This paper introduces DevGPT, a dataset curated to explore how software developers interact with ChatGPT, a prominent large language model (LLM). The dataset encompasses 29,778 prompts and responses from ChatGPT, including 19,106 code snippets, and is linked to corresponding software development artifacts such as source code, commits, issues, pull requests, discussions, and Hacker News threads. This comprehensive dataset is derived from shared ChatGPT conversations collected from GitHub and Hacker News, providing a rich resource for understanding the dynamics of developer interactions with ChatGPT, the nature of their inquiries, and the impact of these interactions on their work. DevGPT enables the study of developer queries, the effectiveness of ChatGPT in code generation and problem solving, and the broader implications of AI-assisted programming. By providing this dataset, the paper paves the way for novel research avenues in software engineering, particularly in understanding and improving the use of LLMs like ChatGPT by developers.

Continuous Integration (CI) is a popular software development practice that quickly verifies updates to codebases. To cope with the ever-increasing demand for faster software releases, CI acceleration approaches have been proposed; however, adoption of CI acceleration is not without risks. For example, CI acceleration products may mislabel change sets (e.g., a build labeled as failing that passes in an unaccelerated setting or vice versa) or produce results that are inconsistent with an unaccelerated build (e.g., the underlying reasons for failure differ between (un)accelerated builds). These inconsistencies threaten the trustworthiness of CI acceleration products.

Implicit gender bias in software development is a well-documented issue, such as the association of technical roles with men. To address this bias, it is important to understand it in more detail. This study uses data mining techniques to investigate the extent to which 56 tasks related to software development, such as assigning GitHub issues and testing, are affected by implicit gender bias embedded in large language models. We systematically translated each task from English into a genderless language and back, and investigated the pronouns associated with each task. Based on translating each task 100 times in different permutations, we identify a significant disparity in the gendered pronoun associations with different tasks. Specifically, requirements elicitation was associated with the pronoun “he” in only 6% of cases, while testing was associated with “he” in 100% of cases. Additionally, tasks related to helping others had a 91% association with “he” while the same association for tasks related to asking coworkers was only 52%. These findings reveal a clear pattern of gender bias related to software development tasks and have important implications for addressing this issue both in the training of large language models and in broader society.

Images are increasingly being shared by software developers in diverse channels including question-and-answer forums like Stack Overflow. Although prior work has pointed out that these images are meaningful and provide complementary information compared to their associated text, how images are used to support questions is empirically unknown. To address this knowledge gap, in this paper we specifically conduct an empirical study to investigate (I) the characteristics of images, (II) the extent to which images are used in different question types, and (III) the role of images on receiving answers. Our results first show that user interface is the most common image content and undesired output is the most frequent purpose for sharing images. Moreover, these images essentially facilitate the understanding of 68% of sampled questions. Second, we find that discrepancy questions are more relatively frequent compared to those without images, but there are no significant differences observed in description length in all types of questions. Third, the quantitative results statistically validate that questions with images are more likely to receive accepted answers, but do not speed up the time to receive answers. Our work demonstrates the crucial role that images play by approaching the topic from a new angle and lays the foundation for future opportunities to use images to assist in tasks like generating questions and identifying question-relatedness.

Existing techniques have succeeded to help developers implement new code. However, they are insufficient to help to change existing code. Previous studies have proposed techniques to support bug fixes but other kinds of code changes such as function enhancements and refactorings are not supported by them. In this paper, we propose a novel system that helps developers change existing code. Unlike existing techniques, our system can support any kinds of code changes if similar code changes occurred in the past. Our research is still on very early stage and we have not have any implementation or any prototype yet. This paper introduces our research purpose, an outline of our system, and how our system is different from existing techniques.

The importance of supporting test and maintenance activities in software development has been increasing, since recent software systems have become large and complex. Although in the field of Mining Software Repositories (MSR) there are many promising approaches to predicting, localizing, and triaging bugs, most of them do not consider impacts of each bug on users and developers but rather treat all bugs with equal weighting, excepting a few studies on high impact bugs including security, performance, blocking, and so forth. To make MSR techniques more actionable and effective in practice, we need deeper understandings of high impact bugs. In this paper we introduced our dataset of high impact bugs which was created by manually reviewing four thousand issue reports in four open source projects (Ambari, Camel, Derby and Wicket).

In the research of Mining Software Repositories, code repository is one of the core source since it contains the product of software development. Code repository stores the versions of files, and makes it possible to browse the histories of files, such as modification dates, authors, messages, etc. Although such rich information of file histories is easily available, extracting the histories of methods, which are elements of source code files, is not easy from general code repositories. To tackle this difficulty, we have developed Historage, a fine-grained version control system. Historage repository is a Git repository which is built upon original Git repository. Therefore, similar mining techniques for general Git repositories are applicable to Historage repositories. Kataribe is a hosting service of Historage repositories, which enables researchers and developers to browse method histories on the web and clone Historage repositories to local. The Kataribe project aims to maintain and expand the datasets and features.

Fault-prone module detection in source code is important for assurance of software quality. Most previous fault-prone detection approaches have been based on software metrics. Such approaches, however, have difficulties in collecting the metrics and in constructing mathematical models based on the metrics. To mitigate such difficulties, we have proposed a novel approach for detecting fault-prone modules using a spam-filtering technique, named Fault-Prone Filtering. In our approach, fault-prone modules are detected in such a way that the source code modules are considered as text files and are applied to the spam filter directly. In practice, we use the training only errors procedure and apply this procedure to fault-prone. Since no pre-training is required, this procedure can be applied to an actual development field immediately. This paper describes an extension of the training only errors procedures. We introduce a precise unit of training, "modified lines of code," instead of methods. In addition, we introduce the dynamic threshold for classification. The result of the experiment shows that our extension leads to twice the precision with about the same recall, and improves 15% on the best F1 measurement.