be a transition period during which the employee has to make a choice: Should I drop what I'm doing and try this new thing or play it safe and keep delivering on my current responsibilities? Only management can make this decision easier and less risky for the employee. Some HR departments have begun allocating extra time for employees to spend seeking knowledge rather than reinventing wheels.

As we've described, an interface that puts barriers in the way of participation or that requires training and practice that consume valuable time can cause potential participants to lose interest early in the process. Using interfaces with which the population is comfortable, even if they lack key features for knowledge exchange, removes these early stage disincentives.

Financial incentives can be structured to reward individuals or to reward work groups, teams, and divisions. They can be offered as bonuses based on amount of participation in conversation, significant contributions to the knowledge pool, or effectiveness of leadership in a knowledge-generating community. There is little experiential evidence by which to judge the long-term effectiveness of such compensatory incentives.

Davenport and Prusak report that a large consulting firm "revamped its performance appraisal system to include contributions to the firm's knowledge database as an important part of compensation decisions."^[80] This may work well in companies that sell knowledge but may not be as practical where knowledge is less central to the business strategy.

New software products record employee activities in ways that can be used to construct incentive reward systems. A product called "Clerity for Enterprise Knowledge Sharing" advertises that its "software and methodologies can create customized, automated incentive systems for your organization that can include points, recognition awards, linkages to human resource and employee review systems or other special programs."^[81]

One of the most powerful and effective incentives is recognition by one's peers as an expert or holder of key knowledge in the group. Being invited to contribute to a conversation is like being invited to speak at a conference; even if you don't get paid, it feels good to be on stage as the center of attention. It feels good to tell people things that you know and they don't. Even being distinguished in the knowledge community's yellow pages directories can be satisfying.

For some, participation in a new or cutting-edge activity such as a knowledge network is reward in itself. The opportunity to be a big fish in a little pond or a pioneer in the company's newest project can be a great attraction. The opportunity to contribute to a community in a tangible way is refreshing to many people whose work seems to disappear into the ether with no response or feedback from its destination. In a conversational community, the likelihood of response is much greater.

Members of a community notice who contributes and whose contributions are most helpful. Reputations can be built in the online social context that may not be accessible in any other form to desk-bound workers. But reputations work in both directions. Those who exhibit selfish or impatient behaviors in a collaborative group can earn themselves social demerits. Redemption is fairly simple, though, because interaction is ongoing, and apologies and reparations are just as visible as offenses and insults. As we pointed out in Chapter 7, software that allows members of the group to rank each other's performance can inhibit participation in the early growth stages of the community. Use such features with discretion because they work best when people can't or don't take the time to form personal relationships.

The nature of a knowledge-sharing conversation is that helpful information and ideas move in both directions: from the participant to the community and from other members of the community back to the participant. Once the reliability of the exchange has been established, contributors of knowledge are likely to return because they expect that they will be enriched in equal measure to what they have given. Reciprocity does not necessarily work throughout the organization. People don't have agreements to reciprocate, and they have limited opportunities to do so in their irregular encounters with each other.

In an online environment dedicated to the exchange of knowledge, it's easier to keep one's books straight on who has contributed and who has not. Because the transactional record is there for everyone to read, a sense of balance and fairness is apparent, at least to regular participants. Once the knowledge marketplace has begun working, members know that they can offer what they've learned and that they can learn from what others offer. The incentive to participate becomes part of the marketplace mentality. Marketplaces are conversations.

The most basic incentive to participate in any community is the sense of community that comes with it: the feeling of being part of something bigger than one's self. The social rewards of membership in a group can be compelling, especially if the group has a purpose and an agreement to collaborate civilly toward achieving common goals. Leadership and role modeling are important in establishing such cultural agreements, and mastering challenges together reinforces the bonds that make teams and conversational communities more efficient.

Thus, the founding members or central interest of a knowledge-sharing community may be the most effective incentives for participation by others. Communities of practice form around common interests, and their motivation to pursue those interests can overcome even powerful disincentives like lack of time and primitive conversational environments.

