A software development process is used by software engineers to guide their activities during all phases of the software product development. When executing a software development process, software engineers may lose time and effort while searching for artifacts or changing contexts. This happens, for example, when they need to search for a specific code file in a list of hundreds of files or when they interrupt an activity to execute another but forget specific details and need to re-execute searches related to the previous activity. This impacts their productivity negatively, because extra time and effort are spent into non-productive work. Therefore, automated assistance is required to mitigate or avoid these issues. The Degree of Interest (DOI) function infers an element’s importance in a context, helping software engineers to handle many artifacts. Mylyn, an Eclipse IDE plugin, uses a DOI function on Java documents to assist programmers when looking for code documents during development. However, Mylyn’s DOI function is limited to the implementation phase of software processes and relies on manual task creation. This paper presents MylynSDP, a software Process-aware extension to Mylyn’s DOI function. MylynSDP’s DOI function infers an artifact’s importance during an activity and filters uninteresting artifacts, reducing the time taken to search items and improving productivity. Mylyn code was augmented, and an evaluation study was performed. Seven subjects executed a software process with many artifacts. Exercise times were recorded for productivity analysis. Subjects answered a Technology Acceptance Model (TAM) questionnaire. New task and artifact creation wizards link tasks and artifacts to specification activities and artifacts, respectively. A new interaction event handles context creation, and the DOI function was extended to other software process phases. Exercise time reduction shows a productivity increase. TAM questionnaire answers show a positive overall willingness to adopt MylynSDP and provide evidence that using a DOI function in different software process phases increases productivity. This work advances the state of the art in software engineering by providing additional methods to support artifact search and discovery, context change management, and artifact relevance mechanisms.

中文翻译：

---

MylynSDP — 基于兴趣的流程感知工件过滤

软件工程师使用软件开发过程来指导他们在软件产品开发的所有阶段的活动。在执行软件开发过程时，软件工程师可能会在搜索工件或更改上下文时浪费时间和精力。例如，当他们需要在数百个文件的列表中搜索特定代码文件时，或者当他们中断一个活动以执行另一个活动但忘记了特定细节并需要重新执行与前一个活动相关的搜索时，就会发生这种情况。这会对他们的生产力产生负面影响，因为额外的时间和精力都花在了非生产性工作上。因此，需要自动化协助来减轻或避免这些问题。兴趣度 (DOI) 函数推断元素在上下文中的重要性，帮助软件工程师处理许多工件。Mylyn 是一个 Eclipse IDE 插件，它使用 Java 文档上的 DOI 函数来帮助程序员在开发过程中查找代码文档。但是，Mylyn 的 DOI 功能仅限于软件流程的实施阶段，并且依赖于手动创建任务。本文介绍了 MylynSDP，这是 Mylyn DOI 功能的一个软件进程感知扩展。MylynSDP 的 DOI 功能在活动期间推断工件的重要性并过滤不感兴趣的工件，从而减少搜索项目所花费的时间并提高生产力。扩充了 Mylyn 代码，并进行了评估研究。七名受试者执行了一个包含许多工件的软件过程。记录锻炼时间用于生产力分析。受试者回答了技术接受模型 (TAM) 问卷。新任务和工件创建向导分别将任务和工件链接到规范活动和工件。一个新的交互事件处理上下文创建，并且 DOI 功能扩展到其他软件过程阶段。锻炼时间减少表明生产力提高。TAM 问卷回答显示了采用 MylynSDP 的积极总体意愿，并提供证据表明在不同软件流程阶段使用 DOI 功能可提高生产力。这项工作通过提供额外的方法来支持工件搜索和发现、上下文更改管理和工件相关机制，从而提高了软件工程的最新技术水平。并且 DOI 功能扩展到其他软件过程阶段。锻炼时间减少表明生产力提高。TAM 问卷回答显示了采用 MylynSDP 的积极总体意愿，并提供证据表明在不同软件流程阶段使用 DOI 功能可提高生产力。这项工作通过提供额外的方法来支持工件搜索和发现、上下文更改管理和工件相关机制，从而提高了软件工程的最新技术水平。并且 DOI 功能扩展到其他软件过程阶段。锻炼时间减少表明生产力提高。TAM 问卷回答显示了采用 MylynSDP 的积极总体意愿，并提供证据表明在不同软件流程阶段使用 DOI 功能可提高生产力。这项工作通过提供额外的方法来支持工件搜索和发现、上下文更改管理和工件相关机制，从而提高了软件工程的最新技术水平。