Highlightdbt Labs

dbt model governance covers access, contracts, versions, and mesh refs

Overview of dbt model governance: public/private access, contracts, versions, namespaces, and Enterprise cross-project dependencies.

The page says model governance controls who can access models, what they contain, how they change, and how they are referenced across projects. Features include public/private access, contracts on column shape, versions for breaking changes, namespaces for ownership, and Enterprise project dependencies via cross-project ref. It also mentions freshness SLAs with State and lag_tolerance, plus caveats about adopting governance too early.

Based on: About model governance | dbt Developer Hub · dbt Labs

Highlightdbt Labs

Model versions treat shared dbt models like versioned APIs

dbt Mesh docs explaining model versioning versus other “version” meanings, and why producers and consumers need graceful change.

The page separates model versions (a Mesh governance feature) from dbt_project.yml and YAML property-file version fields. It compares shared dbt models to APIs where producers must change logic while consumers need stable queries. Model versioning is offered as a deliberate way to handle breaking changes without pretending the tension disappears. It also warns that governance features can harden rollbacks if adopted too early.

Based on: Model versions | dbt Developer Hub · dbt Labs

Highlightdbt Labs

dbt Semantic Layer APIs put MetricFlow metrics behind governed query interfaces

dbt docs on Semantic Layer APIs: define metrics in code with MetricFlow and query governed assets from downstream tools, including via GraphQL.

The page argues that growth in the modern data stack fragments business logic across teams and tools. The Semantic Layer lets you define metrics in code with MetricFlow and dynamically generate and query datasets from dbt-governed metrics and models. It lists use cases such as BI, data quality, governance, discovery, and ML, and notes a GraphQL API for metrics and dimensions.

Based on: Semantic Layer APIs | dbt Developer Hub · dbt Labs

HighlightOpenMetadata

OpenMetadata docs cover discovery, lineage, contracts, and MCP for agents

OpenMetadata documentation hub for metadata discovery, lineage, governance, data contracts, and AI/MCP integrations.

OpenMetadata’s docs present a platform to document, discover, and govern data assets, with quick start, production, and upgrade paths. Guides span discovery, lineage, observability, data contracts, and governance; highlights include Context Center, MCP server and AI SDK for agents, new connectors, and dimensional data-quality validation.

Based on: OpenMetadata Documentation - OpenMetadata Documentation · OpenMetadata

Highlightdbt Labs

dbt exports materialize saved MetricFlow queries as warehouse tables

dbt docs on exports: run saved Semantic Layer queries via the job scheduler and write results to tables or views.

Exports run saved MetricFlow queries and write output to a table or view in the data platform, using the dbt job scheduler. They give SQL and non–Semantic Layer tools access to metric definitions as ordinary relations; running an export counts toward queried metrics usage, querying the result does not.

Based on: Write queries with exports | dbt Developer Hub · dbt Labs

Highlightdbt Labs

dbt measures are column aggregations, now migrating to simple metrics

dbt documentation for measures as aggregations on model columns, with deprecation toward type: simple metrics.

Measures aggregate columns and can stand alone or feed complex metrics. Parameters include name, agg (sum, max, min, average, median, count_distinct, percentile, sum_boolean), expr, non_additive_dimension, and related fields. The new spec deprecates measures in favor of simple metrics under metrics.

Based on: Measures | dbt Developer Hub · dbt Labs

Highlightdbt Labs

MetricFlow centralizes metric specs and SQL construction in dbt

dbt intro to defining metrics with MetricFlow as the Semantic Layer component that builds SQL and enforces consistency.

MetricFlow in dbt centrally defines metrics and builds SQL from semantic models and metric specs. It aims to cut duplicative coding, support governance of company metrics, and keep consumer results consistent; defined metrics can be queried in development and, on higher plans, in downstream tools.

Based on: Build your metrics | dbt Developer Hub · dbt Labs

HighlightConfluent

Confluent Schema Registry centralizes Kafka schema validate-and-evolve rules

Confluent documentation overview of Schema Registry for managing, validating, and evolving schemas on Kafka topics.

Schema Registry is a centralized store for topic message schemas and for serialize/deserialize over the network. Producers and consumers use it for consistency and compatibility as schemas change; Confluent frames it as a data-governance component covering quality, standards, lineage visibility, audit, and collaboration, on Cloud and Platform.

Based on: Schema Registry for Confluent Platform | Confluent Documentation · Confluent

HighlightBitol (LF AI & Data)

ODCS v3.2.0 standardizes the sections of a producer–consumer data contract

Bitol’s Open Data Contract Standard defines a YAML-oriented structure for agreements between data producers and consumers.

The Open Data Contract Standard (ODCS) v3.2.0, under Apache 2.0 from Bitol at LF AI & Data, describes how to structure a data contract. Contracts cover schema, quality, SLA, roles, infrastructure, and related sections, with JSON Schema for YAML validation and media type application/odcs+yaml;version=3.2.0.

Based on: GitHub - bitol-io/open-data-contract-standard: Home of the Open Data Contract Standard (ODCS). · Bitol (LF AI & Data)

HighlightGoogle Cloud

LookML puts join structure and query content in one reusable semantic model

Google Cloud docs introduce LookML as Looker’s modeling language for dimensions, aggregates, joins, and SQL generation.

LookML (Looker Modeling Language) is a dependency-style language for building semantic data models over SQL databases. Models and views define joins and calculations once; Looker generates dialect-independent SQL so analysts avoid repeating expressions and business users query without writing SQL.

Based on: Introduction to LookML | Looker | Google Cloud Documentation · Google Cloud

HighlightDatabricks

Unity Catalog: one governance layer under Databricks data and AI assets

Databricks docs on Unity Catalog—access control, lineage, and auditing for data and AI via a three-level object model.

Unity Catalog is Databricks’ unified governance layer for data and AI. When enabled, it enforces access control, tracks lineage, and logs activity under workspace interactions. Assets are securable objects in a catalog.schema.object namespace; tables and volumes may be managed or external. An open-source implementation also exists.

Based on: What is Unity Catalog? | Databricks on AWS · Databricks

Highlightmartinfowler.com

Data mesh: domain-owned data products instead of a central lake monolith

Zhamak Dehghani’s 2019 essay on moving from monolithic data lakes to a distributed data mesh.

Dehghani argues enterprise data lakes often fail at scale through centralization, coupled pipelines, and siloed ownership. The proposed shift treats domains as first-class, applies platform thinking for self-serve infrastructure, and treats data as a product with discoverability, addressability, trust, self-describing semantics, interoperability, and secure access.

Based on: How to Move Beyond a Monolithic Data Lake to a Distributed Data Mesh · martinfowler.com