Abstract Nowadays, systems containing components based on machine learning (ML) methods are becoming more widespread. In order to ensure the intended behavior of a software system, there are standards that define necessary qualities of the system and its components (such as ISO/IEC 25010). Due to the different nature of ML, we have to re-interpret existing qualities for ML systems or add new ones (such as trustworthiness). We have to be very precise about which quality property is relevant for which entity of interest (such as completeness of training data or correctness of trained model), and how to objectively evaluate adherence to quality requirements. In this article, we present how to systematically construct quality models for ML systems based on an industrial use case. This quality model enables practitioners to specify and assess qualities for ML systems objectively. In addition to the overall construction process described, the main outcomes include a meta-model for specifying quality models for ML systems, reference elements regarding relevant views, entities, quality properties, and measures for ML systems based on existing research, an example instantiation of a quality model for a concrete industrial use case, and lessons learned from applying the construction process. We found that it is crucial to follow a systematic process in order to come up with measurable quality properties that can be evaluated in practice. In the future, we want to learn how the term quality differs between different types of ML systems and come up with reference quality models for evaluating qualities of ML systems.

This paper reflects on a decade-long journey into the industrial practices developed for requirements inspection in the Japanese software industry; it proposes an ELFIEP (Evolutionary Lifecycle Framework for Industrial Engineering Practice). We published two papers in RE’13 and RE’14. The first paper introduced the industrial practice of SRS (Software Requirements Specifications) inspection system of a software company. The system is a set of inspection items: inspection processes, techniques, and supporting tools. In the second paper, to encourage wider industrial adoption of requirements inspection systems, we proposed a methodology for designing requirements inspection systems. Called RISDM (Requirements Inspection System Design Methodology), it has been recognized by professional communities in Japan. A number of software companies have adopted the methodology, and are now operating their individual inspection systems. This paper provides a retrospect of the journey to RISDM. Learning from the journey, we propose ELFIEP as a generic framework for the evolution of engineering practice, and discuss insights gained from the journey and the framework.

In agile software development, there are several well-known frameworks by integrating rules and methods. Among them, Scrum-based agile frameworks for large organizations need much knowledge and experience to start it, although Scrum has fewer rules and is easy to start. Among other known frameworks, Scaled Agile Framework (SAFe), which has complicated rules, has lots of case studies than others. However, Large Scale Scrum (LeSS) and LeSS Huge are also known and have lots of case studies in Japan, so enhancing LeSS and LeSS Huge framework is one of the well-developed ways to solve the organization's issues quickly. This paper aims to propose a new method, named Multi-Products Agile Method, based on LeSS Huge for developing multiple products in parallel at a large organization. We applied the proposed method to three products based in Japan. We had been tracking the velocity that is a total number of about 10 teams for one year. Our results prove that the proposed method is an effective way to scale agile organizations for multiple products. The method also makes it possible to use the knowledge and experiences of LeSS or LeSS Huge. We conclude Multi-Products Agile Method is an extended framework of LeSS and LeSS Huge.

As the applications of machine learning algorithms in various fields are widely demanded, the development of machine learning software systems (MLS) is rapidly increasing. The quality of MLS is different from that of conventional software systems, in the sense that it depends on the amount and distribution of training data in a model learning and input data during operation. This is a major challenge in quality assurance of MLS development for the enterprise. In this paper, we propose a requirements-driven method to determine the quality characteristics of the MLS. Major contributions of this paper include: (1) Extending the quality characteristics of ISO 25010, which defines the conventional software quality, to those unique to MLS; this paper also defines its measuring method. (2) A method to identify requirements, i.e., issues to be determined in the requirements definition, in order to derive the quality characteristics and measurement methods for MLS, since the quality characteristics and the measurement method depend on the goals of the system under development. In order to evaluate the proposed method, we carried out an empirical study of the quality characteristics and measurement methods related to functional correctness and the maturity of the MLS for the enterprise. Based on the study, we compare the quality characteristics and measurement methods derived by the proposed method with those suggested by developers, and demonstrate the effectiveness of the proposed method.

Among the requirements elicitation activities, the stakeholder analysis is the main source of requirements. In this article, we propose a new model of data-driven stakeholder analysis, named SIG (Stakeholder Intention Graph), a semantic extension of property graph model that can represent the stakeholders' intentions and their relationships. To elicit the stakeholders' intentions from the speech data during meetings, we developed a system of structural analysis and SIG generation method from speech data. Based on the graph theory, we also propose an analysis methodology of stakeholders' intentions and their structure with both global and local graph analyses. We implemented a speech data-driven stakeholder analysis system on the graph database Neo4j. As the output, the analysis system automatically generates the stakeholder matrix from the speech data at the meetings. We applied the analysis method and system to the speech data of actual development meetings on the public service systems, and demonstrated the effectiveness of the proposed method.

