How Can Organizational Memory Theories Contribute ... - Springer Link

10 downloads 0 Views 328KB Size Report
information systems supporting organizational memory has to be seen in this very specific context which leads to different perspectives contrasting traditional ...
Information Systems Frontiers 2:3/4, 277±298, 2000 # 2000 Kluwer Academic Publishers. Manufactured in The Netherlands.

How Can Organizational Memory Theories Contribute to Organizational Memory Systems? Franz Lehner and Ronald K. Maier University of Regensburg, Chair for Business Informatics III, Universitaetsstrasse 31, D-93053 Regensburg, Germany E-mail: [email protected], [email protected]

Abstract. Organizational memory systems (OMS) support organizations to ensure organizational learning, ¯exibility and ef®ciency and the management of change. The paper describes the term organizational memory and reviews a set of theories contributing to this ®eld. As organizational memory is considered interdisciplinary, theories are drawn from research ®elds as different as organization science, psychology, sociology and arti®cial intelligence. The consequence is that the design of information systems supporting organizational memory has to be seen in this very speci®c context which leads to different perspectives contrasting traditional software development. These different perspectives are integrated into our view on organizational memory systems. The paper is concluded by an outlook on the main research questions which we intend to address in our research. Key Words. knowledge management, knowledge management system, organizational intelligence, organizational learning, organizational memory, organizational memory system

1.

Introduction

1.1. Innovative use of information systems For some time now under the rubric ``organizational memory'' (OM), both innovative and familiar concepts as well as highly promising systems have been proposed and tested (cf. e.g., Bannon and Kuuti, 1996; Buckingham Shum, 1997; Lehner, 2000; Morschheuser, 1997; Poveschi, 1998; Stein and Zwass, 1995; Wargitsch, 1997). The topic has taken on an intense sense of immediacy given the worldwide processes of restructuring in both economy and society. Relevant projects are already being carried out, especially in large corporations. Environmental

dynamics and the pressure of competition that necessitate the development or the activation of new capabilities are paving the way for change. These adaptations occur automatically only in the rarest of cases, but presuppose (learning) processes. Important goals include elevating organizational ef®ciency and ¯exibility, overcoming growth limits and integrating organizations after mergers and acquisitions. The concentration on qualitative dimensions of organizational design gains increasing importance. One could look at this as an ``expansion inwards'', wherein new or previously unused potential and strengths should be developed. Example 1. In 1996 the Chase Manhattan Bank installed an Intranet-based knowledge management system for approximately 16 million USD. This system enabled the bank's employees to access the knowledge base of the entire business. Customerspeci®c data such as credit history, account balance, investment pro®le, as well as ``soft'' information such as personal preferences, or characteristics of the borrower could be called up directly from the work place. Already in the ®rst year it was introduced, the system brought in 11 million USD through cost reductions and revenue increases. The system's pro®tability results from an increase in employee productivity: the employees can devote more time to speaking directly to the customer because they need Author for correspondence: Ronald K. Maier, University of Regensburg, Chair for Business Informatics III, Universitaetsstrasse 31, D-93053 Regensburg, Germany. E-mail: [email protected]

277

278

Lehner and Maier

not ask for information that had already been obtained at some point in the past. The system was originally developed for the market segment with a turnover per customer between 3 and 500 million USD. Due to its success, the system is now being adapted for other customer segments. Example 2. Under the direction of a chief knowledge of®cer, the consulting company Ernst & Young (worldwide app. 40,000 employees) employs 400 fulltime employees for knowledge management. They are responsible for documenting the existing knowledge and know-how in the company, making it accessible for new employees and for protecting it when employees leave the company. The project's goals are: to introduce knowledge management, to generally improve the exchange of knowledge within the company, and to promote corporate culture in relation to the distribution of knowledge (according to Madey and Muzumdar, 1997). The example of the Chase Manhattan Bank shows that new information systems that aim to improve organizational ef®ciency in no way necessarily have to mean a break with the past; rather they build upon it. In this case, a customer information system serves as the basis. With Ernst & Young the background and the goals are different. This case deals with supporting a worldwide operation during a phase of extreme growth. The driving force is mostly the need for quick and easy access to data, information and knowledge (often in multimedia form) that are becoming increasingly important for corporations. For a long time, data bases served as the most important means to achieve this task. Given the developments of the last few years, completely new structuring possibilities have arisen which, on the one hand, are leading from isolated data base concepts to company-wide information models and, on the other hand, are leading to a renaissance and evolution of existing business management concepts. In order to achieve the aforementioned goals it is necessary to develop an understanding of the effected frameworks and management concepts that are already available. In addition, a more comprehensive understanding of the use of conventional database technology has to be developed and integrated within the broader concept of ``organizational knowledge''. This paper is meant to take into consideration the framework and content of these goals.

The development of organizational memory systems (OMS) is substantially more complex than the development of conventional information systems because existing models and planning methods must be developed further, and a relationship to management applications (e.g., business process re-engineering, process organization, enterprise modeling) must also be developed. Fig. 1 shows how this can be achieved via the convergence of various structural approaches, whereby especially organizational development should be stressed. It should also be emphasized that important developments arise at the juncture where systems and available technologies meet. Organizational memory systems and knowledge management systems1 can either be conceived as a special class of information system, or as a characteristic of conventional systems. They are particularly important for innovation and the retention of organizational ¯exibility. The ways and means in which information technology is actually used for particular corporate goals depend mostly on which concept of ``organization'' the business has as a model. The systems introduced into a corporation should correspond to the prevailing fundamental views of the time (technology-culture ®t). A business that evolves according to the plan of an autopoietic system requires other software-technological solutions than a business that is run in a stable environment according to bureaucratic rules. Differences will also occur depending how the information systems are used, even if it is the same system.

Fig. 1. Framework and conditions for organizational memory systems.

Organizational Memory Systems

By using different organizational measures, most modern managers attempt to create a climate in which learning in and about organizations is encouraged. Fig. 2 shows the relevant components (cf. Goodhue and Thompson, 1995). Most business management applications (see Section 3) concentrate for the most part on the relationship represented by arrow 1, i.e., they attempt to create a ®t between tasks or, in other words, between labor organization and the (organizational) culture. ``Traditional'' information management is mainly concerned with trying to bring tasks and the technology that has been introduced into line with each other (arrow 2). With the help of organizational memory systems, the attempt is made to additionally improve the relationship between the technology introduced and organization's culture (arrow 3) in order to achieve organizational ef®ciency.

1.2. Changes in economic structuresÐgrowing importance of technologies and information Worldwide changes in economic structures are often attributed to the development or introduction of new technologies. Communication technologies and multimedia information systems currently appear to be playing a key role for the organization and competitive edge of businesses. Bullinger (1995) summarizes the leading technological developments and related periods of growth of certain industries over the last few decades as follows (see also Kock, 1999): * * *

*

production industries in the 1950s and 1960s electronics and microelectronics in the 1970s consumer electronics and the computer industry in the 1980s telecommunications, information technology, media and entertainment (TIME) in the 1990s

Fig. 2. Corporate ef®ciency through ®t between tasks, culture and technology.

