The study introduces a measure of software economic complexity based on the geography of programming languages used in open-source software. This approach helps capture digital capabilities that are largely invisible in traditional data sources such as trade, patents, or research publications.The results show that software complexity complements existing measures in explaining cross-country differences in GDP per capita, income inequality, and emissions, while also revealing a distinct global geography of digital capabilities. The paper also shows that software specialization follows the principle of relatedness: countries tend to expand into programming languages that are close to their existing capabilities. The figure below from the paper illustrates how software complexity compares to traditional measures and highlights the differences in the global distribution of capabilities.