面向互联网的开源软件自动化评估证据框架

An Automatic Evaluation Evidence Framework for Open Source Software

如何从海量的、公共可获取的互联网资源中快速定位和获取高质量、满足自身特定需求的开源软件资源是当前软件开发技术核心竞争力的一个重要体现.本文研究了互联网上开源软件各种可信证据的存在及分布形式,给出了相关数据的定位和自动化获取方法;提出了面向互联网的开源软件自动化评估证据框架,用于解决互联网数据到软件可信证据的映射问题;提出面向互联网上海量软件资源的开源软件可信证据查询平台实现方法,用于解决软件可信证据的自动化获取及使用问题;基于该系统能够极大的提高评估效率,使得用户准确、快速、全面的了解相关软件项目的各种信息.最后,通过两个典型的开源软件实例证实了上述证据框架和方法的可行性.

How to locate and choose open source software which is the most relevant m a particular user＇s need from massive and public resources in Internet is an important performance of software development technology today. In their whole life cycle, OSS produced a very large amount of data with various types. These data may provide some new ideas for software trustworthy assessment, yet up to now they still have not been made full use of in the field of software trustworthy assessment. Based on existed software quality evaluation models, some kinds of information related to software trustworthiness are summarized and an automatic evaluation evidence framework for open source software is proposed in this paper, by which the data related to open source software resources in Intemet is mapped into trustworthiness evidence. The evidence framework consists of two parts： one is the properties of software product, the other is about software development team. All types of the trustworthiness evidence in this framework are described in detail in this paper. The trustworthiness evidence query platform is constructed based on this automatic evidence framework. With the platform, the efficiency of open source software resources evaluation could be greatly improved and users could get many kinds of information about one open source software project accurately, quickly and completely. By the method improved in this paper, the flexibility of open source software evaluation can be proved very much, and which can meet the demand of trustworthiness evaluation to the large amount of open source software in the Intemet. By taking the case of two famous OSS projects, the feasibility of the evidence, framework and automatic query system is confirmed.