Learning in Knowledge Communities:: Managing Technology and Context

European Management Journal Vol. 22, No. 1, pp. 1–11, 2004 2004 Elsevier Ltd. All rights reserved.Pergamon

Printed in Great Britain0263-2373 $30.00doi:10.1016/j.emj.2003.11.019

Learning in KnowledgeCommunities:Managing Technology andContextMICHAEL BARRETT, Judge Institute of Management, University of CambridgeSAM CAPPLEMAN, Hewlett Packard Global AlliancesGAMILA SHOIB, School of Management, University of BathGEOFF WALSHAM, Judge Institute of Management, University of Cambridge

In contemporary organizations, significant empha-sis is placed on the processes of knowledge sharingand learning, which are increasingly seen as crucialto organizational success. Information and com-munication technologies play an important role inthese areas, but to many there is a lack of clarityregarding how such technologies can be bestdeployed. In this article, we provide a wide rangeof examples of where technology has been used tosupport learning in knowledge communities, withvarying degrees of success. We use this material todevelop specific characteristics of effective know-ledge communities, and detail ways in which boththe context and the technology should be managed.A key message we derive is that the maintenanceof a supportive culture and context for learning andknowledge sharing is crucial, and that an integratedapproach to technology deployment and use needsto be developed in conjunction with this. We arguethat action in this arena is important for all levelsand functions of management, not just senior man-agers or IT staff, since the support of effectivelearning and knowledge sharing in and betweencommunities involves everyone in the organizationin all job roles. 2004 Elsevier Ltd. All rights reserved.

Keywords: Knowledge community, Informationand communication technologies, Organizationalcontext, Learning

European Management Journal Vol. 22, No. 1, pp. 1–11, February 2004 1

Introduction

The importance of knowledge in organizations hasreceived much emphasis over the last decade follow-ing pioneering work in the early 1990s (e.g. Nonaka,1994). However, unlike some management initiativesthat soon become unfashionable, the knowledgetheme has persisted until the present time with nodecrease in interest. Indeed, the literature on the topiccontinues to develop, with increasing recognition ofthe role of communities in knowledge sharing(Wenger and Snyder, 2000). Empirical evidenceshows that these communities are formed withinorganizations as would be expected, but alsobetween people in different organizations, comingtogether across boundaries to learn through sharingknowledge on particular topics (Anand et al., 2002),or to collaborate collectively on the development ofartefacts such as software (Markus et al., 2000).

An influential body of work focuses on learning andknowledge sharing in communities of practice (Laveand Wenger, 1991; Wenger, 1998). We will, however,use the more general term ‘knowledge communities’in this article, since we are concerned with learningin communities which are both voluntary in terms ofparticipation, and those with a more managed mem-bership (Storck and Hill, 2000). We are also con-cerned with learning through the sharing of know-ledge between communities, in what has been calledthe constellation of communities which exist inorganizations (Wenger, 1998; Ward, 2002).

LEARNING IN KNOWLEDGE COMMUNITIES

Writers on the subject of knowledge sharing haveoften noted the importance of a supportive context,for example in terms of a collaborative culture andappropriate incentives for sharing (Davenport et al.,1998). It has also been emphasized that this contextcannot be managed in some simple top-down way,but that learning in communities needs to be culti-vated through encouragement and facilitation, forexample in allowing new ideas to develop and circu-late within and between communities (Brown andDuguid, 2000). In addition to context, most writerson knowledge communities see a key role for infor-mation and communication technologies(McDermott, 1999; Walsham, 2001). However, thisrole is normally not spelt out in any detail. In thisarticle, we go into specifics on the role of technologyin supporting knowledge communities, and weemphasize the need for the management of both tech-nology and context in order to provide effective sup-port for learning and knowledge sharing.

For instance, we may all be aware of technologiessuch as e-mail, groupware, e-learning systems andteleconferencing, and they clearly have a potentiallyvaluable role in supporting knowledge communities.However, we also know that there are constraints tothe usefulness of these technologies. For example,work on virtual teams has shown that periodic face-to-face contact can be necessary to develop andreinforce trust relationships between team members(Maznevski and Chudoba, 2000). The main messageof this article is that information and communicationtechnologies (ICT), whilst providing a foundationalinfrastructure and environment to support learning,may not, by themselves, be sufficient to stimulateeffective learning in knowledge communities. Softerissues such as motivation and the learning contextare crucial in forming a supportive ‘climate’ forknowledge sharing. However, if the climate is good,then technology has a central part to play in provid-ing the media and infrastructure for learning in andbetween knowledge communities.

In what follows, we develop these arguments usingexamples, firstly, of where ICT have provided highlyeffective support for learning in knowledge com-munities. Secondly, we describe examples of whereICT learning support has been less effective. Drawingfrom both sets of empirical examples,1 we derivesome key characteristics of effective knowledge com-munities, and then some management lessons as tohow to promote, support and manage such com-munities.

Although a key theme of our article concerns ICT usein knowledge communities, we want to emphasizethat our messages are aimed at all managers and theirstaff. The role of ICT in supporting effective learningis not something that can be left to technologists orsenior managers, although both these groups have amajor part to play. It is something that should con-cern everyone engaged with organizational life. We

European Management Journal Vol. 22, No. 1, pp. 1–11, February 20042

are all members of different knowledge communities,and we all need to think carefully about how to makethese communities effective, including an analysis ofthe role of ICT. We aim in what follows to providesome examples and concepts to help managers tothink this through in their own contexts.