In order to appropriately define and share the requirements among stakeholders involved in software development, a model of Software Requirements Specification (SRS), which defines requirement items and their organization, is required. There exist a number of models of SRS such as the one proposed by ISO 29148. However, they are generic and not enough to represent domain-specific requirements. In this paper, we propose a systematic method to design domain-specific models of SRS by extending the existing models of SRS with requirement items for domain-specific requirements. The proposed method derives domain-specific requirement items that should be added to the reference model from the stakeholder concerns about the domain characteristics. We apply the proposed method to the automotive software domain and evaluate its effectiveness.

As Representational State Transfer (REST)-based Web APIs are spreading to enterprise information systems, software development for the use and provision of Web APIs is rapidly increasing. The quality of the Web APIs significantly influences the application quality and development productivity. However, no quality model for Web APIs has been established yet, because Web APIs differ from conventional APIs in that they execute remotely on different servers and may be changed independently of their users. These unique characteristics introduce new problems in the software engineering of Web APIs, and impose risks to the users, especially those using enterprise Web APIs, whose numbers are increasing. To solve these problems, in this paper, we propose a quality model for Web APIs that reflects their unique characteristics. As the main characteristics of this quality model, we propose the concept of Web API learnability to use and stability to change, from the perspective of Web API users. Based on this quality model, we also propose a set of measures and a quantitative evaluation method. In this study, we applied the proposed quality model and evaluation method to four types of actual Web APIs, including Uber, WordPress, OpenStack, and Media Processing. To validate the proposed model, we also conducted an empirical study of the usability of the Web APIs. Our comparison of the proposed quality statistics with those from the empirical study validates the effectiveness of the proposed quality model and its associated measures of the learnability and stability of Web APIs.

Network structures of OSS (Open Source Software) development communities are becoming more and more complicated. Various mining techniques have been applied to the repositories of OSS communities. However, structure analysis of OSS development community evolution has not been established. In this article, we propose SCGM (Software Community Graph Model), a new class of graph models to define the OSS development community. Based on the SCGM, we propose a structural analysis method of OSS development community evolution. To automate the analysis method, a prototype system is implemented with the graph DB Neo4j. We applied the proposed method and prototype system to four major machine learning OSS communities, Caffe, Chainer, Jubatus, and Tensorflow, for over five years on GitHub. From the analysis, we discovered three novel characteristics of community evolution, 1) three layered community evolution models consisting of the Core, Semi-core and Non-core members, 2) three developer growth patterns in terms of contribution behavior, and 3) evolutional changes according to the interaction among developers, which is a major contribution of this work. Based on the experiments, we demonstrate the validity and effectiveness of the proposed method.

This article proposes a multiple product line development method based on variability structure analysis. In product line development, the problem area is divided into the domain engineering and application engineering for delivering diverse products. Now, the development of automotive software requires to meet both agility and extreme diversity, which is a big challenge. We developed a structural analysis method of variability for multiple product lines using an extended model of OVM (Orthogonal Variability Model). Together with the variability analysis method, we propose an agile application development method to refine development items according to variability dependency based on the analysis, and develop them incrementally. We applied the proposed method to the development of the multiple product lines of automotive software systems, and demonstrated to reduce the volatility of the test efforts and usage of the test environment, and higher velocity and better manageability of the value stream.

Safety requirements have been critical to the automotive software systems, and various advanced driving assistance systems and automated emergency braking systems have been developed with complex software systems. In this article, we propose a modeling and quantitative analysis method of safety requirements which integrates the safety patterns, extended misuse case analysis and evaluation method based on the Bayesian networks. In automotive software systems, the threats to the safety include not only external factors from outside the automobile, but also internal factors such as a driver's erroneous operation. The proposed method enables to analyze both internal and external factors of the system as the hazards to the safety of the systems. First, we define safety a set of patterns as a pair of a cause and the mitigation use cases in order to prevent hazards. With the safety patterns, we can identify a set of mitigation points. Then, we analyze extended misuse cases, which enables to identify hazards to the system failures, and mitigation use cases from the misuse case scenarios. Finally, with the Bayesian networks, we quantitatively evaluate the effect of safety requirements by comparing the accident probability of before and after the application of mitigation use cases. We applied the propose method to the actual autonomous emergency braking systems of passenger vehicles of different model years, and demonstrated the validity and effectiveness of the method.

