In one of those internet coincidences, I have encountered (or re-encountered in some instances) a number of assertions today that we need to distinguish knowledge management and information management. Largely for my own benefit I have synthesised these in the following post.
I don’t agree that Information Management should be primarily backwards looking. The use of BI tools like Cognos et al are squarely IM but they are just as useful for forecasting as analysis. More generally, effective IM should always be done with a view to enabling KM process improvements.
I define the difference in this way: Knowledge Management is practised through activities that support better decision-making. IM is practised by improving the systems that store, capture, transmit etc information.
In this sense, a librarian neatly captures both sides of the coin. The act of building and making a library catalogue available is covered by IM. But the transaction by which a person can approach a librarian and leave with a relevant set of data to make a better decision is covered by KM.
Stephen’s post builds on a comment he made to a blog post of Nick Milton’s, in which Nick gives vent to a self-confessed rant:
If, as many people claim, Knowledge Management is “getting the right information to the right people at the right time” then what on earth do they think Information Management is?
Management of X is not concerned with delivery of Y.
Interestingly, although I have had similar experiences to Nick’s of people muddling knowledge and information, many of the links from the linked Google search use the quoted phrase to highlight the same error. One of the clearest of those rejections is that provided by Joe Firestone in one of a series of posts exploring US Governmental Knowledge Management.
If to do KM, we must understand problem seeking, recognition, and formulation, and knowledge production (problem solving), in order to know what is “knowledge,” and what is “just information,” then why not simply recognize that a First generation KM program based on “Getting the right knowledge . . . “ is not a clean alternative that allows one to forget about problems, problem solving, and innovation, but that since it also requires knowledge of these things, we may as well pursue a version of Second Generation KM that seeks to enhance not only “Getting the right knowledge . . . “, but also how we make that “right knowledge,” in the first place.
And as long as we’re at it, let’s also make that distinction between “doing” and “managing” that is at the very basis of the field of Management, and say KM is not primarily about Knowledge Managers “making knowledge” or “Getting the right knowledge to the right person at the right time,” but rather is primarily about enhancing the ways in which knowledge workers do these things. If we do that, we in KM won’t be stepping all over the turf of other managers, who, from a point of view distinguishing managing “knowledge processing,” from “doing knowledge processing,” are some of the primary knowledge workers part of whose job it is to actually make and integrate knowledge into organizations.
Independently, and most freshly, John Bordeaux has revisited an aspect of his critique of KM in the US Department of Defense. Specifically, what is the difference between Information Management and Knowledge Management. His answer:
The difference between IM and KM is the difference between a recipe and a chef, a map of London and a London cabbie, a book and its author. Information is in technology domain, and I include books (themselves a technology) in that description. Digitizing, subjecting to semantic analysis, etc., are things we do to information. It is folly to ever call it knowledge, because that is the domain of the brain. And knowledge is an emergent property of a decision maker – experiential, emotional framing of our mental patterns applied to circumstance and events. It propels us through decision and action, and is utterly individual, intimate and impossible to decompose because of the nature of cognitive processing. Of course, I speak here of individual knowledge.
John’s position is especially interesting for his assertion that knowledge is distinct from information in part because of its location. If I understand him correctly, once knowledge is captured, stored, or manipulated outside the brain, it ceases to be knowledge — it is information.
This makes sense to me, but it is at odds (I think) with Joe Firestone’s position, as expressed in a paper elsewhere: “My Road to Knowledge Management through Data Warehousing” (pdf).
[T]he desire to get beyond “arid IT-based” concerns and to take the human-side of decision support into account, is about a view of KM that sees knowledge as subjective and personal in character, largely “tacit” or “implicit”, and as distinct from codified expressions, which are really not knowledge, but only information. Knowledge is frequently viewed as “justified true belief” in this approach, a definition that has been the dominant one in philosophy since Plato, but which has been under vigorous attack since at least the 1930s. People who take this road to KM, view it as primarily an applied social science discipline, whose role is to “enable” better knowledge creation and sharing by facilitating the “conversion” of tacit and implicit knowledge to codified expressions.
The problem with this road to KM is that (a) in viewing knowledge as “justified true belief” it makes it dependent on the “knower” and therefore basically subjective. And (b) in restricting knowledge to beliefs in the mind, it neglects the role of management in providing a framework of rules and technology for testing and evaluating codified expressions or knowledge claims and thereby creating a basis for producing objective knowledge. In a number of other places, I’ve specified two types of knowledge found in organizations: surviving beliefs and surviving knowledge claims. In restricting attention to facilitating expressing surviving beliefs alone, this road to KM misses one of its major objectives: to enhance Knowledge Production and, in this way, indirectly improve the quality of surviving knowledge claims used in future decisions.
I am not sure that I understand Joe’s position completely, especially as his comprehension of the philosophical foundations far exceeds mine. However, the final sentence of the first paragraph above appears not to fit John Bordeaux’s position, although I think the first part of the paragraph does fit. I also struggle with the second paragraph. Even if one can separate knowledge from the ‘knower’, there remains the possibility that what is known depends on the context. As Nick Milton puts it in a comment on his original post:
I could give you a whole stack of information about the rocks below the North Sea – seismic sections, maps, core samples – but could you make an effective decision about where to site an oil well?
I think this comes to a practical problem. Capturing what is known in an objective sense would require a correlative capture of enough context to make it comprehensible by anyone at any point in the future. How much effort would that take, and at what point would it be more economical just to ask the relevant person (or even to start again from scratch)?