When you begin building a computer music collection, it’s only natural that most thought turns to how you will acquire all the music that you want to own. However, it’s worth investing time in making sure your library can continue to grow while still remaining flexible to cope with changes in the music players you use and how you want to structure your library.
Time invested in the early stages of your computer music library can help its future expansion and flexibility in a few ways. Broadly speaking, I would encourage retaining data where possible, and that’s where the three ways of investing time in your collection are concentrated. Retaining data means your collection is more ready to meet the demands of the future, whether it be in the form of the actual audio you listen to or metadata sitting alongside the audio. This is because, importantly, you never know what data you will need in the future.
On the flipside to this, technical and practical necessities can encourage practices that lead to data loss, often implicitly and surreptitiously. The ripping of CDs to MP3 to support music players that do not play alternative formats and the re-tagging of music to support music players’ sometimes inconsistent interpretation of tagging rules are examples. Typically it’s the music players that force this upon us. This is unsurprising because the music player is king; it’s the mechanism by which music is played after all, and that’s the entire point of building a music collection.
The key is to balance the construction of your ‘pure’ music collection with the practicalities of music player support. First, I’ll cover three ways to invest in your music collection.
All your music should be purchased lossless. CDs were invented in the 70s and became available in the early 80s. You don’t want to step back thirty years do you? So if you are purchasing digital music and downloading it you want audio quality of at least that level. Buy FLAC, or Apple Lossless, or some other lossless codec.
Now, with ripping, let’s look at it from a value perspective. You’ve purchased a CD. You paid £5, $9, €12 for it… whatever. You want to get your money’s worth right? So it makes no sense to rip it and reduce the audio quality, literally throw away those bits and bytes you paid for.
It’s the future flexibility that retaining data, audio data in this case, buys. If you have that audio data you can always change music file format later, should you purchase a new music player with different format support. You can convert from lossy formats (such as MP3) but the quality will be far worse. And it’s not just music players; audio editors work better with more data.
Most tagging is fairly straight forward. Metadata (or ‘tags’) are associated with well-known fields. Album name, year, genre are all fairly self explanatory. But there are some tags that invite confusion.
Artist tagging is an example. The standard artist tag for a track should denote the artist of the contained audio, i.e. the artist of the track. This might be different to the artist for the release in which the track appears. For example, ‘featuring’ artists may appear in tracks for an album by a different artist (e.g. “Beyoncé featuring Sean Paul” on an album by “Beyoncé”) or, on a compilation, the release artist can be recognised as “Various Artists” but each track have a specific track artist. This release artist is known colloquially as album artist.
What’s important is that you treat each tag field consistently and the data entered obeys both syntactic and semantic correctness. By syntactic correctness I mean the form of the data is consistent with the same tag fields elsewhere. An example is track numbers which may or may not include a track total e.g. “1/14″ denoting the first of fourteen tracks in a given release. By semantic correctness I mean both the same field is used for the same conceptual idea throughout your collection and also the data entered obeys your own standards. For example, you may want to limit the number of genres in your collection, rather than grappling with a morass of high and low level genres.
Naturally, some of these approaches only become clear over time and with experience. Worse, when a change is required it can be quite a lot of work to implement it. Take the genre example: if you became a fan of Heavy Metal and you had previously categorised all such work as the more generic Rock alongside other albums you will need to examine each Rock album and recategorise those that apply to Heavy Metal. It’s for this reason that the most important investment you can make in your tagging is learning enough to be able to adapt to change. For this I suggest learning how to use your favourite tagging software and trying out different software to see if they allow you to exhert the control you need over your collection.
The world appears to be moving toward ever larger displays for playing music. In the early days of computer based music, the only large display you had when playing music was on your computer. Mobile MP3 players had tiny displays, as did dedicated networked music streamers.
Now we are seeing larger and larger displays. Smartphones, which by and large have replaced mobile MP3 players, have high resolution displays. Tablets are used to browse and select music. And music players are finally making their move onto the big screen too; integration into television sets are getting more and more common. Before this revolution, artwork sizes of around 300×300 were standard. Now, it feels like 900×900 is becoming a minimum.
The simple way to support this move is to strive for larger, high resolution, high quality album art. High resolution art can always be downsized, but stick with smaller art and upscaling it looks poor.
There are some music players, mainly old mobile MP3 players, that refuse to show artwork above certain resolutions. I still suggest you use larger format art in your ‘pure’ library, but coping with this is an example of a general question. How do you balance the theoretical ideals with actually making your music library work for you?
These recommendations could be identified as ‘purist’; that is, they don’t take notice of the foibles of your particular music player. For instance, ripping to FLAC is ideal in theory, but if you can only use iTunes, which doesn’t support FLAC, you’ve got a problem. I’ll be looking at ways of bridging the gap between music collection purism and practicality in future posts.