To take it one step further, we recommend without qualification that people use the Net to tell stories to each other. This is no longer a belief held by an intellectual fringe; stories are now being adopted as the format for everything from organizational change to product branding. Storytelling is a skill worth learning and could become a distinguishing feature of leadership in a knowledge-oriented culture. People react to stories; we seem to be wired internally to follow from beginnings to middles to ends and to interpret deeper meaning from the tales we hear. Stories elicit emotional responses in us and awaken the human part that wants to connect with others. And some stories make us think.

Within an organization, especially one driven by business values and profit, the humanizing effect of sharing experiential stories builds trust. Trust, as we've emphasized throughout this book, is powerful stuff. A knowledge network can be effective without fancy software, but it can't be effective without trust. People must feel safe to pour out what they know and believe to be true. Stories not only provide an easily interpreted explanation of knowledge, but they also reveal the storytellers as fellow humans, making them more real and trustworthy to the virtual audience.

The storytelling style should be modeled and encouraged in all online conversations aimed at sharing knowledge. Elements of effective storytelling should be part of the organization's training for knowledge community leaders and facilitators.

Describing how one learned something not only imparts the knowledge, it wraps that knowledge in an applicable lesson. True knowledge sharing is not the simplistic activity of asking and answering questions; it's a process of raising the collective intelligence and problem-solving capacity of the group or organization and providing the group with better handles for putting knowledge to work and implementing new ideas.

Stories also are more interesting to read and write than dry facts and information; they make participation in an online conversation more enjoyable and engaging, retaining more participants. In the pursuit of practical knowledge, stories provide context. They serve as a framework for delivering one's experience to others in ways that teach and inspire as well as inform and entertain.

Stephen Denning has a special view and purpose for telling stories. In The Springboard,^[86] he describes the story as "a launching device aimed at enabling a whole group of people to leap—mentally—higher than they otherwise might, to get beyond mere common sense." This is more than a teaching function or simple knowledge exchange. It is more akin to intellectual stimulation or a call to action. Denning's stories are short and tend to describe possibilities based on actual events. They are carefully assembled and delivered. The springboard story is meant to catalyze organizational change.

Springboard stories share some common elements. They generally begin with a protagonist and what he calls "the predicament" that describes a problem the audience can relate to. They then move on to "the resolution of the predicament, embodying the change idea." This is meant to stimulate audience thinking, where they apply the change idea to their organization. The story wraps up either with "extrapolating the story to complete the picture" or with "drawing out the implications." Sometimes both elements are included. They help the audience connect the change idea with practical implications for their specific situations.

Denning's book assumes its readers will use his lessons in face-to-face situations. Drawing from his recommendations, we've adapted them to the virtual meeting place and present those adapted recommendations as follows:

Stories fit well within all three of the community types we are about to describe. One story may be the seed that launches a conversational community, or storytelling may be the preferred style for a knowledge exchange community.

Change is a dire need in many organizations and an aspired-to value in others. The springboard story—brief, illustrative, relevant, sometimes inspiring, sometimes unsettling—is a valuable instrument for catalyzing and directing change. It can be used by managers and team leaders but can just as easily be applied by frontline workers. As an igniter of innovative conversation or as an initiator of a new knowledge-sharing community, a springboard story can be quite effective.

The tone and wording of the first content in a new online discussion are very influential on the subsequent interaction. When the words and tone are inviting and communicate openness and enthusiasm, they are more likely to elicit response than dry, businesslike remarks. It's okay to overdo the emotional content somewhat to offset the impersonal nature of the medium.

A conversation space for knowledge exchange has a purpose. The question is: How specific is the purpose? This should be clarified at the outset of every new conversation or collection of conversations. A reminder on the entry screen to the discussion space is a good idea, and attention to the labeling of topics helps remind participants of their focus of interaction. If there is a deadline or time limit on the conversation, a facilitator should be minding the clock (or calendar) and leading discussion toward resolution in advance of the deadline.

Facilitation may also entail regularly summarizing lessons learned, points made, and issues still left unresolved. A facilitator, if there is one, has specific responsibilities that also should be made clear to participants. In some cases, a facilitator will serve in the role of debate moderator or even referee, charged with helping to resolve or settle arguments. The classic role of a facilitator is to help meetings and discussions stay on track toward a goal by getting people involved and clarifying their contributions to the conversation. If there are disagreements of substance, the facilitator points out the relevance of the disagreement to the group and attempts to lead it to consensus or resolution.