ICT Support for Learning inKnowledge Communities

We provide five areas in this section of effective ICTsupport for learning in knowledge communities. Foreach area, we discuss the ICT application, the focusof the support for learning, important contextualelements that mediate the community interaction,and we provide a case example. Table 1 summarizesthe material in this section.

Virtual Interaction Focused on Products or Issues

Hewlett Packard’s IT Resource Center (ITRC) (Raths,2001) brings together engineers, internal IT staff andcustomers through intranet or extranet-based com-munities focused on specific products or issues.There are thousands of members in these inter-organizational communities covering topics such asbusiness recovery planning and operating systemssoftware. Community participants can ask questionsand receive answers within a short period of time.For example, when systems administrators haveproblems, they can post symptoms electronically andreceive detailed help on how to proceed withinminutes.

In order for communities of the above type to flour-ish, participants must trust the responses they receiveand find them effective in practice, or the approachwill rapidly fail. Hewlett Packard’s approach to thisis based on a system of user profiles and ratings.Community members each create a short biographypage and rate each other’s responses from 1 to 10.The response from a particular person comes, there-fore, with some ‘credit rating’, making it easier forthe questioner to assess the likely value of the answ-er.

Many writers have noted that one of the difficultiesof purely virtual interaction is a potential lack of trust(Kirkman et al., 2002). Members of the Hewlett Pack-ard online communities do not normally meet oneanother face-to-face, but the development of trust issupported by the credit ratings described above, andreinforced if the advice that is received actuallyworks. For example, a systems administration con-sultant and frequent user of the online communityapproach said that there were members of his onlinecommunity whom he would trust on technical mat-ters more than his closest colleagues. Hewlett Pack-


Table 1 Examples of Effective ICT Support for Learning in Knowledge Communities

ICT application Focus of ICT support for learning Contextual elements to facilitate Case exampleseffective knowledgecommunities

Intranet/extranet Asynchronous virtual interaction on Backed by user profiles and Hewlett Packard ITproduct or issue-related training ratings to increase value and trust Resource Centermodules and questions and answer by community membersforums

Groupware — incident Sharing best practices in a Supported by financial incentives, Customer supporttracking support system community through collaboration a cooperative culture and careful department in Zeta

tools and knowledge bases change managementInternet-based interaction Learning through voluntary Individuals’ identification with the Open source softwareand sharing of software participation in a virtual community- community culture, and gaining of movement e.g. Linux

of-practice peer reputationE-learning content delivery Simulations as a virtual medium for Closely mirroring the practice of Airlines, learning

learning community members content providers e.g.SkillSoft

Lotus Notes — email and Communication between different Gatekeepers/translators between Pharmaceuticalelectronic sales database knowledge communities communities company — Compound

UK

ard’s asynchronous ITRC community, which is alsoa repository for thousands of e-learning modules, issupplemented by a synchronous e-learning environ-ment called the HP Virtual Classroom.2

Sharing Best Practices through CollaborationTools and Knowledge Bases

Many organizations have made significant invest-ments in groupware technologies to support collab-orative learning, and to create knowledge basesaimed at sharing experience. The market leader interms of technology has been Lotus Notes. A goodillustration of a successful Notes system is in the cus-tomer support department of a software company,called Zeta, with headquarters in the Midwest of theUnited States (Orlikowski, 1996) An incident-trackingsupport system was developed based on Notes. Itenabled an individual specialist to keep track of aparticular customer problem and attempts at its sol-ution, but it also facilitated the sharing of suchknowledge with others so that a wider group couldcontribute to problem solving. The system thus cre-ated a knowledge base of all past incidents to aid thesolution of similar customer problems in the future.

Systems like this make the work of employees morevisible, and such visibility can be a cause of employeeresistance. This was not the case in Zeta, and therewas considerable acceptance by most specialists thatwork monitoring was a natural aspect of their job.Indeed, work monitoring may be welcomed byemployees who feel that they are performing well,since they are likely to gain financial and pro-motional rewards. However, it was also argued thatthe acceptance was, in part, attributable to the cus-tomer services department’s cooperative culture.


There is an extensive literature describing the appli-cation of Notes in a wide range of organizations tosupport collaborative working. Not all of the appli-cations have been as successful as that in Zeta, andthere have been some clear failures. A key conclusionfrom this literature is that there is no ‘collaborationinducing’ property of the technology itself (Karsten,1999). The outcomes from a specific applicationdepend on the ‘formative context’ (Ciborra and Patri-otta, 1996) within which the technology is intro-duced, and the care with which change managementprocesses are handled. The introduction of new tech-nologies such as Notes normally has a major effecton the work practices of employees, and the manage-ment of the emergent nature of the change process isa crucial factor in facilitating a successful transition.

Voluntary Participation in a Virtual Community

One of the more remarkable stories about learning inknowledge communities in recent years has been thedevelopment of the open source software movement.From its hacker origins in the 1970s and 1980s, thedecade of the 1990s saw the flourishing of softwareproducts such as the Linux operating system, whichwas able to challenge the power of Microsoft in itsparticular market. The core of Linux was developedby Linus Torvelds, originally when he was a 21-yearold computer science student at the University ofHelsinki. However, the debugging and enhancementof the software has taken place through the voluntaryparticipation of members of an online community.This contrasts with examples such as those of Hew-lett Packard and Zeta given above, where the collab-oration took place in more conventional organiza-tional settings.

