Translate 
Abstract
Although software is essential to modern society, its development process remains poorly understood. It is easy to find tools/techniques/methods promoted to help reach ever-higher production volumes and great improvements, even though this contrasts previous academic observations. This study aims to expand on previous research regarding the evolution of software development efforts by analyzing a large dataset of contemporary open-source projects. We examined 875 popular and large GitHub repositories and found a strong correlation between accumulated effort and quadratic functions of time, in 73% of projects, with linear models applicable in 41%. Our analysis of longterm trends suggests that no major external events (e.g., economic downturns, technological shifts) have impacted significantly all projects over the past 25 years. These findings challenge the perception of rapid, exponential improvements in software development efficiency. While further research is needed, our results suggest a disconnect between software development’s perceived and empirical realities.
Published Version
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
