Key technology
Apache Ossie
An incubating project defining a vendor-neutral standard for semantic metadata, so metrics are defined once and used everywhere.
Ossie is an incubating project at the Apache Software Foundation working on a standard for semantic metadata. It describes itself as a universal standard for semantic data, aimed at how semantic definitions are exchanged across analytics, business intelligence, and AI systems.
It takes the form of a declarative specification, expressed in YAML, for describing semantic models: the datasets, fields, metrics, dimensions, and relationships that turn physical tables into something people and software can reason about. The stated goal is to define these once and have every tool work from the same definitions.
Worth being straightforward about maturity: Ossie is incubating, meaning it has been accepted into the Apache Incubator and has not graduated to a top-level project. It appears here because the problem it addresses is real, unavoidable, and currently solved badly almost everywhere, not because adoption is settled.
The gap it addresses
The industry solved portability at the physical layer thoroughly. Open file formats, open table formats, and open catalog protocols mean the same bytes can be read by many engines without conversion or permission.
Meaning did not follow. The definition of a metric, the relationship between two entities, and the judgment that one table is authoritative all live somewhere, and that somewhere is usually a proprietary modeling layer inside a BI tool, a transformation project’s configuration, or nowhere.
The result is that an organization can move its data anywhere and cannot move its understanding of that data anywhere. Every new consumer re-derives the definitions from column names and guesswork.
The project frames this as semantic fragmentation, and names symptoms practitioners recognize immediately: metric drift, manual translation errors, and unreliable AI grounding from conflicting data logic.
The building blocks
A small set of concepts, valuable mostly for being named consistently rather than for being novel.
Semantic models are the container, versioned as a unit. Datasets are the physical tables or views being described, which is the join point between meaning and storage. Fields are columns with a declared role. Dimensions are the attributes work is grouped and filtered by, which is a modeling decision rather than a column property. Metrics are the quantities being measured, with their computation stated explicitly. Relationships record how datasets connect, including join keys and cardinality.
Expressed as YAML, these become artifacts that live in version control, get reviewed like code, and carry a history. That property is worth as much as the format. A metric definition that changes through a reviewed commit is governable. One that changes because somebody edited a dashboard is not.
Write once, query anywhere
The project’s phrase for its goal mirrors what table formats did one layer down.
Before open table formats, moving between engines meant reloading data. After them, a table is a table and engines are interchangeable. The semantic argument has the same shape: before a shared standard, moving between BI tools means re-modeling; after one, the model is an artifact and the tools are consumers of it.
The consumer list is what made this urgent rather than merely desirable. A modern stack has dashboards, notebooks, embedded analytics, data applications, reverse ETL, and now agents, all needing the same definitions. Six copies of a metric was tolerable at two consumers. It is not at six, and it is actively dangerous when one of them answers questions autonomously.
What a standard cannot do
Worth being clear, because semantic layers are sometimes sold as though the format were the hard part.
A standard makes definitions portable and reviewable. It does not produce agreement. If three teams genuinely disagree about what monthly active users means, writing all three in YAML gives you three definitions in a shared format. That is a real improvement, because the disagreement becomes visible and nameable, and it is not resolution.
It also does not maintain itself. Semantic models drift from the tables underneath as schemas change, and a model claiming a column exists when it does not is worse than no model, because it is confidently wrong. Validation against the physical layer has to be part of the process.
And it does not decide who owns a definition. The most common organizational failure is not an absent format but an absent owner. Metrics with no owner accumulate variants regardless of how they are expressed.
Primary sources
Documentation and specifications move faster than any summary. Treat these as the authority and this page as orientation.
- Apache Ossie project Project description, specification material, and current incubation status.
- Apache Incubator Background on what incubating status means and what graduation requires.
Apache Iceberg, Apache Polaris, Apache Parquet, Apache Arrow, and Apache Ossie are trademarks of the Apache Software Foundation. This site is independent and is not affiliated with or endorsed by the ASF. Project names describe subject matter only.