This chapter explains the structure of the BCS. It is not necessary to read it in detail in order to use the BCS, but you should get a little bit familiar with the terms dataPoint, weight, and weightPosition.
The BCS is implemented as a new node in Maya. It has a structure similar to the built-in Maya Blend Shape node, but in order to allow for the flexibility the BCS has, some core concepts of the Blend Shape node had to be abandoned. The biggest change is certainly the separation of what you know as Blend Shape Targets into weights, weightPositions and dataPoints.
In a Blend Shape node, each target has its own weight. This weight can be animated to make the object come to life. In the BCS, the targets are called dataPoints and they can be associated with one or more weights using weightPositions. More about this later.
As a first step, you can imagine the rough structure of the BCS using the following diagram:
There are the animatable weights, the dataPoints with geometry, and the input geometry (i.e, the neutral, not-yet-deformed object). There is also the envelope and any weight painting applied. They all have to be mixed together to result in the deformed output geometry that you can see in the scene.