bbl

improvements compared to the old biblio ecosystem

• combines the whole biblio ecosystem in one small binary (authorities, backoffice, discovery, oai, api's, ...)
• split generic (bbl) and ugent specific parts (biblio)
• rich and consistent database model for all entities and their relations (works, work representations, files, projects, organizations, people, users)
• work types, fields used are configurable
• detailed change history for all entities
• handle duplicate people records robustly
• job engine for long running or recurring tasks (gathering candidate works, large exports, ...)
• remove limits on import, export, upload and download sizes
• direct to s3 uploads and downloads
• seamless index switching
• own query language
• avoid deep paging with cursors/search after
• push notifications to users and api's
• semantic search (TODO)
• use the ORCID v3 api (TODO)
• ORCID sync (TODO)

to be decided

• resumable file uploads and side loading via tus.io

support libraries developed

• bind
• vo
• catbird
• orcid
• oaipmh
• sru
• opensearchswitcher
• httperr

dependencies

• Postgres
• Opensearch
• S3 compatible store
• OIDC provider
• Centrifugal