15 May 2003

Jim McGee has selected a number of recent posts focusing on weblogs as a tool for knowledge management. Dave Pollard's posts on weblogs in business were particularly interesting, including the weblog as filing cabinet and finding the right niche. In the latter, he identifies six knowledge problems faced by organisations with conventional KM systems: You're still hoarding: More knowledge is needed, can't be bought, and isn't being voluntarily contributed by the company's experts. It's out of date: A shortage of current, accurate knowledge is exposing the company to unacceptable risk. We don't know what this means: More background or context is needed to make the knowledge in the company's intranet useful. What have you shared this year?: A more formal process is needed to assess individuals' contribution to knowledge-sharing. We needed it over here: Employees aren't sharing what they know beyond their immediate business circle. We couldn't find it: Knowledge isn't getting to the people who urgently need it to make management decisions or succesful sales presentations. And then articulates the value proposition for weblogs in business and KM: They make contributing knowledge simpler, easier, and more automatic They make it easier to update knowledge on a timely basis They make knowledge more context rich They allow the authors of key business knowledge to build and retain 'pride of ownership' They make contributing knowledge more fun, since it becomes more like 'publishing' Each individual's 'collection' of shared knowledge is easy to define and assess at performance evaluation time They make knowledge easier to route, to 'subscribe' to, to canvass and to 'mine' Simple contribution is without doubt a key enabler for broad engagement with KM. I also increasingly think that "pride of ownership" is important. It's usually implemented by giving the author a byline in the corporate knowledge or content management system. But that's easily missed and the author's contribution becomes part of a larger, anonymous knowledge repository. A weblog on the other hand is clearly identifiable as the work of a specific author (or possibly a small group) and supports the development of an individual voice. And when you're talking in your own voice, you're more likely to convey the all-important context and nuances of knowledge in action. 11:54:39 PM    comment []

Developed by Eric Promislow, The Amazing Baconizer uses Amazon's "customers who bought this book also bought..." functionality to map a path between two books (or CDs or DVDs). Give it a couple of titles and it will show a trail of books linking the two. The further apart the topic of the books (ie the less likely that purchasers of each will have bought the other) the longer - and typically more interesting - the trail. As more organisations make structured information available through web services and other APIs, it will be fascinating to see what unexpected tools spring up to make connections that the originating organisations have never conceived. [Thanks to Jon Udell.] 7:23:30 PM    comment []

Jim McGee, writing about trust, security and organisation design, references a book by Bob Keidel called Seeing Organizational Patterns: A New Theory and Language of Organizational Design and in particular a tradeoff triangle balancing elements of control, autonomy and cooperation. "Typically we tend to think only in terms of the tradeoff between control and autonomy. His [Keidel's], richer, model introduces a third point of cooperation and suggests that organization design problems can be treated as looking for a spot somewhere inside the triangle instead of somewhere along one of its edges. The trend has been northward towards more recognition of cooperation and, hopefully, away from stale debates about control or autonomy." This looks like a useful diagram. It's similar to the features/cost/schedule tradeoff triangle that's widely used in web/software development and serves the same purpose: to help an organisation realise that focusing on one area has implications elsewhere and that balancing different priorities means understanding the consequences of the tradeoffs. 5:14:15 PM    comment []