Why do people actively participate in a task-focused

Members of open-source

software communities can

gain peer reputation and

esteem through their

collaborative activities


activity such as the development of an operating sys-tem when they receive no direct financial rewards fortheir endeavours? It is true that participants may gainfuture financial rewards through the development oftheir skills and knowledge, and thus enhancedemployment opportunities for example. However, adeeper explanation, apart from the sheer stimulationof software development for some people, is thatopen-source projects like Linux confer highly valuedbenefits such as reputation and peer esteem. It hasbeen argued that open-source is an example of a‘reputation culture’ (Markus et al., 2000) with otherexamples including the whole of academia. Well-known professors are normally able to negotiatehigher salaries, but much of their ‘reward’ for highachievement comes from reputation in their ownparticular academic sub-communities. Similarly,members of open source software communities cangain peer reputation and esteem through their collab-orative activities.

Software developers participating in online com-munities provides an excellent example of acommunity-of-practice, distinguishing this term fromthat of a group of people performing some commontask or with some common job title. A group ofpeople becomes a community-of-practice if individuals con-tribute in a voluntary way,because they feel that theylearn a lot through communityparticipation, and because theyidentify themselves closelywith what they perceive to bethe community culture. Thereare lessons here which go wellbeyond the particular task-focused activity of softwaredevelopment. Organizationsshould want their employees to be full members ofknowledge communities, extending both inside andoutside of their own organization, in which they par-ticipate willingly because they learn effectively.

Simulations as a Virtual Medium for Learning

There has been much talk in recent years about e-learning, often without a clear definition of what ismeant by the term. All of the above examples couldbe classified as e-learning, if we define that to meanthe use of electronic media for learning purposes.However, the term is often used for the narrowerconcept of delivering e-content or learning modules,which are available to an individual for self-study ontheir own PC or workstation and often accessedthrough a corporate learning management system. Akey organizational motive for the introduction ofthese e-learning modules is normally cost. It is seenas a cheaper way of training staff. However, theresults from this approach have been very mixed interms of success, leading to the perception that e-


learning hype has not necessarily been justified byresults (Rosenberg, 2001).

From the knowledge community perspective of thisarticle, e-learning modules are often seen byemployees as ‘artificial’, being divorced from theiractual work practice. For example, there is a largedifference between studying an e-learning module onleadership and exercising leadership in practice.‘Blended learning’ has been one approach to thisproblem (Rossett, 2002), mixing self-study with moreconventional classroom work in groups, and thentrying to relate what is learnt to practice. This is apromising way forward, although it needs very care-ful design and facilitation to ensure that the ‘blend’is not just a set of loosely related parts, with thelearner not able to make effective connectionsbetween them.

To increase the linkage between e-learning and workpractice, there has been an increasing focus on gamesand simulations as learning interventions, madepossible in part by the decreasing cost of technology.A positive example of success in this area, which hasbeen implemented for many years, is the use of simu-lations for the training of airline pilots. The technical

quality of the simulations ishigh, and it matches veryclosely the work practice of thepilot community. The blend isa natural one, in that the simu-lated activity closely mirrorsthe practice of communitymembers when flying planes. Itis interesting to note that thelearning content providerSkillSoft is now promoting‘skill simulations’ as part of itslatest generation of e-learning

content (http://www.skillsoft.com). These aredesigned, like pilot simulations, to be much moreinteractive and realistic. For example, for trainingsalespeople, the ‘virtual’ desk has phones that ring,people who interrupt, e-mails that arrive, and iratecustomers. Such simulations offer the promise of acloser link between online learning and actual com-munity practice.

Communication Between Different KnowledgeCommunities

The examples given above of successful learning inknowledge communities have focused primarily ona single community, such as a group of customersupport engineers or an online software community.However, there is also a need for learning andknowledge exchange between communities. Anexample of this (Hayes and Walsham, 2000) is a phar-maceutical company, called Compound UK, whichwanted to bring together the knowledge of its sales-people and its medical experts in order to be more

http://www.skillsoft.com


effective in the selling of its drugs to hospitals. Thecompany implemented a Lotus Notes system, onepurpose of which was to support communication andknowledge sharing between the sales-force and themedical experts.

Knowledge sharing between communities is oftenharder than within communities. Attempts to com-municate meaning will generally be more difficultdue to the lack of shared symbols such as pro-fessional language, job purpose, and norms of behav-iour. Technology may not necessarily be a good sol-ution, since it can potentially reinforce boundariesbetween different groups, behind what has beencalled ‘electronic fences’ (Newell et al., 2000). One sol-ution to this problem has been suggested as ‘organi-zational translators’ (Brown and Duguid, 1998),namely individuals who can frame the interests ofone community in terms of another community’s per-spective.

Compound UK had a particular medical advisor whofulfilled this role well. This individual was enthusi-astic about trying to cross community boundaries inorder to provide effective support for the salespeoplein terms of medical advice. His approach to this wasto form relationships with individual salespeople bymeeting them in person, for example at training andinduction sessions. This made the salespeople morecomfortable in contacting him, as well as providinghim with a deeper appreciation of what was involvedin being a medical salesperson. As the salespeoplebecame more familiar with him, they would increas-ingly send him e-mails and complete more detail onthe electronic sales database. The actions taken by the‘translator’ were thus to open up a forum, both face-to-face and electronic, within which the knowledgeand perspectives of experts from different communi-ties could be more effectively exchanged.

