Using digital traces to analyze software work: skills, careers and programming languages
This research addresses the problem of understanding skill dynamics and career paths in the global software industry for policymakers and educators, though it is incremental as it builds on existing digital trace analysis methods.
The study analyzed tens of millions of Stack Overflow posts to create a fine-grained taxonomy of software skills, revealing that real-world software jobs require highly coherent skill sets and that programmers learn through related diversification, often acquiring lower-value skills, but Python usage correlates with targeting higher-value skills, explaining its rise as a dominant language.
Recent waves of technological transformation are reshaping work in uncertain and hard-to-predict ways. However, jobs at the forefront of the digitizing economy offer an early glimpse of these changes and leave rich activity traces. We exploit such traces in tens of millions of Question and Answer posts on Stack Overflow for the creation of a fine-grained taxonomy of software skills to analyze human capital in the global software industry. Constructing a software skill space that maps relations among these skills reveals that real-world software jobs demand highly coherent skill sets and that programmers learn through a process of related diversification. The latter process often leads to the acquisition of lower-value skills. However, when programmers use Python they preferentially target higher-value skills, offering a potential explanation for Python's successful rise as a dominant general purpose language.