Requirements to automotive software is ever increasing its size and complexity due to rapid advancement of internet connection and advanced driver assistance system (ADAS). To assure the quality of software requirements specification (SRS) is a challenge. The quality assurance of SRS mostly relies on peer-review, which is error prone. To improve the quality of SRS, numerous techniques have been proposed, including a variety of review and inspection techniques. However, previous works indicate the diversity and confusion of requirements at different abstraction levels in automotive SRS makes it difficult to apply conventional review techniques. We propose a two-stage inspection method in order to separate perspectives in inspection. We define the reference SRS and the associated document quality characteristics, and mapping method from the reference SRS to multiple domain SRSs. We also develop a method to quantitatively measure the document quality of an SRS by third-party inspection. The contributions of this research include: 1) proposal of two-stage inspection process, 2) method of mapping the reference automotive SRS to project SRSs in multiple different product domains, 3) method for quantitative measurement and benchmark of SRS quality, 4) evaluation of the effectiveness of the method through applications to five actual automotive SRSs in four different product domains, and 5) adequacy validation of the IEEE Std. 830-1998 SRS template for automotive SRSs.

This paper proposes an approach to concurrent feedback development method in the software development of automotive systems. The automotive software is increasing its size and complexity. Therefore, it has become more difficult to ensure quality. At the same time, automobile manufacturers demand to shorten the development time. Conventionally, we developed the software in the two phases of "prototype development for developing requirements significations" and "product software development based on the requirements specifications". In this article, we propose the concurrent feedback development method using the prototype in the specification development. The proposed concurrent feedback loop model consists of three patterns of feedback loop to evolve the product concurrently. We demonstrated a reduction of the development time and improved the internal quality of the product in our automotive system development.

As the size and complexity of automotive software increases, software requirements specification (SRS) is gaining importance in automotive software system development. However, the diversity in SRSs, which comes from the difference between product characteristics, is becoming an impediment for systematic inspection. We propose a two-stage inspection that combines first-stage inspection by third-party inspectors and second-stage inspection by the project team, which would allow SRSs written in different formats to be inspected in a uniform way. The scope of this paper is limited to third-party inspection. The proposed methodology leads to the following three benefits: 1) systematic inspection of SRSs from different product domains 2) analysis of the inspection results from multiple viewpoints 3) benchmark of inspection results and SRS quality. We applied the proposed method to two actual products' SRSs and evaluated the effect of the method.

Summary form only given. This talk will high-lighten the diverse research opportunities and challenges arising from the software engineering on the connected automobile. The undergoing convergence of automotive and information technology makes automobile fundamentally change. Now, the automobile is not only a vehicle of transportation, but also a vehicle of information, communication, and services. Under the hood of an automobile, more than 50 computers and more than 50 sensors are working together through in-vehicle networks for making the automobile safer, greener and smarter. The software in an automobile is highly-complicated safety-critical, real-time distributed. The size of the software well exceeds ten million lines of code, and is rapidly evolving. Now autonomous, or driverless, automobile is a big challenge in automotive software engineering, and many areas in software engineering. Through communication networks, the automotive software can send the big data on the driver behavior, location, traffic, and weather to cloud computing for mobile crowd sensing. It is considered as an IoT (Internet of Things). Now, automotive software engineering is an exciting research subject for a wide area of research and development in software engineering.

We propose a method of continuously managing the stakeholders and changes in requirements throughout the life cycle of information systems. In the conventional methods of stakeholder analysis, the stakeholders are identified when capturing the requirements with the main objectives of identifying the source of the requirement and the degree of impact. However, stakeholders change along with the progress of the life cycle and cause the change of requirements. In this article, we propose a continuous stakeholder management method. The proposed method enables to: 1) model the interactions of stakeholder requirements that change along with life cycle, 2) to evaluate the interactions with a contribution/risk evaluation matrix, and 3) to identify conflicts between stakeholder requirements. We applied the proposed method to a development of a large-sale e-Government system, and demonstrate the validity and effectiveness of the method.

In this paper, we propose a Requirements Engineering Ontology (REO). We reviewed three well-known Bodies of Knowledge (BoKs) on requirements engineering: BABOK, SWEBOK, and REBOK. Based on the review, we compiled the REO as an organized set of knowledge of the requirements engineering. With the REO, we propose "knowledge spectrum analysis", a technique to analyze and visualize the distribution of knowledge in technical documents. As a practical application of the REO, we conduct two case studies on the knowledge spectrum analysis. First, we evaluate three BoKs (i.e., BABOK, SWEBOK, and BABOK) using the REO. Second, we utilize the REO to survey a set of technical documents for corporate-wide methodology for business modeling at NTT DATA. Based on the results of the case studies, we demonstrate the effectiveness of the REO.

Information technology is introduced to agriculture in order to improve all the cultivating process and the quality of the products. In this article, we propose a method to organize the cultivated knowledge with a focus on the life cycle process, based on the concept of lightweight ontology. We also propose a method for acquiring the knowledge easily by applying the technology of software engineering. We interviewed the farmers cultivating high quality mandarin orange and organized the ontology using the proposed method, and show the effectiveness of the proposed method.