279

Other authors speak of the so-called quaternaryhypothesis (deployment of a fourth economic sector) and point out that, along with the three main economic sectors, a fourth independent sector is beginning to establish itself: the information sector (cf. e.g., Lenk, 1989, 24). Above all it includes the production of ``information'' as well as services that deal with computer technology. Besides that, commerce and the exchange of information or information products are becoming increasingly important. Moreover, telecommunications enables business processes that render geographic proximity irrelevant: electronic business. In all processes in which the aforementioned economic sectors can be observed, production factors are introduced and combined with each other. These production factors are included in practically all goods, or play a role in their production or extraction. However, their importance and combination varies from product to product. Moreover, the value of a production factor is determined largely by the dynamics of supply and demand. In this process it can be observed that the importance of information along the value chain is growing steadily. Information costs already account for a considerable portion of the total cost of the production process. Already in 1963, information costs accounted for more than 50% of the U.S. gross national product (Wild, 1971). In traditional production enterprises the portion of actual production costs as part of total product costs has fallen to 20% on average (cf. Pulic, 1996, 149). This observation has been well supported by a number of recent studies (see e.g., SchuÈppel, 1996, 49; North, 1999: Bullinger et al., 1997, 16, ILOI 1997). The main reasons for this development are due to the elimination of product de®ciencies and the reduction of productive, routine labor in favor of ``intellectual'' work. Schneider (1996) points out a few examples that powerfully illustrate this trend. At ABB, for example, only 3% of the time used for ful®lling a contract is devoted to production; the remaining 97% is used for ``other''. Only 6% of IBM employees work in factories, and for a 700 U.S.D. camera, a mere 8.5% results from material costs; the rest is spent on microprocessors and software, i.e., for the ``intelligence'' of the product (Peters, 1994, 13, quoted from Schneider, 1996, 13). The huge fraction that information and knowledge contribute to production is the main argument for introducing technical support of knowledge

280

Lehner and Maier

management. Additionally, information is the most important prerequisite for decision making and goaloriented negotiating. According to estimates, despite the increasing importance of knowledge, only about 30% of the actual knowledge available to an organization is really utilized (cf. Zucker and Schmitz, 1994). Studies have also shown that due to inaccessible information, expensive but avoidable mistakes occur, and that the danger of losing knowledge is especially great when people leave the company (see e.g., Spek and de Hoog, 1994). Thus, the volatility of knowledge represents one of the most prevalent problems in the management of an organization's knowledge base. In the following the development of information processing is laid out to show the corresponding shift of focus from data via information to knowledge. 1.3. Developmental steps of information processing: from data processing to knowledge management Over the last few years an evolution has occurred that, among other things, has been de®ned by continual and very important improvements in information technology. As a direct result of this development, a broad spectrum of variants of this technology are available for implementation for business applications today. Information and communication systems are not only introduced to serve as a more effective and ef®cient production factor to be used in business processes, but they are also increasingly regarded as a corporate success factor. Of special interest in this respect is the increasingly important role of the ``organization''. It also surfaces in the development of information processing that has occurred in several stages over the last few decades. These stages can be summarized from various standpoints, such as: technological development, corporate applications, importance of data and information, etc. Ortner describes this development from the perspective of the growing importance of data as business resources (Ortner, 1991). By considering increasing organizational integration, he differentiates between the following developmental stages (see Fig. 3): isolated applications, data base systems, data modeling, data management, and information management. According to him, the formation of business integrated information and communication systems is only feasible economically when based on a corporation-wide, consolidated data

architecture or a corporate data model. The decisive contribution to the increasing and necessary integration of corporate processes results accordingly from the data, or rather the function of the data model (cf. Ortner, 1991). A similar system whose focus is corporate application of information technology comes from Hanker (1990). He breaks down the developmental stages of introducing computers into business as follows: 1. Support of operations. 2. Support of management (e.g., decision support systems). 3. Support for competitive strategy ( ˆ computer as strategic weapon). 4. Support for organizational strategy ( ˆ comprehensive standpoint, e.g., knowledge management, organizational memory). The graduated model of Hanker points to a transformation in the way computers are used in business over the course of time. It can be used for diagnosis, i.e., which stage is the business at currently? However, the observation of the new dimension of information processing discussed therein is far more important: knowledge management and organizational memory. Fig. 4 summarizes this development from the perspective of data storage and application. Each individual level supports speci®c tasks in the corporation and serves to meet corresponding base technologies and methods in the development of concrete systems. The dif®culties at the higher levels not only lie in overcoming technical complexity (heterogeneous systems, various norms and standards, differences in languages and interfaces, discretion in system purpose and user groups). More and more, completely new perspectives arise such that con¯icts occur between the aims and the underlying levels. At level 4 (knowledge management), matters are compounded because the technology implemented in no way has to be limited to data or data bases; rather this aspect might be relegated completely into the background. Along with the classical modeling of data structures (e.g., a data model in entity-relationship notation), the design and support of dynamic processes (e.g., the process of creating information or the process of changing knowledge) and the support of organizational learning processes are gaining

Organizational Memory Systems

281

Fig. 3. Development of information processing with the focus on data (steps 0±4 after Ortner, 1991).

importance in the realm of computer science. Practice clearly shows that it is precisely here where the chances and potentials lie, and that innovative businesses meet this challenge with groundbreaking projects that are already underway. No direct hierarchical relationship exists between the levels shown in Fig. 4. Rather it represents an idealized conception that results from chronological development. Hence at least partially, most concepts simply build upon each other.

2. Organizational Memory 2.1. What is meant by ``organizational memory''? Origin and context of the topic Today it can be assumed that organizational memories existÐwhether as construct or concept. Hence, the question is to what extent is an organizational memory accessible as an active structure and which instrumental results can be achieved with it? Of special interest is what the contribution of information

Fig. 4. Developmental levels regarding the handling of data, information and knowledge in businesses.

282

Lehner and Maier

technology means or could mean: Can cultural changes and the mobilization of creative forces be supported by technology? This question can be addressed along with the question of the function and meaning of OMS and will be addressed in more detail in the next section. The necessity of addressing the question of knowledge, know-how, etc., as well as dealing with the issue of information and knowledge management in business results from the intricate relationship of knowledge and negotiation or, in other words, knowledge and decision making. For the following context, a very broad concept of knowledge is assumed. By applying and developing the existing theories further, the foundation for a system capable of ``evolving'' into actual practice should be established. Organizational memory systems are put into place to cope with related tasks. They are the (technical) realization or support of those segments of an organization that are labeled ``organizational memory'', and are always present in principle, even without technical support. Talk about the existence of organizational memory has a longer tradition than one might at ®rst suspect. Traces can be found as early as the end of the 19th century. However, at the beginning of the 20th century interest was lost. The topic was almost completely forgotten, despite psychology's close relationship to organizational memory and the spectacular theories being developed in the ®elds of cognition, memory, information processing and arti®cial memory (cf. Wegner, 1986, 185). This will be discussed in greater detail later. Human memory is often used as model and metaphor for the organizational memory. Comparing characteristics of individual and organizational memories the following differences seem to be important: Metaphor: Machine; Storage *

* *

*

Erasing the contents of memory requires external actuation Data can be called up, duplicated, or recovered The structure and form of storage is established; the behavior of the system determined Growth of the knowledge bases occurs primarily quantitatively, i.e., as an increase in the amount of data

Metaphor: Organism, Memory *

*

*

*

Erasing occurs as ``forgetting'' and is an automatic process Knowledge is reproduced or reconstructed as needed (i.e., when it is used) The structure and form of storage is not established; the behavior of the system not determined The growth of the knowledge base does not happen via accumulation, but rather through reorganization or structural transformations in the knowledge base respectively.

Various management approaches and scienti®c disciplines played and continue to play a role in the development of the theory of organizational memory, some of which enjoy a long and respected tradition of their own. *

*

*

* *

* *

* *

* * *

Organizational learning (OL) and learning organization (LO). Organizational intelligence (OI), competitive intelligence (CI). Knowledge management (KM) as well as the concept of organization as knowledge and/or information processing system. Organization and personnel development (OD). Organizational change, management of change, innovation management. Organizational culture. Theory of the evolution of organizations, organized chaos. System theory and system dynamics. Arti®cial intelligence (AI) and cognitive psychology. Organizational psychology. Social psychology. Organizational sociology.

