Microsoft and Google have joined Apache Ossie, a project aimed at creating an open specification for exchanging semantic models across data, analytics, and AI platforms. This initiative could make enterprise data models more portable, but governance gaps and evolving specifications may limit interoperability.
TL;DR
- Microsoft and Google are backing Apache Ossie to standardize semantic models across enterprise platforms.
- The project aims to reduce engineering overhead and metric drift when moving analytics workloads between platforms.
- Despite potential benefits, governance gaps and evolving specifications may limit interoperability and long-term adoption.
What happened
Apache Ossie, initially known as Open Semantic Interchange (OSI), was accepted into the Apache Incubator in June. The project uses JSON and YAML to represent semantic models, making them interoperable across platforms like Snowflake, Databricks, and Tableau. Microsoft has developed a two-way converter between Power BI semantic models and Ossie, allowing semantic context to be represented in a common format. Google has joined Apache Ossie and listed BigQuery/GoogleSQL as a supported dialect, though details on its contributions are yet to be disclosed.
Over 60 companies, including Databricks, Informatica, Mistral AI, Nvidia, Oracle, Salesforce, and Snowflake, support the project. Microsoft's involvement includes pushing for the Ossie specification to include more support for its ontologies and recognized query languages like DAX. Google's support is still under development, with early community contributions recognizing BigQuery's footprint across enterprise data stacks.
Why it matters
Support from Microsoft and Google could reduce the burden of repeated semantic engineering for enterprise teams, helping to avoid metric drift and improving productivity. Developers can define metrics once and treat them as versioned, reviewable code artifacts, reducing the risk of different agents interpreting the same metric differently. This could give CIOs greater confidence to scale agentic deployments.
However, portability does not necessarily mean complete interoperability. The degree to which enterprises can move semantic models between platforms will depend on how much vendor-specific logic and functionality Ossie can represent and how accurately the destination platform can interpret it. Governance gaps and the evolving specification may also limit interoperability and long-term adoption. Enterprises will still need to validate whether converted models preserve the intended calculations, business logic, and results before putting them into production.
Key facts
- Apache Ossie is an open specification project for exchanging semantic models across data, analytics, and AI platforms.
- Over 60 companies, including Microsoft, Google, Databricks, Informatica, Mistral AI, Nvidia, Oracle, Salesforce, and Snowflake, support the project.
- Microsoft has developed a two-way converter between Power BI semantic models and Ossie.
- Google has joined Apache Ossie and listed BigQuery/GoogleSQL as a supported dialect.
- The Ossie specification uses JSON and YAML to represent semantic models, making them interoperable across platforms.
- The project aims to reduce engineering overhead and metric drift when moving analytics workloads between platforms.
- Governance gaps and the evolving specification may limit interoperability and long-term adoption.
- Enterprises will still need to validate whether converted models preserve the intended calculations, business logic, and results before putting them into production.
Context
Apache Ossie is part of a broader trend in the AI industry to create open standards and specifications for interoperability. As enterprises increasingly adopt AI and analytics platforms, the need for portability and interoperability between these platforms has become more pressing. Projects like Ossie aim to address this need by creating common formats and standards that can be used across different platforms.
However, the success of such projects depends on widespread adoption and support from key players in the industry. While the backing of Microsoft and Google is a significant step in this direction, the long-term success of Ossie will depend on its ability to address governance gaps and evolving specifications. Enterprises will also need to carefully validate and test converted models to ensure they preserve the intended calculations, business logic, and results.
