Data Management

An archive is usable when its files can be found, trusted, attributed, and understood — and those virtues are designed, not hoped for. Data Management is the practice’s custody of a production’s information, from the first file to the final package.

Its governing idea is dual use: a database built for production is already structured for archive. The same system that tracks cues, props, versions, contacts, and schedules during the run becomes — because it was built to — the metadata backbone of the preservation package after it. Nothing is entered twice; nothing has to be reconstructed later from memory.

In practice the service covers the design and operation of production data systems; file and media management under controlled naming, versioning, and integrity checking, so that every asset’s history is traceable; metadata that makes each item findable and correctly attributed; rights documentation gathered in parallel with the work; and, at the end, archival packages aligned with the OAIS reference model (ISO 14721) — self-describing, complete, and ready for an institution to ingest without a phone call. The test applied throughout is the practice’s oldest: could someone who was never here understand this, with no one from the practice in the room?

Together with Framework Design, this is the structural pair of the six instruments: the framework decides what deserves to survive; data management keeps custody of it while the production is alive, and packages it when the run ends.