Preservation of digital data with self-validating, self-instantiating knowledge-based archives

Digital archives are dedicated to the long-term preservation of electronic information and have the mandate to enable sustained access despite rapid t echnology changes. Persistent archives are confronted with heterogeneous da ta formats, helper applications, and platforms being used over the lifetime of the archive. This is not unlike the interoperability challenges, for wh ich mediators are devised. To prevent technological obsolescence over time and across platforms, a migration approach for persistent at-chives is prop osed based on an XML infrastructure. We extend current archival approaches that build upon standardized data for mats and simple metadata mechanisms for collection management, by involving high-level conceptual models and knowledge representations as an integral part of the archive and the ingestion/migration processes. Infrastructure i ndependence is maximized by archiving generic, executable specifications of (i) archival constraints (i.e., "model validators"), and (ii) archival tra nsformations that are part of the ingestion process. The proposed architect ure facilitates construction of self-validating and self-instantiating know ledge-based archives. We illustrate our overall approach and report on firs t experiences using a sample collection from a collaboration with the Natio nal Archives and Records Administration (NARA).