Acquisition
Source discovery and retrieval
Repeatable harvesting patterns for APIs, structured pages, public registers and other inspectable source material.
Infrastructure
The machinery behind the public projects.
One project creates a system. Reusable infrastructure lets the next project start further ahead, with proven patterns for acquisition, provenance, validation, review and controlled publication.
Shared architecture
Messy sources in. Reviewable evidence out.
AcquireIdentify sourceNormaliseValidateReviewPublish
The implementation varies by domain. A biomedical evidence workflow should not use the same validity rules as an events directory. The reusable layer is the control architecture: provenance, structured records, source-health checks, deterministic gates, review points and release discipline.
Provenance
Trace the evidence path
Source identity, retrieval context, dates, transformations, manifests and evidence lineage retained where practical.
Validation
Fail before publication
Deterministic checks, field rules, completeness gates and domain-specific publication constraints.
Review
Keep people in the loop
Human review remains explicit where interpretation, allegations, scientific verification or other consequential judgements are involved.
Release
Public and protected layers
Separate publication surfaces when sensitive, provisional or governance-constrained material should not be exposed publicly.
Re-use
Start the next project further ahead
Extract working patterns into bounded engines instead of duplicating bespoke code and process in every project.
Current reusable engines
Local discovery as a proving ground.
The local-information projects make the reusable layer visible because they repeatedly solve the same ingestion, normalisation, source-health and publication-safety problems.
Local Events Engine
Discovers events from multiple source types, normalises records, tracks provenance and source health, rejects weak records and produces bounded downstream exports.
Local Directory Engine
Applies the same architecture to local services and business listings, separating discovery from publication and retaining source context for review.
Why it matters
Reusable infrastructure is evidence of engineering maturity.
The aim is not to claim one universal engine solves every domain. It is to reuse disciplined patterns while preserving the evidential rules that make each domain different.