For people who are used to think and model logical data models in UML and query with SQL it is often hard to model an ontology in e.g. OWL 2.0. The open world assumption is especially challenging. I would like to focus on best practices for breaking out of the object oriented/SQL mindset to become a master of OWL and DL Query.
Triggered by the OWL 2.0 proposal I would take the idea of discussing the different modeling paradigms from a user's perspective. We have seen different implementations, each with their own rationale. Why do you actually want to use OWL-Lite, and RDFS vocabulary, OWL2 QL or any of the other flavours. Can you show why you did this in practice and convince the others?
Such a discussion forum will significantly add to the understanding of knowledge engineering (= what we are doing) and practice (=what we also (should) be doing).