Roadmap¶

This roadmap shows where Provisa is headed. Groupings and timing reflect current plans and may shift as priorities develop. Items listed here are things a user will see—internal engineering work is not shown.

2027-Q1¶

Connector catalog¶

Process columns in batch with a Python or Node function, reducing round-trip overhead on large tables.
Attach an OWL reasoner to a SPARQL source so Provisa reads inferred triples alongside the asserted ones.

Glossary, tags and ontology¶

Filter the ontology graph with SPARQL directly from the Ontology page controls, or open it as a query.
Glossary terms derive their OWL type automatically from what they are grounded in—no manual classification.
See a glossary term's related terms and column bindings as a small interactive graph in its detail view.
Export your governed graph as Turtle or N-Triples and load it into any triple store you already run.
Join your Jena, GraphDB or Stardog triples with Provisa's governed data live, with no copy needed.

Catalog and discovery¶

State the business meaning of each data quality check in plain language, not just its technical predicate.

Approvals and query limits¶

Refer a pending data registration to another domain steward for confirmation, keeping the request live.

Federation and routing¶

Export server logs with secrets and physical addresses redacted, ready to attach to a support email.
Add a new data source, query engine or secrets vault by writing a driver, using the same kit the built-ins use.

Change events and streams¶

A preflight check can call Provisa's own services—LLM, vectors, cache, data quality—before a source refresh.

2027-Q2¶

SPARQL¶

Write back to your model through SPARQL Update, with governance enforced on every triple change.