Potential Disjoints in ICT LearningSupport

We move now to some general areas where disjointscan occur when trying to use ICT to support learningin knowledge communities. These relate to situationswhere the combination of features of the technologyand the context do not combine together in a waywhich best supports effective learning. We identifyfive such areas, and we provide a description of theproblems that can occur, and a case example for each.Table 2 summarizes the material in this section.

Virtual Interaction in a Traditional Face-to-faceKnowledge Community

Many traditional knowledge communities are for-med by personal interaction, with work practices


reflecting and reinforcing this close contact. Theintroduction of virtual ways of working has potentialefficiency gains, but resistance is often encountered.Some of this resistance can be overcome by goodleadership and change management practices, but itshould not be assumed that all such resistance isnecessarily negative. In some cases, it may reflect theview that electronic interaction seems inappropriatefor particular tasks or for sharing certain types ofknowledge between community members.

An example of this (Barrett and Walsham, 1999) isprovided by the attempt in the London InsuranceMarket to introduce an electronic data interchange(EDI) system to support the critical area of the negoti-ation and agreement of insurance business betweenunderwriters and brokers, known as ‘placement’ inthe insurance industry. An electronic placing systemwas developed in the early 1990s, but low levels ofadoption and use were still being experienced severalyears later. The key reason for the resistance in theknowledge community of underwriters and brokerswas that placement for large insurance risks, forexample a ship, often involves a complex and delicatetrust-based interaction. Brokers and underwriters feltthat this should continue to be conducted in a face-to-face manner in order to carry out the processes ofnegotiation in an effective way, emphasizing certainissues, handling objections, and reaching a mutuallyacceptable agreement.

It should be noted that EDI systems in other workareas of the London Insurance Market were success-fully implemented. These included systems forclaims management and settlement, and accountingsystems. Resistance to electronic systems was nottherefore an inherent property of the community. Thegeneral lesson here is that virtual interaction cansometimes be a poor choice for particularly complex,political or trust-based communication and inter-action. In this particular case, the system was alsoineffective in facilitating learning for new brokers inthe London Market community, who have tradition-ally learnt their trade largely through face-to-faceapprenticeship practices. This was a further reasonwhy many underwriters and brokers were resistantto the transfer of all work practices to an electronicmode.

Sharing Creative Ideas and Problems ofIncentives

The earlier discussion of the use of a virtual onlinelearning community to develop the Linux operatingsystem provided a good example of the sharing ofideas in an environment calling for creativity, in thatcase in software development. There are many othersituations where this need for sharing creative ideasis important, and computer-based systems seem tooffer good opportunities for rapid and effectiveexchange. However, this presupposes that there are


Table 2 Examples of Potential Disjoints in ICT Support for Community Learning

ICT application Focus of ICT support for learning Contextual elements and Case examplesdisjoints in knowledgecommunities

Electronic data interchange Supporting virtual interaction in a Where face-to-face is more London Insurance(EDI) services traditional knowledge community appropriate to delicate trust-based Market

negotiationGroupware — shared Sharing ideas in a creative Where there are no good Advertising agencydatabase knowledge community incentives to shareGroupware — discussion Learning from others about issue or Deterred by hierarchical monitoring Pharmaceuticaldatabases product-based topics of performance company — Compound

UKIntranet Stimulation of a culture of learning Where this is seen to disrupt International non-

in knowledge communities traditional power structures governmentalorganization

E-learning system Providing learning modules for Where content and/or technology Airlineindividual use are poor, and access issues are

problematic

significant incentives for individuals to share, and wediscussed some of the incentives, such as peer repu-tation, which applied in the case of the open sourcesoftware movement.

Such incentives are not always present. Indeed, insome cases, there are positive deterrents to sharing. Alarge advertising agency3 implemented a groupwaresystem that was designed to enable creative advertis-ing ideas to be shared between staff members rap-idly, through shared databases, with the goal of pro-ducing quicker and better advertising products fortheir customers. The system was hardly used. Aninvestigation of the reasons for this established thatincentive systems were crucial. Staff members wereconcerned that their pay and promotion prospectsdepended, to a large extent, on their creative ideas,and the sharing of these ideas with others was notdirectly rewarded. Thus, they wished to put theirstamp on any creative ideas before they shared themwith colleagues.4

More generally, the danger of putting out creativeideas on any electronic system is that they can betaken up and used by others without clear attributionto the originator of the ideas. It can be argued thatpeer reputation for an individual within a particularknowledge community is enhanced if colleaguesknow that he or she is the source of valuable creativeideas. However, in the hard world of advertising,financial rewards are likely to be more dominantmotives than they are in the arena of software devel-opment. One approach to the issue of appropriateincentives is to reward team behaviour as well asindividual creativity. The Hewlett Packard examplegiven earlier provides one model for this, where thepoints rating of an individual is a reflection of theperson’s perceived value to other community mem-bers.


Learning and Hierarchical Monitoring ofPerformance

Learning from others through issue or product-basedinteraction is potentially a highly successful appli-cation of ICT to support knowledge communities, aswe have shown through some of our earlierexamples. However, the climate for learning andsharing has to be right. There is increasing awarenessthat electronic systems provide data for performancetracking and evaluation purposes. This has given riseto general citizen concerns about privacy of personaldata for example. In the business context, people car-rying out the monitoring are generally supervisoryor more senior managers. Lower level staff can befearful of this if they feel that the electronic data arenot a fair representation of their performance, or ifthey are concerned that honesty may be perceived asorganizational disloyalty.

