Leveraging development and field knowledge to assist developers and operators of large software systems

挖掘软件存储库（MSR）领域已经影响了软件研究和实践的各个方面。这条研究路线源于对开发人员、操作人员甚至软件系统留下的数据痕迹的观察，这些痕迹可以被挖掘出来，以提高软件质量和开发实践。国际MSR社区继续展示从静态记录保存库（如源代码控制和bug跟踪系统）中恢复有用和可操作信息的创新方法。如今，MSR的创新融入了一些世界上最大和最成功的组织的日常实践：微软、黑莓、AT&T实验室、思科、Mozilla和Avaya。拟议研究计划的长期愿景是将MSR建立为贯穿软件系统生命周期的关键战略工具，从开始到运行。相比之下，MSR目前主要关注软件开发阶段，这是一个相对较小的时间段，因为今天的软件系统的生命周期太长且变化太快。未来5年的短期目标是与包括黑莓和SAP在内的行业合作伙伴合作，重点展示现场日志的战略价值，将其作为一种低成本的高价值机制，融合开发和现场知识。我们的短期研究将沿着两个主题进行：(1)通过利用现场知识来帮助开发人员改进测井；(2)利用开发知识协助运营商应对操作复杂性。拟建项目将培养11名HQP: 4名博士、2名硕士和5名本科生。所有HQP将与工业合作伙伴密切合作，了解从业者每天面临的挑战，以及软件数据解决这些挑战的潜力，以及提高他们的专业技能和扩展他们的专业网络。

The Mining Software Repositories (MSR) field has impacted all facets of software research and practice. This line of research arose from the observation that developers, operators, and even software systems leave behind traces of data that can be mined to improve software quality and development practices. The international MSR community continues to demonstrate innovative approaches to recover useful and actionable information from static record-keeping repositories such as source control and bug tracking systems. MSR innovations are nowadays integrated into daily practices at some of the world's largest and most successful organizations: Microsoft, BlackBerry, AT&T Labs, CISCO, Mozilla and Avaya. The long-term vision of the proposed research program is establish MSR as a critical strategic tool throughout the lifetime of a software system, from inception to operation. In contrast, MSR currently focuses primarily on the software development phase, a relatively small time period given the excessively long and rapidly changing lifetime of software systems today. The short-term goals over the next five years, to be realized in collaboration with industrial partners including BlackBerry, and SAP, will focus on demonstrating the strategic value of field logs as a low-cost high-value mechanism to fuse development and field knowledge. Our short-term research will be carried out along two themes: (1) Assisting developers by leveraging field knowledge to improve logging; and (2) Assisting operators by leveraging development knowledge to cope with operational complexity. The proposed research program will train eleven HQP: 4 PhDs, 2 Masters and 5 undergraduate students. All HQP will collaborate closely with industrial partners, gaining an appreciation of the challenges facing practitioners on a daily basis and the potential for software data to address those challenges, as well as improving their professional skills and expanding their professional network.