Naturally, organizational memory is not the only concept that plays a part in these disciplines and approaches, the aims and purposes of which often differ markedly from one another. Nevertheless, organizational memory has a common and metastatus for all these concepts, but especially for organizational learning and knowledge management. It is essential for the understanding of organizational memory systems, and it represents simultaneously an important bridge and common denominator for the other disciplines. These concepts are discussed in

Organizational Memory Systems

Fig. 5. Organizational memory as it intersects with the disciplines and concepts in question.

greater detail in the literature (see Lehner, 2000). Fig. 5 illustrates the interrelationships of these disciplines and approaches in question. 2.2. Terminology and theories on organizational memory In general one de®nes the term memory as a system capable of storing things perceived, experienced or lived beyond the duration of actual occurrence, and then retrieving them at a later point in time. Learning is not possible without memory. Accordingly, organizational memory is repeatedly proposed as a prerequisite for organized leaning in this context. Thereby, however, the term ``organizational memory'' should in no way be considered analogous to a ``brain'' to which organizations have access. The term is simply meant to imply that the organization's employees, written records, or data ``contain'' knowledge that is readily accessible (cf. Oberschulte, 1996, 53). Several functions of ``memory'' are inherently present in every organization without softwaretechnical support of any kind (e.g., in the form of search and recall processes as carried out by telephone surveys or brainstorming during meetings). Just as there are terms for the individual level (e.g., individual psychology, the psychology of learning), there are various terms concerning memory and de®nitions of knowledge for the collective and organizational levels. Intensive scienti®c investigation of this topic began at the end of the 1970s and beginning of the 1980s and was mostly carried out in the USA. A very popular article in this ®eld was

283

written by Duncan and Weiss (cf. 1979, 86±87). Hedberg develops this knowledge-based approach further. In 1981 he is the ®rst to introduce the term ``organizational memory'' (Hedberg, 1981): Organizational memory establishes the cognitive structures of information processing, the theory of action, for the entire organization. Parallel to this development, Kirsch coins the term ``organizational knowledge base'' (organisatorische Wissensbasis) for the German-speaking realm (in unpublished papers in 1974 according to GuÈldenberg and Eschenbach, 1996). Building on this, Pautzke in 1989 identi®ed several layers of the organizational knowledge base that can be seen in his model (see Fig. 6), which provided an important starting point for further research and conceptualizations. Hence organizational memory was really not a terminus technicus, but rather part of a tradition of organizational development, organizational learning, and knowledge management. In the beginning, the scienti®c explorations within the realm of organizational memory were marked by rather abstract and theoretical considerations. Turning these theories into general and practical applications for business purposes had to wait. In the following, some well-known concepts and theories will be discussed.

Fig. 6. Horizontal layered model of the organizational knowledge base (GuÈldenberg and Eschenbach, 1996, adapted from Pautzke and Kirsch).

284

Lehner and Maier

2.3. Concept of the organizational knowledge base according to Kirsch/Pautzke ``The organizational knowledge base represents the sum total of knowledge to which an organization has access. Organizations do not only require scienti®c/ technical knowledge for their negotiations and decisions; rather their knowledge base encompasses a broad spectrum of knowledge of all kinds. . . . The organizational knowledge base does not represent an homogenous picture, but rather points to one that is multi-dimensional'' (Pautzke, 1989). Kirsch and Pautzke suggest the following de®nition: Basically, the organizational knowledge base is the collection of knowledge which is either accessible to (horizontal layered model) or, in a larger sense, at the disposal of (vertical layered model) an organization's employees. The horizontal layered model, which is intended to represent the structure of the organizational knowledge base, subsequently became increasingly wellknown. At the center of the model, which will be explained in more detail later on, is the assumption that the chance of applying information for organizational decisions increases as one moves from the outer layers toward the inner ones. Fig. 6 shows this layered model in the form in which it was originally presented. The ®rst two layers represent the knowledge currently available to the organization, i.e., the actual organizational memory. The latent organizational knowledge base (third and fourth layers) encompasses knowledge that is potentially available to the organization. To the ®fth layer belongs, ®nally, all other information available in the cosmos. The individual layers and their traits will be discussed next.

the corporate identity is barely perceptible from the outside, it plays an eminent role in the decision making process within the organization. It acts as a kind of knowledge ®lter: In principle, accessible knowledge will only be accepted by an organization when it does not con¯ict with its organizational paradigm. Example: For over a century, the dominant global leader Mercedes Benz AG represented a technological zenith to be emulated. Until recently, its motto was, ``the best or nothing'', which was readily observable. In the meantime, however, the market could not rectify price differences that bring little advantage for technologies that are barely noticeable. Indeed, price became the decisive argument for selling, and extreme technological solutions unnecessary ballast. Based thereupon, the new Mercedes Benz program for success focussed rigorously on an ``uncompromising customer and market orientation'' as the deciding principle in everyday business proceedings. This behavioral orientation was accepted by most employees within a short period of time and thus became part of the shared knowledge base. Layer 2. Accessible individual and collective knowledge The most obvious differentiation in Pautzke's layered model is between knowledge available to the organization (layer 2), and knowledge not accessible to the organization but to the individual (layer 3). This presents the question of how individuals make their knowledge available to the organization. Basically, one can delineate between three strategies (see Fig. 7): *

Layer 1. Knowledge shared by all employees of a corporation The general knowledge base of an organization or its employees consists of a common language (which takes the form of stories, anecdotes, myths, sagas, rites, rituals, and ceremonies) as well as common values and system of norms that arose historically and that manifest themselves as negotiation and behavioral guidelines, manner of thinking and paradigms. Thus, many authors assign organizations a kind of cultural identity, from which the term ``corporate culture'' arose. All shared values, visions, unwritten rules, etc. are parts of this corporate culture. Although

*

An employee implicitly makes his knowledge available to the organization by participating in decision making processes within the organization (negotiation). An obvious problem here is that the organizational memory of a corporation is diminished whenever an employee leaves the corporation. Another possibility is to distribute the knowledge of individuals among several employees (collectivization), e.g., as a result of group discussions among colleagues. In this way the danger of losing intelligence of an organization can be reduced. However, this presupposes that the person originally in possession of knowledge is willing to

Organizational Memory Systems

285

tion of the knowledge base have already been the subject of numerous scienti®c inquiries. By interviewing people involved, GuÈldenberg and Eschenbach (1996) arrived at the following principle causes: *

*

*

Fig. 7. Organizational circle of learning (according to MuÈllerStewens and Pautzke, 1996, 195).

The following factors are less important according to those employees interviewed: * *

*

participate in ``information sharing''. Moreover, organizational structures and processes must be in place that promote persuasive confrontations. Finally, complete autonomy of individual employees can still be achieved through the preservation of individual knowledge via standard procedures, systems, rules, etc. (institutionalization of knowledge). For example, a data-processing specialist can make his specialized knowledge available to the corporation via documentation and the publication of manuals, or in the form of software. Nevertheless, institutionalization of knowledge is only possible within limits. For one thing, non-formalized knowledge (e.g., the creativity of someone employed as a designer) cannot be stored technologically because it simply does not make sense from an economic point of view to store all knowledge that could be formalized.

Layer 3. Non-accessible individual and collective knowledge The organization's ability to obtain employee knowledge that is not made available to the organization is more problematic. Besides irrelevant knowledge from the private sphere (e.g., leisure activities), knowledge that might well be of use to the organization might be withheld because of various possible information or communication pathologies in the organization. The blockades against access to expansion or transforma-

Fear of change that might result from passing on knowledge (e.g., weakening of one's own position). Inept ability to recognize the organizational relevance of one's own knowledge (e.g., departmental blindness). Power struggles (e.g., mobbing or consciously supplying colleagues with disinformation).

*

Lack of willingness to document knowledge. Lack of ability to present knowledge in an optimal form. Poor management of the organizational knowledge base in the corporation.

It would make sense here to develop appropriate methods (e.g., introduce a corporate suggestion box of sorts, quality-control circles, resource labs, job rotation) to discourage or at least reduce the mental barriers that arise in an organization so that the employee knowledge that is not originally accessible can be incorporated. Finally, the in¯uence of power structures should not be overlooked (cf. Duncan and Weiss, 1979, 95). The probability that new knowledge will be accepted at all largely depends on if it con¯icts with the knowledge or opinions of those employees in positions of power. However, power structures also have very positive aspects in this context in that they contribute to monitoring or reducing uncertainty for certain employees. Layer 4. Meta-knowledge of the environment The fourth layer in Pautzke's model includes knowledge that lies outside the realm the organization has access to but to which the organization might possess meta-knowledge. Meta-knowledge is always present when someone knows something about something. The layered model just described subsequently served as a prototype for numerous models and approaches that followed (Fig. 8 shows an extension of Pautzke's model), especially for organizational

286

Lehner and Maier

theories on group memory and group remembering, which are described in the following sections. They emphasize the importance of interaction and communication as structural components of OMS.

