ChatGPT for Tailoring Software Documentation for Managers and Developers

Abstract

Abstract In many agile software development projects, documentation is often missing, outdated, or written with only a technical perspective. Existing literature recognizes the importance of documentation quality, especially when it comes to its readability for diverse audiences. While recent advances in Large Language Models (LLMs) offer the potential to tackle these issues, the use of LLMs for software documentation remains unexplored. This paper investigates the use of ChatGPT to improve and adapt documentation to specific audiences. We apply ChatGPT-4 for alternative documentation production and measure the resulting text characteristics and readability. Twenty-five experts from management and development rate these different versions. Results show the suitability of ChatGPT for generating high-quality text for both audiences, with managers benefiting more from an adapted version.

Ähnliche Arbeiten