Keep catalogue and evidence references behind a data-readiness dashboard
A ready label becomes interpretable only when it points to a particular dataset and a defined confirmation.
Before selecting ready
A first data-readiness dashboard can begin with completion flags and totals. It is more useful to read one entry’s catalogue identity and the confirmation behind its state. Truthound can check references and conditional requirements in that inventory. It does not assess the source dataset on the owner’s behalf, authorize publication or establish that every activity described by a readiness label has actually been performed correctly.
The new portal describes evidence for readiness
India’s MoSPI launched its data-harmonisation QPR portal on 8 September 2026. The introduction covers cataloguing, metadata, standards, identifiers and quality-check progress, alongside validation and cross-verification features. It is not a finding that every ministry’s data quality has been established. The relevant distinction is between a reported state and the evidence the institution needs to explain that state.
Prepare a small starting inventory
A proposed introductory exercise records catalogue identity, standards version, check state and reporting quarter for one dataset. The owner defines the evidence required for completion. Approved metadata relationships can be explained without copying the actual source data externally. If the catalogue has multiple versions, identify which entry supports the present report instead of treating a familiar dataset name as a sufficient reference.
Know what the first result can establish
Truthound Core Python SDK 3.3.4 (open-source SDK, Polars backend) documents ForeignKeyValidator for supplied readiness and catalogue tables. ConditionalExpressionValidator could check evidence required by a particular state. The operator supplies keys, conditions and null handling. The expected benefit is identification of exceptions to that declared contract, not certification of complete data readiness. A populated field still needs an owner able to interpret its source and scope.
Use a completed-label counterexample
If an entry says complete but has no matching catalogue record, first clarify the relationship. Even an existing record may point to old quality evidence or another release. A person must inspect that distinction. Do not fill values merely to make every row complete; retain a record of which confirmation is missing so the next worker can understand the correction required.
Start with an existing small join
pandas 3.1.1 (general-purpose DataFrame library with project-owned checks) merge(validate=...) can retain a correct readiness-to-catalogue join. Cardinality, full reference existence and state-specific evidence are different questions. Consider Truthound according to the obligation still missing. Maintaining an explanation of the readiness state matters more than replacing a familiar interface, and neither documentation set demonstrates that its library makes a government programme’s source data genuinely ready for use. pandas can match null keys to one another, so do not assume ordinary SQL null-join semantics.
Expand after one entry makes sense
Compare one state and one catalogue reference with its owner. Hold missing identities or unexplained versions. Extend to the next dataset after the relationship and exception are understood. A trustworthy readiness report starts by examining what complete means in one defensible case, rather than expanding dashboard totals before the underlying confirmation can be explained.
Conditions
Define catalogue identities, readiness states and the evidence each state requires.
Alternatives
Retain pandas 3.1.1 joins and separately check catalogue references and evidence conditions.
Limitations
A ready label does not establish actual source quality or suitability for release.
Evidence
- Event checked
- Subject checked
- Modified
