Conservation of Complexity

An interesting debate between Mike Garrett (a member of the SAP practice at the Data Management Group) and Andy Hayler (founder and chief strategist of Kalido).

In an article on SAP Business Intelligence and the Conservation of Complexity, Mike advocates reducing the amount of complexity presented to the broader end-user base. He bases his argument on the principle of conservation of complexity, which he expresses thus:

"The total amount of complexity in an information system remains constant. In other words, complexity can be transferred from one party to another, but it cannot be created or destroyed."

In a blog posting on Complexity Conservation, Andy accepts this principle, but disputes Mike's conclusions. He argues that complexity can be managed through an appropriate BI platform.

However, I think the principle is mis-stated. There is indeed a constant level of complexity, but this is not because complexity is conserved, but because of a kind of homeostasis. As some forms of complexity disappear into the platform, new forms of complexity are continually created. Or rather, we create new forms of complexity for ourselves.

Most people and organizations are permanently operating at levels of complexity at or beyond their ability to manage it. But whenever new technology offers to shoulder some of the burden of complexity, we say thank you very much and immediately take on new and more difficult stuff, thus restoring the status quo.

So I see the job of the technologists (above all the platform designers) differently to Mike and Andy. It is not to reduce the total amount of complexity visible to the end-users, but to increase the end-users' capacity to handle complexity. Doubtless the BI technologies and practices discussed by Mike and Andy can make a useful contribution to this outcome.