Knowledge over time
Judith Meskilled came across an article in Fortune's Fast Forward colum in which David Kirkpatrick talks about knowledge management as something new to the enterprise content management market. Fortune.com - Fast Forward - Document Overload: Managing the Digital Paper Chase:
The latest hot area in ECM is what's called Knowledge Management— KM to those in the know, of course. What that's about is figuring out how or why something happened in an organization. Government investigators, for instance, may need to know exactly what decision-making process a company went through before launching a drug or building an aircraft part in a certain way. Says Jenkins: "This becomes, in a catch phrase our industry uses, the 'corporate memory.' Knowing how things were decided is absolutely critical." He describes knowledge management as "like a Google search that comes with a video of how the document you found was actually created."
While I don't think KM is new to ECM, this is an interesting perspective on the goals of KM. Rather than capturing what people know or recording everything that is known in the company, this idea focuses on capturing a trail through history within a company. Having come out of the pharma world, this aim is quite high on the radar from the aspects of intellectual property (where did an idea orginate) and for tracing the trail of information through long sequences of clinical and laboratory trials.
But having the capability to see through an organization and through its history is an important aspect of governing. The more a company can see these paths and learn from them the better. As the article suggests, not only does it require massive amounts of information, but also the ability to tie it together and make sense of the resulting knot.
I'm not sure enterprise content management, as I understand it, can achieve this aim.
Enterprise Content Management, well implemented, will involve capturing a great deal of information. Not just multiple versions of a given piece of content, but also who changed it, what was changed, when it was changed (including what was merged when multiple people are changing the same item on different tracks and then merge changes, and most importantly - ideally WHY it was changed as it was.)
This last piece is what is critical, and also what is very difficult, about good content management. The automated features of capturing when (and who) changed files can be handled fairly trivially by automated systems - that instead of replacing files, version control them in some manner and manage keeping track of what is the current version to be used). What is difficult is getting people to document WHY they make a change.
As well, ideally a tool will help track a set of related changes. Early change management systems would handle single files at a time - check it out, make a change, check it back in. However, many types of content today are not a single "file" but a series of related items - an HTML page can have dozens of files that make it up.
A separate, and very challenging aspect for many types of content management is to track changes that are not to a single file (word document, HTML file, part of the source code for an application etc) - but are rather changes within a database (or databases) - here it can be very, very difficult to build such systems to track changes over time - it requires a vastly more complex type of database technology than most companies and applications employ today.
Another related aspect of Enterprise Content Management is to keep track of the state of things when a change happens. Specifically for many types of content it is important to capture information about the state of systems (or people) when changes are made.
Consider in a non-technical capacity, it might be very valuable (and important) to keep track of what jobs people had when they made given changes - it might further be useful and important to know who worked in what office and who reported to whom. For that matter, who within a set of change reports worked for the firm and who worked for an outside consultant, or was a summer intern etc.
To bring intelectual property to this, you might then further want to associate people with specific employment contracts and terms underwhich they have (or have not) assigned intellectual property to the firm...
i.e. Enterprise Content Management's toughest challenges is tracking state and getting people to add to the information that is needed to after the fact understand the set of changes.
Shannon
For each