Class meeting #5 – Collaborative music-making – Monday 10/1

Readings:

Topirceanu, A., G. Barina, and M. Udrescu. 2014. “MuSeNet: Collaboration in the Music Artists Industry.” In 2014 European Network Intelligence Conference, 89–94. https://doi.org/10.1109/ENIC.2014.10.
Glowinski, Donald, Maurizio Mancini, Roddy Cowie, Antonio Camurri, Carlo Chiorri, and Cian Doherty. 2013. “The Movements Made by Performers in a Skilled Quartet: A Distinctive Pattern, and the Function That It Serves.” Frontiers in Psychology 4 (November). https://doi.org/10.3389/fpsyg.2013.00841.
Rose, Stephen, Sandra Tuppen, and Loukia Drosopoulou. 2015. “Writing a Big Data History of Music.” Early Music 43 (4): 649–60. https://doi.org/10.1093/em/cav071.

Response:

In a short (150-200) response to these three papers, think about how of the methodologies that are described could be used to study some context for collaborative musical creativity (e.g. group performance, online musical collaborations, mashups, classroom music making, music therapy) of your choosing. In your answer, assess the applicability of the models that are described in each paper to your chosen area.

Whiteboards:

7 thoughts on “Class meeting #5 – Collaborative music-making – Monday 10/1

  1. yh2825

    The three papers discussed various analytical aspects into collaborative music making – including gesture analysis on performers, network of music producers (even across genres) and a historical perspective of music publication. A modern example that I could think of is a music product called Poputar (https://www.poputar.com/en/). It is a hardware product resembling guitar or ukulele which helps amateurs to learn playing the instrument, while supplementing the user experience with a software app – essentially a social network for music amateurs, professionals, and influencers to share their work, make music educational content, and even collaborate with one another.

    Through this product, it becomes more intuitive to see how the methodologies mentioned in the papers could be applied in real life – the software could of a network of collaboration, or expressed interests among the users of the product. For example, it is possible to generate a “genre” map through the data provided by the software. The guitar instrument has its place among a variety of music genres, and through studying some user level behavior metrics across different genres (time spent on learning music pieces, following of key influencers), we can have an overall view of the genre preferences of guitar amateurs – an example would be by looking at the time spent on learning different music pieces, we can have a sense of the compositional complexity of guitar across different genres.

    The gesture analysis could be even made easier – the product itself has LED lights installed on the hardware which are controlled by the software indicating where and when the player should press on the strings. If the data is recorded repeatedly, first of all we can easily have a sense of how the guitar learner improves over time. And with that, combined with the methodologies mentioned in the third paper, we can ask even more interesting questions – how does the performer’s gesture change when they improve along the way, do they become more relaxed and freeform? Or what are the gesture differences between performers that are very accurate on beat and those who don’t?

  2. clj2142

    The methodologies described in these three texts could all be applied to the study of sampling in hip hop music. Songs sampled in hip hop span across many different genres and decades. Big data analysis could be used, for example, to chart the frequency of samples of any genre (ex: electronic) over the course of many decades. It could also do the same for the frequency of samples from a particular song or a particular artist. Social network modeling could be used to examine the relationships between artists who use the same sample, which could bring more understanding to why certain songs have been sampled hundreds or thousands of times by hip hop artists. Often, hip hop artists will perform songs without the presence of the sampled artist, and sometimes, usually depending on the length or musical importance of the sample, sampled artists will perform songs they are sampled on without the original artists. The movements artists perform in these performances can be compared to performances where both artists are present, and their movements can be analyzed to see what movements serve as communication between the two artists versus expressiveness.

    1. clj2142

      EDIT: The methodologies described in these three texts could all be applied to the study of sampling as well as features in hip hop music. Songs sampled in hip hop span across many different genres and decades. Big data analysis could be used, for example, to chart the frequency of samples of any genre (ex: electronic) over the course of many decades. It could also do the same for the frequency of samples from a particular song or the frequency of features of a particular artist. Social network modeling could be used to examine the relationships between artists who use the same sample, which could bring more understanding to why certain songs have been sampled hundreds or thousands of times by hip hop artists. It can also be used to represent collaborations between artists, drawing connections between different artists that have featured other artists. Often, hip hop artists will perform songs without the presence of the featured artist, and sometimes, though sometimes hip hop artists. and sometimes, usually depending on the length or musical importance of the feature, featured artists will perform songs on which they are featured without the original artists. The movements artists perform in these performances can be compared to performances where both artists are present, and their movements can be analyzed to see what movements serve as communication between the two artists versus expressiveness.

