James, Thanks for a great overview of the approach to ISO20022 and SEPA! If one can regard ISO20022 as an ICD(Interface Control Document), and in many ways it is useful to do so, then I am a little disappointed that those of us who were around when EDI was being introduced cannot remember or have not been asked to remember the issues that faced us then and the manoeuvres we employed to avoid or accommodate them. James is absolutely correct in SEPArating (no pun intended) the specifications of the data being exchanged from the behaviours of the systems at each end of that exchange. By it's very definition, the ICD exists to allow the behaviours of the system at each end of the interchange to vary while not introducing necessary change in the other. So, we need to defuse discussions or arguments involving behaviours and our parochial need to accommodate them from definition of the data arising from those behaviours or causing unwanted processing on the recipient. Of course it is true that after discussions, we may need to alter the definition of a data element in the ICD, or to add or remove elements. So, let's remember how many, and which of, the 7 ISO levels we are involved in in each discussion and bound our comments within that scope. Here arises the need for a strong, fair, objective and knowledgable architect to monitor and guide participants, but the recognition of the problem that James introduces is an excellent start. Well done and thanks!!