Music releases and album editions
A music release identifies an issued product and its media, track occurrences and release context rather than only the album’s general title.
An album name often describes a family of releases: an original CD, a later reissue, a download and an expanded edition. A release identifies a particular issued product within that family. It provides the context for which tracks, artwork and edition information belong together.
This is the identity needed when the question is “Which album edition is in my library?” rather than “Which recording is playing?”
Two products called Harbour Songs
Consider these fictional products:
| Property | Release P1 | Release P2 |
|---|---|---|
| Display title | Harbour Songs | Harbour Songs |
| Edition | Original | Expanded |
| Medium | One CD | Two-disc set |
| Disc 1 | Three original track occurrences | The same three track occurrences |
| Disc 2 | None | Two live tracks and a demo |
| Release year | 2020 | 2025 |
| Cover | Original design | Revised edition design |
The shared title is meaningful. So are the different track list and product contexts. If the library combines all six track occurrences into one unexplained album, it becomes difficult to determine which release the viewer is browsing.
MusicBrainz models a release as an issued product with identifying properties and one or more media. Its release group describes the broader album concept. The products above can belong to one group while retaining distinct release identities.
A track is an occurrence in that product
Imagine the studio recording of River Lights appears as P1 disc 1 track 2 and also as track 7 of a compilation. Those positions belong to track occurrences on releases. The recording's identity does not encode “track 2.”
A release track can have a printed title or credit that differs from a general recording description. The occurrence retains what this product presents. The audio can still refer to the same recording.
The relation can be pictured as:
release P1
-> medium 1
-> track position 2
-> studio recording R-A
-> composition River Lights
Each step adds context. The playable file is a representation of that audio with embedded tags describing the desired occurrence and product. Its path does not replace the entire chain.
Product evidence needs interpretation
A catalogue can use release events, label/catalogue information, barcodes, track lists and explicit release IDs to distinguish products. Each clue has a scope and limitations.
A barcode can be useful product evidence, but reissues or catalogue practices can make it less decisive than expected. A printed catalogue number needs its label context. A region/date pair can distinguish an issuance while leaving other properties unresolved.
A complete track-list match is stronger than a single title, but an expanded edition may intentionally reuse the original disc. Matching only that disc can therefore leave both P1 and P2 plausible.
A recording fingerprint has a similar limit. It can match R-A accurately while both products contain R-A. Release matching requires evidence about the product, not only about its constituent audio.
What the local tags need to preserve
| Descriptive field | Job in the library |
|---|---|
| Album title | Human-readable product/album label |
| Album artist | Album-level credit or grouping context |
| Disc and track numbers | Position within the represented release |
| Release identifier | Explicit catalogue product identity, when supplied |
| Recording identifier | Identity of the recorded audio |
| Release-group identifier | Broader album concept |
The exact storage names depend on the tag format. Picard's mapping documentation distinguishes these identifier and position fields across formats. A recording ID placed into the album-ID field does not become a valid release identifier merely because both values resemble UUIDs.
Emby's Music Naming documentation explains its tag-based music organisation and identifies album IDs as relevant when same-named albums need distinction. That does not define the server's complete private album-key algorithm. It does explain why consistent product-level evidence matters.
An edition mismatch can hide behind good audio tags
Suppose disc 1 is tagged as the original album while disc 2 retains the expanded release ID and a slightly different album-artist spelling. The audio and individual song titles may all be correct. The album relationship is inconsistent.
A useful investigation asks which product each file is supposed to represent, then compares the fields that carry that meaning. Renaming every directory to Harbour Songs leaves the contradictory embedded identity intact.
Conversely, appending “Disc 2” to every album title can create two album labels when the intended product is one two-disc release. Multi-disc albums explains how the disc coordinate supplies that distinction without replacing release identity.
Keeping P1 and P2 separate lets the original artwork and three-track list describe one product, while the revised artwork and extra disc describe the other. The shared album concept remains available through the release-group relationship.
See Also
- Music release groups
- Music recordings and track occurrences
- Multi-disc albums
- Album artists and track artists
- MusicBrainz identifiers
- Release dates