If some people hold special roles in the community, as moderators, facilitators, hosts, or leaders, their identities should be made clear to the rest of the community. New joiners won't necessarily recognize which members hold positions of authority, and for the sake of convenience and orderliness, they should have special user names or other obvious designations in the online environment.

As important to smooth knowledge flow as any other factor is the building of trust and relationship. These may happen offline already, but if there are no opportunities for community members to meet in person and spend time with each other informally, the opportunities must be provided in the online meeting space. The kind of interaction that builds trust and relationship is not the same as knowledge sharing, so it's wise to make a clear partition between knowledge exchange areas and social hangout areas as the site is designed and organized. In a simple email conversation list, participants should take the time to correspond with each other individually rather than overload list recipients with social chitchat.

Some veterans of the activity believe that online communities are good for extending conversations but bad for resolving them. Because in most cases the conversations happen without moderation or facilitation, they are subject to an effect that could be called the last word obsession. Everyone wants to get in the last word, and if the conversation has been left open to additional posts, the issue being discussed may never be settled. The cure for this is to include a moderator or facilitator in any discussion that must be resolved or in any community that shows tendencies to argue rather than reach agreement. Responsible participants can control their urges to get in the last word by simply paying attention to the sense of the conversation. Every conversation has a natural ending, and learning what that means is part of learning to be a good online conversationalist.

In the same vein, a knowledge-sharing community should learn the value of avoiding unproductive arguments. These not only distract other members from pursuing the knowledge they need, but they can poison the social well, eroding trust and making participation unpleasant enough to overcome whatever incentives drew people to the community in the first place. As the old Western sheriffs used to say, "Check your guns at the door." The purpose of the community is to get smarter, not dumber.

Some people avoid argument by posting something controversial and then logging out, without returning to deal with the responses they have provoked. This, too, can be deadly to a culture based on goodwill and trust. Accountability is important in professional communities, unlike some consumer-based forums where anonymous participation is invited. If you post something, take responsibility for it and be around for follow-up discussion. Be yourself; don't assume roles that will confuse people about your true identity, what you really think, and how you really behave. If the purpose is to exchange knowledge, don't mix fact with fantasy.

All participants in an ongoing online knowledge-sharing community have equal responsibility to be clear in their communication. This means being a good reader (the text-based analog to being a good listener) and taking part in coherent discourse. Good writing skills should be developed for the sake of the reading audience, and attention should be paid to details of spelling, grammar, composition, and the organization of thoughts. It all makes a big difference, not only in the regard in which the person will be held but also in the overall quality of the knowledge exchange. You'll appreciate good writers when you've been subjected to enough bad ones.

Remember that an online discussion community is a postocracy where those who post their comments and responses get more attention than those who don't. As Woody Allen said, 80 percent of success is showing up, and in an online community, that's how recognition works. If you make your presence known, you're part of the conversation. If you don't, then you aren't. And where reciprocity is a big incentive for participation, it's important that every member contributes his or her share to the knowledge pool.

Spontaneous communities that provide value to the organization are more likely to be approved, supported, and promoted. How do they provide value? Here are several ways:

Spontaneous communities form (and dissolve) constantly within organizations whether the management is aware of them or not. Many of them make use of the internal resources such as paid time, office space, and bandwidth on the intranet. It's in the organization's interest to get some value out of them.

Some grass-roots groups come together spontaneously to form true communities of practice. We'll describe their special case later in this chapter. Other groups are less focused on pursuing common interests and more interested in having convenient access to an online forum where they can get reliable and expert answers to their questions. Such forums may form spontaneously, using whatever communications tools are available. Or they may be provided by an organization on its intranet as part of its knowledge management strategy. Often, though, spontaneous communities appear to fill a vacuum, becoming the organization's first experiments in online community.

Spontaneous communities may be the explorers and pioneers in an organization's migration to new cultural practices and new customer relationships. Working behind the scenes, they take chances and occasionally suffer setbacks, but at minimal cost to the organization. They gain experience in managing the social processes of online conversation and relationship building. They learn to deal with the shifting dynamics of group interaction over time. They become the organization's local experts in virtual community. Their accumulated experience is an asset that the sponsoring organization should both realize it has and put to good use.

