The term metadata has existed as aphantompresence in the music industry for a while now, but the exact nature of this data is still a mystery to many musicians, labels, and consumers. What exactly does metadata mean, what is this information for, and where isit stored? We decided to take a closer look at the issue.
What is metadata?
In general, the term metadata (or meta information) refers to data which shows the characteristics of a specific piece of data, but is not actually the data itself. It is „data about data“, which sounds more complicated than it really is. In music, for example, a piece of audio data is accompanied by metadata giving additional information like the artist name, the song title, the composer, the rights holder, BPM data, or an ISRC. In this context, there are both primary and secondary (i.e. less important) data sets.
The term metadata can cover large collections of data like documents, books, and databases. The features of a single object (like a personal name) are defined as this object’s metadata.
Although the use of the term in this way is fairly new, meta information already existed in early libraries for hundreds if not thousands of years. It is simply a process of categorization, which enables people to sort through data, and find things more quickly.
Strangely enough, even indie labels often don’t know what ‚label copy’ is. You should make sure that you find out about this when you first licence your music as a label or an artist (e.g. for a compilation with other works), otherwise you’ll start to feel pretty stupid! Label copy is a collection of important metadata, including the following information:
Artist, song title (mix), ISRC, composer/author, publisher, producer and rights holder (defined by C and P – see here for more information on composition and phonogram).
Without this information you can’t really enter the system and start to sell your product. The distribution of profits has to match this data, and be correctly assigned to the corresponding rights holders. This means that the information you give has to be as complete as possible, and filed properly. Missing data is a no go!
If you’d like to see what label copy looks like, just check out any CD booklet or old vinyl sleeve.
The digital age has made it easier to attach metadata to data. You can automaticallycode meta information to be an essential component of the whole. However, data must also be saved in a system somewhere. This is not just for accounting purposes, but also to ensure that you have an overview of everything. Labels and publishers have to have some kind of corresponding software or databank for this purpose. Metadata has to be prepared for inspection at any time, and correspond with international standards.
When it comes to commercial CDs, all information is coded into the CD text („red book format“). Information about all interpreters and titles, ISRCs (International Standard Recording Code, see: http://de.wikipedia.org/wiki/ISRC) and catalogue numbers (UPC/EAN) must be included. An ISRC code is uniquely assigned to each track, meaning that every recording has a clearly verifiable owner. The IFPI grants ISRCs to labels after a one-off payment. The label then numbers their releases independently. Every song must also contain an ISWC (http://www.iswc.org/), describing the composition and not the recording. These codes are primarily used by copyright collection societies like GEMA, and music publishers.
Data in the World Wide Web
There are certain standards which apply to digital databases in the world wide web, which are either decided by the affiliation of different companies or individual organisations, or defined by particular companies like Google or Apple. Collection societies and publishers are trying to create a standardized system in this area. The fact is, there was too much self sufficiency in the previous methods of operation. This led to a data flood of chaotic proportions, which then had to be fixed as quickly as possible. This is why we need to have these common markers.
Google also plays an important role via YouTube: the video platform has become the top streaming service for music, and it’s not only ‚kids’ who use YouTube specifically to find songs. In YouTube’s CMS, fingerprinting is used on audio recordings and synchronized with other recordings. Other data (like composition information) is combined with sound recordings. In the best case scenario, all of a song’s data is included at the sound recording and composition levels. However, due to the fact that these different levels are usually separated and originate from different parties, there are often holes or missing links. It is therefore important that labels, distributors, publishers, copyright societies and technology firms work together with standardized levels to create complete and correct records, which are visible to all. Musicians who want to make a digital release should also try to provide data that is as complete and accurate as possible. This is of course in the musician’s own interest too!
The Invisible Contributors
One negative side effect of modern digital data is the fact that people like producers, arrangers and musicians become practically invisible. Something that was impossible in the era of CD booklets and vinyl sleeves has now become a problem in the digital era. “Credits” for audio data have practically disappeared. The film industry has not been affected by this problem, as credits are still shown on digital copies of movies, and are still prominently placed. The opposite is true for audio files these days, which usually just show the interpreter, the song title, and the label. Voices will have to be raised in protest in order to address this situation!