The changes to #19 persist a certain amount of rigidity with SF fields. In another implementation, we can simply ignore changes to the Siegfried structure - other's might say knowing about changes is useful, but we shouldn't have to react to changes that don't effect the overall functionality of the code.
We can:
- Ignore issues with ID imports by default.
- Have a strict mode that measures conformance with a schema.
Really, ID tools change so little, and this code is used only by myself, so it's not a huge priority.