Catherine Mullen is a PhD Candidate in the Department of Folklore & Ethnomusicology and a 2017 graduate of the MLS program at Indiana University. Her research focuses on popular music preservation and the affective dynamics and equitable potentials of community-based music archives. She has worked in a number of ethnographic and audiovisual archives across the US including the Archives of Traditional Music at Indiana University, the University of Washington Ethnomusicology Archives, and the American Folklife Center at the Library of Congress.
Explore My Dissertation
My dissertation, The Power of the Popular, examines the practices of Manchester Digital Music Archive (MDMA), a volunteer-run and user-generated community archive that aims to celebrate and preserve music from Greater Manchester, the large metropolitan area including and surrounding the city of Manchester, England. In each chapter of the dissertation, I focus on a different aspect of MDMA’s work, from its website and programming efforts, to the affect of the archival process and the place within larger networks of heritage work and popular music preservation in Manchester.
Chapter 2 looks at the purpose and usage of MDMA’s digital archive by engaging in an analysis of the MDMA website. Through a manual analysis of website’s characteristics and functionalities and a digital humanities-based analysis of the website’s content, this chapter weaves together ethnographic investigation with data-based visualizations to describe the archive’s structure and how it is utilized by its network of users. On this webpage, you can explore some of the interactive visualizations that contribute to this analysis.
Archival Content and Growth
The extent of MDMA’s archival content and its growth since launching as a user-generated archive help to understand what subject matter is prevalent in the archive and how archival expansion has occurred over time.
Figure 3, a bubble chart showing the popularity of the tags included in the data set, illustrates the proportion of tagged items associated with particular tags and tag types in the archive. Click through to read more about the bubble chart.
Figure 4, a stacked bar graph comparing the sum of artifacts uploaded to the dates of upload, demonstrates how and where the archive’s holdings have grown by examining tag types and separating tags into popularity segments within type. On this page, you can learn more about how growth occurs and the methods for determining popularity within type.
Examining Archival Content through User Networks
Archival content can also be examined by how it connects to other content through users. In part of Chapter 2, I look at how certain subject matter is emphasized as “popular” by highlighting eigenvector centrality statistics as a measure of popularity and illustrating how different groups of users interact with and contribute to MDMA’s database through network visualizations. Divided into five user groups determined by how much users upload to the archive, these networks show how users demonstrate their interests by creating their own networks of tags that represent varying subject matters.
Clicking through the links below, you can explore the different user groups and how certain tags are connected to particular users. The five user groups make up distinct user personas within the archive, demonstrating differing identities of archival participation. Each user group was determined by a box and whisker plot measuring the amount of uploads per user, with the final user group being split in two to better represent those only uploading a few items to the archive. The following video walkthrough explains how to navigate the networks.
Figure 5 represents User Group 1, made up of 8 users who each uploaded 304 or more artifacts to the archive. This group includes a select few archival megausers who have uploaded in bulk over time.
Figure 6 represents User Group 2, made up of 20 less extreme users who each uploaded between 118-303 artifacts.
Figure 7 represents User Group 3, made up of 66 moderate users who each uploaded between 28-117 artifacts.
Figure 8 represents User Group 4, made up of 133 occassional participant users who each uploaded between 10-27 artifacts.
Figure 9 represents User Group 5, made up of 572 users who each uploaded between 1-9 artifacts. This group includes users who can be characterized as sporadic or single-entry participants who have uploaded the least amount of artifacts, but make up the bulk of MDMA’s userbase.
How to contact me
Catherine Mullen (she/her)