Difference between revisions of "User:LordSputnik/Recording Style Guidelines"

From MusicBrainz Wiki
Line 10: Line 10:
 
|jira=STYLE-159
 
|jira=STYLE-159
 
}}
 
}}
 
 
__NOTOC__
 
__NOTOC__
 
__NOEDITSECTION__
 
__NOEDITSECTION__
 
 
=Introduction=
 
=Introduction=
 
This proposal is designed to improve the guidelines on merging recordings. It is felt that the current guidelines are too vague, and that they don't cover all cases where recordings might be merged. This proposal seeks to expand these guidelines and make them clearer.
 
This proposal is designed to improve the guidelines on merging recordings. It is felt that the current guidelines are too vague, and that they don't cover all cases where recordings might be merged. This proposal seeks to expand these guidelines and make them clearer.
Line 21: Line 19:
  
 
==What should and shouldn't share recordings?==
 
==What should and shouldn't share recordings?==
A recording is intended to represent a unique audio source. This means that generally, two releases should only share recordings if the audio source used to create them was the same. An audio source is considered unique as soon as any changes make it different from a previous source.
+
A recording is intended to represent distinct audio. This means that generally two tracks should only share a recording if the audio that they contain sounds the same. If it isn't clear whether to merge two recordings or not, keep them separate. Where two recordings of the same work are separate for a reason, a meaningful disambiguation should be provided for one or both of the recordings.
  
 
===Specific Cases===
 
===Specific Cases===
#If multiple performances of a single work are recorded and released then each performance is a separate audio source, and each released performance should be given a separate recording.
+
#Tracks containing different performances of a single work should each use a separate recording, since each performance will result in distinct audio.
#If there are multiple releases of a single live performance and each was recorded by a different device, each device is a separate source and each release should have its own set of recordings.
+
#Releases of live performances that are recorded by different recording devices should each use a separate set of recordings.
#If a CD or other digital release contains audio that was originally recorded and released in an analogue format, the digital release should be given a new set of recordings. The original analogue source was changed when it was converted from analogue to digital.
+
#Versions of a release that are mastered in audibly different ways should use separate recordings.  
#If a release is recorded digitally and released in both analogue and digital formats then the releases should share recordings. The audio sources for all releases are the digital recordings.
+
#If audio restoration is used to create remastered audio from an earlier recording, a track containing the remastered audio should use a recording specific to that remaster.
#If two versions of a release are mastered in audibly different ways, then the source has been changed for each mastering, and each release should have its own set of recordings.
+
#A track containing edited or remixed audio should always use a recording specific to that particular edit or remix.
#If a release is remastered from an earlier release, all the tracks on the new release should be new recordings. The source was changed when the recording was remastered. If another remaster is made, then the source has changed again, and another set of recordings should be made.
+
#Tracks which contain different numbers of audio channels should always use separate recordings. For example, a mono recording and a stereo recording should never be merged.
#If a recording is edited or remixed, the new version should have a separate recording. The exception to this case is for changes that only affect the length of the fade at the beginning or end of a track, where a new recording should only be used when the difference in length is greater than 10 seconds.
+
#If a track begins or ends in silence and a second track changes the length of the silence, but is otherwise identical, these two tracks should share a recording, so long as the silence isn't a significant part of the audio.
 
 
  
In addition to these cases and the general guidelines, recordings with multiple, visibly different AcoustID fingerprints should not be merged. Instead, the recordings and AcoustIDs should be [[Guides/AcoustID|examined for correctness]].
+
===AcoustIDs and ISRCs===
 +
Both of these identifiers can be incorrectly attached to recordings. For this reason, neither AcoustIDs nor ISRCs should be used as the sole reason for merging two recordings. Different AcoustID fingerprints and ISRCs may be used as a reason for not merging, but bear in mind that they can sometimes be wrong.
  
In theory, recordings which have different sources should be assigned different ISRCs. In practice however, this doesn’t always happen. Recordings with different ISRCs should be examined closely to determine whether they are actually different. Similarly, recordings with the same ISRCs may in fact be different, so care should be taken before merging them.
+
Before attempting to work with AcoustIDs, please read the [[Guides/AcoustID|AcoustID Guide]].

Revision as of 14:37, 5 November 2012


Status: This page describes an active style guideline proposal and is not official.



Proposal number: RFC-Unassigned
Champion: LordSputnik
Current status: RFC
Initial Discussion


JIRA ticket STYLE-159


Introduction

This proposal is designed to improve the guidelines on merging recordings. It is felt that the current guidelines are too vague, and that they don't cover all cases where recordings might be merged. This proposal seeks to expand these guidelines and make them clearer.

Proposal

Edit http://musicbrainz.org/doc/Style/Recording to change the section "What should and shouldn't be merged together?" to the following.

What should and shouldn't share recordings?

A recording is intended to represent distinct audio. This means that generally two tracks should only share a recording if the audio that they contain sounds the same. If it isn't clear whether to merge two recordings or not, keep them separate. Where two recordings of the same work are separate for a reason, a meaningful disambiguation should be provided for one or both of the recordings.

Specific Cases

  1. Tracks containing different performances of a single work should each use a separate recording, since each performance will result in distinct audio.
  2. Releases of live performances that are recorded by different recording devices should each use a separate set of recordings.
  3. Versions of a release that are mastered in audibly different ways should use separate recordings.
  4. If audio restoration is used to create remastered audio from an earlier recording, a track containing the remastered audio should use a recording specific to that remaster.
  5. A track containing edited or remixed audio should always use a recording specific to that particular edit or remix.
  6. Tracks which contain different numbers of audio channels should always use separate recordings. For example, a mono recording and a stereo recording should never be merged.
  7. If a track begins or ends in silence and a second track changes the length of the silence, but is otherwise identical, these two tracks should share a recording, so long as the silence isn't a significant part of the audio.

AcoustIDs and ISRCs

Both of these identifiers can be incorrectly attached to recordings. For this reason, neither AcoustIDs nor ISRCs should be used as the sole reason for merging two recordings. Different AcoustID fingerprints and ISRCs may be used as a reason for not merging, but bear in mind that they can sometimes be wrong.

Before attempting to work with AcoustIDs, please read the AcoustID Guide.