Knowledge-sharing networks initiated by workers are fertile incubators for new ideas and practices that can enhance the organization in many ways. Collaborative networks build intellectual and social capital for the company. The initiators and leaders of productive spontaneous communities may help the organization to expand knowledge-sharing practices as internal consultants. Experienced knowledge networkers can author training programs to disseminate new collaborative skills throughout the organization.

Where do grass-roots efforts fit within an organization's hierarchy and management processes? Who owns these bottom-up projects and who makes use of their output? Who harvests the knowledge they generate through their conversations and the social techniques they invent? Who decides whether they are a valid and productive use of resources? How involved should a company be in providing and maintaining the technical means for their formation and operation?

Peer-to-peer networks link individuals and offer perhaps the optimum in personalization through the network. They encourage and support each user's decisions about who is included in their personal network and how to interact with each member of that network. The individuality of the interface in a program such as Groove may work both for and against fluid collaboration. The exclusivity that each individual can enforce can eliminate extraneous communications, but it can also block what could be valuable input. In practice, a P2P knowledge-sharing network should be managed in much the same way as an email list. If used as the operational base for a spontaneous knowledge network, the members of the network must pay attention not only to the focus of their interest but to what they might be missing by using an interface separated from the intranet.

There's a romanticized idea that cabals of workers, meeting in secret and independently developing projects below the radar of the organization—so-called skunk-works teams—are good things for the organization. Because of a few notable successes, legend has it that skunk-works teams are made up of geniuses, and their inventions invariably revolutionize the company, which looks dumb for having forced its best employees into secrecy.

In fact, skunk-works teams are not commonly composed of geniuses and do not often invent successful products. We don't hear much about their failures, of course. And because of their invisibility, whatever good things they do discover in the way of new knowledge and process are not made available to the organization. So although spontaneous communities might warrant some protection from the effects of overexposure and overpopulation, they should remain visible and active contributors to the organization.

That said, here are five recommendations for realizing value from spontaneous knowledge-sharing communities:

In objective and dollar terms, the returns from spontaneous networks are difficult to measure. Anecdotal evidence is useful, though, because it's also difficult to measure the time lost and frustration experienced in seeking vital knowledge through database searching and not finding relevant answers. Spontaneous networks don't cost the company much to support and administer, but to make them as productive and useful to the company as possible, there are things the company can do to bring out the best in them.

Self-starting, self-governing, self-sustaining innovative teams that provide new knowledge to the organization are valuable assets. The same can be said about employees who communicate and collaborate productively without the need for micromanagement or IT handholding. A company can build these assets for the relatively cheap price of some meaningful cultural support. Creative people aren't as likely to initiate spontaneous networks if they suspect the company doesn't approve of such activities. They're much more likely to start networks if they know it's not only okay but appreciated by the company. So an unambiguous statement by the leadership of the organization, as we described a few paragraphs earlier, is important.

As we all know from our office lives, most people are satisfied to use whatever means are convenient and available to meet with and learn from each other. We meet in the company's cafeteria, in hallways and lobbies, or in temporarily vacant offices. It's not the environment that matters; it's the contact, the content, and the effectiveness of the communication.

In the online realm, people who simply want effective communication with a small group may be satisfied using email or the commercial messaging clients they download from Microsoft or AOL. They don't think in terms of asking the company to provide them with a new software platform. People intent on spontaneous group communication don't look for elegant solutions; they tend to adopt what is familiar to group members.

If the organization is aware that such groups are forming and recognizes potential benefits from their collaboration, it can provide support to help them succeed. It may be in the form of technical help in setting up email lists or custom configuring a private discussion space on the company intranet. It may be in providing authorization and entitlement for restricted access to the group's online home. It's important that the culture make people feel safe in proposing new knowledge-sharing groups and in selectively supporting them when they come forward with their ideas.

