{"title":"Report on the workshop on the state of the art in automated software engineering","authors":"Y. Ledru, D. Redmiles","doi":"10.1109/ASE.2002.1115038","DOIUrl":null,"url":null,"abstract":"The solution is provision of various forms of automation in the software tools developers use - the tools carry out perhaps a wide range of activities for the developer at appropriate times and inform the developer of results of actions in appropriate ways. We have focused in recent years on two areas of automation in software tools: (1) generating code from high-level software specifications; and (2) utilisation of high-level software information by agents to support collaborative work, change management and component testing. From our experiences developing a number of software tools using these automation approaches, we have learned a number of lessons for further research in these areas. These include: the need to support software tool meta-model extension; the need for on-the-fly enhancement of tool notations, event processing and code generation facilities; support for software artefact change propagation and annotation; the need to have reflective, high-level information to running software system components; and the continuing challenges of enhancing COTS tools with these kinds of automation facilities, including the need for sharable, extensible software information models for software tools and open tool infrastructure. We describe 3 software tools generating code from high-level descriptions (performance test-bed generator; data mapper; and adaptable user interface designer), and 3 tools providing event-driven agents (plug-in collaborative work components; requirements management tool; and deployed component testing agents).","PeriodicalId":163532,"journal":{"name":"Proceedings 17th IEEE International Conference on Automated Software Engineering,","volume":"12 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2002-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings 17th IEEE International Conference on Automated Software Engineering,","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ASE.2002.1115038","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
The solution is provision of various forms of automation in the software tools developers use - the tools carry out perhaps a wide range of activities for the developer at appropriate times and inform the developer of results of actions in appropriate ways. We have focused in recent years on two areas of automation in software tools: (1) generating code from high-level software specifications; and (2) utilisation of high-level software information by agents to support collaborative work, change management and component testing. From our experiences developing a number of software tools using these automation approaches, we have learned a number of lessons for further research in these areas. These include: the need to support software tool meta-model extension; the need for on-the-fly enhancement of tool notations, event processing and code generation facilities; support for software artefact change propagation and annotation; the need to have reflective, high-level information to running software system components; and the continuing challenges of enhancing COTS tools with these kinds of automation facilities, including the need for sharable, extensible software information models for software tools and open tool infrastructure. We describe 3 software tools generating code from high-level descriptions (performance test-bed generator; data mapper; and adaptable user interface designer), and 3 tools providing event-driven agents (plug-in collaborative work components; requirements management tool; and deployed component testing agents).