Safari Books Online is a digital library providing on-demand subscription access to thousands of learning resources.
We've taken a look at the arguments for avoiding unnecessary complications, especially the introduction of "features" before they are needed. Conversely, we've also (hopefully) accepted that there are times when a little engineering that reflects a measure of foresight might be a good investment. Now we move to the central theme of this chapter – when you are developing a data architecture for just one domain within an enterprise, should it try to align with the overall enterprise information architecture? And wouldn't doing so be blatant over-engineering for the problem at hand?
This question is based on the premise that the organization already has an enterprise-wide information architecture in place. If it doesn't, the next chapter will give some advice as to how you might be able to construct at least a framework for an enterprise model that is surprisingly robust, yet quick to construct.