Analyzing Knowledge Management Systems: A

Veritistic Approach
Palash Bera1, Patrick Rysiew2
1

MIS division, Sauder School of Business, 2053 Main Mall, University of British Columbia,
Vancouver, V6T 1Z2, Canada
2
Department of Philosophy, 1866 Main Mall, University of British Columbia, Vancouver,
V6T 1Z1, Canada
{palash, rysiew}@interchange.ubc.ca

Abstract. Knowledge management systems (KMS) are increasingly becoming

popular and important in managing organizational knowledge. This motivates a
closer inspection of the degree of usability of various types of KMS. This paper
is an analysis of KMS from a philosophical angle: with the help of veritistic social epistemology we analyze which KMS are likely to be used more in comparison to others. Veritistic social epistemology is oriented towards truth determination; it seeks to evaluate actual and prospective multi-person practices in
terms of their tendency to produce true beliefs (versus false beliefs or no belief)
in their users. We distinguish between KMS that manage structured knowledge
and those that manage unstructured knowledge. It is argued that structured
knowledge is more credible to the users than unstructured knowledge and that,
because of this, KMS that manage structured knowledge bring more veritistic
gains than those that manage unstructured knowledge.

Introduction

The aim of this paper is to explore the question, Which knowledge management systems (KMS) bring more veritistic gains to users in comparison to others? Veritistic
social epistemology seeks to evaluate actual and prospective practices in terms of how
well they tend to promote the acquisition of true beliefs (versus false beliefs or no belief) on the part of their users. The practice of Knowledge Management (KM) is realized by using IT products or information systems (IS) called KMS. This paper analyzes two types of KMS and finds which type is veritistically superior. The paper is
organized as follows. The background of the paper is set out in the next two sections,
on knowledge in organizations and veritistic social epistemology, respectively. Veritistic analyses of KMS, with the interpretation of the findings are done next followed
by the conclusion.

Knowledge in Organizations

A commonly held view among IS researchers is that data is raw numbers and facts,
information is processed data, and knowledge is authenticated information [1]. Another common, perhaps complimentary, way of thinking about knowledge in an organizational context, is as information in action [cf.2]. Nonaka [2] discusses two
types of knowledge in organizations: tacit and explicit. Tacit knowledge is rooted in
action, experience and involvement in specific context. An example of tacit knowledge is the artful oral skills of an experienced sales-person selling a not so useful
product. Explicit knowledge is articulated, codified, and communicated in symbolic
form and/or natural language. An example is a product manual that accompanies a
product, describing what it does and how it works. (The distinction between explicit
and tacit knowledge corresponds to the distinction between propositional and procedural knowledge, respectively.) The tacit form of organizational knowledge is related to the production of knowledge by social means. According to Stahl [3], individuals generate personal beliefs from their own perspectives, but they do so on the
basis of socio-cultural knowledge, shared language and external representations.
Thus beliefs come to be accepted as knowledge through social interaction, communication, discussion, clarification, and negotiation. So understood, knowledge is considered as the product of social processes.

2.1

Knowledge Management Systems

Organizations are collections of humans, their skills and resources. The resources
consist of human skills (e.g. expertise, experience), physical resources (e.g. building,
machinery) and vast collections of information and knowledge. Knowledge in this
context includes both the experience and understanding of the people in the organization and the information artifacts, such as documents and reports, available within the
organization and in the world outside [4]. Organizational knowledge is difficult to
manage, as the volume of knowledge (both tacit and explicit) increases day by day
and tacit knowledge embedded in human minds disappears from the organization
when employees leave. In order to capitalize on organizational knowledge, organizations must create an environment where knowledge is captured, shared and transferred effectively and efficiently. KM helps to capture, share, and transfer knowledge
and thus manage organizational knowledge. The practice of effective KM typically
requires an appropriate combination of organizational, social, and managerial initiatives, along with the deployment of appropriate technology. Technology can help to
capture explicit knowledge (e.g. in databases), identify sources of tacit knowledge
(e.g. using extranet applications), share tacit and explicit knowledge among users (e.g.
using groupware) and transfer knowledge (e.g. preparing best practices documents). A
KM system is a specific type of IS or IT product, applied to managing organizational
knowledge [1].

