What do we do with knowledge?

Every now and then, I discover a new way in which my assumptions about things are challenged. Today’s challenge comes in part from the excellent commentary on my last post (which has been so popular that yesterday quickly became the busiest day ever here). I am used to discussions about the definition or usage of ‘knowledge management’, but I thought ‘knowledge sharing’ was less controversial. How wrong can one be?

The first challenge comes from Richard Veryard. His comment pointed to a more expansive blog post, “When does Communication count as Knowledge Sharing?” Richard is concerned that the baggage carried by the word ‘sharing’ can be counter-productive in the knowledge context.

In many contexts, the word “sharing” has become an annoying and patronizing synonym for “disclosure”. In nursery school we are encouraged to share the biscuits and the paints; in therapy groups we are encouraged to “share our pain”, and in the touchy-feely enterprise we are supposed to “share” our expertise by registering our knowledge on some stupid knowledge management system.

But it’s not sharing (defined by Wikipedia as “the joint use of a resource or space”). It’s just communication.

I agree that if people construe sharing as a one-way process, it is communication. (Or, more accurately, ‘telling’, since effective communication requires a listener to do more than hear what is said.) In a discussion in the comments to Richard’s post, Patrick Lambe defends his use of ‘sharing’ and Richard suggests that knowledge ‘transfer’ more accurately describes what is happening. I also commented on the post, along the following lines.

I can see a distinction between ‘sharing’ and ‘transfer’, which might be relevant. To talk of transferring knowledge suggest to me (a) that there is a knower and an inquirer and that those roles are rarely swapped, and (b) that there needs to be a knowledge object to be transferred. (As Richard puts it, “a stupid knowledge management system” is probably the receptacle for that object.)

As Patrick’s blog post and longer article make clear, the idea of the knowledge object is seriously flawed. Equally, the direction in which knowledge flows probably varies from time to time. For me, this fluidity (combined with the intangible nature of what is conveyed in these knowledge generation processes) makes me comfortable with the notion of ‘sharing’ (even given Richard’s playgroup example).

In fact, I might put it more strongly. The kind of sharing and complex knowledge generation that Patrick describes should be an organisational aspiration (not at all like ‘sharing pain’), while exchange or transfer of knowledge objects into a largely lifeless repository should be deprecated.

I think Richard’s response to that comment suggests that we are on the point of reaching agreement:

I am very happy with the notion of shared knowledge generation – for example, sitting down and sharing the analysis and interpretation of something or other. I am also happy with the idea of some collaborative process in which each participant contributes some knowledge – like everyone bringing some food to a shared picnic. But that’s not the prevailing use of the word “sharing” in the KM world.

This was a really interesting conversation, and I felt that between us we reached some kind of consensus — if what is happening with knowledge is genuinely collaborative, jointly creating an outcome that advances the organisation, then some kind of sharing must be going on. If not, we probably have some kind of unequal transfer: producing little of lasting value.

Coincidentally, I was pointed to a really interesting discussion on LinkedIn today. (Generally, I have been deeply unimpressed with LinkedIn discussions, so this was a bit of a surprise.) The question at the start of the discussion was “If the term “KM” could get a do-over what would you call the discipline?” There are currently 218 responses, some of which range into other interesting areas. One of those areas was an exchange between Nick Milton and John Tropea.

Nick responded to another participant who mentioned that her organisation had started talking about ‘knowledge sharing’ rather than ‘knowledge management’.

Many people do this, but I would just like to point out that there is a real risk here – that sharing (“push”) is done at the expense of seeking (“pull”). The risk is you create supply, with no demand.

See here for more detail: http://www.nickmilton.com/2009/03/knowledge-sharing-and-knowledge-seeking.html

The blog post at the end of that link is probably even more emphatic (I will come back to it later on). John had a different view:

Nick you say “sharing (“push”) is done at the expense of seeking (“pull”). The risk is you create supply, with no demand.”

This is true if sharing is based on conscription, or not within an ecosystem (sorry can’t think of a more appropriate word)…this is the non-interactive document-centric warehousing approach.

But what about blogging experiences and asking questions in a social network, this is more on demand rather than just-in-case…I think this has more of an equilibrium or yin and yang of share and seek.

People blog an experience as it happens which has good content recall, and has no agenda but just sharing the raw experience. Others may learn, converse, share context, etc…and unintentionally new information can be created. This is a knowledge creation system, it’s alive and is more effective than a supply-side approach of shelving information objects…and then saying we are doing KM…to me KM is in the interactions. We must create an online environment that mimics how we naturally behave offline, and I think social computing is close to this.

Nick’s response was interesting:

John – “But what about blogging experiences and asking questions in a social network, this is more on demand rather than just-in-case”

Asking questions in a network, yes (though if I were after business answers, i would ask in a business network rather than a social network). Thats a clear example of Pull.

Blogging, no, I have to disagree with you here. I am sorry – blogging is classic Push. Its classic “just in case” someone should want to read it. Nobody “demands” that you blog about something. You are not writing your blog because you know there is someone out there who is waiting to hear from you – you write your firstly blog for yourself, and secondly “just in case” others will be interested.

Blogging is supply-side, and it’s creating stuff to be stored. OK, it is stored somewhere it can be interacted with, and there is a motivation with blogging which is absent with (say) populating an Intranet, but it is stll classic supply-side Push. Also it is voluntary push. The people who blog (and I include myself in this) are the ones who want to be heard, and that’s not always the same as “the ones who need to be heard”. Knowledge often resides in the quietest people.

