Posted by Mahesh Venugopalan on November 29, 2001 at 22:46:16:

In Reply to: Creating a content managment structure posted by Colleen Fedders on November 29, 2001 at 11:57:17:

Hi Colleen,
I can share with you some lessons learnt. Ours is an IT services company and we decided to adopt a purely democratic approach for identifying the content tags. We went to the users and asked them how they would like seeing the content relevant to their subject categorized. The mobile technologies group did the categorization for their domain, the insurance guys did it for their area and the CAD/CAM guys mapped the categories relevant to them. Based on the inputs we finally arrived at a 4-level tree structure comprising around 1500 content nodes.
Some examples: Technology>Databases>Oracle>Pro *C or
Application Domain>Insurance>General insurance>Claims
or
Methodology>Knowledge Management>Cultural Issues in Knowledge Management (stops at three levels)

Now, we feel that we have over engineered and have built a separate mini-search tool to retrieve these paths. The users can select appropriate paths to retrieve relevant content.

For instance, they can select paths Technologies >> Java >> J2EE/Swing and Domain >> Insurance >>General Insurance to retrieve documents that talk about building insurance applications using J2EE/Swing.

While this is a powerful tool, usage of the same is seen to be a bit tiring. The tagging of paths at the time of content submission has to be accurate and the inputs from the author of the content are critical at this stage.

There is a user group that feels that the stuff is over-engineered and the other group complains that 1500 nodes is only a fraction of what is required.

My advice would be to keep it simple.
Regards,

---

Follow Ups:

---

Click Here to Post Follow Up in New Forums

    Knowledge Management Think Tank (New)

  

 Subject: 

Message: