The PREMIS Working Group has published the Data Dictionary for Preservation Metadata: Final Report of the PREMIS Working Group. Find the report and its components on the PREMIS project web page.
This group was chaired by Priscilla Caplan and Rebecca Guenther and supported by Brian Lavoie and Robin Dale. They have completed a major piece of work which deserves the attention of all interested in the practice and theory of digital preservation going forward.
One of the interesting points about the report is its description of a data model which includes events. The report defines an event as: "an action that involves at least one object or agent known to the preservation repository". It is important to note that I am not talking about an 'event' as the subject of a work here, as in a book about the Galway Races or the American Civil War. Traditional library cataloging practice has largely been concerned with published material. Our central event is 'publication'. Another involves copyrighting, closely tied to publication, or translation. Other events tend not to figure so much. For example, we do not typically record in our bibliographic records the date the item was acquired by the library.
As resources become more fluid, we will become much more interested in events: actions on objects that we may wish to record. Resources may be digitized, tranformed, excerpted, migrated, and so on. And events in this sense are integral to other models, for example the Ontology-x work reported by Godfrey Rust at the FRBR workshop held here recently [ppt].
This raises a question I will return to: how simple is simple enough when it comes to metadata.