Fig. 8. Model of the organizational knowledge base.

learning. Organizational learning is de®ned in this context as the transformation of the organizational knowledge base. A big problem thereby lies in measurement, in making it operational and completing this transformation. Along with applying the theory of organizational learning, great signi®cance is attached to the layered model of the organizational knowledge base in knowledge management. Many authors see the main functions of knowledge management as increasing explicit, documented knowledge (codi®cation strategy) and bringing knowledge workers together so that knowledge is shared by everyone ( personalization strategy, see Hansen et al., 1999). Thereby it is implicitly assumed that this knowledge is useful and qualitatively adequate. In the ideal case, it would be open for further use in various forms to all employees of the organization. The layered model makes the task of knowledge management visible so that organizational and infrastructural conditions are created to insure that the knowledge base can be used, maintained and improved. It indicates what the goal of the externalization and collectivization of knowledge is, but it provides no information as to how to reach this goal. In later discussions, it will be shown that this goal is viewed somewhat differently nowadays. The original pragmatic aim of preserving knowledge and illustrating existing knowledge structures using computer technology is adapted to these knowledge structures with ever increasing acceptance. This development can be seen, for example, in

2.4. Organizational memory according to Walsh and Ungson Walsh and Ungson base their concept on a thorough analysis of the known approaches to organizational memory and its related ®elds (e.g., Argyris and SchoÈn, 1978; Cyert and March, 1963; March and Simon, 1958; Kantrow, 1987; Douglas, 1986; Weick and Gil®llan, 1971). Therein they point out a few contradictions and uncertainties that had not been tackled by the research. According to Walsh and Ungson (1991, 61) organizational memory is composed of several components and, at the least, contains one retention facility for informationÐthe information contained therein as well as the processes for gathering and searching for information. Oriented toward decision making, information is understood as a stimulus or reaction to decision-making. The purpose of organizational memory is seen in connecting past and present decision-making situations. The authors repeatedly point out dif®culties in de®ning and localizing organizational memory. Nevertheless, they consider their construct useful and take steps to de®ne organizational memory more precisely. It must be recognized, however, that they do not use a comprehensive term for knowledge as a point of departure, rather they only draw upon information relevant for decision making. Hence, this approach can be categorized within the tradition of information terminology in a classical business management sense. Walsh and Ungson postulate the existence of ®ve storage components that, together with sources external to an organization, make up the structure of organizational memory (cf. Fig. 9). Organizational memory has a decentralized character and can be distributed throughout the entire organization. The individual components will now be discussed in more detail. Individuals. The employees of a corporation have their own memories about activities and events in and around the workplace. Thereby, experience and observation play an important role. Among other means, information is stored either directly or in more

Organizational Memory Systems

287

Fig. 9. Structure of ``organizational memory'' according to Walsh and Ungson (1991, 64).

subtle forms such as principles, assumptions, values, etc. In order to support and expand one's own capacity, technological aids are installed (e.g., data bases).

Administrative systems per se serve as mechanisms that embody and preserve information.

Culture. For now culture is de®ned as a learned way of perceiving, thinking and feeling about problems that is transmitted to members in the organization. Of central importance in this respect are the terms ``learned'' and ``transmitted''. Past experience manifests itself in the culture of the organization and is of direct importance for current decisions. In this way culture actually assumes the role of a kind of storage mechanism. Manifestations of culture include language, common conceptions of framework, symbols, stories, myths, and rumors. By repeatedly passing on these aspects of culture, details can be lost or new interpretations may arise. Nevertheless, it is understood as a part of the collectivization of information.

Structures. Structures are to be viewed in how they in¯uence the behavior of employees. At the same time they represent a connection to the environment. Organizational information is established in the form of roles. The concept of role was de®ned by sociology and explains behavior in respect to social expectations. The role conditions and limits individual behavioral freedom in favor of predictable behavior or safe interaction. Security arises through ``correct'' behavior (conformity) by ful®lling reciprocal, established expectations. Patterns develop out of social roles that in¯uence distribution of work and the entire structure of the organization. Without discussing this further, it can be said that information is coded in roles and the resulting structure that in¯uence decisions in the organization.

Transformations. Information is also embedded in the numerous processes that are always taking place in an organization. For example, this can be knowledge about data processing (i.e., how is a certain input transformed to output), a standardized procedure, or a self-ful®lling prophecy. It does not matter whether this information is overt or covert, the impulses from the past direct current procedures and decisions.

Ecology. The physical composition and arrangement of the workplace and environment also contain coded information about the organization (e.g., of®ce furnishings, lighting, number of people). Behavior is in¯uenced to a greater or lesser degree by these things. The status of a person is also often re¯ected by his appearance at his work station. In large of®ces it is, for example, more dif®cult to make friends, and con¯icts

288

Lehner and Maier

occur more frequently. An in¯uence of norms and behavioral patterns is a consequence of this, hence also in this case, a component of organizational memory can be spoken of. External archives. As one can easily determine, the organization itself is not the only place where information about the organization and its past is kept. Every organization is surrounded by others that follow its activities. Thus, when information is lost or cannot be recovered when needed, this information can often be obtained from other sources. Previous employees are one such source, regardless of how they left the organization. Records can also be found at the competition, government agencies, advertising agencies, societies, accounting of®ces, news agencies and the media, etc. Interestingly, Walsh and Ungson do not refer to archives, data base systems, etc., even though in theories on memory these storage facilities have traditionally been labeled as external storage facilities. In fact, these particular labels are missing completely; nevertheless it can be assumed that they can be subsumed under the component ``individuals''. 2.5. Group memory and transactive memory system according to Wegner Group memory refers to the ability to store knowledge in a group over time and to retrieve it at a later point in time. Among other things, one differentiates between procedural and declarative group memory. The indication of a procedural group memory can be arrived at through experiments on behavior in socalled dyads, among others. In this way, persons somehow use other persons to expand their own memory by noticing what the person with whom they are interacting notices and what he or she does not. Accordingly, their own attention and their individual memory is in¯uenced, either consciously or subconsciously. At ®rst, all members in the group are in a position to remember a particular event in its entirety (cf. Oberschulte, 1996, 571). A good introduction to group memory and also to collective memory can be found in Wegner (1986, cf. also Hartwick et al., 1982 or Clark and Stephenson, 1989). The author concentrates on so-called ``transactive memories'' and also on its implications for organizations. More recent articles on this topic often build upon this article and usually describe experiments that bring in another factor or analyze a speci®c claim.

Research projects on group memory reach as far back as the end of the 19th century and the beginning of the 20th century. Especially psychology dealt with this and related issues, which was later almost forgotten. The new theoretical versions and approaches to individual psychology are attempting to carry over the realizations to groups via analogy, whereby the terms ``transactive memory'' and ``transactional memory'' are now commonly used (cf. Wegner, 1986, 185±186). The transactive memory system (TMS) is comprised of a multitude of individual memory systems as well as communication between individuals and persons together. Through closer examination, one can subdivide TMS into three components (cf. Wegner, 1986, 186 ff, out of Maier and Kunz, 1998, 12±13): *

*

*

Individual memory: To characterize individual memory, Wegner uses the well-known phaseoriented view of human information processing (coding, storing, recall). Single information elements are stored as linked sets. Memory also includes a meta-memory that contains knowledge about the stored knowledge. It has a signi®cant in¯uence on how well one can use one's own memory. External memory: This has to do with information that is stored outside individual memory (e.g., in books, CD-ROM, computers). In individual memory, only the ``access pathway'' to the external memory is stored. Along with the key to relocation (information marking), another site is needed to access information in the external memory. Unquestionably, there can be several keys and locations. Wegner believes that a signi®cant part of our individual memory is used for storing information related to sites. Transactive memory: Besides books and other storage media, people can also serve as external memory. In this case, persons act reciprocally as external storage media. In this way, a reciprocal and dependent system of storing information arises in groups that is larger and more complex than each individual memory. The term ``transactive'' describes the feature of this system to process information as transactions that several people take part in. This involves coding, storage and recall. For example, during coding the attempt is made to reach a consensus about the information that is to be stored, i.e., at ®rst the information is collec-

