Methodology & Trust

Methodology & sourcing

Liner treats music knowledge as a research problem, not a wiki of unsourced opinion. Every claim, credit, and placement is a structured record with a source, a confidence label, an optional editorial note, and a publication status.

Confidence labels

Verified

Confirmed against primary or multiple independent sources.

High

Strong secondary sourcing; very likely accurate.

Medium

Plausible, commonly reported, not independently confirmed.

Low

Weak or single-source; needs verification.

Disputed

Sources conflict, or the claim is actively contested.

Source reliability

Sources are typed (interview, liner notes, database, article, book, primary) and graded:primary secondary tertiary needs verification.

Prototype data policy

This is a demo prototype. It mixes fully fictional sample content with one sourced real-data test set (Hootie & the Blowfish — Cracked Rear View). Real-world records use cautious labels — “source-backed test record”, “secondary source”, “needs verification”, “prototype data” — and no unsupported real-world claim is asserted as Verified.

Editorial workflow

Public read views are separated from editorial write workflows. Content moves throughdraft → published, and can be marked needs review or archived. Community submissions land in a review queue as pending and are never auto-published.

How Liner differs

Platform	Focus	What Liner adds
Genius	Lyrics & annotations	Structured, sourced claims + confidence
Discogs	Releases & credits	Song-first meaning + placements + lineage
MusicBrainz	Open metadata	Editorial narrative + trust model
Tunefind	Screen placements	Placements joined to credits & sources
WhoSampled	Samples / covers / remixes	Lineage joined to the full song record
Liner	Song intelligence	All of the above, source-backed, in one page