A Methodology for Analyzing Uptake of Software Technologies Among Developers

Motivation: The question of what combination attributes drives the adoption a particular software technology is critical to developers. It determines both those technologies that receive wide support from community and which may be abandoned, thus rendering developers’ investments worthless. Aim Context: We model by developers provide insights on specific are associated with better visibility among alternative technologies. Thus, our findings have practical value for seeking increase rate their products. Approach: leverage social contagion theory statistical modeling identify, define, test empirically measures likely affect adoption. More specifically, we large collection open source version control repositories (containing over 4 billion unique versions) construct dependency chain set R language source-code files. formulate logistic regression models, where library choices modeled, investigate technological drive competing data frame (a core concept science languages) implementations in language: tidy xmlns:xlink="http://www.w3.org/1999/xlink">data.table . To describe each technology, quantify key project might (e.g., response times raised issues, overall deployments, number defects, knowledge base) also characteristics making selection (performance needs, scale, network). Results: find quick larger high-score StackExchange questions higher Decision makers tend adopt closer them technical network author collaborations networks while meeting performance needs. gauge generalizability proposed methodology, spread two popular web JavaScript frameworks xmlns:xlink="http://www.w3.org/1999/xlink">Angular xmlns:xlink="http://www.w3.org/1999/xlink">React , discuss results. Future work: hope methodology encompassing captures rational irrational preferences elucidation collections provides general path toward increasing visibility, driving informed decisions, producing more sustainable widely adopted software.