As noted above, KM has traditionally been about building and populating databases with useful content, creating portals -- generally, making more information readily available. The consequence has often been to drown workers in hard-to-find information of dubious value just in case they should find themselves in a position to use it. We have actually made workers' intellectual activities harder rather than easier, by presuming, top-down or back-office-to-front-lines, to understand what information they need, and how, when and why they need it. In a world where jobs are more and more specialized, and everyone's information needs are increasingly unique, it's not surprising that KM has failed to live up to its promise. How to Save the World: CONFESSIONS OF A CKO: WHAT I SHOULD HAVE DONE
There's something missing in here about serendipity -- sometimes "non-value added time" can be very hard to identify. But overall, this is a very good post from Dave Pollard.