Michael Daconta | A heretical view of knowledge management

For a long time, I have felt that there are multiple, significant problems in how the information technology industry has branded and sold knowledge management.

First, given that we barely know how to reliably produce information, I felt it was arrogant to assume we knew how to reliably produce knowledge. Second, knowledge representation as a field has been going on since Aristotle and is still undergoing widespread change and innovation (as described in my April 6 column). Thus, for industry to claim to have products to solve that problem seems a bit disingenuous.

Finally, what the knowledge management industry has latched onto — probably because of its limited success in this area — are collaborative applications such as Web portals, blogging, discussion groups and wikis as the mainstays of knowledge management. The idea is that logging and sharing experience, in any medium, equate to capturing and managing the knowledge of your organization. But although informal discussions can contain knowledge, it is equally likely that they do not — and there’s the rub.

In my opinion, it’s best to think of raw content as content that can potentially link to knowledge but only after you understand and define the knowledge you want such instances to match. A good example is the distinction between an unmoderated discussion forum and a list of answers to frequently asked questions. FAQs typically consist of previously collected, unmoderated discussions or e-mail traffic arranged into something useful. They usually follow a pattern whereby multiple questions about a single issue are combined and then answered. There is a mapping process between problems and symptoms.

Creating FAQs requires a type of knowledge management because the creator is exercising superior knowledge of the underlying process or product to explain the root cause of specific symptoms, how those symptoms map to a root cause and then provide the solution — if known — to the problem.

However, even FAQs, because they are based on sometimes ambiguous natural language, do not reach the ideal of knowledge management, nor do they reach the ideal for understanding the crux of the knowledge being shared. On any particular topic, there is a subset of knowledge that forms a framework upon which all the other topical facts and information rest.

It is that underlying framework that we wish to unearth when questioning experts and those deep knowledge secrets that are the gold nuggets of knowledge management. Getting to those knowledge nuggets has almost nothing in common with capturing blogs, wikis or discussion threads. And therein lies the core of my heresy.

In this new light, knowledge management is all about curation over content. Both are necessary, but the curator should exist before the collection and have an outcome-based process to perform the curation. For example, how do you ensure skills retention if you don’t know how to model skills and thus don’t formally know what skills the organization possesses and what skills are about to walk out the door? Should you differentiate mission-critical skills from others? Are there nuances and variations to your skill sets that distinguish higher-level skills from lower-level skills?

If your organization faces mass retirements in the coming decade, do you kick off your knowledge management program with a Web 2.0 blog or an agency wiki? If you are banking on more than dumb luck, you don’t use either. If you want to be effective, you first model what skills retention means to the organization and then methodically implement processes and tools to transparently link to the individual expression of those skills. In a nutshell, define curation first and then capture content. That is what I mean by curation over content.

Is that heresy? Let me know what you think via e-mail or at 1105 Government Information Group's upcoming Knowledge Management conference (tinyurl.comcw8ww9).

Reader Comments

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above