2.2

Categories of KMS: Structured and Unstructured Knowledge

Hahn and Subramani [5] propose a framework for classifying KMS, based on where
the knowledge (to be managed) resides and the extent to which it is structured. They
distinguish between KMS in which knowledge is structured, and those in which it is
unstructured. Structured knowledge either has an inherent structure (as does, e.g.,
an electronic database) or the structure is imposed upon it (as, for example, when
documents are tagged with keywords) [5]. Structured knowledge is generally stored in
the organization such as in the corporate websites. Unstructured knowledge has no
prior structure and is generally dynamic in nature for example information in an electronic discussion forum. It is difficult to impose structure on unstructured knowledge.
For example, tagging text obtained from an electronic discussion forum is difficult as
the content is highly context dependent. Organizational knowledge that is structured
can essentially be coded and thus classified as explicit knowledge. Tacit knowledge,
which is generally unstructured (such as expertise or experience), is created over a period of time in organizations.
KMS have been developed that manage both structured and unstructured knowledge. Some KMS use classification mechanisms, tags or meta tags to structure
knowledge and then manage them. An example of a KM system managing unstructured knowledge is the collaborative filtering systems. These systems predict browse
and search behaviors by analyzing past behavior of other users when they performed a
similar activity [5].

Veritistic Social Epistemology

Epistemology is the branch of philosophy that deals with the study of knowledge.
Traditional or classical epistemology is concerned with the pursuit of truth, typically
on the part of individuals considered in isolation from other agents or any broader social setting. Social epistemology focuses on the social dimensions of knowledge
and/or knowledge-production; it acknowledges, where traditional epistemology
largely ignores, the important role that social factors play in the knowledge-forming
process. This is not to say, however, that social epistemology must jettison the traditional epistemological concern with truth (true belief): veritistic social epistemology
(VSE). VSE is concerned with the role of social factors in the production of knowledge, where knowledge is understood in the weak sense of true belief [6] (p. 5).
VSE is intended to be evaluative or normative rather than purely descriptive or explanatory: the task of the theorist is to identify and evaluate actual and potential social
processes/activities/institutions in terms of their tendency to promote the acquisition
of true belief (versus false belief or no belief) in their users [7].
The motivation for adopting a veritistic approach is straightforward: both for
practical reasons and because they are spontaneously curious, humans across cultures
and throughout history commonly seek the truth. Moreover, epistemic notions such as
knowledge are properly conceived in terms of truth: when we discover that a belief is
false, we cease to consider it to be a candidate for knowledge -- something one might
know -- in any non-figurative sense. From this perspective, when knowledge is

used to refer to what is accepted within an organization, this must be taken with a
grain of salt: if what is accepted turns out not to be true, then it is not knowledge,
properly so-called, even though users within the organization might continue to consider it to be such. Hence, understanding and evaluating the role of inter-personal, social factors in knowledge acquisition/production requires that we look at their tendency to promote true belief. In a word, institutions and practices that foster true
belief are epistemically good and should be promoted; institutions and practices that
result in false belief (error) or the absence of true belief (ignorance) are epistemically
bad, and should be avoided or corrected. (This is assuming that there are not overriding, non-epistemic reasons which speak against a given veritistically good practice
or institution, or in favor of some practice/institution which, while less good epistemically, is judged to be better overall.)
The main question for VSE is thus, which practices have a comparatively favorable impact on knowledge as contrasted with error and ignorance? [6] (p.5). Once
again, the rationale for taking up a veritistic perspective is that in everyday life a certain value is placed on having true beliefs rather than false beliefs or no opinion. This
type of value is veritistic value (or V-value). In order to understand the concept of
V-value we cite an example from Goldman [6]. Suppose that a person S has an interest in a yes/no question as: Is it the case that P has occurred? V-values can be assigned for three possible states. If S believes that the proposition is true then the Vvalue is 1.0, if he rejects the true proposition P then the V-value is 0, and if he withholds the judgment then the V-value is 0.5. The first case constitutes knowledge, the
second error and the third ignorance respectively. Veritistic analysis focuses on
change of V-value over time. Over a period of time if a person changes his state of
belief from no-opinion to rejecting P, then the V-value either improves or worsens
depending on whether P is true or false.
A high V-value of beliefs indicates an increase in the level of knowledge of the
user. If the V-value increases of an entire community then the knowledge of the entire
community increases. The increase in V-value can yield veritistic profit to the users
whose beliefs have been modified. If a user moves from false belief to true belief by
receiving correct information, and the receiver is able to draw true conclusions from
the information, then there is an increase in veritistic profit of the user. Finally, it is
important to note that veritistic analysis is always to be assessed relative to the questions of interests. Though they may be true, answers to questions that are of absolutely no interest to the user do not qualify as properly having V-value.
Knowledge in organizations is formed, shared and practiced by processes that
are essentially social. The practice of managing knowledge in organizations is done
by KM, and this practice is realized by KMS. KMS help to bring changes to individual beliefs by allowing individuals to use KMS. Any tool such as a KM system will
have an impact on modifying the beliefs of users, resulting in change of V-value.
However, we predict that not all KMS will have an equal impact on modifying users
belief. This is because different KMS manage different types of knowledge (structured and unstructured). Users might be influenced by the credibility of these knowledge and modify their beliefs accordingly. In the next section we analyze the credibility of these different types of knowledge, and their consequent impact on change in
V-value.

