The present talk emerges from the work of an international research group that seeks to make nineteenth-century newspaper collections more applicable for scholarship by interfacing and connecting national archives, and by testing various means of cross-linguistic computational text analysis. Specifically, we trace the news coverage of the eruption of Krakatoa in 1883, one of the most impactful natural disasters of the century, transnationally. This is done through a comparison of various news cycles on the volcanic eruption and its aftermath in six national, digital newspaper archives. In order to study the geo-temporal expansion of coverage and to compare particular elements of the reporting (natural occurrences, fatalities, navigation, etc.), automatic reprinting detection algorithms are used. After clusters of news reporting and reprinting have been identified, sentiment analysis tools help to analyze the content and style of coverage across national and linguistic boundaries. Such a digital humanities-informed approach aims to advance our understanding about the relationship between distance and sympathy in, and the ideological underpinnings of, nineteenth-century news reporting.