Examples of both of these fears can be given fromthe Lotus Notes implementation in the pharmaceut-ical company, Compound UK, discussed earlier.Sales staff were monitored by more senior staff onthe number of contacts that they had made with doc-tors. A ‘league table’ was drawn up by a particularsenior manager showing salespeople in terms ofdescending contact numbers, and messages were sentto middle managers praising those with a high num-ber of contacts. This approach was perceived nega-tively by many of the sales staff since a more appro-priate measure of a salesperson is volume of salesachieved, rather than number of sales contacts. Theexistence and ease of production of electronic meas-ures should not be used by senior managers as anexcuse for lack of thought about appropriate metrics,or what level of performance monitoring is accept-able or desirable within the organization.

Salespeople in Compound UK had access to elec-


tronic ‘discussion databases’, on which they couldraise concerns about particular drugs or other issuesin selling to their clients. However, these electronicforums were open to all levels of management, andthis resulted in many of the salespeople not partici-pating in them. The reason was that raising a queryabout a particular drug, for example that it washigher-priced than that of a competitor’s drug, couldbe perceived as a negative attitude by senior manage-ment. Even though the query might be aimed at ask-ing sales colleagues how to handle this issue whendealing with doctors, the safe approach was to avoidthe visible electronic domain. This argues for elec-tronic ‘safe enclaves’ (Hayes and Walsham, 2000)where staff can interact without higher-level moni-toring of comments, or perhaps for a more enlight-ened senior management who see the value of con-structive questioning of their organization’sproducts.

Creating Learning Communities and ExistingPower Structures

One of the major benefits that ICT offers is a way offacilitating rapid and widespread interaction acrosstime and space. This enables the creation of newknowledge communities, which would not have beenpossible previously, and new opportunities for learn-ing from a wider group of people. The potential‘social capital’ (Anand et al., 2002) that can be drawnupon is much bigger than in previous eras. However,new forms of interaction can sometimes be perceivedas threatening to existing power structures, parti-cularly if those structures are highly top-down andcentralized. Instead of orders flowing down andinformation flowing up, there is the possibility of lat-eral communication and grass-roots development ofideas.

We have discussed this issue with a large non-governmental organization that operates in mostcountries of the world, with a headquarters in a west-ern country, and regional offices in different worldareas.5 A particular team in the organization is look-ing at their knowledge management practices, withthe aim of trying to create new knowledge communi-ties which transcend national barriers and do notoperate in the traditional top-down manner. Thesecommunities will be facilitated by appropriate elec-tronic systems, such as a corporate wide intranet forknowledge exchange. This is an exciting prospect inan organization that up to now has operated in arather traditional and hierarchical manner. However,it would be foolish to underestimate possible resist-ance, not this time from lower level workers as insome of the previous examples in the article, butfrom more senior managers who perceive theirpower and status as threatened by new approachesto learning in a more lateral way.

The World Bank is another international organization


with a wide global agenda. The Bank has explicitlystated its goals to include the creation of knowledgemanagement systems open to staff, but also to clients,partners and other stakeholders (http://www.worldbank.org/ks/km—overview.html). They aim tobecome the clearing-house for knowledge aboutdevelopment, encouraging communities of practicearound particular development themes. All of thissounds admirable, although a word of caution is inorder. There are some critical commentators whoargue that the Bank is thereby trying to take over thecentral role in all debates about development, ther-eby suppressing or weakening alternative voices(Thompson, 2002). We do not wish to adopt a parti-cular position ourselves here, but merely to note thatthis provides an excellent example of the way thatICT development and support for learning cannot bedivorced from the political context within which itis located.

E-learning Systems, Technology Quality andAccess Issues

We raised some problematic issues with respect to e-learning systems earlier, particularly when definedas stand-alone learning modules with only a looseconnection to actual work practices. However, wealso noted the scope for valuable approaches in thisarea, giving an example of simulations. There seemslittle doubt that e-learning is here to stay, althoughthe precise details of its future are unclear. Tra-ditional educational institutions, technology sup-pliers, content providers, and client organizations areall investigating ways to harness the power of tech-nology more effectively for individual and com-munity learning (Cappleman, 2001).

We had the opportunity to look at a particular air-line’s efforts in the e-learning arena.6 They wereamong the more progressive organizations in theirapproach, aware of both stand-alone modules andwider agenda of supporting knowledge communi-ties. However, two broad problems are worth men-tioning here, which apply more generally to e-learning systems. The first of these is the quality ofthe technology itself, and in particular the accesstechnology. Providing all employees with a fastbroadband connection so that they can downloadand interact with graphic-intensive e-learning sys-tems can be costly, especially outside the environ-ment of a corporate local area network. Since the goalof many e-learning systems is partly to save cost,there can be a tendency to cut back on technologycosts, which may be detrimental to the overall usageand acceptance of the material, and the quality of thelearning experience. The right management approachis to understand the business and economic modelwhich will underpin the e-learning intervention andinfrastructure, both from a cost and a benefit perspec-tive.

A related problem for the use of e-learning systems

http://www.worldbank.org/ks/km_overview.html

http://www.worldbank.org/ks/km_overview.html


is that organizations often encourage their employeesto access the system from home. However,employees often find that they do not have the unin-terrupted time, either in the workplace or at home,which would be necessary for them to engage pro-ductively with e-learning systems. In the home, therecan be the additional pressure of balancing leisureand work time harmoniously. As in the office, goodbroadband access may not be available, and tryingto use e-learning systems on slow dial-up lines is notgenerally a good experience. Organizations need toface up to the reality of the need to provide highquality systems and good access if they wish to takee-learning for their employees seriously.

