[{"data":1,"prerenderedAt":4},["ShallowReactive",2],{"raw-en-articles\u002Fopen-data-and-public-reuse-provenance-licensing-and-collaboration-models":3},"---\ntitle: Provenance, Licensing, and Collaboration Models\ndescription:  Provenance, Licensing, and Collaboration Models\nlang: en\nnavigation:\n  enabled: false\n  section: articles\n  order: 30\ntags:\n  - data\n  - mlops\n  - open-data\n---\nWhen data concepts are defined consistently, engineers reason about data safely.\n\n## Why this matters\n\nData governance fails because teams define \"quality\" and \"lineage\" differently.\n\n## What this looks like in practice\n\n- Data lineage is trackable from systems, humans, and AI pipelines consistently.\n- Data quality thresholds mean the same thing whether applied by automation or humans.\n- Metadata is reusable because schemas describe the same concepts consistently.\n\n## How teams use it\n\n- implementing data governance across clouds, on-premises, and third-party stores\n- automating discovery and classification without rebuilding business logic\n- proving provenance for audit, debugging, and compliance systematically\n\nData quality is a team property—it requires agreement on meaning.\n",1776235588116]