This post continues my analysis of Ray Sims’s presentation to the Boston KM Forum on April 9, 2008. In approaching “Knowledge Management,” Ray again asks what the term “means to me.” He begins by describing his own research on definitions of KM. His initial effort produced 43 separate definitions which he reported on in his blog. Aided by comments from others, his list soon grew to 60 definitions. Ray’s research provides a comprehensive picture of the diversity of conceptualizations of KM and their relationships, while explicitly rejecting the idea of arriving at a single synthetic definition.
In his presentation, Ray transitions from the diversity of KM definitions to three key definitional questions that are important to answer to understand the kind of KM one is doing. The questions are: what is the scale of KM? What is the domain of KM? And what are the metaphors we can use to understand or define KM?
Beginning with “scale,” I very much agree with the idea that focusing on scale is important. Ray distinguishes the individual level of analysis with its personal learning environments and personal KM, from the team, community, and project levels with their knowledge management and learning processes, and the organizational level with its learning and KM processes and activities. In my own work, I’ve made similar distinctions, specifically referring to levels of nesting of knowledge processes in organizations, and to KM targeting based on both knowledge sub-processes and agents (individuals, teams and projects, friendship and other self-organizing groups, organizational sub-divisions, committees of “experts,” communities of practice, and the organizational level). I think distinctions among agent levels are rather fundamental to KM since the complexity and expense of interventions tends to be correlated with the level of targeted “agents.”
Ray’s next definitional question relates to the “domain” of KM. Here Ray begins by asking the question, “where does KM start or stop?” His slide on this presents a number of categories in white, gray and blue ellipses. The slide is effective in displaying the broad range and diversity of categories associated with KM in one way or another, and therefore in highlighting the boundary or scope problem, and the fuzziness of its boundaries. In fact, it may be the best slide I’ve seen yet on this problem. Ray’s treatment continues with some of the results of Stephen Bounds’s evaluation of Ray’s survey of definitions of KM. Stephen performed his evaluation when Ray’s list had grown to 53 definitions. Here is Stephen’s analysis (also quoted by Ray) of the attributes identified by the definitions.
KM Activities
— Knowledge Distribution (26)?
— Knowledge Creation (21)?
— Learning (14)?
— Knowledge Classification (9)?
— Collaboration (7)?
— Knowledge Harvesting (6)
— Knowledge Discovery (2)
KM Outcomes
— Improved Execution (28)?
— Value from Knowledge Assets (12)?
— Knowledge Codification (11)?
— Knowledge Retention (3)?
— Continuous Improvement (2)?
— Cultural Change (2)?
— Improved Knowledge Processing (1)?
— Improved Resource Assignment (1)?
In this analysis the numbers in the parens indicate the number of definitions out of the 53 that include the named attribute. Ray highlights “improved knowledge processing” in this slide, and in the next one discusses this definition as an “outlier,” though not pejoratively. Before I move on to that slide however, I’d like to point out a couple of things. First, while all of the “KM activities” are performed by knowledge managers in the course of their work, they are also performed by “knowledge workers” in general. That is, none of these activities are exclusively or primarily knowledge managerial in nature. Rather, it is more accurate to call them “knowledge processing activities.” Second, note that if we take all the “KM activities” in Stephen’s categorization and place the word “enhanced” in front of them, then they look just as much like KM Outcomes as the listed items under this last category.
These comments bring us to the point made by both Ray and Stephen that the definition of KM as activities and processes intended to enhance knowledge processing is unique in making a distinction between KM and knowledge processing activity. Of course, that definition, as indicated by both Ray and Stephen, is the one developed by Mark McElroy and I, over the years, since 1999. Ray proceeds to amplify the distinction by presenting our three-tier model which distinguishes among KM and its outcomes, knowledge processing and its outcomes, and business processing and its outcomes.
I think Ray and Stephen are right to recognize the difference between our definition and others, but I’m not sure they draw out the full implications of our definition. Specifically, if one acknowledges that all the “KM activities” identified in other definitions by Stephen, are really all knowledge processing activities, then it follows that our definition recognizes all of these activities as targets of Knowledge Management, or, if enhanced, as KM outcomes, rather than as KM activities. Further, only four of the KM outcomes listed under that category would be viewed as direct KM outcomes namely: enhanced “knowledge codification,” enhanced “knowledge retention,” “improved knowledge processing,” and “cultural change.” The other “KM Outcomes” are only indirect impacts of KM from the viewpoint of the three-tier model.
Why accept our definition of KM and the three-tier model, rather than some of the other definitions? I think the answer to that question is that the scope and boundary problems Ray highlights in his presentation are greatly alleviated by this definition, because KM starts and stops with a much narrower class of activities which are clearly management activities targeted at enhancing knowledge processing activities. With such a definition, KM efforts are much less likely to overlap with general management efforts, or with content management or customer relationship management.
The orderly development of KM as a discipline is much more likely if the field is defined in such a way that its activities minimize overlap and turf battles with other management disciplines. KM already has the difficulty that the distinction between it and information management is very hard to make. To define KM in such a way that the term means the broad range of activities we do when we are making and integrating knowledge (and in some definitions, even when we use knowledge, which is all of the time), is disciplinary suicide because it is a continuous source of conceptual confusion for the field.