Key Characteristics of an EffectiveKnowledge Community

We can draw from the above experiences, both posi-tive and negative, to identify key characteristics ofan effective knowledge community, and the ways inwhich ICT can contribute. Our examples show thata supportive context and culture for learning andknowledge sharing is a necessary condition for theflourishing of knowledge communities, and ways toachieve this are not technology-related. If a support-ive context is there however, integrated ICT develop-ment and use can be vital to effective learning andknowledge sharing. We discuss the non-technologyissues first, and then the role of ICT, although wewish to emphasize the need to manage both contextand technology together. Table 3 summarizes thematerial in this section.

Table 3 Characteristics of an Effective Knowledge Community

Supportive context Clear reasons for the community to want to ❖ Product or issue-related questionsfor learning and interact ❖ Interest/stimulationknowledge sharing ❖ Aim for community of voluntary or willing

participationAppropriate incentives to share knowledge ❖ Financial

❖ Peer status❖ Rating systems

Perceived freedom to learn from others ❖ Lateral communication❖ ‘Safe’ enclaves

Willingness to learn from sources inside and ❖ Social capital networksoutside of the organization ❖ Translators for between community interaction

❖ Technology valuable for interaction across timeIntegrated ICT Balanced mix of media in supporting the and spacedevelopment and use community ❖ But face-to-face often necessary for complex,

trust-based interactions❖ Need for continuous evolution of media blend

Careful use of electronic monitoring ❖ Often accepted where felt to be ‘part of job’❖ But only if metrics deemed appropriate

ICT focused on cost effectiveness and ❖ Not simply cost reductionrelevance ❖ Need to blend e-learning with practiceGood ICT infrastructure ❖ Speed

❖ Quality❖ Access


Supportive Context for Learning and KnowledgeSharing

A first issue for any potential knowledge sharingactivity within a group is the existence of clearreasons for that interaction. Obvious business-relatedreasons include the need to focus on product orissue-related questions. However, we have also seenthat interest and stimulation can be other reasonswhy communities form. In all cases, the aim shouldbe for thriving communities, where individuals con-tribute voluntarily or willingly, because they identifywith the concerns, issues and interests of the com-munity.

In addition, although individuals may contribute wil-lingly to a knowledge community, such activity isnormally related to their perception of the rewardstructure, and individuals need appropriate incen-tives to share. As we have seen from the casematerial, these incentives can take a number of differ-ent forms, including financial gain, peer status, andthe desire to do well on a rating system. This is notto argue that individuals are driven solely byimmediate tangible rewards. For example, altruismin helping others is quite common. However, if thereward system works entirely counter to the ethos ofknowledge sharing, for example with highly individ-ualized recognition of creative work, then it isunlikely that this will provide a supportive contextfor knowledge exchange.

A further condition for a supportive context is thatindividuals perceive themselves as having the free-dom to learn from others, for example through lateralcommunication of issues and problems with their

%the support of effective

knowledge communities

involves everyone in the

organization in all job

roles


peers. Excessively hierarchical organizations maytend to discourage this form of communication. Inaddition, people are very aware of who is ‘watching’or monitoring when they share ideas with others, andthere is a need for ‘safe enclaves’, whether electronicor not, within which people feel free to express theirviews in an honest way. Again, this can be discour-aged by management or staff perceptions that hone-sty equates to negative thinking or even disloyalty.

Sources from which to learn are much wider nowthan in the past, and it is important that organiza-tions encourage a willingness on the part of theiremployees to learn from social capital networks bothinside and outside of theorganization. For example, pro-duct-related communitiesshould necessarily include bothsuppliers and customers so thatmutual learning can take placeeffectively. When trying toshare knowledge betweenknowledge communities, wehave seen that there can be animportant role for ‘translators’,individuals who are able to straddle the boundariesof different communities, acting as a conduit forknowledge exchange between them.

Integrated ICT Development and Use

Even within a supportive context for learning, con-siderable thought and careful action is necessary tointegrate ICT effectively in knowledge communities.A first issue is the mix of media that is appropriate.There is no simple answer to this because, as westress, it is which is context-specific. Technology isvery valuable for contact across time and space, usingasynchronous media such as e-mail, and synchron-ous media such as video conferencing or the use ofvirtual meeting or teaching rooms. However,electronic-based contact will often require sup-plementing with direct face-to-face meetings for com-plex, delicate trust-based interactions. A final pointon media mix is that it is not static. New media comealong, such as the exponential growth in mobile com-munication devices in recent years, and othersdecline. Choosing an appropriate media balance,using an increasing range of access and deliverydevices, should be a continuous process of learningfor knowledge communities.

ICT offer new opportunities for electronic feedbackand the monitoring and tracking of performance.This is often acceptable to employees when it isregarded as ‘part of the job’. Most of the customersupport personnel in the Zeta case accepted elec-tronic monitoring of their performance in dealingwith customer queries, and indeed some of themactively welcomed it as making their good perform-ance visible. Nevertheless, electronic monitoring is


problematic when the measures produced by it aredeemed inappropriate by employees in terms of beinga fair reflection of their actual performance. We sawthe case example of measuring sales performance bythe number of sales contacts as an illustration of this.

ICT should be focused on cost effectiveness, not sim-ply cost reduction. E-learning systems provideexamples here, where motives of reducing the costsof learning may result in that learning being highlyineffective. Simulations for airline pilots are expens-ive, but it is widely agreed that they are highly effec-tive in improving pilot performance. Such simula-tions also illustrate the importance of the perceived

relevance of ICT. There needsto be a close connectionbetween an ICT application andthe actual work practice ofemployees, so that the idea ofblended learning has meaningfor them.

