I probably sound like a broken record, but the fundamental keys to success for a knowledge repository are collecting knowledge and finding knowledge when needed. And the fundamental obstacle to success is changing employee behavior. I think the answer lies in understanding what documents are important.

Important to the employees that is, not what management thinks is important. That means empirical measurement of how documents are used, which is where the knowledge weblog comes in. The k-log builds a web of references which allow us to evaluate what information is important and what it's really about. And that gives us leverage to change employee behavior.

I have to admit that I belong to the "As long as I pay your salary, I make the rules" school, but that only applies when choosing between roughly comparable options. To establish new behavior, it's necessary to frame it as a true win-win. The k-log web of references allows us to evaluate employees as content authors and content synthesizers. And if it can be evaluated, then it can be rewarded.

And behavior can be changed by framing the desirable behavior as being valued by the company and as furthering one's career.