Veritistic Analysis of KMS

4.1

Credibility of Knowledge Sources

We make a claim that: structured knowledge is more credible to the users than unstructured knowledge. This claim is based on analyzing a similar proposition of
credibility between structured and unstructured information. Information is normally
seen as that which has meaning, in that it reduces uncertainty for the seeker [8]. Information can, however, increase uncertainty and create more dissonance [9]. Information can also be viewed as something that describes fact. Hicks et al. [10] mention
that individuals exposed to structured information may infer the same knowledge
from it and majority of unstructured information is either personal or developed
through interaction between two or more individuals. For example, an engineering
drawing is a structured piece of information containing text, numbers and symbols
and drawn for a specific context. This information can also be evaluated for its accuracy (thus measurable). Different engineers will infer similar knowledge by studying
the drawing. The primary differences between structured and unstructured information are outlined in Table 1.
Table 1. Information category [10]

Category
Structured

Unstructured

Characteristics

Textual (e.g. numeric, alphabetic) and pictorial (visual image)

Generally context dependent
Consistent by producing similar knowledge from structured information
Textual (e.g. personal note), verbal (conversation) and memory
Generally context independent
Inconsistent as individuals infer different knowledge from unstructured
information

Trust is an important factor in belief formation. McDowell [11] points out that trust
can have a crucial epistemic impact on social epistemology, which assesses the epistemic value of social practices. Foley [12] points out that our most fundamental assumptions from where opinions are formed and not self-generated. They are
passed to us as part of our intellectual inheritance. For example, we do not verify the
assumptions mentioned in the elementary science text books, as (we think) the assumptions are already verified by some experts and these assumptions often become
the basis of our or others beliefs. Thus, formation of our fundamental beliefs depends
on the trust we place in their sources. Hardwig [13] (p. 694) mentions that trust is often epistemologically even more basic than empirical data or logical arguments: the
data and the arguments are available only through trust.
Trust and credibility are closely related. Credibility is defined as "believability"
and trust is regarded as "a positive belief about the perceived reliability of, dependability of, and confidence in a person, object, or process" and in this sense trustworthiness of information is a synonym for credibility [14]. Credible sources are described as trustworthy and having expertise [15]. McDowell [11] points out that
an epistemic effect of social trust is that people can be less willing or likely either to

speak or to listen to socially untrustworthy sources of information, and more willing

