schemasthat

schemasthat is a term used to describe a design principle and, in some contexts, a software approach to treating schemas as first-class artifacts in data processing. It advocates creating, sharing, and applying schemas alongside the data they describe, with the goal of improving validation, interoperability, and governance across distributed systems.

Core concepts include declarative schema definitions, support for multiple schema formats, and tooling that links schemas to data through validation and transformation. Proponents emphasize schema evolution and backward compatibility, including versioning strategies, migration trails, and clear deprecation policies to manage changes over time.

Architecturally, schemasthat often relies on a registry or meta-schema that stores schemas and their relationships. Data payloads may carry a reference to a schema or embed the schema inline, enabling consumers to validate and interpret the payload without prior knowledge. Runtime validators, serializers, and cross-format transformers are common components.

Typical use cases are API contract enforcement, data integration and ETL pipelines, data governance and lineage, and automated documentation generation. By standardizing how schemas are defined and discovered, schemasthat seeks to reduce schema drift and improve interoperability between heterogeneous data sources.

Relation to existing standards is broad, intersecting JSON Schema, XML Schema, SHACL, and database schema design. Critics warn of potential performance overhead, added complexity, and the risk of overfitting data models to schemas at the expense of flexibility.