Data Collection

As previously mentioned, the data sonification is part of a larger project that examines how data breaches have been constructed as crises from 2005 to the present. The year 2005 was chosen because it marks a significant shift from using the term “virus” to “breach” to describe cybersecurity threats, reflecting both technological and linguistic changes. Additionally, 2005 predates the widespread adoption of smartphones, with Apple launching the iPhone in 2007 and reaching 1 billion smartphones in use by 2012. Studying data breaches over this period highlights the transition from desktop computing to widespread cloud computing and data collection.¹

We examined 16 data breach cases from 2005 to 2021. In the early years, it was often difficult to assess the geopolitical significance of data breaches, as these events were not yet considered highly significant. Therefore, we selected high-profile cases that received extensive coverage, shaping the discourse around data breaches. Our analysis shows that in 2007, data breaches were seen as inevitable consequences of technological progress. In contrast, contemporary sources focus on identifying the causes or malicious actors behind breaches. This shift from viewing data breaches as inevitable to seeing them as abnormal or exceptional reflects a change in how data breaches are understood. We understand this shift as a normalizing discourse of the data breach as security crises.¹

Our sample comprises primary sources from various publications, collected using the Dow Jones search engine Factiva. Although Factiva is mainly a business news database, it offers extensive global coverage of newspapers, trade journals, blogs, and websites, making it ideal for our study on data breaches. We developed a framework to identify key terms and phrases, using search terms like “MyFitnessPal” AND “data breach” for specific case studies. We limited our searches to the year when information about a data breach was publicly released and examined the five most frequently publishing sources for each case in that year. Our sources often included major publications such as the Associated Press, BBC, and The New York Times. We also incorporated relevant international and local news sources as needed. We filtered each publication’s results by relevance, ensuring the articles were primarily about the given case, contained sufficient information for analysis, and were not republished from other sources. We collected approximately 10 articles from each source that met these criteria, resulting in a total of 288 primary sources across 26 cases.¹

---