Automating the Specification Process

自动化规范流程

• 批准号: 8804085
• 负责人: Stephen Fickas
• 金额: $ 33.61万
• 依托单位: University of Oregon Eugene
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1988
• 资助国家: 美国
• 起止时间: 1988-08-15 至 1993-03-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=8804085&HistoricalAwards=false
• 关键词: Automating; Specification; Process

This project concerns itself with a means of both formalizing and automating the specification process. It proposes a model of specification design that a) is evolutionary, b) is based on collaborative problem solving by both analyst and client c) represents conflicting goals, and d) is knowledge-based. Efforts are divided along two broad fronts: 1) the definition of a process model of problem-acquisition and specification-construction, and 2) a characterization of the form and content of the knowledge underlying such a model. The long range goal is to approximate the type of expertise seen in a competent, experienced, software analyst as he or she interacts with a client to construct a software specification. The project continues a) the experimental study of expert, human, software analysts and b) the construction of a system for specification critiquing derived from such studies, and proposes new work in the areas of c) conflict and compromise in specification construction, and d) models of design that extend that of traditional top-down refinement.

该项目涉及一种既正式化， 自动化规范过程。 它提出了一个模型， 规范设计a）是渐进的，B）是基于 分析师和客户的协作问题解决c）代表 （1）知识的基础;（2）知识的基础。 努力是分开的 沿着两个大的方面：1）过程模型的定义， 问题获取和规范构建，以及2）a 知识的形式和内容的特征 这样的模型。 长期目标是近似于 在一个有能力的，经验丰富的软件分析师， 她与客户交互以构造软件规范。 该项目继续a）专家，人类， 软件分析师和B）规范系统的构建 批评来自这些研究，并提出了新的工作， c）规范构建中的冲突和妥协领域，以及 d）扩展传统自上而下的设计模式 精致。