We propose a design methodology of IT-driven business architecture to realize the business model utilizing information technology. As a component technology, they propose a business architecture design method using the goal model. A business goal to be met in business and a system goal to be met by system for business goal meet are defined. This study proposes a design methodology of business architecture and system architecture to meet business goal. The authors apply the proposed method to the case of mobile music delivery business and evaluate its effectiveness.

To make a business competitive advantage with the information system, it is necessary to validate the system in order to assure the system performs the business as the users expect. While the business context is becoming diverse, conventional acceptance test only confirms that the system meets the requirements specifications, and hard to assure the quality of use in the business context. This article proposes a methodology to design a set of test scenarios effectively in diverse usage context based on the model of business context. We propose a method to identify and analysis the business context with an extension of Empathy Maps. To model the business context, we propose the BC (Business Context) matrix, which represents the dependency between business process and properties of the business context identified. With the BC matrix, we propose a method to design a set of test scenarios from the perspective of Personas. We applied the proposed methodology to a large-scale human resource management system shared by 29 ministries and agencies of the Japanese government, and evaluated the effectiveness of the methodology.

PROMCODE (PROject Management for COntracted DElivery) is an innovative research consortium to change the way of project management on the resource-oriented services platform. It was launched by Nanzan University and six major Japanese IT companies including Fujitsu, Hitachi, IBM, NEC, Nomura Research Institute, and NTT DATA in May, 2012. Since then, we have developed a common data model to exchange project management data across organizational boundaries along with software supply chains, and supporting service platform to facilitate the data exchange. This article reports on the resource-oriented service model based on a common data model. We also developed a resource-oriented platform for exchanging data based on the OSLC (Open Services for Lifecycle Collaboration), an open standard for tool chain compliant to Web standards, namely RDF/Linked Data and RESTful Web services. To study the feasibility of the developed technologies, we, six companies, conducted intensive experience with real project management data. We discuss the effectiveness of the developed technologies and lessons learned from the experiences.

The quality of the SRS (Software Requirements Specification) is the key to the success of software development. The inspection for the verification and validation of SRS are widely practiced, however, the techniques of inspection are rather ad hoc, and largely depend on the knowledge and skill of the people. This article proposes RISDM (Requirements Inspection Systems Design Methodology) to design the RIS (Requirements Inspection System) to be conducted by a thirdparty inspection team. The RISDM includes a meta-model and design process of RIS, PQM (Pragmatic Quality Model) of SRS, and a technique to generate inspection question set based on the PQM and PBR (Perspective-Based Reading). We have been applying the RIS designed by the proposed RISDM to more than 140 projects of a wide variety of software systems in NTT DATA for five years. By analyzing the statistics from the experience, we discovered some key quality characteristics of SRS reveal strong correlation to the project cost and level of quality to be used for evaluating the maturity of the SRS and predicting the risk.

A metadata-driven architecture for multitenant cloud services is an important class of cloud computing architectures. It enables dynamic variability of services on a single instance for embracing diverse requirements among multiple tenants. It does not require to develop applications, but to generate applications through metadata Web APIs at runtime according to individual tenant's requirements. The variability among requirements makes the design of multitenant services complicated. To design the variability of multitenant services, this article proposed an extended model of OVM (Orthogonal Variability Model) to model variability and their dependencies at two levels: meta-level and base-level. We demonstrate the effectiveness of the extended OVM through a case study of the development of an application service on Force.com from Salesforce.com.

We introduce a whole view of current status of requirements engineering based on REBOK (Requirements Engineering Body Of Knowledge) and its practical use with case studies.

This article reports on the developing and practicing of software professional education program for ten years at DENSO. Through the experience in fostering professionals, we focused on solving challenges in order to create an effective solution, and introduced the problem solving technique, advanced software engineering and engineer' philosophy to the education program. The performance of participants demonstrated significant improvement in the quality of their solutions. Particularly, the courses in problem solving and engineers' philosophy contributed to motivating the participants and improving the performance. This article reports the improvement of the education program with statistics.

To deliver a business information system, it is necessary to not only verify the system, but also validate the system in order to assure the system can provide the business value which the users expect. However, conventional acceptance testing only confirms that the system meets the requirements specifications. Therefore, it is difficult to detect lack of user requirements. This article proposes business-driven acceptance testing methodology in order to validate the system by confirming the system meets the users' business requirements. We explain the testing process, test case selection technique, test scenario design method, and discuss the experiences in the application of the proposed method to a large-scale human resource management system shared by multiple ministries and agencies of Japanese government.

