I am new to BPM and trying out a scenario. I have a few questions regarding the way that BPM works.
I understand that BPM can only work with abstract message types. So lets say that I wanted to take an IDoc and run it through a BPM to split it into different messages and then process them in a certain order. Does that mean that I first have to make individual maps for each message type (abstract) that I want to work with in BPM?
To put this in another way, it seems to be the case that XI can't simply bring an outbound message into XI and stick it straight into the BPM process. It seems as though first the outbound message must be mapped and transformed into abstract messages before it can enter the BPM. In this case, how does the transformation step in BPM play in?
Could someone clear this process up for me? I looked at a few weblogs and documents but I'm still having a hard time wrapping my head around the process from a high level.