Organizing a Digital Music Library

Since I was 12, I’ve maintained a fairly impressive iTunes collection. What began as a place for me to digitize my CDs has grown into a fifty-plus gigabyte collection that is, frankly, a bit unwieldy to browse through. I’ve been thinking about the differences between having a digital musical library versus having a physical music library–one where you can browse through CDs or other methods of music storage, such as vinyl or cassette tapes–and I’ve considered a few differences overall.

  1. In a digital music space, organization is more fluid and more customizable.
    David Weinberger notes in his article “The New Order of Order” that iTunes taught us “the natural unit of music is the track” (para. 5). When you input thousands of songs into a digital repository, you can easily organize them however you wish–by name, by artist, by album, even by genre. This runs counter to having a physical music collection, where everything must be taken as a unit. If you have a copy of Coldplay’s A Rush of Blood to the Head, for example, you can’t just remove one song from the album and place it with others. Likewise, if you have a compilation album of various artists, such as the charity compilation Dark Is the Night, you can’t put songs by individual artists with other work by those artists; the album must be consumed as a unit. Both approaches have their advantages: digital organization helps users find individual similar items fairly easily, while physical collections allow users to see tracks as part of a whole, separate product.
    In addition, digital music libraries allow users to customize playlists–i.e. different groupings of songs based on how the user wishes to group them. I can have a “Workout” playlist that features songs I find good for running. What I think is good running music, however, might not be what someone else considers to be good running music. Playlists end up being hugely subjective and sometimes require guidance from the person who created them; a casual user who is not familiar with the library may need certain groupings explaining to them, as the logic may not be terrifically obvious to begin with, making the library difficult to navigate by themselves.

    Screen Shot 2018-09-24 at 22.36.54
    I know why I’ve grouped these songs together–for exercising–but a stranger would not be able to find a connection inherent between these songs.
  2. Audio quality can vary wildly.
    Like most young people who grew up in the 2000s, I did a good share of illegal downloading as a broke teenager with no money. (I don’t advise resorting to piracy the way I did.) If I couldn’t buy materials off of the iTunes store, I would end up resorting to Limewire, a filesharing service, with mixed results. Limewire was a true crapshoot in terms of what you could download, and as a result I have files that are .MP3, .WMA, .MP4, .AAC, and .WAV. To say they’re not consistent is definitely one way of looking at it. Some of these tracks are lossier than others; you can hear digital artefacts, like pixellated noise, or the audio gets tinny. In addition, some tracks that have been part of my music library for over ten years which were converted into Apple’s “”lossless” format (.AAC) have begun to glitch and erode over time. By contrast, the CDs that I have, for the most part, sound as good as they did the day I got them. I’ve run the risk of scratching them if I played them too many times in my stereo, but otherwise the audio quality has remained the same. It seems that digital music libraries may require constant upkeep and re-updating of files to ensure they do not become lossy over time in a way that physical forms of music may not require.
  3. Metadata becomes more customizable–and potentially nightmarish.
    When you digitize music, there’s usually a good deal of metadata that arrives with the music itself. If you buy from the iTunes store, for instance, the names of songs will already be filled out, as will the name of the artist, the album title, any relevant composers, and even genre. There is usually a space for album artwork and lyrics as well, should you choose to add those. But is there a good, coherent way to organize music digitally by genre in a way that will make sense for casual users?
    Musical genre can sometimes be a minefield. For music aficionados, it becomes a question of whether or not music should be classified broadly (i.e. “rock”) or by narrower subgenres (i.e. “alt-folk,” “acid jazz”). Genre distinctions can be difficult to narrow down, especially in a formal sense. Kulczak and Lennertz Jetton (2011), for instance, note that MaRC (a formal library cataloguing structure) is geared towards classifying classical music rather than pop music, and that LCSH subheadings for newer music genres, like EDM, may not exist yet (218). In addition, users may not agree as to what genre(s) a piece of music may belong to–there’s usually a general consensus among the broader category, but the level of detail is subjective to the user (230). This presents an additional challenge for users who may be navigating an iTunes library with little knowledge of genre, especially if the person who has organized the music library decides to use narrower terms for genres. The controlled vocabulary for a person’s individual music library is going to differ based on their preferences and how much of a music nerd they are; however, it can also make searching within the library difficult if a casual user has no idea what to look for.
    In contrast, a physical music collection only has the metadata that is presented on the CD/vinyl case–nothing more, nothing less. To some degree, that makes it standardized; a hard copy of an Adele album will always have the same tracks listed in the same order and will probably include the composition credits in a leaflet inside the CD without additional frills. Browsing through a physical music library may help someone who is fairly new to music and who does not want to be overwhelmed with information all at once, while browsing a digital music library may be better suited for those who are more familiar with the collection.

