Paper: KM Technologies 1Paper | Introduction | Categorizations |
[KM strategy] | [Taxonomy] |
[Discussion]
Title: A Taxonomy of Knowledge Management TechnologiesAbstract: Here comes the abstract.
Keywords: some keywords.
IntroductionEven being
regarded by many as just an instrument for the management of knowledge,
technology is still a critical enabler of organizational knowledge
processes. However, it has been challenging to properly understand the many ways technology can
support the creation, organization, transfer and use of
knowledge, and to select adequate solutions for particular knowledge
management (KM) initiatives. This is due not only to the the dynamics of technology,
which is rapidly developing in many areas, like connectivity and
mobility,
communication and collaboration, and even semantics and intelligence,
but also due to the diversity of the KM field itself, which presents
diverse and sometimes conflicting perspectives of knowledge and of what
consists its management. Authors studying the subject follow, for instance, either technically oriented
approaches focusing on codifying
knowledge and building repositories, or people-oriented ones focusing
on collaboration and the nurturing of communities, or even
assets-oriented ones, focusing on intellectual capital measurement
and management (Earl, 2001).
There
have been many attempts to explain how technology supports KM, and
several categorizations of KM technologies have been proposed. For one
reason or another, however, those categorizations do not provide
adequate guidance to practitioners’ decision making. Some studies, for
instance, concentrate in illustrating how technology can support
KM and do not aim at a comprehensive coverage of existing alternatives
(Nonaka, Reinmoller & Toyama, 2001; Alavi & Leidner, 2001,
Marwick, 2001). Others
describe technology commercially available as KM products and provide
such comprehensiveness (Hoffmann, 2001; Wenger, 2001; Lindvall, Rus &
Sinha, 2003), but the groupings are numerous and
additional guidance is needed for adequate selection. Another set of
studies facilitate selection by classifying KM technologies according
to their
application to business (Binney,
2001; Tsui, 2003), but they sometimes group technologies
too
different from one another without further consideration (e.g.
intelligent agents, relational databases and push technologies put in
the same category), rendering them insufficient for adequate
analysis.
An
indication of useful criteria for organizing and explaining the range
of KM technologies may be found in the KM literature itself. A
recurrent advice has been the
need to link KM programs to business strategy. It is argued that the
approach to KM must be consistent with the organization's strategy, and
that KM must support strategic goals in order to be successful (Hansen,
Nohria & Tierney, 1999; Zack, 2001; Horwitch & Armacost, 2002).
This recommendation is also present in many methodologies proposed for
the implementation of KM. Existing frameworks usually include
considerations about strategy in order to plan and prioritize
initiatives (Rubenstein-Montano et al., 2001; Mentzas, 2001; O'Dell et
al., 2003). Moreover, the study of information systems (IS) and
information technology (IT) has been concerned with IS/IT strategic
alignment for a long time (Scott Morton, 1991; Earl, 1996). Therefore,
consistently with these studies, to analyze KM technologies according to
their relation to strategy may provide valuable insights.
In this paper, we
aim to develop a taxonomy of KM technologies that helps their
understanding and facilitates their selection in support to KM
programs. We explore the connections between KM technologies and KM
strategy, and propose a categorization based on elements emerging from
that relation. We begin, in the next section, with a critical review of existing
categorizations of KM technologies, eliciting their main contributions
and shortcomings. Then, we proceed with a content analysis of relevant
literature on KM strategy and build a conceptual map to identify key
criteria to be used for developing the taxonomy. In the following
section, we present the proposed taxonomy and survey the existing
technologies, classifying them according to those criteria.
Finally, we describe how the taxonomy could be used in typical
situations and make some considerations about its usefulness and
limitations.Existing categorizations of KM technologiesThere are many works that shed light on how technology supports
knowledge management and knowledge processes. The topic, however, remains complex, and any attempt to categorize
KM technologies faces varied difficulties, which have been
diversely addressed by existing studies. Before examining the proposed
categorizations and how they have dealt with this challenge, though,
we will discuss an issue that is needed for further analysis.
