【24h】

Analysis beyond UML

机译:超越UML的分析

获取原文

摘要

In spite of being a de facto standard for analysis and design, UML has some obvious shortcomings: the UML definition is at best semi-formal, the set of result types is far too large and heterogeneous, the tool-support is not satisfactory. If this is true - how do people get on with UML? How do they use it in their every day work? At sd&m (a medium size software company in Munich, Germany) we conducted a study of best practices. The study was restricted to analysis; we considered neither requirements specification (gathering requirements) nor design issues. The aim of the study was to answer the following questions: · What does a typical analysis documentation contain? · Which parts of UML are really used? · What kinds of non-UML documentation is used? Here is the surprising result: We identified 17 modules a typical analysis documentation consists of. These will be briefly described in the following sections. Only 3 out of 17 modules use UML at all. For the other modules, projects have defined their own standards, sometimes supported by cute home-made tools, while text takes up the major part of the documentations.
机译:尽管是分析和设计的事实标准,但UML有一些明显的缺点:UML定义是最佳半正式的,结果类型的结果类型太大而且异构,工具支持不令人满意。如果这是真的 - 人们如何与UML继续?他们如何在每天工作中使用它?在SD&M(德国慕尼黑的中型软件公司)我们进行了一项关于最佳实践的研究。该研究仅限于分析;我们既不考虑要求规格(收集要求)也不是设计问题。该研究的目的是回答以下问题:·典型分析文件包含什么? ·UML的哪些部分真正使用? ·使用哪种非UML文档?这是令人惊讶的结果:我们确定了17个模块,典型的分析文档包括。这些将在以下部分简要描述。只有17个模块中只有3个使用UML。对于其他模块,项目已经确定了自己的标准,有时由可爱的自制工具支持,而文本占据文档的主要部分。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号