In this paper1, we investigate and discuss the underlying nature of the requirements traceability problem. Our work is based on empirical studies, involving over 100 practitioners, and an evaluation of current support. We introduce the distinction between pre-requirements specification (pre-RS) traceability and post-requirements specification (post-RS) traceability, to demonstrate why an all-encompassing solution to the problem is unlikely, and to provide a framework through which to understand its multifaceted nature. We report how the majority of the problems attributed to poor requirements traceability are due to inadequate pre-RS traceability and show the fundamental need for improvements here. In the remainder of the paper, we present an analysis of the main barriers confronting such improvements in practice, identify relevant areas in which

: Researchers and practitioners are looking for systematic ways of comparing domain analysis (DA) methods and related technologies such as systems modeling. Less attention has been paid to comparing DA methods in terms of certain core methodological issues, including problems of scoping, contextualizing, descriptive vs. prescriptive modeling, and formalized models of variability. This paper presents key aspects of Organization Domain Modeling (ODM), a systematic domain analysis method structured in terms of a core domain modeling life cycle directly addressing these methodological concerns. 1 Keywords: domain engineering, domain analysis, domain modeling, variability modeling, organization domain, stakeholder model, feature modeling 1.0 Introduction Since the seminal work of Neighbors and other researchers in the early 1980's, domain analysis (DA) has emerged as a distinct area of research and practice within the software reuse research community [9,10]. There are now a number of pu...

cornelia.boldyreff,david.nutter,stephen.rank¡ We describe a software artefact repository that provides its contents with some awareness of their own creation. “Active ” artefacts are distinguished from their passive counterparts by their enriched meta-data model which reflects the work-flow process that created them, the actors responsible, the actions taken to change the artefact, and various other pieces of organisational knowledge. This enriched view of an artefact is intended to support re-use of both software and the expertise gained when creating the software. Unlike other organisational knowledge systems, the meta-data is intrinsically part of the artefact and may be populated automatically from sources including existing data-format specific information, user supplied data and records of communication. Such a system is of increased importance in the world of “virtual teams ” where transmission of vital organisational knowledge, at best difficult, is further constrained by the lack of direct contact between engineers and differing development cultures. 1

This paper first explores the knowledge management problem in more detail and discuss challenges to acquiring, maintaining, and disseminating design knowledge. We then describe a framework for integrating a design memory tool Human-Computer Interaction, , 10, 1 (1995), 1-38.

6 2 Introduction 7 3 Distributed Architecture Description Language 11 3.1 Overview : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 11 3.2 Backus-Naur Form (BNF) Semantics : : : : : : : : : : : : : : 12 3.3 Conversations : : : : : : : : : : : : : : : : : : : : : : : : : : : 13 3.3.1 Conversation Identifier : : : : : : : : : : : : : : : : : : 13 3.4 Connections : : : : : : : : : : : : : : : : : : : : : : : : : : : : 14 3.4.1 Connection Identifier : : : : : : : : : : : : : : : : : : : 15 3.4.2 Connectivity : : : : : : : : : : : : : : : : : : : : : : : : 15 3.4.3 Order : : : : : : : : : : : : : : : : : : : : : : : : : : : 15 3.4.4 Delivery : : : : : : : : : : : : : : : : : : : : : : : : : : 15 3.5 Dagent : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 16 3.5.1 Dagent Identifier : : : : : : : : : : : : : : : : : : : : : 16 3.5.2 Dagent Connections : : : : : : : : : : : : : : : : : : : 16 3.6 Alphabet : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 17 3....

: This paper focuses on experiences and lessons learned to date in applying STARS technologies on the STARS Army/Unisys Demonstration Project. 1 The paper provides background on demonstration project context and objectives, and on methods and technologies being applied, including Organization Domain Modeling (ODM), a systematic approach to domain analysis, the Conceptual Framework for Reuse Processes (CFRP), and the Reuse Library Framework (RLF). A general strategy for technology transition and coevolution in the context of "early adopter" pilot DA projects is described. Keywords: domain analysis methods and techniques; domain modeling; systematic reuse; re-engineering for reuse; reusable architecture; organizational and management techniques for implementing reuse. 1. Introduction The Early Adopters' Challenge. Despite recent progress in reuse techniques, there is still need for a comprehensive, integrated set of methods for achieving systematic reuse [Ara91]. To apply in diverse orga...

Abstract – In this paper, we present our recent effort on using a feature technology and ontology for embedded systems modeling and design. We present an overview of embedded system design and propose an object-oriented UML modeling approach to representing embedded systems, i.e., open embedded system model (OESM). OESM supports models of embedded system artifacts, components, features, configuration/assembly, and embedded system platform and family, design rationale, etc. Our focus is on modeling of feature semantics in embedded systems. We call this open embedded system feature model (OESFM). We also present a semantic web environment for modeling and verifying feature models using ontologies, in which the Protégé-OWL is used to precisely capture the relationships among features in feature diagrams and configurations. The OESFM models and ontologies provide a feature-based component collaborative framework. This allows the designer to develop a virtual embedded system prototype through assembling virtual components in which the platform-based hardware/software (HW/SW) codesign is supported and the design rationale is captured. The collaborative co-design framework can not only provide formal precise models of the embedded system prototypes but also offers design variation of prototypes whose members are derived by changing certain virtual components with different features.