Please indulge me momentarily and pardon this esoterica. By the time you finish reading this, I hope I will have shown the need for this “scenic detour.”
For a long time thinkers and practitioners in the area of knowledge management have made a distinction between knowledge that is tacit and knowledge that is explicit. Knowledge and expertise in your head that you use nearly effortlessly and sometimes unaware is called tacit. Knowledge that is written down so that others can understand it is called explicit knowledge. Tacit knowledge is hard to communicate. Explicit knowledge is easier to communicate. Thus we need to convert tacit knowledge to explicit knowledge.
I wonder if this distinction isn’t more of a mental shortcut than a meaningful way of thinking about knowledge. I also wonder if this popular division has handicapped IT people and the knowledge management industry. Because tacit knowledge is viewed as “hard to convert” into explicit forms, tacit knowledge has often been confused for unstructured knowledge. This leads to the following mental elision that direct person-to-person interactions are unstructured and hence an example of tacit knowledge transfer. In other words, conversations between people are a form of tacit knowledge.
To me, the notion that tacit knowledge resides in one’s mind is the safest, albeit minimalist, way to think about tacit knowledge. Many writers consider tacit knowledge as non-codified or unstructured. From here many make the mistake of assuming the reverse — that non-codified knowledge is therefore tacit. I believe tacit knowledge which resides in one’s head can vary in its codification from high to low. Just because the knowledge is in someone’s head doesn’t mean it’s unstructured.
The safe dualism of tacit/explicit knowledge has mutated into an illusory dualism of tacit-noncodified/explicit-codified knowledge and from here all hell breaks loose. In my research into visualization, I have had to foray into psychological research concerning neurological structures of the brain and more functional (not biological) descriptions of how the brain works. These research tracks are vitally important.
To help out with this discussion, I’m going to define three concepts borrowed from the knowledge management field I find useful for thinking about knowledge: diffusion, codification and abstraction.
- Abstraction — how generic, removed from concrete reality, or high level the information is.
- Codification — how much of the information is assigned known categories or aligned with structured metadata.
- Diffusion — how many people possess the information.
Highly codified and highly abstract information requires minds who share the codification scheme (what terms and concepts mean) and who have learned the abstraction hierarchies. For example, I can say the term “ISO OSI 7 layer model” which is a useful abstraction that requires one to know not only the terms and concepts within the seven layer model but the relationships between the terms and why they are grouped into one single concept. Much of IT knowledge is highly codified and abstract which is why geeks infuriate non-geeks in their use of language.
Knowledge management is still largely woeful in practice. The knowledge management researchers and practitioners need to take into account psychological theories about how we know things and consider the following:
- That knowledge inside our heads can vary in its structure from highly codified and abstract (and hence complex) to very uncodified and concrete (and hence simple). Knowledge is not a unitary concept.
- That all knowledge, regardless of its complexity, must pass through our limited working memory in order for it to be processed and made available in the future. The transfer of knowledge from medium to mind is not a “black box” process to be set aside because we think it is opaque but one that desperately needs our attention.
- That language is the primary means for conveying knowledge from mind to mind and that language itself, spoken or written, can vary in its level of codification and abstraction. Direct personal dialog is not entirely comprised of flows of unstructured knowledge. It all depends on who is doing the communicating and what they are communicating about.
- That people with superior knowledge are called experts and that in many critical problem domains experts process information better than non-experts, thus generating better decisions and judgments. Experts codify and abstract knowledge differently in their minds than non-experts. This different codification and abstraction takes years of effort to develop. Firms have to find ways of building expertise, not simply storing knowledge.
- That processing knowledge has two sets of costs: the costs of making knowledge explicit (transferring knowledge from mind to medium) and the costs of absorbing explicit knowledge (transferring knowledge from medium to mind). The former is often thought of as converting tacit knowledge into a document and the latter as individuals reading documents. One should think of it as the cost of communicating knowledge from one person’s head into linguistic codes (written or spoken) and from vice versa, absorbing knowledge embedded in linguistic codes into one’s head.
You might be thinking, “OK, nice theorizing. What is the significance for IT?” Here are some points to ponder.
- All knowledge in our heads needs to be communicated in linguistic form in emails, voice mails, face-to-face meetings, documents, video conferences, databases, etc. The level of codification and abstraction can vary significantly both within and between these media categories. When thinking about IT and knowledge, don’t assume that face-to-face dialog is unstructured or that documents are structured. Examine the specific task and communication needs in the organization. Knowledge in one form may not be easily substituted for knowledge in another form. The type of knowledge must be matched to the type of task.
- IT organizations tend to focus on the costs of abstracting, codifying and diffusing tacit knowledge, which is only half of the problem. Knowledge must be transferred into people’s brains for the knowledge to be useful. Barriers such as politics, hoarding for personal gain, improper incentives, inability to access experts, poor firm culture, insufficient individual mental capacity or time for absorbing complex information and poor teamwork all loom large in preventing firms from absorbing knowledge.
- Face-to-face interactions provide us a way to manage each other’s working memory when absorbing complex information, which is why face-to-face meetings are still critical despite the abundance of information and communication technology. In the middle of a dialog, people can react to each other’s gestures, emotions, body language and questions within seconds, which is critical for our working memory to absorb the knowledge. High-definition video conferencing may still not be a substitute. IT is not entirely up to the task of letting us human beings manage each other’s limited working memories.
- While the Internet and technology has drastically shrunk the cost of diffusion, it still costs money to codify and abstract information, as these require human experts to do the codifying or abstracting, or human experts to configure the IT system which does the codifying or abstracting. Since experts absorb highly codified and abstract information best and since experts are fewer in number and more likely to hoard knowledge, the best and most profitable knowledge has some of the highest production and absorption costs thus limiting its diffusion. IT is not poised to fully help, even with crowdsourcing, social tagging and other Web/Enterprise 2.0 techniques.
- Those who practice knowledge management should spend time to better understand current cognitive science and learning research. There is so much hand-waving, even by knowledge management experts, over the so-called “black box” for absorbing knowledge, that I would skeptically question all knowledge management initiatives.
- CIOs should not assume that a repository of documents provides a firm with any advantage. Even with Enterprise 2.0 technologies and techniques, investments in knowledge management can seriously miss the mark, mainly because firms often misunderstand the nature of knowledge and handle knowledge management in a crude, Neanderthal way. The Enterprise 2.0 wave is causing many to slide backward into optimistically delusional thinking.
Competitiveness of firms depends so much on knowledge. Here at the end of 2007, we have so much more to learn about knowledge that I seriously call into question whether any technology can have a significant and repeatable impact and whether there are any firms that can reliably manage their knowledge for advantage year over year.