Next, I’m not sure Ray’s use of the term “domain” in the above context is a good choice. Certainly discussion of KM activities and outcomes is important, but I think that his “scale” and “domain” categories are really about how to cut up the world in such a way as to distinguish the targets, or foci, of KM. I do think the term “domain” can be usefully applied to the description of targets. Specifically, I think the issue of business domains is an important one. KM needs to be focused on specific levels of interaction (agents), and on specific activities or processes, but, also, I think it needs to be focused in specific business domain areas since the quality of knowledge processing in organizations may often vary greatly across different business domains in the same organization.
Moving on, Ray’s next definitional question focuses on “metaphors.” He mentions “stocks” and “flows” as important metaphors in defining KM. “Stocks” are static and are associated with codifying, capturing, harvesting, and storing knowledge, while “flows” are “fluid” and “dynamic” and are associated with “conversations,” “fragments,” and “connections.” The idea that the “metaphors” one uses to talk about, describe, or analyze “knowledge” may have important effects in how one defines KM is an idea I agree with. In KM it is an idea that has been advanced by Daniel Andriessen who is referenced by Ray in his presentation.
However, though I agree with the general idea of the impact of metaphors about knowledge on our ideas about KM, I don’t see it as very different from the idea that the conceptual frameworks or “paradigms” we use to talk about, describe, or analyze “knowledge” will affect how we define KM, or how we view it for that matter. So, I’m afraid I don’t see what is novel or particularly important about the idea. After all, why shouldn’t our epistemology, i.e. our theory of knowledge, affect how we define KM?
In any event, Ray’s highlighting that many view “knowledge” as a stock, is surely correct. And a stock at any point in time is certainly static. However, I don’t see what’s static about activities such as “codifying,” “capturing,” “harvesting,” and “storing.” That is, it seems to me that such activities are always changing from moment to moment, and therefore are dynamic in character.
Next, is the metaphor of viewing knowledge as a flow, as dynamic and as associated with connections, conversations, and fragments. I must confess that I think the idea that “knowledge is a flow” and therefore is not a “thing,” is a misuse of ordinary language. It’s not that I doubt that knowledge claims can be expressed in messages and sent from person to person, or store to store, through connections, so that such claims can be said to “flow” through a network in the form of communications. Certainly this is a valid and sensible way to look at the exchanges between and among human beings, and the exchanges of bits and bytes that occur in organizations. However, I think that such flows of knowledge claims don’t in any way contradict the idea that what the knowledge claims assert is linguistic content with its logical and semantic content and that this content is comprised of abstract objects. On the contrary, the idea that “knowledge flows” from node-to-node in social networks, presupposes the prior existence of static knowledge claims that can be communicated through flows of various kinds.
Of course, the fact that knowledge claims occur in conversations, and may be expressed in fragments, and that both are exchanged through connections, doesn’t mean that the knowledge claims themselves are dynamic. In fact, the fragments expressing knowledge claims and connections through which they are communicated, are static at any point in time, while the conversations are dynamic and continuous even while the knowledge claims made in the course of such conversations express static content objects.
Ray does ask the question “why flows?” and lists the following points:
— Speed of change versus speed of codifying
— Continuous versus something that happens at the end of the project
— Small pieces loosely joined, context preserving
— Broader participation, with more connections
— Weak signals perception
— Results: innovation and better decision-making
He also quotes Dave Snowden to the effect that:
“If you ask someone, or a body for specific knowledge in the context of a real need it will never be refused. If you ask them to give you your knowledge on the basis that you may need it in the future, then you will never receive it.”
I’m afraid I don’t see the connection between these points and whether knowledge ought to be viewed as a flow. Again, my basic problem is as above. Just because we can make knowledge flow, doesn’t mean that “knowledge” is a flow. And, if we define it that way, it will lead to a definition of KM which is quite misguided, because that definition will overlook the idea that it is our knowledge processes that produce change in the things we call knowledge.
Ray takes the idea of “flow” further by using two of the metaphors analyzed by Daniel Andriessen, “knowledge as water,” and “knowledge as love.” He points out that in Andriessen’s panel study, “knowledge as water” was favored most as a metaphor for diagnosing the KM problems in a situation, while “knowledge as love” was favored most as a metaphor for developing KM solutions. And to illustrate this he quotes some diagnoses of problems and some solutions developed by Andriessen’s panelists. Without quoting the detail here, I’ll simply say that most of the diagnoses are not very relevant to common sense KM concerns, and even though the solutions are a little more relevant, one of them, namely “Don’t try to manage and systematise knowledge,” is in direct contradiction with the others since they are all clearly prescriptions for KM activity.
In general, while I liked Ray’s emphasis on scale, and domain, I thought his emphasis on “metaphor” in defining knowledge should have been replaced with a more formal consideration of alternative theories of knowledge and their implications for how KM is defined. In my view, these too, involve metaphor, but given their more explicit and elaborated nature, they are more subject to illuminating critical analysis and fair comparison in arriving at the theory of knowledge that will best complement one’s own view of KM.
Reviewing Ray’s presentation up to this point, we see that his treatment of Web 2.0 is vague in specifying actual Web 2.0 configurations and that his treatment of Knowledge Management emphasizes the diversity of conceptualization in the area, but lacks an assessment of the various definitions and a decision about which definition he will use in relating Web 2.0 and KM. Let’s keep this in mind as we review his treatment of how they are related.
To Be Continued