You can choose whatever suits your need. You are basically gonna catch your
calls to steps A & B, and in the catch clause, are going to use some
business logic to decide to make it fail, and just continue with a simple
if condition.
When you're ready, just write those, and ask for review here if
Well, either you didn't give enough info, or seems quite straightforward.
The human interaction is the input step. Make sure to use it outside a node
{ } to not occupy an agent while waiting for user input.
I don't see the actual difference you mean between your 1st and 3rd case.
2016-05-31 20:02