Thanks to David I had the opportunity to participate in this forum yesterday. One point that had been stressed by the presenters yesterday was that knowledge management is about whom you know and not what's in your database. The real life examples cited are well convincing. This is an experience knowledge management pundits learned the hard way in the last two decades.

But I wonder if this is still true. Consider my examples: I have a programming problem (say an SQL statement) and ask a programmer I know - what will he say Google it and voila there is an answer for almost any problem you may have. Just ask programmers you know - it is the way of choice to resolve 99% of their problems - nobody mails support it takes ages.
Consider basic health care - our German Weblog ("Der Gesundheits Weblog") gets swamped everyday with search requests for health problems people have or are interested in.
Consider legal problems - there are huge open databases and even more weblogs given you good advice for many basic and not too hard problems.

So in my perspective the social web is creating a "knowledge management" already for the humanity. It works because people use these knowledge management as part of their social interaction (email, forum, weblog etc.) and not as a separate closed database. If we extrapolate this into the future and taking the lifelog project (Search) for real we might create a data super store for social interaction.

As a conclusion for myself I see the key in delivering and using media that enables distribution, interaction, storability. Stored social interaction combined with search technology can provide very valuable information also without additional interaction with the producer of the information.

With Creative Weblogging we are working on giving great brains a chance to interact with their audience. With 21Publish we bring the social power of weblogs and their ease of use into an enterprise.