Data models and semantic definitions

Structure.
Meaning.
Or both.

MeaningGraph is an open registry for data models and semantic definitions. Describe how your data is structured, what it means, or both.

What a number means · intro film, 1:43 · plays from YouTube

Try an example:Browse the Registry

Explore the graph

Connected domains

Focus a reference to see where it leads
◈

Commerce

Customers, orders and payments.

Explore domain
♧

Identity / Contacts

People, organizations and contacts.

Explore domain
◇

Assets / Finance

Ownership, value and obligations.

Explore domain
Start with structure, meaning, or both.See how it works

01 / Understand it

From isolated data
to a living graph.

A name in a table is only a start. MeaningGraph connects a field to what it represents, what it references and where that definition came from.

Explore Invoice
A meaning has context
DomainCommerce
EntityInvoice
FieldcountryId
↓ references
Geo / EntityCountry

A semantic reference connects definitions even when their data lives in different systems.

02 / Map it

Same meaning.
Different shape.

A semantic field is independent of its storage type. Map it to a SQL column, JSON property or CSV heading without redefining what it means.

Inspect Invoice.customerId
Transparent layers align separate patterns into one connected view
Semantic definitionPhysical representation
Invoice.customerId↔chinook.Invoice.CustomerId
Country.alpha2↔countries.csv → ISO2
Customer.contactId↔customer.contact.id

Illustrative mappings. ChinookDB.com provides an example database.

03 / Connect it

Sources that never
met can make sense together.

Definitions provide a common layer for SQL databases, APIs and files. Follow the meaning of an identifier across sources before writing a join.

SQLHTTP / JSONCSVOpenVaultDB
SQL ServerOracleHTTP / JSONCSV
MeaningGraphshared meanings
SearchQueryValidateAnalyze

Public registry

Browse data models and semantic definitions.

Registration is optional. Describe logical record types and fields, define meanings and relationships, or connect the two later. Registration adds discovery and pinned provenance. Database implementations and deployments are listed in the OpenVaultDB Directory.

Supported by CodeGrapher

Explore meanings and their model bindings.

CodeGrapher recognizes MeaningGraph concepts and their declared ModelSpec bindings in indexed source. Registered graph and concept pages link to the exact source revision and concept line pinned by this registry. Code relationships appear where the indexed repository declares them.

04 / Own and extend it

Open by design.
Yours by default.

Keep definitions in your own Git repository. Reference an immutable version, extend a shared concept, or propose a change upstream. Alternatives can coexist with clear provenance.

How to contribute
⌁
community / commerceInvoice @ 1.2Reusable definition
extends ↓
⌁
acme / finance-modelAcmeInvoice @ 3Narrows id to integer · adds approvalCode

Repo, version and extension remain visible wherever a definition is used.

One registry and a physical directory

Logical models, shared meanings, physical databases.

MeaningGraph · you are hereLogical models and semantic definitions

The shape of the data (record types, fields, types and links), plus domain concepts and vocabularies in one open registry. Models can stand alone; links to meanings are optional. ModelSpec remains an independently usable model format.

Browse the Registry ModelSpec format
OVDB DirectoryWhere it is

Physical databases, engines and deployments: where the data is, who publishes it and how to reach it.

directory.openvaultdb.com

AdventureWorks OLTP has a physical database listed in the OVDB Directory, with logical model AdventureWorks OLTP model and semantic definitions in MeaningGraph.

05 / Contribute

Start with structure,
meaning, or both.

Register a logical data model, add semantic definitions, or connect both. Registration adds discovery and pinned provenance.

Author a DataModel: example, validation and JSON Schema ↗

01Describe structure, meaning, or both

Logical models and semantic definitions can be registered independently, then connected when useful.

02Keep control of the source

Register a repository and immutable revision for discovery.

03Connect optionally

Meanings can map to model fields without changing either definition.

Used in product development

Find meaning in your data with DataTug.

DataTug helps developers explore database schemas, build and run queries, and understand results. DataTug can reference versioned DataModels from MeaningGraph and consume their structure independently of semantic mappings.

Explore the tools behind DataTug
⌁
DataTug developmentModel structure ↔ shared meaningConnecting definitions across datasets

A shared language for people and software

Let's build a more
meaningful tomorrow.

Explore. Define. Connect. Together.

Explore the graph

Start with meaning

Propose a definition

This local draft is for a semantic meaning only. Register logical data models or meaning files through the registry guide.

A local JSON draft downloads to your device. Nothing is published.