Organizational Memory Systems

tively interpreted and brought into relation with or adapted to other information. When recalling information, the role communication plays is even greater. Working together, the individual memory systems are negotiated or navigated through and informational elements are located and combined. During this process for those outside the group, a group member can act as a kind of guide to lead one through the group's memory. A TMS is a system that is constructed out of individual memory systems and communication between the group members. Overtime it gradually grows (cf. Wegner, 1986, 191). Through the presence of a TMS and the way it is realized, the information processing of the group as well as the individuals involved are in¯uenced. As a general rule, the construction (or better still, the evolution) of a TMS begins when the individual group members learn something about the sectors and depths of knowledge of the other group members. The longer a group exists, the more everyone knows, including who is the expert in certain areas. It often happens that those identi®ed as experts become responsible for coding, storage and recall of knowledge about the respective domain. If this ``distribution of labor'' is not formally described, then the circumstances or situation determine who is responsible for a particular subject area. The willingness of group members to assume responsibility for particular subject areas or information units is a deciding factor for the establishment of a TMS. Advantages of the TMS for the individual are: enhancement of personal experience, specialization of knowledge and the related increase in ef®ciency, access to new information, creation of new knowledge, and turning one's attention toward information that might be relevant for a group member. Besides creative combinations, transactive processes can naturally produce errors, too. Such disadvantages result from group processes that, for example, can create confusion when the responsibility for a particular subject area is not clearly de®ned, or when the formal organizational structure interferes with passing on information to the respective experts. In the end, it depends on the group and group-internal processes as to whether or how information is integrated or through the combination thereof new knowledge is created, etc. Problems also

289

arise when groups are broken up because the transactive part of the individual memory is suddenly rendered super¯uous. This can lead to disorientation among the group's members (cf. Maier and Kunz, 1998, 14±15). The concept of TMS can be applied to the organization's layout. Wegner differentiates between two basic variations, namely integrated and differentiated TMS (cf. Wegner, 1986, 204). The border between the two is ¯uid. Integrated TMS are marked by having identical information stored in several different individual memory systems among the group members. Redundancy and overlapping are consciously taken as given. In differentiated TMS, information is stored only once whenever possible. What is shared is merely the labeling of knowledge elements and their location. Wegner assumes that the applicability of a certain kind of TMS depends on the task at hand and other factors. However, he already names explicit information and communication technologies as critical factors for making groups more ef®cient. The more differentiated the TMS is, the more this is true. Within a corporation, TMS occur in multifarious relational contexts. The structural interrelationships between individual TMS can be seen in Fig. 10. The aggregation and in¯uences at the level of the organization as a whole have not yet been studied. Nevertheless, one can assume an overlapping through ``linking pins'' and the existence of similar mechanisms at higher organizational levels. A TMS is de®ned accordingly by the group af®liation and not necessarily by corporate af®liation. In principle, it can be independent of corporate boundaries. Since most people belong to several groups simultaneously, they are also parts of several TMS and provide the basis for a ¯ow of information between the various groups. The overlapping of teams, i.e., of TMS, can occur based on the people who belong to the various groups. This can also result from using other common resources (e.g., common data bases or being connected to the same computer network) that are not necessarily under the control of any one individual or group. Since the relevance of one type of knowledge and its importance as a stimulus for negotiation is not easily predicted, a visual conceptualization is hardly possible. If one stresses the exchange of information as an especially important kind of relationship, then three different types of relationships can be seen (see Fig. 10).

290

Lehner and Maier

Fig. 10. Structural interrelationships between transactive memory systems.

*

*

*

Type 1: Information exchange between groups (contact at the institutional level). Type 2: Information exchange between a group and a single ``group component'' (usually a person, but it may also be an agent or a data base). Type 3: Information exchange between members or components of two different groups.

The relationships themselves can be either unidirectional or bi-directional, formal or informal, personal or electronic, at one place or over a distance, etc. The communications infrastructure present (e.g., telephone, computer network, postal distributor) enables a nearly inexhaustible potential of interactive forms and patterns to exist. Because of the huge number of combination variables, it appears to be practically impossible to identify super structures or a higher order of TMS in this environment. When addressing the concept of TMS, only a partial aspect is focused that can increase our understanding of speci®c work within a group. What appears relatively clear and understandable in theory quickly becomes exceedingly complex and formally nearly inde®nable when one attempts to apply it in practice. Possibly much can be explained about the structural similarities of social networks by drawing on theories of the neural net and soft computing. In spite of this, many questions remain open, especially on how the performance of organizational memory can be understood and evaluated in regards to individual TMS.

Wahren's speculations (1996, 179ff ) are of interest in relation to TMS (especially on transactive memory). According to Wahren, organizational memory cannot be localized, but rather lies in personal relationships that produce, preserve, and permanently must adapt to a social system. The level of knowledge thus stored is in¯uenced by communication. In this manner, individually stored knowledge is made accessible and understandable to other members of the organization. Within the organizational memory itself, many processes of change take place, which are represented by arrows in Fig. 11. In order to use the knowledge that is stored, knowledge management is necessary in a business. Thereby it does not matter how much knowledge has been accumulated and created, but rather how ef®ciently the speci®c knowledge needed can be recovered at the right place and time (cf. Wahren, 1996, 168±169).

Fig. 11. The structure of the organizational knowledge base according to Wahren (1996).

Organizational Memory Systems

2.6. Group remembering according to Hartwick et al. In his studies, Hartwick et al. (1982) concentrate on the role of group memory within the framework of human information processing. By summarizing the research results in this area, the authors attempt to identify the various theoretical approaches and their empirical evidence. Thus, it does not deal directly with an original concept of memory, but rather with an ordered compilation of models and knowledge in relation to the performance of memory. The inquiries of Hartwick et al. center on: *

*

*

the in¯uence of other persons on the individual person's remembering; the in¯uences of social interaction on information processing in groups; and the combination of individual memory performance and social interaction.

The presence of other people can either hinder or promote remembering. This in¯uence is attributed to how well the other people present pay attention and how they react, i.e., it is dependent for the most part on normative social in¯uences that apply to groups (cf. Hartwick et al., 1982, 50). Another reason might be that individuals are, because of their own limited capacity for information processing, drawn to other persons that can provide additional informationÐ information that expands their own memory capacity, but that can also be used to verify or correct one's own memory. There appears to be no doubt that large groups can remember more than individual persons. Hartwick et al. (1982, 52ff ) propose four models that roughly apply to groups of various sizes. The Truth-WinsModel proposes that a group can remember an informational element whenever one of the group's members can remember it. Experiments indicate, however, that this model of remembering for groups is probably overestimated and actually represents the upper limit of performance, because it is reduced by the effects of social and psychological forces. The Truth-Supported-Model proposes that an element is recalled when at least two group members remember it. It is thereby assumed that it is often dif®cult for the individual to analyze the correctness of information, and thus either an implicit or explicit rule for decision making is utilized in groups. The Simple-MajorityModel is based on a similar standpoint. According to

291

this model, an informational element is only remembered by consensus when it is remembered or at least recognized by a simple majority. The fourth model is merely seen as a combination of the ®rst three whereby the process begins by having one or more group members remember something, which is then over time submitted to the process of group discussion. After possible reorganization through social interaction, a group opinion is formed. With a rather high degree of probability, group members tend to accept the group opinion as their own. As can be shown empirically, group discussions lead to an improvement of individual memory performance. From this it can be concluded that interaction in groups helps to consolidate information and makes information acquisition more ef®cient. The in¯uence of persons present on memory performance has already been alluded to. One assumes that this in¯uence varies depending on group size. Hartwick et al. attempted to combine the individual experimental results in one model. The test results, however, did not turn out as expected. This led Hartwick et al. to assign co-responsibility to factors such as power, group conformity, or the development of behavior and roles. Models that re¯ect this ®rst have to be developed, though. The research here still has not made any breakthroughs, and many hypotheses are waiting now as then for empirical veri®cation. Because of the proven in¯uence of interaction and communication on memory performance in groups, it is highly likely that computer support will at least have some effect. In any case, the results of the research must take into account the development of organizational memory systems. The ¯ow of information can then be made much more ef®cient through the help of electronic communication media. Discussion and voting processes remain just as dif®cult as before. Here new questions arise that can be analyzed within the framework of groupware research. For example, how large should groups be? How can groups be modi®ed that arise independently in discussion groups? Which persons should be connected to which topics? In the studies of Hartwick et al., only highly speci®ed partial functions of memory are examined, namely memory performance or recall functions in relation to explicit knowledge. However, this only accounts for one part of organizational memory, and the research results are relevant for the structuring and

