A Feature Defintion is composed of two parts:
- The "feature defintion" which defines the nature of the feature being defined. For example, it is an edge of pavement, or it is a curb. These are rather ordinary, I suppose since there is not a lot to descirbe them other than "I am a curb". But, coming soon, you will see examples such as catch basin, where the feature defintion part will not only decribe what I am but also contain the hydraulic characteristics of the catch basin. Or, a Pipe, where the feature defintion describes the various available pipe sizes.
- The element template defines the symbology. More specifically, it describes its presentation. For example, I am blue. Or I am on level whatever. By assigning different element templates for different views (plan, profile, 3D) then the same feature can be presented differently in each context. In the future, you might imagine that the in addition to plan, profile and 3D context, you might have context based on user roles. A road designer wants to see it one way, a drainage designer wants it another.
This last aspect (user defined roles) is one of the reasons that the feature defintion is in two parts. What a thing is (edge of pavement, curbe or catch basin) is always the same. But presentation can vary.