Finally, it is important to men-tion good ICT infrastructure.Information and communi-

cation technologies are massively superior in qualityand technical performance than they were even adecade ago. This of course provides a major opport-unity for improved ICT support of knowledge com-munities, but we should not forget that access andother related technology shortfalls do still persist inplaces. Trying to use a slow dial-up line from homeas a means of accessing e-learning systems, parti-cularly where the systems were not designed withthis delivery mechanism in mind, provides one illus-tration of this. More generally, many of the poorercountries of the world still have a relatively poor ICTinfrastructure, and their efforts to use technologyeffectively are seriously hampered by this. Goodtechnology is a necessary but not sufficient conditionfor its effective use.

Messages for Management

A key message from this article is that managers needto be aware of what ICT can and cannot achieve withrespect to effective learning in knowledge communi-ties. The generation and maintenance of a supportivecontext is not primarily a technological issue.Employee perceptions of freedom to learn and theappropriateness of incentives, or their willingness toshare knowledge with others inside and outside ofthe organization, reflect the quality of the manage-ment of the enterprise as a whole. An integratedapproach to ICT development and use is veryimportant, but it is only likely to be successful in acontext which encourages and nurtures learning ingeneral, and knowledge communities in particular.

A second important message is that this does not


happen by chance, or by the actions of a small num-ber of senior managers or IT people. It requires theparticipation of all managers in an organization, sincethe support of effective knowledge communitiesinvolves everyone in the organization in all job roles.This is not something that can be left to the technol-ogists, or delegated to others. All managers shouldask themselves questions about the knowledge com-munities which are important in their area ofresponsibility, whether the context is supportive totheir learning processes, and how ICT can best sup-port these processes.

In terms of more specific messages, the characteristicsof an effective learning community discussed in theprevious section can be used to develop advice here.The reasons for a community’s existence change overtime, so allow communities to form, emerge, and diein an organic way. Recognize that multiple styles andmodes of learning exist, and that different strategiesare needed dependent on the activity being sup-ported. Analyse reward systems and incentives toshare knowledge, and change these if they areinhibiting effective community interaction. Encour-age lateral communication both inside and outside ofthe organization, rather than solely up-down hier-archical communication.

With respect to ICT itself, use a dialogue with eachcommunity to consider the best blend of electronicand non-electronic modes of interaction and knowl-edge sharing. Evolve the technological support con-tinuously, learning from success and failure. Be verycareful in choosing the measures for electronic moni-toring to ensure that they are suitable for the purposeand are accepted as fair by employees. Make surethat ICT investment is focused on cost effectivenessand relevance, not simply on cost reduction. Con-sider whether the ICT infrastructure provides goodsupport for employee learning and intra- and inter-community knowledge sharing, in terms of issuessuch as speed, quality and access.

We have aimed in this article to clarify and detailsome of the ‘soft’ issues of context and the ‘hard’issues of technological systems when trying to sup-port effective learning in and between knowledgecommunities. In our increasingly globalized organi-zations, knowledge sharing and the ability to learnquickly and effectively are key sources of competitiveadvantage or, more generally, of organizational effec-tiveness. Thriving knowledge communities, sup-ported by appropriate technology, are of fundamen-tal importance to achieving these organizationalgoals.

Notes

1. Several of the case examples are taken from our own work,but we also draw on examples from other authors in orderto increase the scope of our discussion to areas where we


have not worked directly ourselves. We reference the rel-evant sources. In a few cases, related to our most recentwork, there is no published source, so we provide a brieffootnote on how the research was carried out.

2. In the HP Virtual Classroom (HPVC), individuals frommany different geographies and time zones can meettogether to exchange ideas and receive tuition in a virtualclassroom format. The HPVC is used by global businessteams to hold meetings, which not only save the companysignificant money, but also makes much more efficient useof participants’ time. See also http://education.hp.com/hpvc/

3. The description here is based on an MBA student projectsupervised by one of the authors.

4. Digital copyright systems would be one way of providingappropriate electronic support in this area.

5. Two of the authors are involved in an action research pro-ject aimed at helping the agency to develop new knowl-edge management practices.

6. This involved a substantial case study of e-learning prac-tices at the airline, including 25 interviews or focus groupsessions at its headquarters and a particular country office.

References

Anand, V., Glick, W.H. and Manz, C.C. (2002) Thriving onthe knowledge of outsiders: tapping organizationalsocial capital. Academy of Management Executive 16(1),87–101.

Barrett, M. and Walsham, G. (1999) Electronic trading andwork transformation in the London insurance market.Information Systems Research 10(1), 1–22.

Brown, J.S. and Duguid, P. (1998) Organizing knowledge. Cal-ifornia Management Review 40(3), 90–111.

Brown, J.S. and Duguid, P. (2000) Balancing act: how to cap-ture knowledge without killing it. Harvard BusinessReview 78(3), 73–80.

Cappleman, S. (2001) Cultured for the Web? Hewlett PackardIT Journal. Hewlett Packard IT Journal Quarter 3, 78–89.

Ciborra, C.U. and Patriotta, G. (1996) Groupware and team-work in new product development: the case of a con-sumer goods multinational. In Groupware and Teamwork:Invisible Aid or Technical Hindrance?, ed. C.U. Ciborra.Wiley, Chichester.

