The next major contribution to the KM 2.0 discussion comes in an interview of Dave Snowden by Jon Husband entitled “The Impact of Web 2.0 on Knowledge Work and Knowledge Management.” The interview was done on October 15, 2007, and then produced and distributed as a podcast.
This podcast was followed by many reactions in the blogosphere over a period of months. In general, the blogosphere comments, are laudatory and either summarize the podcast (See Luis Suarez’s contribution on November 23), or comment on it in a general way (See John Tropea’s on November 28). I may have missed something, but thus far, I haven’t seen very much specific value-added commentary on the views expressed by Dave in the podcast, anywhere. So, I’d like to offer an analysis of some of his statements in my next two blog posts.
1) Dave states that the social computing tools constituting Web 2.0 “effectively self-assemble, self-organise and deal with informal connectivity learning” in contrast to earlier tools which were about structuring data and content. It is correct, of course, to say that Web 2.0 tools are focused on “connectivity,” though I’m not sure what the phrase “connectivity learning” is meant to convey. It is also true that their use enables self-organization. However, I don’t think that what Dave means by the tools self-assembling and self-organizing is very clear. The people using the tools may self-organize through their use, but I don’t think the tools by themselves do any of the self-organizing.
Nevertheless, as Dave says, the important point here is that social computing tools are about people voluntarily linking to one another and creating newly self-organized networks. And it is clear, I think, that such tools present new possibilities for collaboration that do not exist without them. However, it’s also true that tools facilitating linking and more intense social interaction also create new possibilities for conflict that did not exist before. It is probably wise to keep this in mind when looking at Web 2.0 tools, as well.
2) Next, Dave emphasizes heavily the role of context in knowledge sharing and the idea that older tools assume “context independence.” He also, correctly points out that “context” is very important to get people to share their knowledge. He ties ‘context” very compellingly to knowledge when he says that “we only know what we know when we need to know it,” and when he also says that “human knowledge is the real-time assembly of multiple fragmented memories in a real-time context to create a new application.”
Dave’s written about “context” and its importance for some time now, in many different blog posts and publications. But I must confess that I’m still not clear about what he means by the term. Of course, I have my own view of context, and it’s represented graphically just below.
In that graphic, the active agent, the decision maker, has a context of action. You can see all the influences of context on the agent’s decisions. And you can see that the agent’s decisions feed into, and are part of the social network that is the CAS within which the agent acts, and that also is the primary context of its action.
Now, let’s assume that Dave’s view of “context” and my own are not that far apart. Then, when he says ““human knowledge is the real-time assembly of multiple fragmented memories in a real-time context to create a new application,” he probably means that the human as a system actively interprets the context by assembling memory fragments to construct knowledge in context, and if this is so, it is certainly very consistent with the graphic I’ve presented. But is that all that knowledge is? Is it only the situational orientation that our systems construct in the course of our interacting with context?
If so, then I think this view is too simple. Here, I’ve given a view of knowledge suggesting that there are three kinds of knowledge: biological knowledge, mental knowledge, and cultural knowledge. I know that Dave doesn’t accept my semantics here. But regardless of the semantics, situational orientation/knowledge that we construct by assembling memory fragments and other elements exists. Whether we include it in a category called ‘mental knowledge,” or use some other label is of little importance. Further, even if one doesn’t accept that cultural objects, and biological synaptic structures and DNA are forms of knowledge, the idea of “mental knowledge,“ raises some questions? For example, are our situational orientations, the result of our assembling memory fragments, all knowledge. Or are some aspects of them just belief, and not knowledge? Are emotional orientations knowledge? Are purposes knowledge? Are intentions knowledge? All of these questions are relevant because all of these seem, in some way, to follow on our assembling memory fragments in context.
Also, the idea that there is mental knowledge, also raises the question of whether there is more to such knowledge than just situational orientations or some aspect of them. In particular what about “tacit knowledge” or “implicit knowledge”? And what about “knowledge predispositions”? Are these just what we assemble from our memory fragments? How can this be, since predispositions are not assembled from memory fragments in context. Yet surely they exist, since more than a century of psychological research says that they do, and also, that they exist in addition to, and alongside of, our memory fragments. The graphic just below suggests that what we assemble in context from our memory fragments is only one aspect of our knowledge, only our “explicit knowledge” in the mind. It is that sort of knowledge of which we can say: “we only know what we know when we need to know it.”
