Tag Archives: Data visualization

From California to New York to Germany: A Brief Look at the Similarities and Differences

It’s been a bit longer than I anticipated since my last post. You can go ahead and blame the magnificent 15 day Australian honeymoon for that, I guess.

The previous post examined the core data that was only for California based festivals. This time around I’d like to look at how the database has begun to expand (there are 3,363 performances recorded in here as of this writing!) and include performances from all over the world.

Currently, there are a total of 29 different festivals represented in RMPDB. In addition to the California ones already mentioned, we now have the following as well:

  • Bang on a Can (New York, NY)
  • Donaueschinger Musiktage (Donaueschingen, Germany)
  • Ear Taxi Festival (Chicago)
  • ECLAT (Stuttgar, Germany)
  • Florida State University Festival of New Music (Tallahassee, Florida)
  • London Contemporary Music Festival (London)
  • London Ear Festival (London)
  • MATA (New York, NY)
  • MusicNOW Festival (Cincinnati, Ohio)
  • New Gallery Concert Series (Boston, Massachusetts)
  • Scale Variable New Chamber Music (Perth, Australia)
  • Wittener Tage für neue Kammermusik (Witten, Germany)

While the majority of data is still comprised of California festivals, there are 735 performances from other locations. The amount of performances from any one location outside of California is fairly small; however, we can begin to see some patterns emerge related to which composers are being programmed.

Note: The graph defaults to only showing 168 nodes. To see all of the data you can simply update the number in the top left of the window. Hover over a composer’s name to highlight the links to the festivals that have programmed their music.

As I examine this network a few different relationships begin to stand out:

  • There are definite overlaps between who is being programmed at any single location and other locations. No festivals (that have been added to RMPDB thus far) exist in a vacuum. Some of them, like those in Chicago, New York, Stuttgart, and Perth, program many composers not seen in other locations, but others, like Boston and Cincinnati (albeit with an extremely small number of performances represented here) fairly well overlap with others.
  • Despite the relatively close proximity of San Diego and Los Angeles (~120 miles) there is a fairly big difference between the composers programmed in each place.
  • Just looking at the graph, there appears to be a fairly large overlap between London performances and those in California (particularly Los Angeles).
  • Tallahassee is also fairly unique. Upon closer examination this is one of only two festivals in RMPD that are sponsored by a university. It appears as though some of my initial assumptions about university sponsored festivals are true in this case, in that it is programming a large number of works by its own faculty and students and only a smaller number by composers outside of their walls.

Implications to the collecting habits of music libraries:

  • There is a difference between which composers are being programmed that is, at least somewhat, related to geography. We should certainly do our best to present a balanced and global collection to our users, but local musicians are programming local musicians. This is the music your students are more frequently being exposed to when they go to professional concerts and what they consequently might want to play/study later. Are you collecting this local music or are you only focused on getting the “big” names?

An overview of RMPDB’s original (California only) data

Welcome to the Recent Music Performance Database’s (RMPDB) inaugural post! Today I’d like to very briefly introduce this project and then take a long look at the California-focused data that forms the initial core of this dataset.

RMPDB began as a personal collection development project to identify trends in the performance of contemporary music in California. Why California? Because I, Scott Stone, am the Research Librarian for Performing Arts at the University of California, Irvine (UCI) and decided that our collection would serve both the music needs of UCI and the greater community best by acquiring music that is making an impact in the local music scene. This certainly doesn’t mean we aren’t acquiring other materials, but that we are consciously making an effort to collect contemporary music that is of local importance. I decided to focus on music being performed at festivals and concerts series (hereafter collectively referred to simply as “festivals” ), because these type of events normally have some sort of Board and/or collective decision making process occurring. In my mind these pieces are more likely to reflect a broader swath of programming, instead of just the tastes of one person. That’s certainly an arguable statement, but it helps to keep the scope of this project defined. Feel free to visit the About page for more background of this project.

The data forming the core of this project all comes from festivals based in California. A quick overview of this data tells us it includes:

What struck me immediately when I began to look at (rather than just compile) the information was how few composers were being selected to perform multiple times (16.1% performed 5 or more times) and at multiple festivals (29.3%). When I think about purchasing sheet music for the library, this helps me begin to narrow my focus down to the composers who seem to be making the most waves and are regularly being selected for inclusion. There are, of course, other possible factors involved, most notably personal (e.g., many festival directors or curatorial board members might be composers who program their own music of that of their friends); however, this does not negate the fact that this frequently programmed music is more likely to be heard by the active contemporary musician who might then want to perform and/or study this music themselves. In my ideal world, they will turn to their local library to scratch this itch.

Analyzing large amounts of data can be difficult. As I’ve been working on better understanding what is being shown here, I’ve begun to experiment with different data visualization programs. Google Fusion Tables, while not extremely robust, do offer a nice network graph view that helps one to more easily discover the patterns and connections. Take a look below:

Note: The graph defaults to only showing 163 nodes. To see all of the data you can simply update the number in the top left of the window. Hover over a composer’s name to highlight the links to the festivals that have programmed their music.

Initial analysis of this data has already changed the sheet music purchasing at UCI. Composers who were not previously on my radar (e.g., Nicholas Deyoe and Christopher Cerrone) are now regularly being purchased and others, who were being purchased (e.g., Stephen Paulus and Vincent Persichetti), have since been removed.

Certainly much more could be said about this information, but let’s leave something for next time. Stay tuned for my next post that takes a look at the data currently in RMPDB after it has expanded to include festivals outside of California!