How Does MusicBrainz Work?

Status: This page is part of the effort of RestructuringTheDocumentation. It is a wikified version of this static page. The text is pretty old (it talks about Napster!) but the concept is still the same. Also the page NeedsIntertwingling.

General Information

The MusicBrainz service is broken into two components: The MusicBrainzServer and the client. The server contains all of the metadata and can be searched and browsed via a normal web browser. The server also talks to MusicBrainz clients. A MusicBrainz client could be an MP3/Vorbis/AudioCD player such as Zinf, WinAmp, XMMS or Sonique. A client could also be a service like FairTunes which could use MusicBrainz in order to properly identify the Artists that should receive money. The MusicBrainzTagger is also a client to the server.

In order to have complete MusicBrainz support each of these clients can use libmusicbrainz (supported on Win32, Linux and various other UNIX variants) which is released under the [ http://www.gnu.org/copyleft/lesser.html GNU Lesser General Public License] which allows closed source applications to link to the binary form of the MusicBrainz client library. There are no royalties or other fees charged for this.

Music Identification

Right now Napster does not have a clue what music people are actually trading. The files are named by the people who ripped and encoded the file in the first place, and each person has a slightly different way of doing things. Consequently, when a user searches for music on Napster, and they are looking for "Billy and the Boingers" they won't find tracks by someone who erroneously named their files after "Billy and the Bongers". However, MusicBrainz is able to provide a unique ID for each track, then a user will be able to search for that unique ID, rather than searching by track name.

• What are these ID's? ReleaseID, ArtistID and TrackID I suppose--DonRedman

Of course there are two sides to the coin. The RIAA will also be able to identify the tracks of its member companies and then be able to go after people who pirate these tracks. This technology will be created, whether or not we like it -- if we can create the technology and create it in such a fashion that it benefits everyone, then the Internet community wins.

Another group of people that will win with this system are the Artists. With the use of MusicBrainz, services like FairTunes will be able to create a precise system that will allow users to compensate artists for the music that they may have downloaded without permission from the record labels. This way the people who love music can give money to their favorite artist, while cutting out the record labels.

Playlist Exchange

As mentioned above, people who encode MP3 files tend to use their own favorite file naming scheme. This results in the fact that playlists cannot be exchanged between users, unless they have exactly the same naming convention and the same music collection.

These facts pretty much prevent a lot of people from swapping playlists. Using MusicBrainz unique IDs, intelligent MP3 players will be able to create playlists that contain all the metadata about the playlist as well as the unique IDs for each one of the tracks. If user A passes an intelligent playlist to user B, user B's player can play the playlist because it will be able to cross reference the items in user A's playlist with the tracks in user B's collection. Furthermore, user B can choose to have the player purchase the missing tracks, or download the tracks from a file swapping service such as Napster or Scour, and then have the player automatically tip the artists using FairTunes.

This creates a powerful system that seamlessly enhances the digital audio experience and actually funnels money to artists. AudioScrobbler already uses the MusicBrainz ID's

MusicBrainz MP3/Vorbis Support

Relatable.com has recently (recently?) integrated their audio track recognition software into FreeAmp 2.1 beta 6 and higher. Anyone that opts in to use the Relatable features will also be automatically contributing their metadata to the MusicBrainz server.

If the MusicBrainz server ever sees the same metadata and track signature twice, it will accept the metadata into its database. With enough people using FreeAmp with the Relatable features enabled, this server should be automatically populated with data!

This process of gathering data will be prone to errors. This process is going to get coupled with a data editing/moderation system in the near future. I believe that if we have the basic data in place, we can appeal to the Open Source community to help clean the data up and build a useful open metadata server that everyone can use to aid our enjoyment and discovery of digital audio.

Audio CD Support

To find out how the CD lookup portions work, please refer to DiscIDCalculation.

CategoryDocumentation