  3. lnl2110

    All three of the methods could potentially be applied to studying music training at early ages, especially music schools that have established training instructions so that pedagogy is standardized across different areas. The graphing methods that Topirceanu uses for different measures in social networks could be used to gather interesting information about how training techniques are spread across geographical areas—within a city, schools form teaching communities that might create certain teaching methods that are significantly different from schools that are further away, for example. However, since teachers move and might have been taught in different places, they might bring different methods from where they were trained. Also, early music training can involve a lot of group activities between the kids, who learn from each other by exposure and because they play together. These types of music communities could also be interesting for “Big Data”—the Suzuki method, for example, comprises a large set of books for different instruments, and their distribution could be compared to other types of training books, or even early classical music training that is done without books, for toddlers who have not chosen instruments. Finally, I believe that a lot of times, affect studies do not have controlled ways of studying movements of professional musicians, because they cannot separate the way they play the same ensemble piece from their previous experiences playing the same pieces in an ensemble. Therefore, it might be more interesting to study the movements of young kids when they play alone, and how their movements develop differently as they are trained to play in large groups like in group classes.

  4. erc2175

    After thinking for a while about collaborative musical experiences of my own that I’d be interested in studying, I settled on marching band as a subject of particular interest. Obviously, marching band is a field of considerably less “personal expression” than “group expression,” which makes analysis of individual movement difficult; here, entropy would mostly result from errors, rather than moments of changing expression. However, it might be interesting to examine entropy over a long series of performances, or between different groups, to determine how marching bands with different “characters” treat or deploy entropy in their movements. I suspect there may be enough marching band music in existence for Big Data treatment; since marching band music is a relatively recent phenomenon (at least not emerging until less than 200 years ago), I would be very interested in studying trends in publishing using Big Data strategies. The social world of marching band is probably somewhat harder to examine in the terms deployed by the MuSeNet team, since each marching band exists as a single unit, but there are ways of systematically examining social connections between members of different marching bands (for example, Facebook friendships); it may be interesting to compare such networks in high school vs. college bands, for example, or amateur vs. professional bands.

  5. ijg2112

    All of these analysis methodologies could be applied to collaboration within the Electronic Dance Music genre. Opportunities for collaboration within EDM can span from ‘flipping’ another artists song and posting it on SoundCloud to the production of a huge live show with a dozen different artists performing simultaneously on the stage.
    The exploration of human movement could be applied to the interactions and body language between DJs during B2B (back-to-back) sets and how this affects the flow of the music as the artists switch off and overall success of the show.
    A complex network analysis could explore either relations between artists in various subgenres of EDM, or the increasingly blurring lines between EDM and genres such as trap, alternative, and pop. Big data analysis of pieces released on large music platforms could add a different form of quantitative data with potential to explore how these collaborations and genre-crossings correlate to published music. Additionally, we could look at changing trends within local music cultures as artists from different geographical locations, which can also relate to different sub-genres of EDM, collaborate and release music.

  6. spn2120

    All three articles discussed musical collaboration between different artists, whether it was through micro gestures in ensembles, variations in social networks between popular producers and artists of “fringe” genres, and the network of a piece of music’s publication history and movement throughout time, even if the composer themselves has passed away.

    I’m interested in how musical collaboration works when one responds to themselves. More specifically, my brother recently purchased a looper pedal, which he’s been attaching his guitar to and has been playing over himself. It’d be interesting to combine the approaches of all three articles. How do we respond to recorded sound data that we’ve produced, in a real time performance? If an instrumentalist or singer is looping their own sound, is there a away to create a network between common directions or pathways during each variation of the loop? During live performances, are artists less likely to show a physical reaction that indicates some sort of emotional response to their own playing?

    While all the methodologies in the articles were pretty interesting, I am still not completely convinced that any of these papers presents anything earth shattering. I found the point in the second article about jazz being more collaborative than other musical genres the most interesting, because that may reveal something about the nature of jazz as improvisatory and not based in musical score/notation that enhances collaboration. Beyond that, however, I am not sure how studying the gestures of musicians, or even networks gives us insight into the music. The language used by the studies even describes musicians as “nodes,” which I find kind of interesting because I’ve always thought the beauty of music is that the visceral response it initiates cannot be studied or objectified. However, I think the point about networks reflects a larger trend about social capital and the music industry, that is separate from the music itself.

Leave a Reply