One characteristic of participants in spontaneous knowledge networks is their sense of independence and freewheeling exchange. As organic entities, they may invent or adopt styles of interaction unique to their groups and very different from the visible artifact level of the organization's overall culture. In an online environment, a very strong sense of unity is possible. That sense may be creative but also rebellious, confident but also vulnerable and wary of any management attempt to control or interfere with its self-governance. The relationship between an organization and the spontaneous knowledge networks it supports should resemble patron and artist: The sponsor supports the artist from a distance with faith that the creative spark will pay the patron back, and then some.

This is not to say that the organization must support every group that wants to use its network infrastructure as a regular meeting place; resources do have limits. But given limited resources, the organization should first support the aspiring communities that are most important to (and aligned with) its strategy. In Chapter 3, "Strategy and Planning for the Knowledge Network," we described "knowledge of maximum leverage," explaining how groups that seek, share, and generate such knowledge should get priority treatment and support.

Successful spontaneous communities advance strategy and cultural change most effectively if their work is transparent to the organization. They should serve as prototypes and learning models to bring maximum value to their sponsors. As Davenport and Prusak write: "it's not terribly difficult to envision ways of using knowledge more effectively in business strategy. The difficulty, of course, is in making changes to strategic programs and adopting the necessary behaviors throughout your organization." Spontaneous knowledge sharing helps push those changes along while modeling those necessary behaviors.

Communities that form spontaneously are just as likely to dissolve spontaneously. Because they are powered by passions and interest, they are subject to the periodic fading of those power sources or to the loss of key participants and leaders. The sponsoring organization should stay aware of the various segments of its knowledge network, not only to learn from them but also to know which of them are vigorous, which are dying, and which are going through periods of activity and quiet.

Networks formed by special project teams fire up quickly at birth and burn brightly, fueled by high focus and motivation until the project is done or the problem is solved. The knowledge sharing required to meet the deadline is intense. After that, their members disband and the activity fades out. Their reason for being has disappeared, and their purpose has been fulfilled.

Other networks form for a purpose that remains vital; the problem doesn't ever get completely solved but changes shape and requires the group constantly to adapt. These groups may show cyclical ebbs and flows in their activity and creative spirit. The organization should be aware of these dynamics in its management of the communications platform and support of the groups that are using it. There should be regular communications between the leaders of spontaneous networks and a clearly identified administrator of the communications platform being used so that communities going through lulls in activity are not mistaken for communities that have ceased to be.

Though we recommend that they serve as prototypes for the company, the success of a spontaneous conversational community can be its worst enemy. When such a group is regarded as a best practice example, labeled by the company with a gold star, it may gain prestige but lose the anonymity that was part of its appeal. Suddenly, waves of people want to join the community or observe what makes it tick. Executives in the organization want to analyze it and break it down for replication. Tech magazines may even want to write an article about it. Its world is changed into a fishbowl, and the uninhibited spirit that was responsible for its initial success is threatened. Most people stop feeling so open and creative when thrust into the roles of cast members in a dramatized online demo community.

External forces can shut down free-flowing conversation, but equally damaging to the creative spirit can be the inflated sense of self-importance that cohesive communities often develop, especially after being recognized as exceptional and successful. Such communities may become isolated from the rest of the organization, too exclusive to accept input from nonmembers. The WELL became that way after it began getting mentioned in the national press in the late 1980s. As new members appeared, they were challenged. Members began to be preoccupied with the value of their words and of the WELL's conversations to the detriment of the quality of the discourse that had attracted the press attention in the first place.

So how can the organization make the best use of a successful community's example without "killing the goose that laid the golden egg"? First, don't allow the online meeting places of such groups to become tourist attractions; keep them closed to all but their natural community members. If the community has something to teach the organization, arrange an online forum, event, or in-person meeting where its members can answer questions and describe their techniques. In general, use discretion when heaping praise on grass-roots conversational groups or on any productive collaborative groups, regardless of their sources. Learn from their success and provide a repository for lessons learned about their techniques, mix of talents, and style of communication, but don't turn the glare of the spotlight on them and distract them from their good work.

Beyond the realm of organizational values, missions, and policies, though, there must be an online medium available to people in the workplace through which they can participate in such communities. One medium remains, after almost 40 years, the most ubiquitous and understood means of group communication over the Net: email.