Different levels of aggregationOne salient characteristic of present technology is its capacity of
being integrated into systems. For instance, spark plugs, valves and
cylinders represent different technologies that can be integrated into
a fuel engine. A fuel engine is a system that can be combined with
other ones like transmission, suspension, and electric, into an even
larger system called automobile. In the KM technology domain, we have,
for instance, agents, indexes and search engines combined into a search
system, and search systems, databases, and ontologies combined into
a content management system. Or we have a knowledge base and an
infererence engine combined into an expert system, and expert systems,
datamining and multi-dimensional analysis combined into a business
intelligence system.
As we can notice, technology may be integrated in
many different levels, and this feature hinders analysis it if we are
not aware of which level is our focus. For
instance, it does not make sense to compare agents with content
management systems, in the same way that it does not make sense to
compare tyres with cars. In this manner, besides the hurdle of defining
KM, along with its many perspectives and approaches, and the challenge
of identifying the whole range of technologies that can support it, we
must be careful to analyse KM technologies only in compatible levels of
integration, otherwise the result may render meaningless. We will
notice in the following analysis of existing categorizations that this
has happened in some cases.
Technologies supporting knowledge processes
The varied categorizations of KM technologies presented in the
literature can be grouped into four main types.
The most frequent ones classify technologies according to knowledge
processes - e.g. creation, storage and retrieval, transfer, and
application; or socialization, externalization, combination, and
internalization. They usually adopt a particular perspective of KM,
identify a set of core processes, and list tecnhologies that can be
used to support them (Nonaka, Reinmoller & Toyama,
2001; Marwick, 2001; Alavi & Tiwana, 2003; Becerra-Fernandez,
Gonzalez & Shaberwal, 2004; Jashapara, 2004). Their objective is
either to demonstrate that technology can actually support KM, or to
illustrate how a particular KM model can be implemented with the aid
of technology.
Those studies provide a good explanation of how technology
can
be used for KM. However, the categorizations themselves are highly dependent on
each author's
particular interpretation of what
consists knowledge management. The processes chosen to
describe the range of activities in KM vary widely, and although it is
possible to perceive a certain degree of similarity (see Table I), a
closer analysis
reveal that they are not equivalent. For instance, Nonaka et al. (2001)
base their work on the well known SECI spiral of knowledge creation:
socialization, externalization, combination, and
internalization, which focus on interactions among people and emphasize
the social nature of knowledge. Alavi and Tiwana (2003), although trying to balance
the social and
technical aspects of knowledge, choose processes that tend to interpret
it as product: creation, storage and retrieval, transfer, and
application. Becerra-Fernandez
et al. (2004) adopt a technical slant and emphasize a knowledge
engineering approach, proposing the processes discovery, capture, sharing, and application. Although technologies listed in those studies are surprisingly similar, the way they are
grouped and organized reflects particular interpretations of KM.
Take in Table I.
What is evident from an analysis of those studies is that knowledge
processes are too complex to be used as a general classification
scheme. Knowledge processes occur at many different levels -
individual, group, organizational - and are deeply inter-related. For
instance, Nonaka's SECI model aims at the creation of knowledge at the
organization level, but there is much capturing, sharing, storage,
retrieval,
application, etc., happening at individual and group levels. In a
similar way, knowledge can be created at the individual, group and
organizational levels, and what is creation in a certain level may be
interpreted as transfer at a different level. This complexity becomes
apparent if we try to associate technologies to knowledge processes.
Alavi & Tiwana (2003), for instance, cite e-elearning as a
technology
for knowledge creation. They are obviously focusing the individual
level, since e-learning is widely recognized as a tool for
disseminating existing knowledge. Nonaka et al. (2001), on the other
hand, cite communication and collaboration technologies like
videoconferencing and groupware, focusing on knowledge creation at the
group level. Therefore, any given set of knowledge processes must be
accompanied by a detailed explanation of what they mean, what are their
focus, and what are the activities they refer to. Otherwise, it is
liable to subjective interpretation.
Alternative approaches to categorization
A second type of studies categorize commercially available KM products
and solutions implemented in existing KM initiatives. Instead of
trying to prove the supporting role of technology for KM, these studies
focus
on cataloguing the wide range of alternatives, aiming at a
comprehensive coverage of KM technologies (Hoffmann, 2001; Wenger,
2001; Luan & Serban, 2002; Lindvall,
Rus & Sinha, 2003; Tsui, 2003; Maier, 2004; Rao, 2005). They
provide a good overview of the range of possibilities. One major
contribution of these works, besides the comprehensive
coverage, is to present technologies in similar levels of integration.