Although I know we will be building digital libraries that are different from an iTunes library, I cannot help but think about how I’ve organized my own digital music library over the past few years. Other people must also have music libraries that only make sense to them. It’s fine for personal use, obviously, but not an ideal way to organize a formal digital library for unfamiliar users to come in and browse. Clearly, I’ll need to take a different approach when I begin designing my digital library for this course.


Kulczak, D. E. & Lennertz Jetton, L. (2011). “Lexicon of love”: Genre description of popular music is not as simple as ABC. Music Reference Services Quarterly, 14(4), 210-238.

Weinberger, D. (2007). The new order of order. In Everything is miscellaneous: The power of the new digital disorder. Retrieved from

3 thoughts on “Organizing a Digital Music Library

  1. Great blog post, Alex! I loved the detail you went into, and how you laid it out. It was understandable and crystal clear. I liked how you used the playlist example to demonstrate the differences between digital music libraries and physical music libraries, it was easy to follow along with, and also made me realize that I’m not exactly a playlist person. I just hit shuffle, and go with the flow. And by go with the flow, I mean skip 200 songs until I find the one I want to listen to. I guess that makes me incredibly lazy? My brother has a playlist for everything, though. We drove up to Nashville this past weekend for an Anderson East concert, and since he was driving, he controlled the music. His current driving playlist is smooth jazz.
    I’ll be honest.
    I felt like I was sitting in a Starbucks. It was so not fun. But hey, he likes to listen to jazz as he drives, I would have subjected him to Queen or Bastille if I’d been driving, so. But like you said, digital libraries give us the freedom to organize music the way we want to organize it, and create these unique combinations that aren’t possible with physical CDs.
    I was surprised when I was l looking for articles to cite to find just how much there was to say about the subject of digital music libraries. Obviously I knew that people would talk about it, and how to organize them, but I didn’t realize how complex the subject really was. Then again, I don’t really think about digital libraries on a daily basis (yet, that will definitely change), so it isn’t surprising. A paper I found quite interesting was Sergio Oramas’s paper. He mentioned a digital music library where one could search for songs that were written by composers in 18th century Vienna. Or any other time or place imaginable. No such library exists yet, but the technology is there. It just needs to be applied to digital music libraries. I think that that could be a really awesome way to find music. It’d introduce people to new music and new artists, which is always good!
    Your blog post was absolutely fantastic, and I loved seeing a sample of your workout playlist (which is awesome). There is a lot to take into consideration when deciding between a physical or digital music library, and you did a great job of showing the pros and cons of both. I am looking forward to reading your next blog post!


    Sergio Oramas, & Mohamed Sordo. (2016). Knowledge Is Out There: A New Step in the Evolution of Music Digital Libraries, (4), 285. Retrieved from


  2. Thanks so much for the article! I would love to see that technology take off and applied to digital music libraries–it would expand the way that you could perform known-item searches. I wonder if/how they could also apply the technology to music recommendation services? I guess Spotify/iTunes might have some of that already covered, but for other repositories I’d love to see that in action.

    (Also I use a lot of shuffling too; technically I have over 700 songs on my workout playlist but I skip some every now and again because I’ll have heard it three workouts in a row, or I’ll also skip through 50 songs until I find the perfect one. How on earth do I have over 7,000 songs in my library but I only listen to the same few over and over again?!)


  3. Listen … I feel this post in my soul. I was thirteen when I first got iTunes, over a decade before I learned what metadata actually was, but I already had the beginnings of my own ‘standard vocabulary’ as I imported my little collection of CDs on my brand new Dell desktop computer. I also did a lot of, uh, nonstandard file acquisition back in the day, and even now, there are some songs I’m not actually able to find purchase links for, or that creators offer for download from Soundcloud or other hosting websites. You’ll appreciate this – a few months ago, I got my entire VIXX and BTS discographies and standardized the way they displayed in my Google Play library, which took me like two hours but felt so good. I made sure that all the titles were set up with the English translation, followed by the Korean title (if applicable) in parentheses; all album titles were the same way. I didn’t include hangul in the artist names, and I got rid of all the other extraneous genre stuff … it looks beautiful.

    Now that you’ve made me think more critically about this stuff, I did some SCOUTing, and I thought you might find this article as interesting as I did – It’s from 1989 (the year, not the album) so it’s not the most up-to-date contribution to the conversation, but it talks about natural-language vocabularies vs controlled vocabularies, which I think is pretty timeless. It’s the balance we’ve talked about in this and a lot of other classes, regarding controlled vocabularies – how do you build a vocabulary that facilitates searches performed by people who aren’t trained in it, but still has the degree of specificity necessary to be a meaningful form of organization? It’s extra difficult with genre, like you talked about, since ‘rock’ means something different to everyone, but ‘acid jazz’ means nothing to uninitiated people such as, for example, me.

    The paper pretty much concludes that Research Should Be Done, and I’m sure that’s happened, but I’d have to dig more to find the results of it. I’m curious whether or not iTunes follows a specific vocabulary when they compile their own metadata. I’m gonna google that later, too.

    Also, your workout playlist looks awesome.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s