Domain data types
Scientific objects are represented as typed values rather than anonymous arrays, files, or dictionaries.
Inspired by practical distributed semantics in ALICE O2 at CERN, Semantiva provides a reusable substrate for typed data, context flow, contract validation, stable graph identity, inspection, tracing, and reproducible scientific workflows.
These primitives support technical evaluation: how typed data, context, contracts, and execution records turn scientific workflow meaning into inspectable structure.
Scientific objects are represented as typed values rather than anonymous arrays, files, or dictionaries.
Algorithms declare the data they consume and produce, making processor contracts inspectable and enforceable.
Metadata, state, provenance, and validity information travel alongside the data instead of living in side channels.
Semantic Execution Records provide evidence of what ran, what was read or produced, and how results were derived.
A Semantiva pipeline can make acquisition, calibration, reconstruction, quality evidence, and archival steps inspectable as executable contracts.
DetectorFrameDataType enters the data channel.
Validity context prevents silent mismatch.
Algorithm contract produces a typed result.
Metrics and uncertainty enter context.
Result and provenance are stored together.
Semantiva is appropriate when the immediate problem is workflow-level semantic control: data product identity, processor contracts, context flow, provenance, testing, and controlled change.
Discuss a bounded assessmentDomain semantics come from scientific modelling and organizational governance. Semantiva provides the executable substrate for making that knowledge operational: typed workflow contracts, context-aware execution, provenance, inspection, and repeatable evidence.