Quote - This is described in the manual:

The Component node is useful for extracting the red, green,
or blue components from an input color...

 

0 = Red

1 = Green

2 = Blue

 

The EZSkin one you described is pulling info from the Red channel - if I remember correctly it's to modulate the way that bump and / or specularity is applied.

Hmmm, so the question is why would they use info from the red channel to control the intensity of the bump & specular? Why not blue or green?