292

Lehner and Maier

support of these partial aspects (e.g., knowledge management). No statements can be found on organizational learning that deal with other processes or other storage forms of knowledge. 2.7. Technological concept of organizational memory according to Watson For Watson (1996) ``organizational memory'' is understood above all as something technological, especially in the form of a comprehensive and corporate-wide data(base) concept that is increasingly enhanced by multimedia and network technologies (e.g., imaging, archive and document management systems). For Watson, the tasks of organizational memory and data management are largely equivalent. The following two questions take precedence for him: Where and how are data stored? How can the data be accessed? (see Watson 1996, 5). Fig. 12 illustrates the most important components of his model. Watson proceeds from the assumption that organizations need stored information in some way, shape or

Fig. 12. Components of organizational memory (Watson, 1998).

form in order to prepare for and make decisions, and to deal with daily operations. The content of organizational memory can increase by adding external information (cf. Watson, 1996, 25). In order to differentiate between the terms data, information and knowledge, Watson suggests describing data as facts that have not yet been analyzed. Information can be produced from this data via appropriate processing. And knowledge is explained as the ability to use the information. Fig. 13 shows the relationship of these terms. Incidentally, this way of viewing things does not represent the prevailing views, but is rather one option among many (cf. especially Lehner et al., 1995). In relation to the concept of ``organizational memory'' Watson points out a few weak points, i.e., critical areas where technological realization should be given special attention (see Watson, 1996, 26): *

*

Redundant data storage, i.e., the same data can be found in several different storage media. Inadequate data management.

Organizational Memory Systems

293

Fig. 13. Relationships between data, information and knowledge (Watson, 1996, 29).

*

*

*

De®ciencies in query tools and user interfaces can make accessing the data dif®cult. Organizational memory does not include the entire complexity of the business's world and hence does not provide adequate support. Data that are spread out over many storage systems lead to a reduction in data integrity.

common level of knowledge and the phenomena associated with it. Examples of terms in use are: *

*

*

3. Integrating and Applying Theories and Concepts for the Development of OMS 3.1. De®ning organizational memory systems All de®nitions and attempts to explain organizational memory can be found along a spectrum that runs from concept to construct. These two dimensions' basic properties are each characterized by one of their proponents. ``OM is a concept that an observer invokes to explain part of a system or behavior that is not easily observed'' (Krippendorg, 1975, quoted from Rao and Goldmann-Segall, 1995, 333±334). Organizational memory ``. . . is the know-how of a business recorded in documents (reports, ideas, concepts, etc.)'' (Morschheuser, 1997, 19). Regardless of the position one assumes, organizational memory has to do with either something abstract (theory, explanatory model, thought schemata, concept) or something concrete (e.g., documents, data bases, knowledge base, repository). The idea of organizational memory in relation to economic and technological developments led to an intense debate in the relevant literature. This discussion is turning more and more toward corporate practice. Not to be ignored are the numerous similar or synonymous terms that can be gleaned from the intensity of the discussion that revolve around a

* *

* * *

* *

*

* *

organizational memory, corporate memory (Walsh and Ungson, 1991; Rao and Goldmann-Segall, 1995) corporate, organizational or enterprise-wide knowledge-base (e.g., Duncan and Weiss, 1979; Pautzke, 1989; GuÈldenberg and Eschenbach, 1996) organizational or corporate knowledge (e.g., Gershman and Gottsman, 1993; Duncan and Weiss, 1979) institutional, collective, or systemic knowledge cooperative memory (Rao and Goldmann-Segall, 1995) social memory (Walsh and Ungson, 1991) collective mind collective intelligence, corporate intelligence (e.g., Brown and Duguid, 1991) corporate genetics (Prahalad and Hamel, 1991) transactional or transactive memory (Wegner, 1986) group memory, group mind (e.g., Nelson and Cooprider, 1996; Dennis, 1996; Mandviwalla and Clark, 1995) meeting memory (Sandoe et al., 1991) team memory (Morrison, 1993)

Further variations found in practice are: * * * * * *

shared knowledge base (SKB) knowledge warehouse (Scheer, 1998, 64±65) corporate repository (Nelkowski, 1996) corporate memory (Jacobsen, 1996, 168) technological knowledge (Bohn, 1994) know-how database (KhDB) (Roithmayr, 1997)

The fact that there is still no clear or uni®ed use of terms is a sign of the liveliness and novelty of the

294

Lehner and Maier

research topic. As the discussions have shown up to this point, the meaning of the terms is also not always identical because they originate in part from different disciplines and therefore also have different aims as far as knowledge is concerned. As far as can be determined at this time, the status of knowledge and analysis is mostly the same. By momentarily disregarding the perspective that is connected with the terminology of a scienti®c discipline, one can somewhat justifyably speak about synonyms, since they address phenomena that are interdisciplinary, and no discipline can be characterized solely as the ``problem owner''. Fig. 14 summarizes the current status of terminology usage as terms are related to each other. Finally, pursuing the concept or construct of organizational memory is not an end in itself. It serves to explain complex developments in organizations and should, among other things, support the active structuring of processes in change. To provide a powerful example, a statement by Matsuda regarding the application of organizational intelligence is helpful: ``In order to maintain the long-term availability of perceived information, it must be stored. In this case, information is divided into institutional

Fig. 14. Relationship of OM-related terms to each other.

information and cultural information. For instance, ®rst technological expert knowledge, organizational and management knowledge, the organizations' standards and norms, and the procedures within the organization are stored (institutional documentation). In addition, the culture of the organization (corporate identity), for instance, is stored along with the fundamental outlook and guidelines of the organization, the traditions, key occurrences, precedence cases, habits, customs and rituals, as well as the organizations' special terminology (cultural documentation). Information and communication technologies are increasingly used to aid the storage process.'' (Matsuda, 1993, 14) OMS are closely connected to the concept of OM presented above. They are easiest understood, if one starts with the functions which are to be ful®lled by these systems as well as from the tasks and procedures which they support. Both are closely connected with the above mentioned approaches which attribute a particularly central role to organizational learning and knowledge management. However, the discussion on the different areas of application has not progressed much in the relevant literature, though there does

Organizational Memory Systems

seem to be a large degree of consent at least regarding organizational learning. The following central functions are stated (cf. e.g., Huber, 1991): *

* *

* *

General support of learning processes within the company (individual and collective)Ðand as a speci®cation of this general function. The support of knowledge acquisition. The distribution of information and knowledge (general and homogenous access to the knowledge base, information service center, active but also foresighted service point). Help with the interpretation of information. (Technical) realization of the OM.

Due to the situation described above and the dynamic development of the application areas, it is not surprising that there are hardly any de®nitions for this new type of information system yet. One of the few attempts at an explanation was made by Stein and Zwass (1995, 95), who de®ne an OM Information System as ``a system that functions to provide a means by which knowledge from the past is brought to bear on present activities, thus resulting in increased levels of effectiveness for the organization''. This de®nition clearly aims at the contribution of such systems to the increase in organizational ef®ciency. In order to be compatible with the considerably widened spectrum of use, the following de®nition on a subset of organizational memory systems, knowledge management systems, is suggested (see Maier and Lehner, 2000): De®nition: A knowledge management system (KMS) is a dynamic system which provides functions to support the identi®cation, acquisition, retention, maintenance, search and retrieval, distribution, selling and logistics of knowledge, which is seen as information plus context, the aim of which is to support organizational learning and organizational effectiveness. A more detailed investigation into the subject and a discussion of multiple perspectives on organizational memory systems can be found in Lehner 2000. Apart from the synonymous term OMIS (organizational memory information system) which was already used above, the term CSOM (computer-supported organizational memory) can be found in the relevant literature.