PROMIS (Project Management Information exchange Services) is a research consortium with Fujitsu, IBM Japan, Hitachi, NEC, Nomura Research Institute, NTT DATA and Nanzan University. It aims at developing a new software engineering platform for managing SSN (Software Supply Network), an inter-organizational model of software development. This article reports on the background, goals, research challenges, our approaches and main results of the PROMIS consortium. Major technical contributions include a management model for SSN, a common resource model for exchanging management data as Linked Data, and a RESTful resource-oriented architecture on top of OSLC (Open Services for Lifecycle Collaboration) over the Web.

This article proposes a new architectural principle for LISA (LInked Service Architecture) based on the Linked Data, ROA (Resource-Oriented Architecture), and meta-level service broker. LISA employs a two-layer architecture separating the concerns into meta-level of linking and base level of provisioning of services. A meta-level broker can dynamically coordinate services, and make the provisioning at the base level truly decentralized.

We have been involved in a number of large-scale software development projects, which might lead to loss of millions of dollars if failed. The quality of SRS (Software Requirements Specification) is the key to success of the software development. Review and inspection are common practices for the verification and validation of SRS. However, verification techniques used in projects might be characterized as ad hoc. In this article, we propose requirements clinic, a third party inspection methodology for improving the quality of the SRS. In order to systematically inspect a SRS, we developed a perspective-based inspection methodology based on PQM (Pragmatic Quality Model) of SRS. PQM is derived from IEEE Std. 830 from the perspective of pragmatic quality. To inspect a SRS according to PQM, we identified 198 inspection points, which lead to a quality score between 0 and 100. The requirements clinic advises to the requirements engineering team by a comprehensive quality inspection report including quality score, benchmark and SRS patterns for improvement. Since 2010, we have been practicing the methodology to a variety of development projects, and revealed an average of 10.6 ROI in 12 projects. We also discuss the feasibility of the methodology and lessons learned from the practices.

Why does it change?

The aim of our research is to empirically analyze the impact of requirements trace ability quality to the productivity of the enterprise applications development. We analyze trace ability links of Requirements Specification Documents (RSDs) collected from the enterprise applications development. These RSDs are classified into two groups in terms of project performance. One group includes RSDs of significant cost overrun projects, another group includes RSDs of normal range cost projects. Our analysis revealed that high quality of trace ability in RSDs significantly reduces the cost overrun in the enterprise applications development.

This article reports our concept and experiences on ACSS (Automotive Cloud Service System) based on SOA (Service-Oriented Architecture) for the next-generation automotive software platform. Along with rapid deployment of cloud computing, we expect, automotive software is evolving to ACSS where vehicles are collaborating with outside cloud computing and a variety of social networks. We propose an ACSS based on SOA named DARWIN, and demonstrate the validity of DARWIN with case studies running on a prototype electric vehicle. This article contributes to reveal key aspects of new software architecture for the next generation automotive software of integrating software in a vehicle and cloud services out of vehicles seamlessly.

Automotive software is required to collaboratively control multiple distributed processors, called ECUs (Electronic Control Units) connected through in-vehicle networks and embedded into different locations in a physical system of a vehicle. The extremely high complexity of interactions among ECUs makes it difficult to design the collaborative control. For such systems, conventional design methodology based on the structure is not able to adequately present the architectural properties. This article proposes a design methodology for collaborative software architecture based on the behavioral properties. We define the property as attributes and states of the vehicle, surrounding environment and users. Based on the property model, we propose the extended DSM (Design Structure Matrix) and a set of associated models for modeling global behavior of systems for meeting NFRs (Non-Functional Requirements). We applied the proposed methodology to the VDM (Vehicle Dynamics Management System), one of the most complicated automotive control software, and demonstrated the effectiveness of the methodology.

Requirements engineering has been extensively developed as a discipline. Many statistics on the software development indicate requirements process is the most influential to both success and failure of software development. However, practitioners are still difficult to learn and apply requirements engineering. As a guideline for practitioners to learn and apply requirements engineering, we developed REBOK (Requirements Engineering Body Of Knowledge). In the development, we found there is no common model of BOKs in software engineering. This article proposes the model and architecture of the body of knowledge of REBOK and its proof of the concept.

We initiate the discussions on the REBOK (Requirements Engineering Body of Knowledge) in requirements engineering community.

Extending the applications of SOA (Service-Oriented Architecture) to embedded/mobile service systems requires the asynchronous and lightweight messaging. However, conventional SOA focuses on the synchronous and heavyweight messaging such as SOAP. Furthermore, the lightweight messaging is diverse. Designing lightweight SOA, a class of SOA based on the lightweight messaging, is complicated. To accommodate the complexity of designing lightweight SOA, we propose ABAP (Attribute-Based Architecture Pattern), a set of architecture patterns of lightweight SOA and mode-based development method of ABAP by focusing on the attributes of the structure and behavior of lightweight SOA. ABAP enables to systematically design lightweight SOA with model and patterns. We demonstrate the feasibility of the proposed method with an example.