The founding participants of the Kraken at Price Waterhouse saw themselves as a group of "self-selected creatives" looking for a way to collaborate more innovatively than they could within the established structure of the large consultancy. To that end, they made use of their Lotus Notes email, a tool that was accessible to all employees and could be configured appropriately without having to submit a formal proposal, wait for funding, or explain their needs to IT.

Their technique was simple: Make a list of email recipients that was shared by all of the original members. Every message reached everyone on the list. Every response to every message reached the same list. Any individual was free to read or ignore each message. Success relied only on regular participation and the perceived value of new knowledge (or of appreciation) that each member realized by participating.

They named their interactive email list after a mythological sea creature of great power (the Kraken) that lived largely out of sight. Word soon got around about this low-profile but exceptionally productive knowledge exchange, and others were allowed to join—to have their names added to the email list. Even after the merger of Price Waterhouse with Coopers & Lybrand (another large accounting and consultancy firm), the Kraken continued to exist and grow to about 500 members. Its success has since spawned new internal email lists for knowledge exchange within the firm, serving different purposes and including different categories of expertise.

The Kraken is notable as an example of spontaneous knowledge exchange for several reasons:

This last reason is especially telling because following the merger, PriceWaterhouseCoopers put in place an elaborately designed interface called KnowledgeCurve to expand its knowledge management activities. KnowledgeCurve was not used as often or with as much enthusiasm as the email-mediated communities. It was not considered by employees to be as effective in disseminating usable knowledge.

The strength of email is that it is so basic and adaptable. Every employee with a networked computer on the desktop now has an email account and knows how to send and receive messages. No new software is required; the IT department doesn't need to install or integrate new software into its systems. People are free to adjust their participation according to their available time. The tool does not drive the worker, so there is less resentment of its use.

Nevertheless, some practices should be learned and adopted to make email productive for knowledge sharing. Because there is so little structure—for example, an email conversation does not live on a specific Web page and cannot be seen as a scrolling list of responses on a series of Web pages—motivation and incentive serve as the only bonding agents to initiate the conversation and keep it going. New messages to the group must be circulating almost constantly or people tend to pull their heads out of the conversation and forget it's even going on.

Because most ongoing email conversations happen without leadership or moderation, it's contingent on the participants to maintain the focus, to make the list readable and worthwhile, and to occasionally reflect on what has been discussed, decided, and accomplished through the interaction. The original sense of purpose must be reconfirmed regularly to keep incentives for active participation alive.

An email list should be set up by having IT provide the initiator with ownership of a listserv run through the organization's mail server. Listservs offer options for managing the correspondence that personal email clients don't have. They offer convenient means for registration and joining the list and allow the owner to control access through the member list. A company portal can even provide the means for individuals to start listservs without help from IT.

One of email's main assets is also one of its main liabilities: the lack of a distinct home on the computer screen. Except for when new messages arrive, there is no location where the conversation can be found and revisited. There is less to keep track of, but when one wants to review the messages that made up a recent conversation, the medium makes such a task clumsy and time-consuming. For many people, an identifiable location on the Net where conversations can happen and be stored for reference makes more sense.

If the organization's culture is friendly to it, spontaneous knowledge sharing will happen when virtual meeting rooms are easily available. Many, if not most, corporate portals feature links to "community" or "collaboration" where discussion boards can be found. These are often dedicated to the business line associated with the employee, but some companies provide discussion space for general use and access, permitting conversation across business lines. As we pointed out in Chapter 7, "Choosing and Using Technology," commercial portal products vary in their ability to integrate applications from different vendors. Many portals bundle the eRooms message board, whereas others allow standalone discussion interfaces such as Web Crossing to be integrated. For spontaneous communities, the feature set is not as important as the easy availability. The motivation of the spontaneous group can overcome many shortcomings in the interface.

