Scientific Workflow Management Systems (SWfMSs) are specialized software frameworks pivotal to modern research, facilitating the design, automation, and analysis of intricate computational experiments. These systems serve as a robust foundation for constructing computational applications, offering essential features like modularity, abstraction, customization, interoperability, workflow composition tools, resource management, error handling, and extensive documentation. The utilization of these frameworks accelerates the development of scientific computing, resulting in more efficient and reproducible research outcomes. Despite their significance, developing a user-friendly, efficient, and adaptable SWfMS poses challenges for developers. This study explores these challenges by investigating open-source projects related to SWfMSs. Through mining and applying topic modeling techniques to a dataset of 444,208 GitHub issues from 124 SWfMS projects, we quantify the prevalence and frequency of these challenges. Additionally, we validate our findings using 12,015 SWfMS posts from Stack Overflow. The results provide valuable quantitative insights into the current state of challenges within the SWfMS domain. They shed light on the community's support available for addressing these challenges and identify crucial areas necessitating further research and development efforts. By understanding these challenges and leveraging the collective knowledge and expertise of the community, we can foster innovation and improve the efficiency and reliability of SWfMSs.
We idnetiy the topics from SO and GitHub data and noticed that many topics are common and we also find out the complexity of each category.