A step-by-step technical path for moving your institution's existing collection data into a fully standardized, Omeka-powered environment.
Inventory what you currently have: file formats, existing metadata (spreadsheets, exports from old DAMS, or none at all), total data volume, and rights/licensing status. This determines how much cleanup is needed before importing.
Install Cabina OS on your laptop. The built-in Omeka instance is pre-configured no manual LAMP stack setup is required. Ensure the local dashboard is accessible before moving data.
Translate existing metadata fields into the 15 Dublin Core elements (Title, Creator, Subject, Description, Date, Type, Format, Identifier, Rights, etc.). Non-standard fields can still be maintained via custom Item Types and Element Sets in Omeka.
Standardize file naming, convert media to preservation-friendly formats, and ensure clean UTF-8 metadata encoding. This step prevents import errors and keeps the archive queryable long-term.
Import a small subset (20-50 items) first using Omeka's CSV/bulk import tools. Verify metadata renders correctly, files attach properly, and search/browse functions as expected before committing to a full import.
Once the pilot is validated, run the complete migration. Spot-check samples against source records, confirm collection grouping, and ensure no data is truncated or improperly mapped during transfer.
Organize migrated items into Collections, then use Omeka's Exhibit Builder to construct public-facing narratives and virtual exhibitions from the structured archive.
Configure user roles for staff, and establish local backup routines — since Cabina OS runs archives on-premise, institutions retain full control over end-to-end preservation.