KM: Art or Science?

Have been watching the actKM forum build up some momentum on the issue of the "science of KM" - which has since moved on to the "traditional approaches of KM". Joe Firestone has been contributing quite a bit to the discussion. His contributions are always really well thought out and articulated, but for some reason, after a few paragraphs, my eyes cross, I lose track of the thinking, and end up having to read the whole thing 2 or 3 times and still feel I'm missing something. I'm as certain as I can be that the problem is mine, and I've been trying to figure out why. Its especially important to me since, in principle, I think I should agree wtih most of what he says - I'm always especially impressed with someone that works to get a PhD, and still asks people to call them by their name. It should be mandatory, and so anyone who does it by choice is someone I think I should listen to.

I think I finally cracked some of my resistance on Sunday morning in the midst of a codeine/ paracetemol/ pseudoephedrine-induced cognitive enhancement (the cold is winning, sadly). I've always found Joe's approach to KM as what I would call "mechanistic" - it looks at knowledge as a discrete, identifiable artefact or object. This artefact can be applied, enhanced, combined with others or otherwise manipulated to form a new object. This means that "knowledge" and its use should be something that can be experimented on - in its own right - as something that has demonstrable and repeatable properties, and the results of the experimentation subject to critical evaluation. My understanding is that this realtes to critical rationalism, but since reading stuff on it induces some kind fugue on me, I can't say for certain.

I call this approach mechanistic in that the idea Joe present's of "knowledge processing" seems to be a component in an overall system (which I guess to be the "Open Enterprise" he advocates, though I can't say for sure, because I can't read it, as much as I want to) that, as defined, is highly analogous to a mechanical system: apply "processed knowledge" to "decision" to produce "outcome X". And I think its this approach that shuts my brain down - I just don't see knowledge in that way. To my mind, it removes - or at least greatly reduces - the complexity of inspiration. Those small epiphanies that we experience when a series of thoughts brewing together in your mind suddenly congeal for no apparent reason into a totally new way of doing things.

One of my favourite epiphanies is one I heard being discussed on the radio once: the epiphany of predictive text. When it was first introduced, I hated predicative text - I could never get the right word, it suggested all sorts of rubbish, and it took far longer than just writing the thing out in more keypresses. Then one day, bang! It just all clicked, and now I curse anything that uses the keypad arrangement for text and doesn't predict anything. I like this epiphany, because a lot of people share it. And it symbolises to me that "knowledge creation" is an organic process. There was no new input, no new way of doing text, no new phone, it just all of a sudden, for no apparent reason, all clicked into place.

On a side note, that's my second most useful epiphany. My most useful epiphany, which may be more germane to this topic, is when, on the brink of failing an exam in software engineering, I suddenly "got" what the lecturer and tutor had been carrying on about for months. I remember the question was on creating a set of Java algorithms to manage dates, including accounting for leap years. That one question was worth nearly half of the total exam value, and for the first 2 hours of the 3 hour exam, it remained almost completely unanswered, because I had no idea how to do it. Once I'd bluffed my way through every other question, I had the option of hanging around for an hour randomly scribbling bits of incoherent answer, or leaving. I chose the former. I used up the next half-hour scribbling here and there on my notepaper, then suddenly, the answer appeared fully formed in my mind. In that last 25 minutes, I came closer to understanding the arcane art of software engineering than I have before or since, and wrote the perfect code for performing the required function.

Since then, I've failed 2 further attempts at more advanced software engineering, before dropping half my double-degree and deciding I'd much rather be "just" a librarian with a bit of tech knowledge. But the experience of having my mind fire up in a novel, unique, and unrepeatable way for just 30 minutes left me with the firm conviction that knowledge isn't something that can be managed in that mechanical input/output way. It defies that view in any way that makes working in KM meaningful.

Of course, this all means that my views on knowledge are difficult to pin down in a way that can be readily refuted or defended. Much like art, I "just know" what is "good" KM and "bad" KM. Which aside from some broad, fairly useless generalities around "sharepoint isn't good KM" but "sharepoint + people might be", doesn't help much in a conversation about KM initiatives and goals. Aside from being able to tell which environments I do want to work in vs which ones I don't. And that's damned useful to me.