Email, as a home for ongoing focused discussion, has many limitations for scaling, expanding, and diversifying the conversation. In Chapter 6, Taking Culture Online, we described the differences between linear and threading discussion interfaces and their effects on social interaction. Either one of these offer advantages over an email list in that they both provide a clear, graphic depiction of the conversational landscape. Coherent conversation in email relies on the conscientious use of the Subject line to distinguish one thread from another amidst the flow of messages, but discussion interfaces display conversations and related threads in stable formats that can be navigated easily by new members of the community. Discussion systems provide a history of the conversations and how they developed. They include member profiles that can be referenced to add context and credibility to the opinions and knowledge expressed. Some of them include the ability to search the content of their discussion databases.

Discussion interfaces offer more options and flexibility than email lists, so spontaneous interaction can take different forms. It may include separate topics for new member introductions or the identification of a core problem to tackle or interest to explore. The initiator may want to conduct one conversation as an interview of an expert while creating a related conversation for open discussion of the expert's specialty.

Thus, open access to a discussion board located on the company intranet may be an improvement over the email approach. The board could provide an area for innovation where any member of the organization could start a collection of conversations, or a site administrator could answer requests by opening new private discussion areas for specific communities of knowledge exchange. Leaders of these areas could be granted the power to invite and provide access to members of a group, or the areas could require new joiners to go through a registration process. Either way, a discrete meeting place would be provided that would hold a record of all conversations and would be available for participation at any time of day.

Within certain populations, moderation or facilitation of the conversation is appropriate. This entails selection of which submitted messages will be passed on to the recipient group in the case of a moderator. Or it may entail regular guidance, clarification, summarization, and direction of the conversation in the case of the facilitator. The group as a whole must decide how much intervention is helpful or how much is distracting. More creative populations will opt to manage the conversation on their own, allowing the topic to drift in exploratory directions at times and then pulling it back on topic when required. Groups that are more interested in finding the missing knowledge held by its members and using it to solve ongoing problems may not care as much about having the freedom to widen the conversation.

When IBM produced its WorldJam event, inviting all of its 320,000 employees to log on to a site that it had spent multimillions of dollars to design and build, it had no guarantee that the results would pay for themselves. It was meant to be a wide-open brainstorming and knowledge transfer opportunity, but its greatest value to the company may have been in the publicity it attracted. The scale of an online event has a lot to do with its success, as does the focus. In IBM's case, both of these parameters were unbounded. It came up with a technical solution that could handle a larger real-time participating audience than had existed up to that point, and it committed itself to months of follow-up analysis of employees' suggestions, opinions, and reactions to the event. This is not the route we would recommend that any organization take to learn what its employees think. But it did illustrate some useful principles.

First of all, WorldJam got the attention of 320,000 people who shared an interest in their company's success. An invitation to submit feedback and ideas is difficult to ignore, even if one has nothing new to offer or believes that their contributions would be lost amidst such an avalanche of feedback. In fact, only 52,600 employees logged on (to the reported delight of event managers), and only 6,000 proposals and comments were posted during the 3-day duration of the site. So one of six IBM employees thought enough of the idea to check it out. If that caused most of them to feel better about their employer or gave them hope that IBM was interested in what its employees— from top to bottom of the hierarchy—had to say, then it may have paid for itself in worker loyalty and goodwill.

Second, WorldJam was probably the first experience in online conversation for thousands, if not tens of thousands, of IBM employees. For them, the idea of online conversation is now a new option, another potential solution, for collaboration in the workplace. It's difficult to get people to try new methods, and by putting on such a widely publicized event, the individual risks were reduced. More people were willing to give it a try because it was not targeted at an elite or advanced group within the company.

And finally, many of those 52,600 people who logged on to WorldJam certainly got their eyes opened by reading opinions from others that either matched their own, conflicted strongly with their own, or exposed them to ideas they'd never seen expressed within the company. To those people, the possibilities of knowledge exchange through online channels became much more significant.

On a more practical scale, Cisco Systems has, for several years, been escalating its use of Webcasts to leverage its expertise around the world. Claiming cost savings of 80 percent over its previous techniques of sending speakers and trainers to hundreds of geographic locations, the company now works with Yahoo! Broadcasts (http://business.broadcast.com) to produce seminars and individual training events to update global employees and business partners and to build a larger population of Cisco-certified technicians.

