Reading map
A useful way through the notes.
Explore the scope of Orabl Data Notes, choose a question, and follow the local article archive.
01
Models & Grain
The identity and meaning of a row before the first aggregate.
02Data Contracts
Explicit rules for completeness, validity and the action on failure.
03Replayable Pipelines
Batch identity, durable inputs and a deterministic replay result.
04Reconciliation & Semantics
Comparable totals with the same grain, time window and precision.
The article index
Search and filter- Data QualityData Quality Contracts With a Defined Response5 min read
- Data ModelingDeclare the Grain Before Writing the Query5 min read
- SQLJoins Without Accidental Row Multiplication5 min read
- ReconciliationReconcile at the Same Grain5 min read
- PipelinesReplayable Batch Ingestion5 min read
- Time & PrecisionTime and Decimal Semantics in Data Reports5 min read
Read the assumptions first
Examples describe an explicit setup. Expected outcomes are checks to make, not evidence that a live service has already passed them. A dated result should record the environment and the observations that support it.