Seb Pacquet points to Meatball Wiki, a resource about online culture and how people online come together in groups. Seb also points to Sunir Shah's ideas about soft security (also in more detail here). Soft security is not the same as weak security - it is a flexible, people-oriented approach to security that underlies the openness of wikis. I think it's worth quoting the basic principles at length because I see them as valid across a whole range of KM and community type applications. "It flows from a few basic principles. Assume good faith. People are almost always trying to be helpful; so, we trust everyone, confident that occasional bad will be overwhelmed by the good. Peer review. Users, rather than software or sysadmins, moderate each other. Forgive and forget. Even well intentioned people make mistakes. They don't need to be permanent. Harm reduction. When mistakes are made, minimize the damage. Fair process. The theory that being transparent and giving everyone a voice are essential management skills. (cf. Chan Kim and Renée Mauborgne, "Fair Process: Managing in the Knowledge Economy." Harvard Business Review January-February 1997 pp65-75.) These principles have inspired a number of solutions to common problems. Almost all solutions rely almost totally on people to do the right thing. Technology is only introduced to assist the people in their work. Because soft security relies entirely on social forces to maintain order, it remains not only adaptable to new threats, but tolerant in its responses. Conversely, encoded, programmatic ("hard") security is incapable of distinguishing attacks from mistakes, nor can it be argued with, nor can it be held accountable. Generally, soft security seeks to be humane and liberalist rather than impersonally technocratic. The most important examples: Audit trail. All changes are logged publically. Reversible change. All changes are reversible by anybody. Even reverted changes are themselves reversible. Based on these two alone, almost all problems can be solved. The audit trail allows peer review. The reversible changes gives people the confidence to do anything knowing they cannot do permanent harm, and that their mistakes can be forgotten. The public nature gives everyone the opportunity to have a voice, as well as enabling the Samaritans to squelch any abuse. Contrasted with normal Internet moderation, where either nothing is done or only some privileged person can unilaterally control the text, we have found that this system is much less vulnerable to malfeasance or human error." Although it's not listed in the basic principles, soft security clearly assumes that people are personally identified in the system (ie not only authenticated by the technology but also publicly identified to other users). This underlies both the individual responsibility and social pressures that soft security relies on. I strongly believe in the principles of soft security in KM systems: keep them open - give people the chance to contribute - let people define their own competencies and knowledge - allow them to choose and pace their own learning - maintain reputation by peer recognition rather than management fiat. But it can be a hard sell in some organisations, even where trust and sharing are proclaimed (but not necessarily enacted) as important values. This only reinforces the need for KM projects to be implemented in the context of a change programme that goes beyond technical solutions to address the more fundamental people and process issues. And it is often supported by starting small with low-profile projects that allow the value of openness and soft security to prove themselves without triggering individual and corporate paranoia. 3:39:13 PM    comment []

A useful article on knowledge continuity management by Anne Field on the HBS Working Knowledge site. "At its base, knowledge continuity management is about communication—about employees understanding just what it is that they know, what others need to know, and what information needs to be shared, and then methodically passing on that knowledge to peers. But while the idea itself is simple, setting up an effective knowledge continuity management program is a complex undertaking, involving a mix of technical, organizational, and management steps, and requiring a major commitment from the top." Getting commitment to the idea of preserving knowledge so that it's not lost when employee's move around or leave the company is critical. It has a longer-term payoff than some other knowledge management practices but you need to put it in place now to reap the benefits later - it's no good starting to worry about it just after your key knowledge resource has left. Like all knowledge management projects, it is inherently difficult due to the fluid, tacit and contextual nature of the knowledge you're trying to capture. "The more sophisticated and complex the knowledge a worker possesses, the more difficult it is to pass on—and the more crucial it is that it be passed on." Anne Field summarises some steps you can take to ensure that mission-critical expertise doesn't leave your company when workers do: Create a knowledge profile. Foster mentoring relationships. Encourage communities of practice. Ensure that passing knowledge on is rewarded. Protect people's privacy. Decide whether you're interested in recorded knowledge as well. Start small. Many of these apply to any KM project, of course. One of the key differences with knowledge continuity management is that you can't point people to an expert to assist them - once that expert has gone, there's no-one behind the profile to expand upon whatever limited knowledge is captured in the KM system. You need to act now to start transferring that expert's knowledge to others and to preserve whatever explicit knowledge you can for future reference. And don't overlook that a knowledge transfer programme, such as a mentoring relationship, has immediate as well as long-term benefits, often to the senior partner in the relationship as well as the junior. [Thanks to Column Two.] 12:05:31 PM    comment []