We propose AORE (aspect-oriented requirements engineering) methodology for automotive software product lines. The AORE methodology enables to model the collaborations among the distributed embedded automotive software systems in terms of aspects, and generate multiple product lines while assuring a set of non-functional requirements including safety, performance and cost. Applying to two case studies of automotive product lines modeled from the specifications, we demonstrate the effectiveness of proposed methodology.

This article proposes a unified methodology for designing asynchronous SOA (Service-Oriented Architecture) based on the asynchronous messaging models and patterns. Conventional SOA focuses on synchronous messaging. Although asynchronous messaging provides much efficient and productive way to coordinate services, design of aSOA (asynchronous SOA) is far more complicated due to the variety of messaging and architecture while assuring behavioral consistency of architecture. This paper proposes a model-driven design methodology for aSOA. The methodology is based on aMEPs (Asynchronous Message Exchange Patterns) identified by classifying the messaging in terms of behavioral concerns. Based on the meta-model of aSOA, a set of aSOA patterns is generated by composing aMEPs. Then, an aSOA pattern is selected and transformed to a platform specific aSOA on top of Web services standards. We successfully implemented an aSOA on Apache Axis, which enables to asynchronous messaging of SOAP over SMTP. We demonstrated that conventional methods are subsets of the proposed methodology, which is the major contribution of this work.

This paper proposes Persona-Scenario-Goal (PSG) methodology as an extension of Persona-Scenario methodology, and its empirical study in the usercentered requirements engineering for a Web application system. The Persona-Scenario methodology demonstrated the effectiveness in capturing requirements of consumer products, such as mobile phones. However, the methodology has limitations to evaluate the multiple, possibly conflicting, requirements. The PSG methodology intended to identify a set of value requirements from multiple and even conflicting requirements by combining persona, scenario and goal-oriented techniques. This paper discusses the model and procedure of PSG methodology, and demonstrates its effectiveness by applying the methodology to reengineering a Web application system.

The middleware market represents a sizable segment of the overall Information and Communication Technology market. In 2005, the annual middleware license revenue was reported by Gartner to be in the region of $8.5 billion. In this article we address the question whether research had any involvement in the creation of the technology that is being sold in this market? We attempt a scholarly discourse. We present the research method that we have applied to answer this question. We then present a brief introduction into the key middleware concepts that provide the foundation for this market. It would not be feasible to investigate any possible impact that research might have had. Instead we select a few very successful technologies that are representative for the middleware market as a whole and show the existence of impact of research results in the creation of these technologies. We investigate the origins of Web services middleware, distributed transaction processing middleware, message-oriented middleware, distributed object middleware and remote procedure call systems. For each of these technologies we are able to show ample influence of research and conclude that without the research conducted by PhD students and researchers in university computer science labs at Brown, CMU, Cambridge, Newcastle, MIT, Vrije, and University of Washington as well as research in industrial labs at APM, AT&T Bell Labs, DEC Systems Research, HP Labs, IBM Research, and Xerox PARC we would not have middleware technology in its current form. We summarise the article by distilling lessons that can be learnt from this evidenced impact for future technology transfer undertakings.

The technology of a home network system (HNS) allows integration of several kinds of home appliances to provide a user with value-added integrated services. Development of the integrated HNS services requires implementation of the appliance components (with APIs) and the services, according to each home-network environment. There are various implementation standards such as DLNA, ECHONET, OSGi and Jinifor the HNS applications. Therefore, even if a developer can choose the optimal one, it's very difficult to develop the integrated services by composing a new HNS implementation. In this paper, we propose a model-driven development of integrated HNS service applications. In our former manuscript, platform-independent design language for verifying HNS service scenarios was proposed. Our model-driven development method uses this design language as a meta-model of integrated HNS services. By model transformation to concrete implementation together with verification by SMV (symbolic model checking), productivity and quality of this kind of HNS applications are improved.

Dynamic composition of Web services encounters many challenges including selecting qualified services, composing services dynamically, and assuring quality of services. This article proposes a value model and its representation language, VSDL (value-based service description language), of Web services, and an architecture of value-added service broker of dynamically composing Web services based on the value model. The value model defines a common model of diverse quality aspects of Web services. We also propose a value meta-model representing relationships among values. The value-added service broker provides value-based service selection, composition and assurance of services. To specify service interfaces with values, the VSDL extends the service interface descriptions of WSDL with the value model. Based on the VSDL, the value-added service broker generates process of composing services and generates WS-BPEL codes for composition. We implemented a prototype of value-added service broker and demonstrated the effusiveness with a case study using three dictionary services.

