← Catalogue
World Models · legacy example

Dataset & Data Register

This meta-model describes datasets as catalogued assets: identifiable collections of data with distributions, declared schemas, measured quality and a listing in a data register. It is its own model because cataloguing semantics (what a dataset is, how it is served, how good it is, where it is listed) apply uniformly to open data portals, research repositories and internal registers, independently of what the data itself is about.

AI YAMLAGENTS.mdSource
IDworld.n3-dataset-and-data-register
Version history0.2.0-legacy
CategoryReference example
IndustryCross-industry
Domainknowledge-information
Tagsdataset, data, register
Data structure

Bundles and layers

Bundle

catalogue

Describing and serving the asset

Layers

description`: titles, themes, holders, cadence · `distribution`: files, endpoints, formats, services

Bundle

structure

What shape the data has and where it came from

Layers

schemaReference`: declared schemas resolved against the model registry · `lineage`: sources and derivation chains

Bundle

quality

How good the data is

Layers

qualityMeasurement`: measured dimensions and scores · `fitnessForUse`: assessments and certifications for purposes

Findings

Catalogue-native findings must describe the information grouped by each layer. This legacy version does not declare them separately.

Questions & artifacts

Questions, artifact requirements and serial naming rules are required by Vercy vNext; they remain unassigned in this reference version.

Service layers

How this specification operates

Dimension & namespace

  • REFERENCE `world.organization` (O1): data holders, publishers and service operators.
  • REFERENCE `world.modelOntology` (N6): schemaReference resolves to schemas and models registered there.
  • REFERENCE `world.identifierNaming` (N8): dataset and distribution identifiers come from registered schemes.
  • REFERENCE `world.officialStatistics` (N10): statistical releases appear here as catalogued datasets while their statistical semantics stay in N10.
  • imports: dcat (EXTEND): the catalogue backbone of dataset, distribution and data service is specialized by this model.
  • imports: w3c-dqv (MIX-IN): the quality measurement facet applied to datasets and distributions.

Canon, patches & standards

  • REFERENCE `world.organization` (O1): data holders, publishers and service operators.
  • REFERENCE `world.modelOntology` (N6): schemaReference resolves to schemas and models registered there.
  • REFERENCE `world.identifierNaming` (N8): dataset and distribution identifiers come from registered schemes.
  • REFERENCE `world.officialStatistics` (N10): statistical releases appear here as catalogued datasets while their statistical semantics stay in N10.
  • imports: dcat (EXTEND): the catalogue backbone of dataset, distribution and data service is specialized by this model.
  • imports: w3c-dqv (MIX-IN): the quality measurement facet applied to datasets and distributions.

Artifact formats & serial data

Format-independent core. Concrete artifact formats and naming prefixes are not declared in this legacy version.

Policies & accepted processes

  • `openDataLicenseContract`: standard public license terms attached to distributions.
  • `dataSharingAgreement`: bilateral terms for controlled sharing of non-public datasets.
  • `apiAccessContract`: service terms for programmatic access through an access service.

Read / add / edit / delete

CRUD procedures and interface bindings are not declared in this legacy version.

Ownership, roles & access

The neutral owner archetype is the data holder, the party accountable for the asset's content and service levels; catalogue operators list but do not own. Access is always granted by the holder through the catalogue's S1/S2 ownership and access models, with grants and harvests audited via S4.

Existing assembly

Objects, relationships, events and projections

Objects

  • `dataset`: an identifiable collection of data managed as one asset; key attributes: datasetId, title, themes, updateFrequency, holderRef.
  • `datasetSeries`: a grouping of related datasets released over time; key attributes: seriesId, title, cadence.
  • `distribution`: a concrete serving of a dataset; key attributes: format, byteSize, checksum, accessUrlRef, licenseRef.
  • `accessService`: an endpoint that serves distributions; key attributes: endpointRef, protocol, authRequirement.
  • `dataSchema`: the declared structure of the data; key attributes: schemaRef, version, fieldCount.
  • `qualityMeasurement`: a measured quality result; key attributes: dimension, score, measuredAt, method.
  • `registerEntry`: the listing of a dataset in a catalogue or register; key attributes: catalogueRef, listedAt, status.

Relationships

  • `datasetSeries` -> aggregates -> `dataset` (1:N): editions or slices of one continuing asset.
  • `dataset` -> distributedAs -> `distribution` (1:N): one asset, many formats and channels.
  • `distribution` -> servedBy -> `accessService` (N:N): services expose distributions.
  • `dataset` -> describedBy -> `dataSchema` (N:1): the schema fixes the data's shape.
  • `dataset` -> derivedFrom -> `dataset` (N:N): lineage across processing steps.
  • `qualityMeasurement` -> evaluates -> `dataset` (N:1): measured quality attaches to the asset.
  • `registerEntry` -> catalogues -> `dataset` (1:1): the register anchors discoverability.

Events

  • `datasetPublished`: a dataset became available through at least one distribution.
  • `distributionAdded`: a new format or channel was opened for an existing dataset.
  • `schemaVersionAdopted`: the dataset switched to a new declared schema version.
  • `qualityAssessed`: a quality measurement or fitness assessment was completed.
  • `datasetDeprecated`: the holder marked the dataset as superseded or discouraged.
  • `datasetWithdrawn`: the dataset was removed from service.

Projections

  • `openCataloguePage`: public descriptive metadata and open distributions; omits internal lineage and holder-only quality detail.
  • `harvestFeed`: a DCAT-shaped serialization for catalogue aggregators; omits non-catalogue attributes.
  • `stewardQualityDashboard`: quality scores, lineage and schema drift for the holder; omits public presentation fields.