>I'd prefer to keep the DBC mainly for the field names because that way they can use the same data dictionary to look fields up. As for the RI, I would be inclined to agree with you even though copying that code from the source DBC to the newer one is easy enough.
>
>Open to any other ideas you may have.

You mentioned efficiency: Inserting into archives with referential intergrity intact implies that those "rules" will need to be evaluated each time a child is added. How much of an impact that has, I don't know.

If you don't have any specific concerns and it works for your users, then there probably isn't a problem.