295

Organizational memory systems (OMS) are generally characterized by the fact, that a whole bundle of technologies is used and not an isolated single technology. A particular obstacle regarding the categorization and order occurs due to the close connection to the concepts of information and knowledge. These terms are usually understood in a very inhomogeneous and wide way, and a homogenization is neither to be expected in the near future nor realistic. That is why it is not surprising that a very wide scope of realization forms is seen. This reaches from a very restricted understanding in the context of specialized systems for the support of speci®c, relatively clearly de®ned tasks (e.g., the admission procedure of new medicines), via the company-wide use (e.g., knowledgeLINK, see Lehner, 2000) to very general information systems which at ®rst have no speci®c connection to the corporate performance (e.g. Answer Garden, see Ackerman and Malone, 1990). Thus, OMS could certainly be treated in the same way as conventional information systems. The justi®cation for a speci®c approach which is emphasized here is based, on the one hand, on the insuf®cient support provided by traditional methods of systems development and, on the other hand, on the particular requirements which are expected from such systems, due to the contribution to the development of the organization and to learning organizations. 3.2. Future research on organizational memory In summary it should be remembered that the terms cited are being used with increasing frequency, but in a relatively broad and general way and with inadequate precision. This then leads to signi®cant dif®culties in understanding or making oneself understood, or at least leads to problems when one attempts to turn theory into practice. The basic concepts discussed here are thus to be stressed once again. Additionally, ample opportunities for confusion arise once again with knowledge bases as they are known from arti®cial intelligence (which can naturally be a part of the organizational knowledge base) and with corporate knowledge bases (e.g., electronic organization handbook) as a part of organizational information systems (this is generally an important part of organizational knowledge). From an historical perspective, another facet arises still, since the terminological context of knowledge management once had to do with the application of

296

Lehner and Maier

expert system technology and knowledge-based systems. Only gradually and rather cautiously then is this technological horizon broadening. In the meantime, through the availability of ef®cient technologies and media, a broad spectrum of possibilities exists that is promoting knowledge management and organizational learning. Further discussions will concentrate on this, whereby the differences documented herein will be given the attention they deserve. There are a lot of unresolved research questions in this area. We want to address the following ones in our studies: Economics of the application of knowledge management systems: How can we measure the success of such an effort? What measures could be programmed into the systems so that they are provided automatically? How should we fund and ®nance knowledge management services? Strategies of knowledge management: What strategies can we distinguish when we look beyond the very abstract differentiation between a codi®cation and a personalization strategy? Which strategies have proven to be successful? Organizational design of knowledge management: What roles can we differentiate? Who uses knowledge management systems (individual characteristics; different organizational levels are considered, e.g., the whole organization, organizational units, e.g., departments, work groups, teams, communities; individual level) and how do people use these technologies? Who should be responsible for what kind of tasks? Which knowledge processes can be distinguished? What is the relationship between business processes and knowledge processes like? Organizational culture; What impact does the application of knowledge management systems have on organizational culture? Is this impact manageable? Content of knowledge management systems: How do we proceed to structure knowledge management systems? What level of detail is best suited for which applications? What is the best size of knowledge elements? What meta-knowledge do we have to consider? Knowledge management systems: What architectures can be distinguished? How can we classify knowledge management systems? What are the differences to other application systems? What kinds of KM technologies exist or what kinds of technologies are

proposed for the use in knowledge management approaches? What functions do knowledge management systems provide? Implementation of knowledge management systems: How do we proceed to implement a knowledge management system? Which (modeling) methods are used for the design, implementation and maintenance of the technological infrastructure concerning KM technologies? Which pros and cons are attributed to these methods? These questions show that much has to be done in this research ®eld. As we tried to point out, there is a strong need for an interdisciplinary approach, which combines research ®ndings from (at least) organizational psychology and sociology, organization science, business informatics and computer science. Our de®nitions and perspective are seen as a ®rst step in the direction of an integration of the various approaches developed in these disciplines.

Notes 1. Knowledge management systems in our view are a subset of organizational memory systems which have a tendency to focus on the more static documentation (retention, maintenance, search and retrieval) and distribution parts of organizational memory systems. This holds true for most of the systems offered on the market which lack functionality to support the dynamics of an organizational memory, that is organizational learning (see Maier and Lehner 2000).

References Ackerman MS, Malone TW. Answer garden: A tool for growing organizational memory. In: Lochovsky FH, Allen RA, eds. Proceedings of the Conference on Of®ce Information systems 25.±27.4.1990, Massachusetts 1990:31±39. Ackerman MS. Answer Garden: A Tool for Growing Organizational Memory. MIT Sloan School of Management. (PhD thesis), Boston, 1994. Argyris C, SchoÈn D. Organizational Learning. A Theory of Action Perspective. Reading/Massachusetts, 1978. Bannon L, Kuuti K. Shifting perspectives on organizational memory: From storage to active remembering. In: Proceedings of HICSS'96, 29th Hawaii International Conference on System Sciences. Hawaii: IEEE, 1996. Buckingham-Shum S. Negotiating the construction and reconstruction of organisational memories. Journal of Universal Computer Science 1997;3(8). Bohn RE. Measuring and managing technological knowledge. In: Sloan Management Review, 1994;35(4):61±73.

Organizational Memory Systems Brown JS, Duguid P. Organizational learning and communities-ofpractice: Toward a uni®ed view of working, learning, and innovation. In: Organization Science 1991;2(1):40±57. Bullinger H-J. Electronic businessÐMaÈrkte, Technologien, Erfolgsstrategien fuÈr Anbieter und Nutzen. In: Proceedings of the Fraunhofer IAO-Forum, 23.11.1995. Stuttgart, 1995. Bullinger H-J, WoÈrner K, Prieto I. Wissensmanagement heute. Daten, Fakten, Trends. Fraunhofer Institute fuÈr Arbeit und Organisation (IAO), Stuttgart, 1997. Clark NK, Stephenson GM. Group remembering. In: Paulus PB Ed. Psychology of Group In¯uence. 2nd. ed., Hillsdale, 1989. Cyert RM, March IG. A Behavioral Theory of the Firm. Englewood Cliffs, 1963. Dennis AR. Information exchange and use in group decision making: You can lead a group to information, but you can't make it think. In: MIS Quarterly. 1996;20(4):434±457. Douglas M. How institutions think. Syracuse, New York, 1986. Duncan RB, Weiss A. Organizational learning: Implications for organizational design. In: Staw B, ed. Research in Organizational Behaviour, Greenwich/Conn. 1979:75±123. Gershman A, Gottsman E. Hypermedia for Corporate Knowledge Dissemination. In: Proceedings of the 26th Hawaii International Conference on System Sciences '92, Los Alamitos: IEEE Press, 1993. Goodhue DL, Thompson RL. Task-technology ®t and individual performance. In: MIS Quarterly 1995;19(2):213±236. GuÈldenberg S, Eschenbach R. Organisatorisches Wissen und LernenÐerste Ergebnisse einer qualitativ-empirischen Erhebung. In: zfo. 1996;(1):4±9. Hanker J. Die strategische Bedeutung der Informatik fuÈr Organisationen: IndustrieoÈkonomische Grundlagen des strategischen Informationsmanagements. Stuttgart, 1990. Hansen MT, Nohria N, Tierney T. What's your strategy for managing knowledge? In: Harvard Business Review 1999;77(2):106±116. Hartwick J, Sheppard BH, Davis JH. Group remembering: research and implications. In: Guzzo RA, ed. Improving Group Decision Making in Organizations. New York: Academic Press, 1982. Hedberg B. How organizations learn and unlearn. In: Nystrom PC, Starbuck WH, eds. Handbook of Organizational Design. New York/Oxford, 1981:3±27. Huber GP. Organizational learning: The contributing processes and the literatures. Special Issue: Organizational learning: papers in honor of (and by) James G. March. In: Organization Science 1991;2(1):88±115. ILOI (International Institute fuÈr Lernende Organisation und Innovation, ed.). Knowledge Management. Ein empirisch gestuÈtzter Leitfaden zum Management des Produktionsfaktors Wissen. Munich, 1997. Jacobsen A. Unternehmensintelligenz und FuÈhrung ``intelligenter'' Unternehmen. In: Technologie & Management, 1996;45(4):64± 170. Kantrow AM. Intelligent enterprise and public markets (An interview with professor James Brian Quinn). In: The McKinsey Quarterly 1994;(2):83±95. Kantrow AM. The Constraints of Corporate Tradition. New York, 1987. Kirsch W. Entscheidungsprozesse. Vol. 1±3, Wiesbaden 1970 and