This exchange puts me in a quandary. I respect both Nick and John, but they appear to be at loggerheads here. Can they both be right? On the one hand, Nick’s characterisation of supply-side knowledge pushing as something to be avoided is, I think correct. However, as I have written before, in many organisations (such as law firms), it is not always possible to know what might be useful in the future. My experience with formal knowledge capture suggests that when they set out to think about it many people (and firms) actually rate the wrong things as important for the future. They tend to concentrate on things that are already being stored by other people (copies of journal articles or case reports), or things that are intimately linked to a context that is ephemeral. Often the information stored is fairly sketchy. One of the justifications for these failings is the the avoidance of ‘information overload’. This is the worst kind of just-in-case knowledge, as Nick puts it.

I think there is a difference though when one looks at social tools like blogging. As Nick and John probably agree, keeping a blog is an excellent tool for personal development. The question is whether it is more than that. I think it is. I don’t blog here, nor do I encourage the same kind of activity at work because someone might find the content useful in the future. I do it, and encourage it, because the activity itself is useful in this moment. It is neither just-in-case nor just-in-time: it just is.

In the last couple of paragraphs, I was pretty careless with my use of the words ‘information’ and ‘knowledge’. That was deliberate. The fact is that much of what we call KM is, in fact, merely manipulation of information. What social tools bring us (along with a more faceted view of their users) are really interesting ways of exposing people’s working processes. As we learnt from Nonaka all those years ago, there is little better for learning and development of knowledge than close observation of people at work. (Joining in is certainly better, but not always possible.) What we may not know is where those observations might lead, or when they might become useful. Which brings me to Nick’s blog post.

We hear a lot about “knowledge sharing”. Many of the knowledge management strategies I am asked to review, for example, talk about “creating a culture of knowledge sharing”.

I think this misses the point. As I said in my post about Push and Pull, there is no point in creating a culture of sharing, if you have no culture of re-use. Pull is a far more powerful driver for Knowledge Management than Push, and I would always look to create a culture of knowledge seeking before creating a culture of knowledge sharing.

Nick’s point about knowledge seeking is well made, and chimes with Patrick Lambe’s words that I quoted last time:

We do have an evolved mechanism for achieving such deep knowledge results: this is the performance you can expect from a well-networked person who can sustain relatively close relationships with friends, colleagues and peers, and can perform as well as request deep knowledge services of this kind.

Requesting, seeking, performing: all these are aspects of sharing. Like Richard Veryard’s “traditional KM” Nick characterises sharing as a one-way process, but that is not right — that is the way it has come to be interpreted. Sharing must be a two-way process: it needs someone to ask as well as someone who answers, and those roles might change from day to day. However, Nick’s point about re-use is a really interesting one.

I suggested above that some firms’ KM systems might contain material that was ultimately useless. More precisely, I think uselessness arises at the point where re-use becomes impossible because the material we need to use is more flawed than not. These flaws might arise because of the age of the material, combined with its precise linkage with a specific person, client, subject and so on. Lawyers understand this perfectly — it is the same process we use to decide whether a case is a useful precedent or not. Proximity in time, matter or context contributes significantly to this assessment. However, an old case on a very different question of law in a very different commercial context is not necessarily useless.

One of the areas of law I spent some time researching was the question of Crown privilege. A key case in that area involved the deportation of a Zairean national in 1990. In the arguments before the House of Lords, the law dating back to the English Civil War was challenged by reference to cases on subjects as varied as EC regulation of fisheries and potato marketing. That those cases might have been re-used in such a way could not have been predicted when they were decided or reported.

In many contexts, then, re-use is not as clear-cut an issue as it may appear at first. My suspicion is that organisations that rely especially highly on personal, unique, knowledge (or intellectual capital) should be a lot more relaxed about this than Nick suggests. His view may be more relevant in organisations where repetitive processes generate much more value.

On the just-in-case problem, I think social tools are significantly different from vast information repositories. As Clay Shirky has said, what we think is information overload is actually filter failure. Where we rely solely on controlled vocabularies and classification systems, our capability to filter and search effectively runs out much sooner than it does when we can add personalised tags, comments, trackbacks, knowledge about the author from other sources, and so on. Whereas repositories usually strip context from the information they contain, blogs and other social tools bring their context with them. And, crucially, that context keeps growing.

Which brings me, finally, back to my last post. One of the other trackbacks was from another blog asking the question “What is knowledge sharing?” It also picks up on Patrick’s article, and highlights the humanity of knowledge generation.

…we need to think laterally about what we consider to constitute knowledge sharing. This morning I met some friends in an art gallery and, over coffee, we swapped anecdotes, experiences, gripes, ideas and several instances of ‘did you hear about?’ or ‘have you seen?’… I’m not sure any of us would have described the encounter as knowledge exchange but I came away with answers to work-related questions, a personal introduction to a new contact and the germ of a new idea. The meet up was organised informally through several social networks.

The key thing in all of this, for me, is that whether we talk of knowledge sharing, transfer, or management, it only has value if it can result in action: new knowledge generation; new products; ideas; thoughts. But I think that action is more likely if we are open-minded about where it might arise. If we try and predict where it may be, and from which interactions it might come, I think it is most probable that no useful action and value will result in the long term.