The challenge of the requirements engineering to digital consumer products lies in that the users are unknown many of wide variety. The major contributions of this article include a technique to identify persona to provide a rich contextual model of a group of primary targeted users based on conjoint analysis theory, and a technique to identify the value and hot spots in the requirements through the interaction analysis between personas and scenarios. We applied the proposed methodology to mobile phone product lines through a series of field studies conducted with more than 150 users from 2003 to 2004 in Japan. The studies revealed the various aspects of nontechnical users, and proved the effectiveness of proposed methodology in the requirements engineering for digital consumer products.

We report our experience of product-line development for a wide variety of e-Government applications based on multilayered frameworks. The key contributions include the modeling a variety of requirements for e-Government applications, software architecture for multiple product-lines framework anticipating the variety, scenario-and-event-based patterns for composing the frameworks and components, and the statistical evaluation of the experience.

We present an overview of the 5th International Workshop on Principles of Software Evolution (IWPSE 2002).

Web services are emerging technologies to reuse software as services over the Internet by wrapping underlying computing models with XML. Web services are rapidly evolving and are expected to change the paradigms of both software development and use. This panel will discuss the current status and challenges of Web services technologies.

We propose the XML-based component commerce language XSCL (eXtensible Software specification and Commerce Language) and JavaBeans Component Player. ESD (Electronic Software Distribution) and ASP (Application Service Provider) are gaining ground in software distribution and provision. However, the representations of software/service specification and commerce information vary vendor by vendor. It is indispensable to provide a common representation of software/service specification and commerce in order to properly understand and compare software, services and software components. Furthermore, it is still difficult to specify the complex dynamic behavior and various non-functional aspects of software. Thus, it is helpful for the potential users of software to provide a tool to remotely exercise the software at his/her computer over the Internet. We developed the SCB (Software/Service Commerce Broker) on which we can provide a digital catalog of software with a HTML-based language SCL (Software specification and Commerce Language) and Component Player. Although it was effective in trial use, the representation capability is relatively weak due to the low-level representation capability of HTML. Thus, we developed the new language XSCL (eXtensible SCL) based on XML and accompanied JavaBeans Player.

To make componentware happen, it needs to explore the collective behavior of components at levels of software architecture. Componentware or CBSE (Component-Based Software Engineering) is promising to change the way we develop software. A variety of software components are available at different levels of software architecture. To explore the matching of components across the different technology groups, we have conducted a number of experiments in part of the Software CALS NGSE (Next Generation Software Engineering Program) in Japan. We developed a single prototype application with multiple combinations of technologies such as ActiveX-and-CORBA, multiple CORBAs, CORRA-Web and CORBA-and-legacy-code. Major problems we encountered include: 1) different interface models; 2) poor interface information such as behavior information; 3) poor interface-centric design methodology and architecture design methodology; 4) poor non functional information of components.

This article proposes an architecture of SCB (Software Commerce Broker) which aims at trading software packages and components over the Internet. Component-based software engineering is coming into real development paradigm due to widespread distributed objects environment and componentware. Since software can be distributed over the Internet, a Web-based commerce of software components and packages over the Internet is emerging. To realize Web-based software commerce, we develop a prototype of SCB, which collects information on software components worldwide over the Internet and provides a set of electronic catalogues of software components in a semi-formal specification language SCL (Software specification and Commerce Language). Furthermore, SCB provides play mechanism on which customers in remote locations can play the component through the Web. SCB is under evaluation in the Next Generation Software Engineering Program of Software CALS (Continuous Acquisition and Life-cycle Support) project in Japan.

This article proposes a new software process model, ASP (Agile Software Process) and discusses its experience in large-scale software development. The Japanese software factory was a successful model in the development of quality software for large-scale business applications in the 80s. However, the requirements for software development have dramatically changed. Development cycle-time has been promoted to one of the top goals of software development in the 90s. Unlike conventional software process models based on volume, the ASP is a time-based process model which aims at quick delivery of software products by integrating the lightweight processes, modular process structures and incremental and iterative process enaction. The major contributions of APS include: a new process model and its enaction mechanism based on time; a software process model for evolutional delivery; a software process architecture integrating concurrent and asynchronous processes, incremental and iterative process enaction, distributed multi-site processes, and the people-centered processes; a process-centered software engineering environment for ASP; and experience and lessons learned from the use of ASP in the development of a family of large-scale communication software systems for more than five years.

The article proposes a new software process model, ASP (Agile Software Process) based on a decade-long evolution of software process models inside a Japanese software factory. The Japanese software factory was a successful model in the development of quality software for large-scale business applications in the 1980s. However, the business climate of software development has dramatically changed in the last few years. Development cycle-time was promoted to one of the top issues of software development in the 1990s and then globalization appeared. A new paradigm of software development is required. The RSP model does not implicate any physical office but a virtual collaboration space over the Internet. It enables multiple small teams, which are geographically distributed, to concurrently develop multiple functions for a family of large-scale software systems.