Davenport, T.H., De Long, D.W. and Beers, M.C. (1998) Suc-cessful knowledge management projects. Sloan Manage-ment Review 39(2), 43–57.

Hayes, N. and Walsham, G. (2000) Safe enclaves, politicalenclaves and knowledge working. In Managing Knowl-edge: Critical Investigations of Work and Learning, eds C.Prichard, R. Hull, M. Chumer and H. Willmott. Macmil-lan, Basingstoke.

Karsten, H. (1999) Collaboration and collaborative infor-mation technology: what is the nature of their relation-ship? In Information Systems: Current Issues and FutureChanges, eds T.J. Larsen, L. Levine and J.I. DeGross. IFIP,Laxenburg, Austria.

Kirkman, B.L., Rosen, B., Gibson, C.B., Tesluk, P.E. andMcPherson, S.O. (2002) Five challenges to virtual teamsuccess: lessons from Sabre, Inc. Academy of ManagementExecutive 16(3), 67–79.

Lave, I. and Wenger, E. (1991) Situated Learning: Legitimate Per-ipheral Participation. Cambridge University Press, Cam-bridge.

Markus, M.L., Manville, B. and Agres, C.E. (2000) What makesa virtual organization work? Sloan Management ReviewFall, 13–26.

Maznevski, M.L. and Chudoba, K.M. (2000) Bridging spaceover time: global virtual team dynamics and effective-ness. Organization Science 11(5), 473–492.

McDermott, R. (1999) Why information technology inspiredbut cannot deliver knowledge management. CaliforniaManagement Review 41(4), 103–117.

http://education.hp.com/hpvc/

http://education.hp.com/hpvc/


Newell, S., Scarbrough, H., Swann, J. and Hislop, D. (2000)Intranets and knowledge management: de-centred tech-nologies and the limits of technological discourse. InManaging Knowledge: Critical Investigations of Work andLearning, eds C. Prichard, R. Hull, M. Chumer and H.Willmott, pp. 88–106. Macmillan, Basingstoke.

Nonaka, I. (1994) A dynamic theory of organizational knowl-edge creation. Organization Science 5(1), 14–37.

Orlikowski, W.J. (1996) Evolving with Notes: organizationalchange around groupware technology. In Groupware andTeamwork: Invisible Aid or Technical Hindrance, ed. C.U.Ciborra. Wiley, Chichester.

Raths, D. (2001) Practice makes perfect. Infoworld,http://archive.infoworld.com/articles/pe/xml/01/11/05/011105pecommunity.xml

Rosenberg, M.J. (2001) E-Learning: Strategies for DeliveringKnowledge in the Digital Age. McGraw-Hill, New York.

Rossett, A. ed. (2002) The ASTD E-learning Handbook: Best Prac-tices, Strategies and Case Studies for an Emerging Field.McGraw-Hill, New York.

MICHAEL BARRETT, SAM CAPPLEMAN,Judge Institute of Manage- Hewlett Packard Ltd, Cainment Studies, University of Road, Bracknell, BerkshireCambridge, Trumpington RG12 1HN, UK. E-mail:Street, Cambridge CB1 [email protected], UK. E-mail: [email protected] Sam Cappleman is Business

Development Director forMichael Barrett is a Faculty Hewlett-Packard’s EuropeanMember in Information Sys- Alliances Programme. Hetems and E-business at the has also worked for HP onJudge Institute of Manage- its Worldwide Education

ment Studies, where he also gained his Ph.D. His cur- Projects and Consulting Business and as Global Clientrent research focuses on the adoption and use of infor- Business Manager.mation and communication technologies in the contextof globalisation and consequences for the nature ofwork.

GAMILA SHOIB, Man- GEOFF WALSHAM,agement School, University Judge Institute of Manage-of Bath, Bath BA2 7AY, ment Studies, University ofUK. E-mail: g.m.shoib@ba- Cambridge, Trumpingtonth.ac.uk Street, Cambridge CB1

1AG, UK. E-mail: g.walsh-Gamila Shoib is Lecturer in [email protected] Systems at theUniversity of Bath’s School Geoff Walsham is Professorof Management. Previously, of Management Studies atshe was Hewlett-Packard the Judge Institute of Man-Senior Research Fellow in E- agement, University of

learning at the Judge Institute of Management. She Cambridge. His latest book is Making a World of Dif-teaches and researches social and organisational aspects ference: IT in a Global Context (Wiley, 2001). Hisof ICT development, implementation and use. teaching and research centres on social and manage-

ment aspects of the design and use of communicationtechnologies.


Storck, J. and Hill, P.A. (2000) Knowledge diffusion through‘strategic communities’. Sloan Management Review 41(2),63–74.

Thompson, M. (2002) ICTs, power and developmental dis-course: a critical analysis. In Global and organisational dis-course about information technology, 2002. Proceedings ofIFIP wg8.2 Conference, Barcelona, Spain, December.

Walsham, G. (2001) Knowledge management: the benefits andlimitations of computer systems. European ManagementJournal 19(6), 599–608.

Ward, A. (2002) Getting strategic value from constellations ofcommunities. Strategy and Leadership 20(2), 4–9.

Wenger, E. (1998) Communities of Practice: Learning, Meaningand Identity. Cambridge University Press, Cambridge.

Wenger, E. and Snyder, W.M. (2000) Communities of practice:the organizational frontier. Harvard Business Review Jan-Feb, 139–145.

Documents

Learning in Knowledge Communities:: Managing Technology and Context