Since they focus on commercial products and implemented solutions, they
avoid listing component technologies like intelligent agents,
ontologies, and OLAP (on-line analytical processing). Instead,
they list search systems, content management and business intelligence,
which are
in a level more appropriate for decision making and implementation,
rendering this type of approach useful from a practitioner's point of
view. The wide coverage of alternatives, though, also represents a
drawback. The groupings listed are usually numerous and vary
significantly in functionality and scope. In addition, the link between
technologies and knowledge processes is not as clear as in the previous
approach. Therefore, additional guidance is necessary in order to
facilitate the selection of technologies suitable for particular KM
programs.
A
third type of study argues
for the development of integrated platforms for KM. They propose
layered architectures that provide the infrastructure required for a
complete set of knowledge processes and activities (Tiwana, 2002; Luan
& Serban, 2002; Lindvall et al., 2003; Maier, 2004).These works
usually provide an accurate distinction among different levels of
integration. A major contribution of this approach is that it provides
a framework for
the implementation of KM technologies. Layered architectures, actually,
are a standard way by which information and
communication technology in general are implemented,
and it is more than likely that
organizations already have some sort of layered infrastructure in
place. A KM system architecture, then, provides indications on how to
integrate different types of KM technologies among themselves and into
the existing infrastructure. One shortcoming is that the architectures
proposed usually provide for a centralized, organization-wide KM
platform. Networked or distributed organizational configurations may
need a decentralized architecture, and independent or ad hoc needs do
not require such considerations.
A fourth type of study focus
on technology selection from a managerial point of view. They relate KM
technologies to business needs, and
classify them according to business applications (Binney, 2001;
Tsui,
2003). This approach is akin to that intended in the present study,
aiming to facilitate practitioner's decision making. One major
contribution of
this approach is to consider technologies in terms of applications,
changing the focus to the function they perform in the organizational
context. Technologies are grouped according to the kind of support to
business: e.g. operations, decision making, asset management, process
improvement, or innovation. This type of categorization is closer to
the usual managerial mindset than those based on knowledge processes or
KM system architecture. One drawback in those studies, though, is the inadequate treatment of different levels of
integration. As mentioned in the introduction, component technologies
like semantic
networks, intelligent
agents and push technologies are listed in the same level as document
management, data analysis and reporting, and on-line training, which
fit a higher level of integration. This confusion hampers the main
purpose of facilitating selection and decision making.
Relevant criteria for the taxonomyThe
critical review of existing studies on KM technologies, we could
identify two criteria that are relevant for building the taxonomy. The
first is the existence of different levels of integration, and the
second is the type of application to which technologies are associated.
In order to facilitate analysis and selection of KM technologies, it is
necessary to keep in mind that specific levels are linked to specific
conditions of use. We will distiguish here only two levels of
integration: component technologies and systems. The basic criteria for
classifying technologies as either component or system is that, in
general, system integrate a series of component technologies. Also,
systems usually have a more specific funcionality, usually being
applied for a more restricted purpose. Component technologies are more
generic, can be integrated in diverse systems, for different purposes.
The line separating both is fuzzy, though, and the classification of a
given technology in one or the other may be questioned.
The type of application refers to what technologies are used for. Since
component technologies are generic, in comparison to systems, they will
not be differentiated according to application. Systems, although, can
be distinguished between general purpose KM applications and specific
purpose business applications. KM applications are systems that can be
applied to a wide range of knowledge domains, being developed to
support generic knowledge processes. Examples are document management,
group support, knowledge portals, and e-learning. Business
applications, on the other hand, are systems that focus on specific
knowledge domains, and integrated KM technologies in the particular
context of that given domain. Examples are customer relationship
management, computer integrated manufacturing, enterprise resource
planning, and business intelligence.Link between KM technologies and KM strategyKM strategy as an approach to knowledge management
KM strategy as a knowledge-based strategy for the organization
KM strategy as a plan to implement knowledge management
KM tools
Explain conceptual map
Proposed taxonomy of KM technologiesExplain criteria and present taxonomy
Component technologies
Knowledge management systems
DiscussionKM technologies require context
Three approaches to KM technologies
Conclusiontext
Notes
Tags: JKMx, paper, KMtech
|