The paper analyzes the characteristics of the design information in a distributed concurrent development of large-scale software systems, and proposes a method for the classification and management based on the concept of structuredness in terms of structural and temporal characteristics of the design information. With the classification method, we discovered that many pieces of design information involved in the upper processes are semi-structured. To share and manage the semi-structured design information, we developed a wide area information network (WAIN) on our intranet. It was also possible to verify the changes in those pieces of design information requiring a strict change management.

Summary form only given. We discuss the methodology and lessons learned in the reengineering and improvement of software process. The following areas are specifically tackled: new demands to and changing paradigms of software development; methods of software process reengineering and improvement; information and software technology for software process reengineering and improvement; best practices and lessons learned; and future directions.

The article addresses the emerging technologies for developing network software. As networking technology has made great progress in its speed and connectivity in the past twenty years (1976-96), network software has also evolved and many technologies have been tried and applied to network software development. Formal methods provide model verification, and some sorts of automatic program generation and artificial intelligence open the new arena of intelligent network software. Object oriented technology is coming into practice and enhances the maintainability and reuse of software. Recently, emerging technologies such as distributed objects, mobile agents, Java and network programming will redefine the concept of network software and its development methodology. Commemorating the twentieth COMPSAC, the article summarizes these twenty years' experience and discusses future directions of network software development.

The article discusses a new software process model, distributed concurrent development, and its experience in developing a family of large-scale communication software systems. We have experienced an evolutionary change of our software development process. The change is spread into two dimensions, that is, from sequential to concurrent, and from centralized to distributed. With the concurrent development process model, multiple functions to a single software system are concurrently developed from the beginning of the development process, i.e. requirement specification, to system testing. In the distributed development process model, the development organization is decentralized into multiple development sites which are geographically distributed. The two dimensional changes require a fundamental re-think of our software process and brought about a significant impact on every aspect of software development activities ranging from individual developers to the entire project. We discuss the model, the issues in implementing the model, and our total system of managing the process, project and products.

The authors propose a development process model and support environment for developing intelligent business communication services and other evolving communication services. To increase customer satisfaction and cut development time, they have developed a development process model, distributed concurrent development, in which multiple services are concurrently developed at a number of regionally distributed development centers. To support the process model, the ICAROS tool was developed, where developers and customers can cooperatively define service specifications with a visual object-oriented specification language. To smoothly migrate existing communication service to assets into intelligent network services, ICAROS supports the cyclic process model, which integrates forward engineering and reverse engineering, together with amplifier concept, which augments human ability on a groupware platform. Experimental use of ICAROS revealed the positive effect of proposed approach.>

The author proposes a development process model: distributed concurrent development. The model has originated from various experiences in the development of large-scale switching software systems. Critical issues in the distributed concurrent development are discussed based on a process programming concept: observations on analogies between the modeling of distributed concurrent development and the disciplines of distributed concurrent software systems. A formalization of software development processes with an object-oriented approach is then proposed. Two process description languages and two behavior description languages are proposed which make it possible to represent the structure and dynamic behavior of development processes, respectively. The author also proposes a process-design process which is unique in that one can design and evaluate a development process before one initiates development activities. The entire process for developing a family of switching software systems has been defined with the description languages.>

The critical issues in the development of real-time software systems are: (1) lack of modeling techniques, (2) lack of design techniques, (3) lack of testing techniques, and (4) lack of awareness of such systems. Each of these issues is examined.>

The authors have designed a system functional requirement (SFR) specification processor using a knowledge-based system approach in conjunction with other artificial intelligence techniques, this processor can be used for a variety of software systems. Domain-specific expert knowledge is defined for each software system application. The adopted specification language can easily be read by a variety of people such as customers, managers, requirement writers, software designers, and the test team. The processor, for a given software system application, not only checks for consistency, completeness, nonredundancy, and nonambiguity, but also uses domain-specific expert knowledge to verify that the SFR specification document is correct from the domain expert's point of view. To illustrate the proposed approach, the authors give an example of a component of a telephone switching system and the processor system methodology.>

Presents a knowledge-based rapid prototyping system that facilitates the documentation, analysis, and development of a software system. The particular application domain to be modeled is represented in terms of objects and activities. A frame-and-rule oriented requirements language (FRORL) is developed to describe the objects and activities of the problem domain. FRORL is designed to capture most of the semantics of the application domain, thus making it possible to translate the requirements specification directly to the Prolog code. The checking for correctness of the specification developed in FRORL using a resolution strategy is also presented.>

The authors explain an integrated support environment for software maintenance. The essential feature is bridging configuration management and quality management. The environment has been successfully applied to the development and maintenance of a family of switching systems for several years.>

No abstract available.