CLASSIFICATION
Data Types
A multi-axis system for classifying data without reducing it to file formats.
CURRENT V1 AXES
What records actually encode
Every published record currently carries only three classification properties: field, representation and structure. The record schema now enumerates every value used by this snapshot.
| Encoded property | Values present in v1.0.0 |
|---|---|
field | 10 IDs listed on Fields of Knowledge. |
representation | binary, coordinate, event, frequency_series, geometry, graph, identifier, image, index, matrix, measurement, record, ruleset, sample, schema, semantic_node, sequence, state_vector, style, symbolic, tabular, temporal, text, time_series, vector |
structure | structured, semi-structured, unstructured, scalar, time_series, raster_or_scalar |
Proposed future metadata axes
The following cards preserve the design roadmap. These properties and values are not encoded in dataset v1.0.0 and must not be assumed by consumers.
By origin · proposed
Observed, measured, entered, generated, inferred, simulated and aggregated.
By time · proposed
Static, event, stream, time series, snapshot, historical and predictive.
By access · proposed
Public, personal, confidential, restricted, anonymous and derived.
By processing state · proposed
Raw, structurally checked, normalized, indexed, vector-ready, analyzed and archived.