or likely either to speak or to listen to socially trustworthy ones. Self presents a summary of early Socratic and Aristotelian ideas about credibility [15]: First, sources are
credible because their messages rightness is perceived by the audience. Second,
sources are credible because they rightly read how to reveal themselves to particular
audiences. And, third, sources are perceived to be credible because of audience characteristics. (p.423) In most cases a messages rightness is more easily perceived by
the users using structured information. Users are able to rightly deduce the meaning
of the structured information more consistently than unstructured information. For example, studying an engineering drawing of an engine (structured information) versus
listening to the functioning of an engine (unstructured information), different users
would be able to interpret the drawing correctly by studying it, but listening to the
functioning may result in different interpretations of the message as the message is
unstructured and its context is unknown to the users. Unstructured information is
dynamic (changes whenever new content is added) and therefore more prone to misinterpretation and error. Thus, structured information is more credible than unstructured information.
We extend the same logic to argue that structured knowledge is more credible to
the users than unstructured knowledge. The differences in structured and unstructured
information (Table 1) are applicable to the structured and unstructured knowledge as
well. We make an assumption in this claim that the sources of knowledge in both
structured and unstructured are equally credible, as long as their sources are within
the organization. We do not consider sources of knowledge that are not produced or
managed within the organization, such as documents obtained from the internet. In
other words, we are treating the credibility of various knowledge sources as the same
when that source is an individual or artifact within the organization.
Other than the differences mentioned in Table 1, we argue that the credibility of
unstructured knowledge is less than that of structured knowledge for two reasons.
First, unstructured knowledge is often possessed by single individuals in the organizations such as experts. For unstructured knowledge the users have to first find the
sources of the unstructured knowledge before accessing it; second, the transfer process of unstructured knowledge is often difficult because of its tacit nature. Unlike accessing structured knowledge by search and retrieval mechanisms, there are no formal
methods established to access unstructured knowledge. The structured knowledge can
be categorized and stored in the organization and therefore easily accessible to the users. But once the unstructured knowledge is converted to the structured knowledge,
the credibility of the structured knowledge increases to the users as they can now access and share the knowledge. For example, experience of an expert can be converted
to explicit knowledge such as best practices, case studies or stories. This knowledge
becomes more credible to the users as it is now available in explicit form and could be
easily accessible. The task of KMS that manage unstructured knowledge resources is
more difficult as the knowledge cannot be accessed easily. From the above arguments
we therefore claim that in most of the cases structured knowledge is more credible
than unstructured knowledge. (It is to be noted that the present discussion is on degree
of credibility that some piece/source of knowledge is less credible than another does
not mean that it will not be believed, much less that it will be as false; rather, it will be
less likely to lead to belief on the part of the user.)

4.2

Veritistic gains of KMS

We claim that KMS that manage structured knowledge bring more veritistic gains to
users than those that manage unstructured knowledge. KMS make available knowledge to the users, which otherwise might have been difficult to obtain. While accessing this knowledge, an individual increases his/her knowledge. Goldman (1999) suggests that the higher an individuals degree of belief in a true proposition, the more
knowledge this individual possesses. In other words, an individual acquires knowledge if this individual increases his or her degree of belief in true proposition. When a
user uses a KM system, he/she uses the acquired knowledge to do certain tasks that
can be stated as propositions. For example, a task could be how to evaluate salesmen
to distribute incentives? A KM system can help in defining evaluation criteria for incentive distribution based on the past information. Users belief for doing the task will
change when he/she obtains the evaluation criteria from the KM system. We therefore
view KMS as tools that modify users belief in doing tasks. If using KMS can increase the state of belief of users from (a) false belief to true belief, (b) false belief to
partially-true belief, or (c) partially-true belief to true belief then we can claim that
KMS bring veritistic profits to the users.
Goldman mentions two cases where communication can yield veritistic profits.
First, when a communicated message contains a direct answer to some question that
interests the receiver. And second, when a communicated message does not contain
a direct answer to a question but contains a report of some evidence that the receiver
uses to answer her question [6] (p. 164). The second case applies more for KMS usage. The users use the knowledge that the KMS manage as evidence in doing tasks.
We previously analyzed that structured knowledge is more credible than unstructured knowledge, therefore beliefs of users would be changed more positively (false
to true, false to partially true or partially-true to true) using structured knowledge than
using unstructured knowledge. When KMS manage structured knowledge, the change
in users belief from false to true, false to partially true or partially-true to true would
likely be high and therefore users will be able to draw accurate conclusions using this
type of KMS. Likelihood of the veritistic gains of the users will be high in this case.
The situation will be reversed when users use KMS that manage unstructured knowledge where likelihood of veritistic gain is low. This finding can also be generalized to
the practice of KM in organization. If on average (average value of individual Vvalue) the practice of KM increases the V-value of the users belief status, then the
over all practice of KM in the organization will bring veritistic profits to the users.
The above analysis is based on likelihood of veritistic gain instead of simply
veritistic gain because there may be some beliefs that are irrelevant or unimportant
to the users. These beliefs, though few in numbers would be viewed as uninteresting
and therefore would not bring veritistic gains to the users.
We made two assumptions in this analysis. First, users need to have genuine interests on propositions or artifacts to be fit for veritistic analysis. KMS that are used in
organizations help the users to take decisions in the organizations and therefore perceived to be useful. Therefore knowledge managed by KMS is of interest to the users.
Second, it is assumed that the technologies used in the KMS are most appropriate and
correctly chosen. The degree of trust users place in the KMS also depends on the
technologies used in the KMS and we assume that users trust the technologies.