297

1971. Kock N. Process Improvement & Organizational Learning. The Role of Collaboration Technologies. Hershey/London, 1999. Lehner F. Organisational Memory. Konzepte und Systeme fuÈr das organisatorische Lernen und das Wissensmanagement. Munich, Vienna, 2000. Lehner F, Maier R, Klosa O. Organisational Memory SystemsÐ Application of Advanced Database and Network Technologies in Organisations. Research Paper No. 19, Department of Business Informatics III, University of Regensburg, February, 1998. Lehner F, Hildebrand K, Maier R. Wirtschaftsinformatik. Theoretische Grundlagen. Munich, 1995. Lenk Th. Telearbeit. MoÈglichkeiten und Grenzen einer telekommunikativen Dezentralisierung von betrieblichen ArbeitsplaÈtzen. Berlin 1989. Madey G, Muzumdar M. Organizational knowledge management: A case study of a leading practice in the professional services industry. In: Proceedings of the SIM Academic Workshop 1997, 12.-13.12.1997. Atlanta, 1997. Maier R, Kunz St. Ein Modell zur organisatorischen Informationsverarbeitung. Research Paper No. 20, Department of Business Informatics III, University of Regensburg, February 1998. Maier R, Lehner F. Perspectives on Knowledge Management SystemsÐTheoretical Framework and Design of an Empirical Study. Mandviwalla M, Clark S. Collaborative Writing as a Process of Formalizing Group Memory. In: Proceedings of the 28th Hawaii International Conference on System Sciences '95. Los Alamitos: IEEE Press, 1995. March JG, Simon HA. Organizations. New York, 1958. Matsuda T. ``Organizational Intelligence'' als Prozeû und als ProduktÐEin neuer Orientierungspunkt der japanischen Managementlehre. In: Technologie & Management. 1993;42(1):12±17. Morrison J. Team memory: Information management for business teams. In: Nunamaker J, Sprague R, eds. Proceedings of the 26th Hawaii International Conference on System Sciences '93, Los Alamitos: IEEE Press, 1993:122±131. Morschheuser St. Integriertes Dokumenten- und Work¯owManagement. Wiesbaden, 1997. MuÈller-Stewens G, Pautzke G. FuÈhrungskraÈfteentwicklung und organisatorisches Lernen. In: Sattelberger T, ed. Die lernende Organisation. 3. ed., Wiesbaden, 1996:183±205. Nelkowski Ch. Corporate repository: Information als Wettbewerbsvorteil. In: Industrie Management 1996;12(6):36± 38. Nelson KM, Cooprider JG. The contribution of shared knowledge to IS group performance. In: MIS Quarterly 1996;20(4):409±433. North K. Wissensorientierte UnternehmensfuÈhrung. 2nd ed., Wiesbaden, 1999. Oberschulte H. Organisatorische IntelligenzÐEin Vorschlag zur Konzeptdifferenzierung. In: SchreyoÈgg G, Conrad P, eds. Managementforschung, Vol. 6: Wissensmanagement. Berlin, New York, 1996:41±81. Ortner E. Ein Referenzmodell fuÈr den Einsatz von Dictionary-/ Repository-Systemen in Unternehmen. In: Wirtschaftsinformatik. 1991:(5):420ff. Pautzke G. Die Evolution der organisatorischen Wissensbasis.

298

Lehner and Maier

Bausteine zu einer Theorie des organisatorischen Lernens. Munich, 1989. Peters T. Jenseits der Hierarchien. DuÈsseldorf, 1993. Peters T. The Tom Peters Seminar. London, 1994. Prahalad CK, Hamel G. Nur Kernkompetenzen sichern das È berleben. In: Harvard Manager 1991;13(2):66±78. U Poveschi R. Information Technology for Knowledge Management. Berlin et al. 1998. Pulic A. Der Informationskoef®zient als WertschoÈpfungsmaû wissensintensiver Unternehmen. In: Schneider U, ed. Wissensmanagement. Frankfurt, 1996:147±180. Rao VS, Goldmann-Segall R. Capturing stories in organizational memory. In: Proceedings of the 28th Hawaii International Conference on System Sciences '95. Los Alamitos: IEEE Press, 1995:333±341. Roithmayr F. Know-how-Datenbank. In: Mertens P et al. (eds.): Lexikon der Wirtschaftsinformatik. 3rd. ed., Berlin, 1997:227± 228. Sandoe KL, Holfman, L., Mandviwella M. Meeting in time: Recording the Workgroup Conversation. In: Proceedings of the 12th International Conference on Information systems. New York, 1991:261±272. Scheer A-W. ARIS - Vom GeschaÈftsprozeû zum Anwendungssystem. 3. ed., Berlin et al. 1998. Schneider U (Ed.). Wissensmanagement. Die Aktivierung des intellektuellen Kapitals. Frankfurt 1996. Schneider U. Management in der wissensbasierten Unternehmung. Das Wissensnetz zwischen Unternehmen knuÈpfen. In: Schneider U, ed. Wissensmanagement. Frankfurt, 1996:13±48. SchuÈppel J. Wissensmanagement: organisatorisches Lernen im Spannungsfeld von Wissens- und Lernbarrieren. Wiesbaden, 1996. Spek Rvd, deHoog R. Towards a methodology for knowledge management. Technical Note Knowledge Management Network, 23. November 1994, URL: http://ceres.cibit.nl/web/kmn/

pospapers.nsf Stein E, Zwass V. Actualizing organizational memory with information systems. In: Information Systems Research 1995;6(2):85±117. Stein E. Organizational memory: Review of concepts and recommendations for management. In: International Journal of Information Management. 1995;15(2):17±32. Wahren H-K. Das lernende Unternehmen. Theorie und Praxis des organisationalen Lernens. Berlin, New York, 1996. Walsh JP: Ungson GR. Organisational memory. In: Academy of Management Review 1991;16(1):57±91. Wargitsch C. Ein Organizational-Memory-basierter Ansatz fuÈr ein lernendes Work¯ow-Management-System. Research Paper No. FR-1997-004, Bayerisches Forschungszentrum fuÈr Wissensbasierte Systeme, Erlangen, Munich, Passau, May 1997. Watson RT. Data Management. An Organizational Perspective. New York, 1996. Watson RT. Data Management. Databases and Organizations. 2 ed., New York, 1998. Weber M. Wirtschaft und Gesellschaft. TuÈbingen, 1976. Wegner DM. Transactive memory: A contemporary analysis of the group mind. In: Mullen B, Goethals GR, eds. Theories of Group Behaviour. New York, 1986:185±208. Weick KE, Gil®llan DP. Fate of arbitrary traditions in a laboratory microculture. In: Journal of Personality and Social Psychology 1971;17:179±191. Weick KE, Roberts KH. Collective mind in organizational headful interrelating on ¯ight decks. In: Administrative Science Quarterly 1993;38(3):357±381. Wild J. Zur Problematik der Nutzenbewertung von Informationen. In: ZfB. 1971;41:315±334. Zucker B, Schmitz C. Knowledge ¯ow management: Wissen nutzen statt verspielen. In: Gablers Magazin. 1994;8(11±12):62±65.