These events pair graphic presentations with streaming audio, video, and chat windows to broadcast information while allowing audience interaction. These are usually asymmetrical knowledge exchanges, with knowledge sources doing the broadcasting and the knowledge recipients attending and submitting their questions through chat, email, or even voice via conference calls. Yet they offer the possibility, like WorldJam, of stimulating audiences to be involved and contribute ideas and suggestions to the forum based on their own experiences or expertise.

Transitory broadband events are too expensive to produce to be used as a company's main format for conversation. They are appropriate for episodic communications, between which their audiences can study on their own or follow up with new practices learned from the latest event. They are also useful for kicking off the opening of new, more persistent online communities. Because they bring broadband communications to the desktop in the form of voices, they are more attractive to many people than having to read comments and interpret their emotional component.

But transitory networks don't have to use broadband media to be effective. Special purpose meetings and symposia can also be staged using asynchronous platforms. They are more attractive to people who want to get deeper into subjects and who don't have the flexibility of schedule to allow them to participate in real-time events. For the same reason, they may be able to attract experts and instructors whose schedules don't match up.

Group Jazz (www.groupjazz.com) is a small consulting company that, among its other services, produces asynchronous "time-bound" meetings for organizations. The viewpoint of its founder, Lisa Kimball, is that virtual meetings that have beginnings, middles, and ends can be more productive than open-ended online conversations that don't feel the need to resolve and wrap up their ideas by a certain time. A time-bound meeting can have many of the features of a face-to-face conference or seminar, with keynote addresses, slide shows, break-out sessions, papers to download, and even the all-important "hallway conversations" where most of the valuable knowledge transfer takes place. Sponsors can advertise their wares, and people can trade business cards.

By working with client companies on the design of online environments for virtual meetings and helping to train their employees in the skills of virtual facilitation, Group Jazz provides comprehensive support and enables the client to manage most, if not all, of its events. It may introduce technologies appropriate to specific events that the client will want to purchase or contract on its own once it has learned how to use them productively. And for organizations that would prefer to enable knowledge-sharing conversations on an as-needed basis rather than support them all of the time, it's cost effective to have these technical platforms and the people who know how to implement them on standby for whenever they are needed.

Transitory networks are purposeful networks that lend themselves to success because of the focus their audiences bring. Once the production and user experience has been mastered (and the technologies themselves have also been learning steadily over the past few years), subsequent events are cheaper to produce and bring more predictable success. The technologies—whether broadband subcontracted services, such as Yahoo!, Broadcast, Placeware, and Webex, or asynchronous platforms, such as Web Crossing and Caucus—can be formatted for future use and modified cheaply for each new event. The skills gained in moderating and facilitating the initial events become assets that can be smoothly put back into action when required. Ideally, a production team—either contracted from consultants or trained in-house—should be kept intact and available for whenever needed. The real cost in convening transitory networks is in the original design, production, and staff training. Once a successful formula has been arrived at in terms of design, promotion, implementation, management, and follow-up, additional costs per event become minimal.

Transitory events can become such regular occurrences that they are no longer transitory but become part of the organization's strategy. Just as Cisco's use of real-time interactive broadcasts have become integral to the training programs that enhance the value of its technologies, a company may find that regular production of online meetings is an important aspect of its branding and product identification.

We've written an awful lot about culture in these pages, and we don't intend to repeat it all here. But in the context of actually implementing strategic knowledge conversation, it must be emphasized that unless the culture is aligned with the overall strategy, the benefits of a technical infrastructure will be severely limited. Design for knowledge networking that will serve an organization's strategic goals begins with an admission that, even at the top of the hierarchy, the organization is starving to learn more.

The motivation to learn and to recognize the value of each person's contribution to mutual learning must be strong throughout the company. The importance of skills such as online community leadership, online meeting facilitation, and online meeting production must be elevated. The idea that an individual's knowledge is more valuable when hoarded than when shared must be debunked through new incentive programs and attitudes. The positive results of a group's collaboration around shared knowledge must be publicized and praised. Training to help all employees understand how to represent their work skills and experiences should be easily available.

A strategic approach means that the entire company must be included in the new knowledge-sharing attitude. Communities may be broken down by project, department, or profession, but all aspects of the organization should be invited and encouraged to engage in the online conversations that will enhance their performance and productivity.