Software evolution concern summarization and explanation

软件演化关注点总结与解释

• 批准号: 194193-2010
• 负责人: Murphy, Gail
• 金额: $ 3.13万
• 依托单位: University of British Columbia
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2015
• 资助国家: 加拿大
• 起止时间: 2015-01-01 至 2016-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=569746
• 关键词: Software; evolution; concern; summarization; explanation

Canadian society has become dependent upon software intensive systems, such as cellphones, online banking, and ultrasounds to name just a few. To keep up with our changing world, software developers must continually evolve these systems. When evolution tasks are performed incorrectly, there are consequences. For example, an incorrectly completed evolution task at the Canada Revenue Agency (CCRA) resulted in the suspension of on-line personal tax filings in early March 2007. As a result of this error, taxpayer refunds were delayed and the CCRA was forced to incur additional costs to fix the problem. To complete evolution tasks with higher quality, software developers need to communicate efficiently and effectively about the intended change and they need to perform the technical work associated with the change precisely and completely. Currently, these activities are needlessly complicated because developers are working with all of the details associated with all the kinds of information needed for the task. This research program aims to improve developer productivity and the quality of a developer's work by enabling developers to work in terms of tasks rather than having the developers mired in the details of each task. This research program will develop techniques and tools to support the representation of the fragments of software project information needed for an evolution task. We refer to this representation as an evolution concern. The representation will include automatically produced summaries and explanations of the information comprising an evolution concern, providing the abstraction that is currently lacking in a developer's work day. We will evaluate whether the techniques and tools we built to support evolution concerns can ease development activities through laboratory experiments on open-source projects and field studies within the context of industrial developments. This research will expand Canada's $100 billion dollar software production industry.

加拿大社会已经变得依赖于软件密集型系统，例如手机、网上银行和超声波等等。为了跟上不断变化的世界，软件开发人员必须不断改进这些系统。当进化任务执行不正确时，就会产生后果。例如，加拿大税务局 (CCRA) 错误完成的演化任务导致 2007 年 3 月上旬暂停在线个人报税。由于这一错误，纳税人退款被延迟，CCRA 被迫承担额外费用来解决该问题。为了以更高的质量完成演进任务，软件开发人员需要就预期的变更进行高效、有效的沟通，并且需要精确、完整地执行与变更相关的技术工作。目前，这些活动不必要地复杂，因为开发人员正在处理与任务所需的各种信息相关的所有细节。该研究计划旨在通过使开发人员能够按照任务进行工作，而不是让开发人员陷入每个任务的细节中，从而提高开发人员的生产力和开发人员的工作质量。该研究计划将开发技术和工具来支持演化任务所需的软件项目信息片段的表示。我们将这种表示称为进化问题。该表示将包括自动生成的摘要和对包含演化问题的信息的解释，提供当前开发人员工作中缺乏的抽象。我们将评估我们为支持进化问题而构建的技术和工具是否可以通过开源项目的实验室实验和工业发展背景下的实地研究来简化开发活动。这项研究将扩大加拿大价值 1000 亿美元的软件生产产业。