Difference between revisions of "History:What Defines A Unique Release"

From MusicBrainz Wiki
Jump to navigationJump to search
((Imported from MoinMoin))
(ARRGG :) (Imported from MoinMoin))
(No difference)

Revision as of 16:13, 24 May 2007

Status: This page is part of the set of MusicBrainzPhilosophy pages, which discuss the rationale behind metaphysical decisions/suggestions in regards to MusicBrainz data.

New Release, or Variant of Another Release?

When does a work actually become a "separate record" from MusicBrainz point of view? This is a very complex issue which relates to some of the other aspects above. MusicBrainz has been widely hailed for its GUID approach to identifying music, but what does this exactly mean? Does the identifier refer to a specific version/mix of a work, to specific printed concept of it or just to a generic song? To what degree is it reasonable to guarantee/expect this to hold?

If a release is re-released with modified track titles but otherwise same content, is a new identifiable work created? This would make things like linking to merchandise databases easy.

If a release is re-mastered with everything else staying the same (Ie. remove tape hiss, normalize volume etc.) is a new work created from MusicBrainz point of view? This might be desirable if you're interested in finding the specific high-quality version.

If a track is cut by taking out some ending or outro, is a new MusicBrainz identifiable track created? Collectors and fans might be very interested in tracking these alternate versions.

If a track is re-performed and recorded with basically same setup and sounding roughly the same, is a new work created? People who're just interested in having the same experience/listening to the specific track (Ie. the average music listener) might find a system where these have the same identifier most convenient.

I believe presently MusicBrainz creates a new unique identifier for each instance of the same track on different releases by neccessity, and the track title is not sufficient to link tracks between releases. This is suggestive of a policy where very small changes create a new work; but on the opposite end of the spectrum merging releases is often encouraged when releases have either same track lengths and number or same titles. This approach may lead to re-recorded tracks having effectively same identifiers.


  • Very interesting. This, along with the concepts of VirtualDuplicateRelease, essentially describes the exact argument I've tried to make in Nirvana recently, as I tried to explain why a bootleg soundboard recording of a concert remastered in 2004 ought not to be merged in with a pressed cd bootleg sourced from a audience recording and published in 1994, just because the tracklists match. -- BrianSchweitzer 16:21, 23 May 2007 (UTC)

Authors: --MatthewExon