Conclusion

The change in beliefs of the users to make certain decisions will depend on the KMS
that they use. KMS manage knowledge, and users use this knowledge to perform certain tasks within the organizations. We have argued that KMS that manage structured
organizational knowledge bring more veritistic gains to their users than KMS that
manage unstructured knowledge. This veritistic analysis of KMS can help users to
identify KMS that they are most likely to use. Users would like to use KMS that bring
more veritistic gains to them than others. In other words, users tend to use KMS that
are seen as increasing their stock of true belief, and they tend not to use KMS which
they regard as not having such veritistic benefits. Our veritistic analysis of KMS usage is consistent with the current design of KMS used in organizations as Marwick
[4] (p. 814) points out, the strongest contribution to current knowledge management
solutions is made by technologies that deal largely with explicit knowledge, such as
search and classification. Contributions to the formation and communication of tacit
knowledge, and support for making it explicit, are currently weaker.

References
1. Alavi, M. and D. Leidner: Knowledge Management and Knowledge Management Systems:
Conceptual Foundations and Research Issues, MIS Quarterly, Vol. 25, no 1, (2001)
2. Nonaka, I.: A Dynamic Theory of Organizational Knowledge Creation, Organizational Science, Vol. 5, no 1 (1994) 14-37
3. Stahl, G.: A Model of Collaborative Knowledge-Building, in Fourth International Conference of the Learning Sciences, Mahwah, NJ: Erlbaum (2000)
4. Marwick, D.: Knowledge Management Technologies, IBM Systems Journal, Vol. 40, no 4,
(2001) 814-830
5. Hahn, J. and M.R. Subramani: A Framework of Knowledge Management Systems: Issues
and Challenges for Theory and Practice, in 21st International Conference on Information
Systems, Brisbane (2000)
6. Goldman, A.: Knowledge in a Social World, Oxford University Press, New York (1999)
7. Goldman, A.: Social Epistemology, The Stanford Encyclopedia of Philosophy, ed. E.N.
Zalta, plato.stanford.edu/archives/spr2001/entries/epistemology-social/ (2001)
8. Belkin, N.J.: Information Concepts for Information Science, Journal of Documentation, Vol.
34 (1978) 55-85
9. Rogers, E.M.: Diffusion of Innovation. 4th edition, The Free Press, New York (1995)
10.Hicks, B.J., Culley S.J., Allen R.D., Mullineux G.: A Framework for The Requirements of
Capturing, Storing and Reusing Information and Knowledge in Engineering Design, International Journal of Information Management, Vol. 22, (2002) 263 -280
11.McDowell, A.: Trust and Information- The Role of Trust in The Social Epistemology of Information Science, Social Epistemology, Vol. 16, no 1 (2002)
12.Foley, R.: Intellectual Trust in Oneself and Others, Cambridge University Press (2001)
13.Hardwig, J.: The Role of Trust in Knowledge, The Journal of Philosophy, Vol. 88, no 12
(1991) 693-708
14.Fogg, B.J. and H. Tseng: The Elements of Computer Credibility in Conference on Human
Factors and Computing Systems, Pittsburgh, PA (1999)
15.Self, C.: Credibility. An Integrated Approach to Communication Theory and Research, ed.
M. Salwen and D. Stacks, Mahwah, N.J.: Erlbaum (1996) 421-441