Towards a transparent and controlled CM Process

In principle create, manage and publish form a one-way content management chain, spreading content management through the entire organization. Sharing the effort was a good idea, leveraging knowledge already available in the organization.

However, many implementations used old fashioned IT techniques to connect the systems. Sharing a data model where each system built its own logic on top of the data model. And after a while each system became connected to all other systems. Thus paralyzing the entire chain for change. But change is vital to content management. Our customers demand more and our websites develop. More content was needed. Both more detailed content and richer content. More planning on content: what will actually be there? Will our website be just as attractive in a month as it is today?

So we needed agility. The rusty chain needed to become flexible. And we introduced two important concepts: de-normalization and demand-driven.

No more sharing of data models. This is content management, so we share content. Edsger W. Dijkstra put it like this: “Avoid duplication of volatile information”. And IT took that as a sign to never duplicate any information, but normalize everything. But what Dijkstra meant was do not manually duplicate information. If you can do it automated, then the restrictions are less.

Communicating entire sets of information together instead of a data model, gives you tremendous flexibility when it comes to accepting changes in your data model. Each system can be optimized for its own purpose: a translation memory system will need to maintain different relations than an authoring environment which on its own needs different logic then a publication system. Combining all of those in one system, or one data model is just not possible.

The second innovation in the new way of content management was the introduction of a demand. Our customers had demands, so our channel managers had demands. Since the internet to regular information was just an afterthought, channel managers were used to simply taking what was there. Information Architecture in combination with the notion of a live chain changed that. Information will ROT over time, become Redundant, Outdated and Trivial, an acronym that was introduced by Peter Morville and Lou Rosenfeld in the Polar Bear book. The need arose to plan content in advance.

So we introduced a demand into our chain and the catalog became the contract between demand (publication channel) and supply (author). A catalog has a start and an end date and between those dates the agreement is that content is a) available and b) constantly updated, and of course c) published.

Having catalogs in place you create a communication flow between channel manager and author. But we need more information flowing back…

This entry was posted on Monday, March 2nd, 2009 at 11:40 and is filed under content management, information quality. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.