.png){kind=icon}
.png){kind=icon}
Return
Open source oriented cross-platform survey
.png){kind=icon}
PRE
AI
DOI：10.1016/j.infsof.2025.107704.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Context: Open-source software development has become a widely adopted approach to software creation. However, developers' activities extend beyond social coding platforms (e.g., GitHub), encompassing social Q&A platforms (e.g., StackOverflow) and social media platforms (e.g., Twitter). Therefore, cross-platform research is essential for a deeper understanding of the nature of software development activities. Objective: This paper focuses on open-source platforms and systematically summarizes relevant cross-platform research. It aims to assess the current state of cross-platform research and provide insights into the challenges and future developments in this field. Method: This paper reviews 69 cross-platform research papers related to open-source software from 2013 to 2024, with a focus on several key areas, including platform interconnections, research themes, experimental design methods, challenges and research opportunities. Results: Through the analysis of 69 papers, we found that cross-platform research primarily involves platforms such as social coding, social Q&A, and social media. Researchers typically rely on information traces, including user personal info, technical info, project/post/bug report metadata, interaction info, to facilitate connections between platforms. Cross-platform research in the open-source domain mainly focuses on problem classification and feature extraction. The predominant research methods include data-driven approaches, qualitative studies, modeling and machine learning, and tool development and implementation. Despite these advancements, common challenges remain, such as subjective evaluation bias in manual data classification, insufficient data source coverage, and inaccurate data recognition. Future research opportunities may focus on increasing the diversity of data sources, improving data recognition accuracy, optimizing data classification methods, and clarifying user skill requirements. Conclusions: Based on our findings, we propose six future directions for cross-platform research in the opensource domain and provide corresponding recommendations for developers, researchers, and service/tool providers.
Keywords:
Open source
Cross-platform
Systematic literature review
GitHub
StackOverflow
Twitter
Journal
IF:
4.3
Papers:
3.7K
Citations:
7.7K