, , , , ,

After several courses of digital humanities, we are all very familiar with idea of transforming miles of archives in Venice’s library into an information system. Somehow you might not find the history of Venice very attractive, but how about the digitization of your favorite songs sounds to you?

The digitization of music is not only just about recording songs in mp3 format. If you think about the idea of digitization, it is about the transformation from an analogue media into a universal computer readable language. In this context, the mp3 format is just one of the languages we used to digitize sounds. We also need other languages to describe musical scores, and more importantly, to convert both scores and sounds into searchable, editable and usable data. Currently, there exist some nice websites where you can view, share, edit and even play (if you pay!) the sheet music online. But these websites are mostly built by music lovers, and both their source and their usage are quiet limited. Therefore, this article will discuss the current trends in a large collection of sheet music as well as how the usage of these data.


A online guitar TAB Player

Apparently, the collection of scores will be very different from the collection of books. The first problem is the identification of music scores. How can we tell that we are scanning a real music sheet instead of a painting or a map? In the abstract “SIMSSA: Towards full-music search over a large collection of musical scores”, researchers of McGill University declared that they have developed a new and efficient algorithm which can identify 98.7% of musical archives during their preliminary study.

After successfully collecting these scores, researchers at McGill University just index the information of where they have found these scores. This might works well at a laboratory level, but a large collection requires a well-organized database to store all the found data. This is especially important if you want make them searchable. Instead of the traditional database model and common metadata schema, researchers of Trinity College Dublin decide to use the Linked Data technologies. In the abstract “Linked Data for Music Collections: A User-Centered Approach”, they introduce this technology as a more efficient structure for the database of music archives as well as a more convenient way of interchange between external data and music information. In fact, they have applied some ideas of this new technology on the Contemporary Music Center in Ireland to make their database more efficiency.

Once the scores are well stored, people are finally able to use these data. If we look back the first abstract, thanks to their new optical music recognition software, those researchers have already developed several web-based prototype systems which can perform musical research and retrieval. But the ideas behind their “Single Interface for Music Score Searching and Analysis” project are more than the collection of musical archives, they want to present a dynamic and interactive environment of digital music creation and edition so that musician will have much more access to massive scores and powerful tool to analysis them. That is why they are working with the Music Encoding Initiative, an international group who are trying to create a commonly-accepted, digital representation of music notation documents. Comparison of different kind of musical coding environments is covered by another abstract “Live Coding Music: Self-Expression through Innovation”. Although this abstract focus more on live coding,  it does illustrate the different environments that codes use during musical coding.