You may actually find the following document useful: http://www.iptel.org/files/db-structure-0.10.pdf In addition to describe the scheme in almost-human-readable form, it shows ER diagram and also a flow-chart demonstrating the usage. Alas, it is all SER/ottendorf based and doesn't presumably apply to openser. The bottom line is that all data elements are properly keyed by unique identifiers to eliminate ambiguities. E.g., in multidomain operation all domain synnomys are linked together by a number, which makes it easy to merge domains, rename domains, and otherwise manipulate them. The same applies to other tables too. -jiri At 11:26 29/11/2006, Daniel-Constantin Mierla wrote: -- Jiri Kuthan http://iptel.org/~jiri/