Qwertyuiopia

Why: An introduction to the benefits of organizing your digital music into a media library

Maintaining a library of hand-sorted categories (such as “Alternative” or “Jazz”) may be applicable to a small, 20GB collection (of 300 albums or so), but when things go up in volume (as in a 200GB hard-disk-ful of music) you just realize that browsing your music through the folder structure, let alone categorizing by hand, would just be too time consuming and inefficient. After all, you’re probably using a computer to browse through your music and those devices are made to help out with such things .

This is the first of a series of articles on building a music library that is suitable for browsing through various library-capable audio players (Such as Winamp, Quod Libet, Rhythmbox, iTunes and a lot more) and consistent, thus capable of having its stucture manipulated with scripts. In this introductory part we cover some of the reasons to want a music library and the work involved.

…yes, we all used to browse that way

In the last years the compression of digital music (starting with the mp3 file format) and the evolution of the internet have drastically changed the way people come by music. A lot of free-to-listen music is available through innovative internet music companies (such as Jamendo, or Magnatune), creative commons material can easily be obtained (such as the music on Archive.org) and pay per download services are also available. Furthermore, those with a taste for piracy can very easily find what they’re looking for, since it’s a very widely supported movement and network speeds are better than ever.

A direct consequence of this rise of availability in music is a geometrical increase in the quantity of material a person ends up with. Put simply, when you start collecting digital music, you end up having a lot (a lot) of stuff and you need a way to organize it so you can find what you are looking for.

Library-based players

A media library is a way of browsing your music. What the “library” part means is that the player reads through all of the music in a particular directory (the one with your music collection) and makes a database of all the information in finds, so you can easily search and browse it. This makes browsing way quicker than going to the folder with your file manager, since your computer doesn’t have to search through a whole hard disk to find the music you are looking for, but rather through a small file containing that information. Furthermore, it makes it possible to quickly search the whole library for any particular bit of information associated with the specific track, such as the artist name, the album or even the date music was released.

The screenshot below shows the paned library browser in iTunes. The top left pane shows artists, the top right albums and the bottom one shows tracks. When you click on an artist name, the other panes filter the results to show albums and tracks by that artist. On the top there’s a search bar. Winamp also uses the same paned browser layout. For linux users, Quod Libet has an even nicer (in my opinion) “Album list”, where you scroll through the albums seeing their covers (as seen on the first screenshot). In the search bar, you can also use advanced search expressions (in Quod Libet, but I guess the other players can probably do it, too).

So, I just feed it a folder?

One downside of this method is that if not all of your tracks have a specific attribute filled in, then searching for that type of information will not be very useful. For example, there’s not much point of searching for albums from 1974 if not all of your music has date information, as you will miss a lot of 1974 albums that just don’t say it anywhere. Even worse, if some songs don’t have artist or album information tagged in the files themselves and not only in the folder structure, they might be near impossible to search for. So, in order for a library-based player to really be of use to you, an important step is tagging your digital music into a music library.

What distinguishes a music library from a collection of digital music is consistency on naming conventions, folder and file stucture and -most importantly- selected fields of information. This means that if a folder is in your music library there’s a record in it (and maybe its album cover(s)), and that this record is whole and must have specific tags filled in. Obviously the tags you’ll want to consistently have vary depending on how much time you’re prepared to spend looking for that information, and a library can be moderately functional even with only as few information as artist, title, album and track number. You can also keep musical categories (if you’re into it) by entering a category in the “Genre” field.

To help you with that task there are many different Tagging applications (like Tag&Rename, EasyTAG, Ex Falso, …) both open source and proprietary, and there will be a whole part about them later in the series. Those applications can commonly write meta-information tags to files, as well as manipulate the folder structure (moving and renaming the files to directories with names corresponding to their tags). Also -very conveniently- they can automatically retrieve that sort of information from free internet databases.

Going the extra mile, though, can make a difference as having a few other important fields and specs in mind enables you automate a lot of the tagging process, as well as grant you considerbly more capabilities in manipulating and maintaining your library. A really nice example is being sure that all of your track numbers are in an xx/xx format. That means that every track contains the information of how many tracks the complete album has, together with its place in that album. If all the files have the field this way, then a script can be written to count the files in each folder (presuming each folder is an album) and compare them to the total tracks. This way you can automatically check which albums are complete, even after you lose files on a hard disk failure (or if your kid brother eats them or something).

This series of articles is intended to walk you through the main capabilities and limitations of the meta-information formats, the players and the hierarchical file system structure as they apply to digital music files, in order to build a really consistent music library that can then be manipulated in various ways.

In the next part of the series we go over some basic concepts of encodings and meta-information formats.

This entry was posted on Monday, December 3rd, 2007 at 2:23 and is filed under Tutorials. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.