Origin as the backbone
Source, owner, system, transfer and processing remain connected.
BronVast Metadata & Context
Files become information when provenance, meaning and relationships remain understandable.
BronVast connects technical metadata with descriptive, structural, administrative and preservation metadata, showing what an object is, where it came from, what happened to it and who is responsible.
2dA BronVast
Dossier, attachment and version stay together
Structures and dependencies are not flattened into loose files.
Access and use remain explainable
Privacy, copyright, confidentiality, retention and responsible role are recorded.
The route from source to lasting organisational capability
Map source fields
existing metadata and system context
Reconstruct structure
series, dossiers, versions and relationships
Enrich context
people, organisations, places and subjects
Apply rules
rights, retention and access
Measure quality
coverage, consistency and exceptions
Example: loose files
A folder becomes an understandable dossier again
Filenames reveal little on their own. BronVast connects objects to dossier, version, sender, process, rights and actions so people and systems can continue to understand the relationships.
Metadata for people and systems
The profile is no fuller than necessary, but rich enough for management, transfer, search and future reuse.
- descriptive metadata for discovery and understanding
- structural metadata for hierarchy, sequence and relations
- administrative metadata for rights and access
- technical metadata for format and properties
- preservation metadata for events, agents and integrity
AI as a proposal engine, not silent truth
Local models can propose descriptions and entities. Source, confidence and human review remain visible.
- controlled vocabularies and taxonomies
- person and place entities linked to sources
- confidence thresholds per field
- review queues for uncertain cases
- corrections as new evaluation examples
What the organisation retains
Clear agreements
Fields, definitions, cardinality, vocabularies, source and quality rules.
Relationships remain useful
Objects, dossiers, people, organisations, systems and events in context.
Coverage becomes measurable
Completeness, consistency, uncertainty, exceptions and human corrections.
Context for the future
New documents receive better proposals immediately
A local model can build on approved metadata and corrections. For new intake it proposes document type, relationships, responsible role and relevant terms while keeping source and confidence visible.
Local by design
Context can be built and queried locally
Metadata extraction, entity recognition, embeddings and relational or graph storage can remain within agreed infrastructure. Search and chat answers link back to sources and responsible context.
01Source metadata02Mapping03Context graph04Review05Search and chat
Technical reference points
We align the route with recognised standards and specifications where they fit the source, destination and customer responsibility. This is never presented as automatic certification.
Questions about this route
Can existing metadata be reused?
Yes. Existing fields are inventoried and mapped first. Values are not replaced without reason; source and transformation remain recorded.
Can AI fill metadata automatically?
AI can propose values. Final application depends on confidence, risk and agreed human review. Source and model version remain part of the result.
Can the context layer connect to chat?
Yes. A local search or chat layer can answer from controlled objects, metadata and relationships and point to the source or responsible colleague.
Next step
Choose the right starting point
Start with one source, one process or a read-only assessment. We will define what should remain local, what evidence is required and which route creates value first.