Villanova
As of 2015-02-26, exploration of migrating a non-Hydra F3 instance to a Sufia-based F4 via the FedoraMigrate 0.0.1 utility has been underway.
The nature of the starting F3 repository includes the following.
Profile of content and size
The collections are comprised predominately of scanned books, manuscripts, and maps from our special collections and archives. Within Fedora, there are models that describe Collections, Resources, and Data objects.
The repository includes roughly 350,000 objects in Fedora3, consuming about 18T of disk.
The preliminary work in the AF9/F4 front has been mostly focused around metadata enrichment and linking, preparing for the native LD architecture of F4 and the eventual abandonment of our XML datastreams.
F3 features in use
- Storage: Legacy storage (or Akubra)
- Akubra
- XML metadata
- Descriptive, provenance, preservation, and technical metadata stored as xml in stored and inline datastreams.
- Content models
- Existing models that describe Resources, Collections, and Data objects.
- Datastream types (inline, managed, redirect, and external)
- Inline and managed
- Identifiers
- Not exactly sure what you are looking for here, but we are currently using the namespaced identifiers minted by F3
- Indexing strategies (GSearch, RI-Search vs. F4 approaches)
- GSearch
- Versions
- Metadata Only