3) Dave also said that the immediate context is very important to get people to share their knowledge. I think he is right about this, and that we can also see it if we reflect on the remarks I’ve just made about our respective views of knowledge. From my point of view, first, we clearly can’t share biological knowledge directly, whatever the context. Second, however, we can share our cultural knowledge, and we do so by creating it in our expressions of what we think, and by exposing those creations to others. But how do we do that? The answer is that we assemble our memory fragments in context, and then proceed to create our cultural knowledge by using what we’ve assembled in expressing ourselves. Third, can we share what we assemble from our memory fragments, i.e. our psychological situational orientations? The answer is no. These beliefs, whether knowledge or not, cannot be shared directly because we are not telepaths. They can only be expressed in cultural form, whether in language, graphics and art, music, and other cultural expressions, and then made available to others.
So, when Dave says the immediate context is very important to get people to share their knowledge, I think he is right, and he is right because people express themselves and create cultural knowledge in context. Without context, no cultural knowledge can be produced, and if none is produced then none can be shared.
But having said all this, it raises some important questions which I don’t think Dave has dealt with. Thus, fourth, if we look at the two figures again, we can see that all our knowledge is produced “in context,” and, not only that, but all our actions are performed in context as well. So, there is no way to produce knowledge outside of some immediate context or other, or for that matter to share it or refuse to share it outside of some context. Since there is no way to either make or share knowledge outside of some context, what does it mean to say that immediate context is necessary to get people to share their knowledge?
I think it means that different contexts produce different kinds of knowledge and varying predispositions to share it, and that certain situational contexts lead to producing and/or sharing cultural knowledge that isn’t very relevant to decision contexts, while certain other situational contexts are better for producing and/or sharing cultural knowledge that is more relevant to action contexts we are interested in.
For example, survey research contexts produce cultural expressions from people that may express aspects of their knowledge, but these aspects are not, in general, highly relevant to any problems businesses or organizations face, because the contexts of generation and sharing of survey research knowledge are not sufficiently similar to the contexts of everyday decision making or learning to make the knowledge shared in survey responses very useful. In addition, in survey research contexts, respondents don’t have very much control over what and how they will share. Questions are presented to them that assume a particular framework that they may not either share, understand, or approve of. Their response choices are often frequently restricted and are also in a framework that is not their own.
Further, when people in organizations worry about “capturing knowledge” of those who are retiring or leaving for parts known or unknown, what they want is to somehow transfer the very relevant “how to” knowledge of the retiring individuals to other individuals who aren’t retiring or leaving. However, this “how to” knowledge is, as implied by Dave’s expression, assembled by people as they perform tasks in context. Such knowledge can’t be expressed in language in a sufficiently detailed way for people to use the results of interviews or other structured language products elicited in artificial inquiry sessions to reproduce the same or functionally equivalent “how to” knowledge in themselves.
On the other hand, when people express themselves in blogs and wikis, in collaborative content tagging, in folksonomies, in digital videos and podcasts, in social bookmarking and in virtual environments, in problem solving communities and collaborative teams supported by web 2.0 software, and using mashups to move across a gamut of applications to express themselves, then the contexts of expression are either much more like natural work contexts or, alternatively are even work contexts that have been re-invented anew using the new tools. Also, what and how people will share in these contexts are more influenced by their own perceptions, understandings, frameworks, and choices. There is a certain freedom in their sharing activities that does not arise in survey or other structured inquiry contexts.
So, the argument in favor of Web 2.0 tools here, should not be that some (specifically Web 2.0) knowledge is generated and shared in contexts following the assembly of memory fragments, while other types of knowledge (associated with older tools) is shared successfully because it is out of context, but rather that the contexts associated with the use of Web 2.0 tools are much more closely associated with the creation and sharing of knowledge that is meaningful in problem solving and decision making contexts.
And another very important point about context, which is made explicitly by Dave in the interview, is that older tools, in the way they produce data and content, seem to assume, erroneously, that the knowledge that is produced, and then shared, is context independent. So, when that data and content is stored, it is routinely stripped of the context in which it was generated, which is necessary both for its interpretation and understanding in knowledge sharing. In contrast, however, the knowledge produced and shared through using Web 2.0 tools is often shared along with much more of the context that generated it, then the “knowledge” shared using the older tools. Therefore, Web 2.0 shared knowledge is much easier to interpret and understand, than “knowledge” shared through the use of tools that share much less of the social and cultural context of knowledge production, and, as a result, knowledge sharing can be much more successful than it was before.
To be continued in Part 5