← Back to catalogue
Published

Dataset, schema and contract

EM-DAT-01 · em-dat-01

Support reliable and permitted data sharing by linking dataset versions and distributions to defined semantics, consumer commitments and verifiable quality acceptance criteria.

Enterprise profiles Enterprise subject EnterpriseDAT

Bundle → Layer → Finding → Questions Filled

2 bundles · 3 layers · 6 findings · 14 questions

Dataset identity Distinguish the semantic dataset from its versions and distributions.

Dataset and versions

Record dataset continuity and distribution references.

Dataset identity

Establish the semantic identity represented by a dataset record.

  1. Which semantic dataset does this dataset record identify?
  2. What identity decision does this dataset record document about continuity with an existing dataset?

Distribution

Connect a dataset version to its available representations.

  1. Which distributions and formats does this dataset version record identify?
  2. Which dataset reference does this catalogue record provide?
Schema and contract Define data semantics, permitted uses and measurable quality commitments.

Schema definition

Associate data versions with structural and semantic definitions.

Schema version

Identify the applicable schema and field meanings.

  1. Which schema version applies to this dataset version?
  2. Where does this schema record reference definitions of field meanings?

Data contract

Record consumer commitments and the criteria for accepting data.

Contract terms

Identify the accountable parties and permitted uses.

  1. Which owner and consumers are named in this contract?
  2. Which uses does this contract permit or prohibit?

Quality commitments

Specify measurable quality obligations and evidence for acceptance.

  1. Which quality measures and measurement scopes does this contract commit to?
  2. Which acceptance thresholds and assessment periods does this contract specify for its quality measures?
  3. Which verification methods, assessment frequency and evidence requirements does this contract prescribe?
  4. Which accountable role and response to failed quality acceptance does this contract specify?

Compatibility

Assess consumer impact and notification for schema changes.

  1. What backward compatibility assessment does this schema change record contain for affected consumers?
  2. What evidence does this schema change record retain of consumer notification before an incompatible release?

Classifiers Filled

Family
Enterprise profiles
Category
Enterprise subject
Entry kind
subject
Domain
EnterpriseDAT
Industry
Cross-industry
Tags
EM-DAT-01W1subjectDatasetDatasetVersionDistributionDataSchemaDataContract

What it is Filled

Dataset, versions, distribution, schema, contract for quality and permitted use. The catalog record is not the dataset itself.

Why it exists Filled

Support reliable and permitted data sharing by linking dataset versions and distributions to defined semantics, consumer commitments and verifiable quality acceptance criteria.

Distinguishing features Filled

  • A dataset keeps its identity across exports; a new file does not by itself make a new dataset.
  • Data versions and schema versions are tracked separately.
  • A data contract binds a named owner to named consumers on quality and permitted use.
  • A catalogue record describes and points to a dataset and is not the dataset.

What robots and AI may and may not do Filled

Must not

  • Create a new semantic dataset for every new CSV export of the same data.
  • Change a schema incompatibly without notifying contract consumers.
  • Use data outside the permitted uses stated in its contract.
  • Treat field names as definitions of field meaning.

Only with a human decision

  • Approving an incompatible change to a data contract.
  • Granting a new consumer access to restricted data.

May

  • Register new distributions under the existing dataset version.
  • Detect schema changes and classify them as compatible or breaking.
  • Report which contracts and consumers depend on a dataset.

Moral aspects Filled

  • Datasets may contain personal data whose permitted use must be respected.
  • Silent breaking changes harm consumers who make decisions on the data.

Who is affected

  • Data consumers
  • Data subjects in the dataset
  • Data owners and engineers

Owners Filled

Steward

The data owner or head of analytics who answers for data assets and their fitness for use.

Master systems

  • Data catalogue
  • Data warehouse
  • BI platform
  • Data orchestrator

Links to other meta-models Filled

neighbor

  • EM-AI-02
  • EM-DAT-02
  • EM-DAT-03
  • EM-DAT-04
  • EM-DAT-07
  • EM-LEG-06

references

  • WM-DAT-001 - conceptual-candidate
  • WM-DAT-004 - conceptual-candidate

What else AI and robots need to interact with it Filled

Identity and identifiers required Filled

  • Identified by dataset identifier in the catalogue plus version, with distributions by format and access URL.

Direct properties not applicable Not applicable

Not applicable

Enterprise record contour: physical properties belong to referenced world models.

Recognition optional Filled

  • A dataset record names an owner, versions, distributions and a schema.
  • Often confused with a single file export or with its catalogue entry.

Capabilities and actions required Filled

  • One dataset can be offered in several formats with shared provenance.
  • Contracts make compatibility of changes checkable before release.

Hazards and failure modes required Filled

  • Dataset sprawl when each export becomes a new dataset.
  • Breaking changes that silently corrupt downstream reports.
  • Use of data beyond its permitted purpose.

Standards and interfaces required Filled

  • W3C DCAT for datasets, distributions and catalogues.
  • W3C PROV-O for provenance of dataset versions.
  • JSON Schema for describing data structures.

Context of use required Filled

  • Used wherever data is shared between teams, systems or organisations.
  • Personal data in datasets falls under data protection law such as the GDPR.

Sources Filled

  1. W3C DCAT: dataset, distribution, catalog
  2. W3C PROV and Data Cube: provenance and observation
  3. SDMX and BI/data catalog practice: indicators, breakdowns, definitions and releases
  4. W3C Data Catalog Vocabulary (DCAT) Version 3 - Checked live (HTTP 200) on 2026-10-07.
  5. W3C PROV-O: The PROV Ontology - Checked live (HTTP 200) on 2026-10-07.
  6. W3C The RDF Data Cube Vocabulary - Checked live (HTTP 200) on 2026-10-07.

Open questions

  • What preserves identity across a new export?
  • How to distinguish a schema from field semantics?
  • When is a contract change incompatible?
  • Установить границу и решение reuse/extend/new по действующим спецификациям.
  • Подтвердить semantic crosswalk, права и source mastership.
  • Выбрать immutable refs; провести проверки fixtures до заявления о публикационной готовности.

Machine files

Provenance

enterprise research programme · published-partial

Built from: enterprise/models/em-dat-01/brief.json, ver-cy/world-models/card-supplements/em-dat-01.json

Still in Russian: suggested owner, blocking decisions, vercy candidates. Translate in research/enterprise/i18n/units.en.json.