CAREER: Aiding Comprehension of Complex Software Dynamics to Support High-Quality Software Development

这项研究旨在帮助软件开发人员理解、诊断和修复软件中的错误。由软件代码中的逻辑不一致引起的一大类bug，即开发人员对程序应该如何运行的期望与实际运行的方式之间的不一致。这种常见的逻辑不一致通常需要开发人员的关注和理解，并且不适合算法定位和修复。因此，本研究的目标是帮助软件开发人员通过促进他们对软件及其行为的理解来执行软件维护和调试任务，最终目标是使他们能够更经济地生产出更高质量的软件。将软件维护和调试的挑战视为对人类理解能力的挑战，该研究旨在帮助理解代码库中的错误，为什么代码行为不正确，谁是最有能力理解和修复问题的开发人员，以及何时进行的更改引入了错误。这些问题是通过以下组合来解决的：(1)自动推荐系统，使用软件工件和分析为工程师提供自动帮助；(2)发现和探索工具，使工程师能够更深入地理解自动化建议；新的实验方案和评估，以确定理解挑战的来源。这项研究结合了软件分析和人机交互的思想。更好地了解软件故障以及如何修复它们将以更低的成本获得更高质量的软件。

This research is aimed at helping software developers understand, diagnose, and fix bugs in software. A large class of bugs that are caused by logical inconsistencies in the software code, i.e., incongruences between the developers' expectation of how the program should behave and the way it actually does. Such common logical inconsistencies typically require developer attention and comprehension, and are not amenable to algorithmic location and repair. As such, the goal of this research is to assist software developers performing software maintenance and debugging tasks by facilitating their comprehension of the software and its behavior, with the ultimate goal of equipping them to produce higher quality software, more economically.Viewing the challenges of software maintenance and debugging as challenges of human comprehension, the research aims to help understand where bugs reside in the codebase, why the code behaves incorrectly, who are the developers best equipped to understand and fix problems, and when were the changes made that introduced bugs. These questions are addressed using a combination of (1) automated recommendation systems to use software artifacts and analyses to provide automated assistance to engineers; (2) discovery and exploration tools to enable engineers to develop deeper comprehension of the automated recommendations; novel experimental protocols and evaluations to determine sources of comprehension challenges. The research combines ideas from software analysis and human-computer interaction. Better understanding of software faults and how to repair them will lead to higher quality software at reduced costs.