OPEN AI GLOSSARY
Back to glossary

Schema

Data schema

A schema describes how data is organized and what its fields mean, with structural or value constraints where specified. State the data model and dialect: the exact obligations depend on the schema language. [16] [18] [19] Editorial interpretation

On this page

Definition comparison

Institutional terminology

CNSSI 4009-2022, reproduced by NIST CSRC; information-security context

CNSSI describes an object through its organization, field syntax, meanings and associated metadata. This is broader than a list of column names. [16]

XML validation

XSD 1.1; W3C Recommendation

XSD separates abstract schema components from their XML schema documents; a validator applies a conforming schema to an XML instance. [18]

JSON description

Draft 2020-12 Core; first-party Internet-Draft

JSON Schema describes an instance using an object or boolean schema. Assertions constrain instances; annotations supply information for applications. [19]

Data-derived summary

Goldman and Widom, VLDB 1997; semistructured databases

DataGuides provide structural summaries generated from a database, serving as dynamic schemas for browsing and querying. Their scope is observed structure in that database. [17]

Scope

Here, schema means a data-description artifact used for interpretation or validation. Psychological schemata and prediction models are outside this entry. [16] [14] Editorial interpretation

Includes

  • Named fields, types, permitted structure and explicit constraints; meanings or metadata may also be documented. A schema need not specify every aspect of domain interpretation. [16] [19] Editorial interpretation

Excludes and counterexamples

  • A particular record is a data instance, not the rules applied to it. For example, {"temperature": 23.5} supplies a value; a schema separately describes the property and its constraints. [19] Editorial interpretation
  • A validator is software that applies a schema. The schema itself is neither that execution nor an entire workflow. [18] [14] Editorial interpretation

Different usages and open boundaries

  • Schema may mean declared requirements or a summary inferred from existing data. A DataGuide follows the database; a declared contract can reject data that fails its requirements. Label which direction is intended. [17] [18] Editorial interpretation
  • The schema/instance distinction is a role distinction: a JSON schema is itself an instance when checked against a meta-schema. JSON syntax alone does not identify that role. [19]

Relation to schema extraction

Schema extraction is the process; a recovered or inferred schema is its output. DataGuides illustrate inference from instances. Recovering an explicitly declared schema is a different evidence task from generalizing observed records. [17] [14] Editorial interpretation

For Repository work, retain source-to-field evidence and label observed, declared and inferred constraints. An absent field in a sample does not by itself establish that the field is forbidden; a structural summary alone does not prove complete intended semantics. [17] [14] Editorial interpretation

Operational criteria

Identify the described population, schema version/dialect, fields and constraints; distinguish declared rules from observed patterns and unknowns. Show how an example instance is interpreted or checked without treating that example as evidence for every possible record. [14] Editorial interpretation

Concept relationships

Only is-a means a strict subtype. Uses and supports/orchestrates describe composition and control.

Schema supports/orchestrates Workflow

Supports a workflow at data-interpretation or validation boundaries; it supplies a description or contract, not the executing control flow. This is a supporting-artifact relation, not a subtype. [18] [14] Editorial interpretation

Repository working definition

Repository · Proposed / not certified

Proposed Repository meaning: an explicit, versioned description of data structure, field meanings and stated constraints. Record whether it is declared, recovered from a specification, or inferred from observed data, together with evidence and unresolved coverage. [14] Editorial interpretation

No consensus decision recorded.

Sources and research gaps

Research notes

The applicable institutional definition was verified through the NIST reproduction; the linked original CNSSI document was not retrievable in this review.

The peer-reviewed DataGuides example supports one semistructured-data approach. This review does not establish a universal definition of schema extraction across paper, database and metadata workflows.

  1. CNSSI 4009-2022: schema (as reproduced by NIST CSRC)

    Committee on National Security Systems; NIST CSRC glossary reproduction. Tier 1 / Institutional terminology; source-attributed glossary reproduction

    CNSSI 4009-2022, NIST term record accessed 10 October 2026. NIST CSRC schema entry, attributed definition and source link. Accessed 2026-10-10.

    Context: Object organization, field syntax, semantics and metadata in information-security terminology.

    Verification: NIST reproduction and its CNSSI attribution checked. The linked CNSS original could not be retrieved; no claim of direct review or a universal NIST-preferred definition.

  2. DataGuides: Enabling Query Formulation and Optimization in Semistructured Databases

    Roy Goldman and Jennifer Widom. Tier 2 / Peer-reviewed conference paper

    Proceedings of the 23rd VLDB Conference, 1997, pp. 436-445. Abstract and section 1, printed p. 436; section 2, DataGuide definitions. Accessed 2026-10-10.

    Context: Data-derived structural summaries for the Lore semistructured database; one method, not a universal definition of extraction.

    Verification: Conference PDF abstract/introduction and proceedings publication record checked. The search engine crawl date is not the publication date.

  3. W3C XML Schema Definition Language (XSD) 1.1 Part 1: Structures

    W3C; Shudi Gao, C. M. Sperberg-McQueen and Henry S. Thompson (editors). Tier 3 / W3C Recommendation

    W3C Recommendation, 5 April 2012. Sections 2.2, 2.4, 2.5 and 2.7: components, validators, validity and XML representations. Accessed 2026-10-10.

    Context: Schemas and schema-validity assessment for XML; not a definition of all research metadata.

    Verification: Dated Recommendation and cited sections checked, rather than an earlier Working Draft.

  4. JSON Schema: A Media Type for Describing JSON Documents

    Austin Wright, Henry Andrews, Ben Hutton and Greg Dennis. Tier 3 / First-party specification; IETF Internet-Draft, not an RFC

    Draft 2020-12 Core, draft-bhutton-json-schema-01, 16 June 2022. Sections 4.2, 4.3, 4.3.1 and 4.3.4: instance, schema, keywords and meta-schema. Accessed 2026-10-10.

    Context: The JSON Schema dialect and instance data model; assertions and annotations have different roles.

    Verification: Dated Core text and definitions checked. The document identifies itself as an Internet-Draft with intended Informational status; it is not relabeled a final IETF standard.

  5. Repository terminology guide

    Repository internal proposal. Tier 4 / Repository internal proposal; not certified

    Repository proposal 1.3 / not certified. Term-specific proposal sections and editorial policy. Accessed 2026-10-12.

    Context: Internal proposals and editorial interpretations, not external consensus.

    Verification: Versions 1.1 and 1.2 add the Schema proposal and the flow/governance experimental example while preserving prior proposal sections. All remain unratified; no adoption decision claimed. Version 1.3 changes branding and certification scope only; it does not certify any proposal.

Revision record

  1. 1.0 / 2026-10-10

    Initial cited Schema entry; separates data instances, validation and extraction. Repository meaning is an unratified editorial proposal.

  2. 1.1 / 2026-10-12

    Scope the working definition and certification labels to this repository under Open AI Glossary. Citations and proposed status remain unchanged.