http://sedici.unlp.edu.ar:80/handle/10915/183319 2026-07-11T05:01:25Z 2026-07-11T05:01:25Z Wong, Felix McGovern, Jim Fernandez, George http://sedici.unlp.edu.ar:80/handle/10915/185118 2025-09-26T20:10:00Z 2003-01-01T00:00:00Z

Objeto de conferencia Simposio Argentino de Ingeniería de Software (ASSE 2003) - JAIIO 32 (CABA, 1 al 5 de septiembre de 2003) An important part of software engineering is the collaborative development of artefacts, such as requirements analysis and design documents. Software engineers can independently work on parts of software engineering documents, but they will often need to work closely with other software engineers and domain experts developing and refining these documents. The Internet and distributed computing tools allow geographically separated stakeholders to participate in the software engineering process, not only asynchronously, but also synchronously in realtime. Real time collaboration for software engineering is technically challenging, requiring support for concurrent editing of text and diagrams. This paper describes a replicated architecture, where copies of software engineering documents are kept at each participating site. While this architecture provides reliability, fast local response, and supports a high level of concurrency, maintaining consistency among copies, in the face of network latency and conflicting operations from independent team members, is a difficult task. Conflicts in software engineering analysis and design are often an indication of a need for further investigation, and are best resolved by discussion between stakeholders rather than by automatic and arbitrary actions. This project proposes a scheme to support the management and resolution of these conflicts, and demonstrates its correctness in a simulation environment.

2003-01-01T00:00:00Z An important part of software engineering is the collaborative development of artefacts, such as requirements analysis and design documents. Software engineers can independently work on parts of software engineering documents, but they will often need to work closely with other software engineers and domain experts developing and refining these documents. The Internet and distributed computing tools allow geographically separated stakeholders to participate in the software engineering process, not only asynchronously, but also synchronously in realtime. Real time collaboration for software engineering is technically challenging, requiring support for concurrent editing of text and diagrams. This paper describes a replicated architecture, where copies of software engineering documents are kept at each participating site. While this architecture provides reliability, fast local response, and supports a high level of concurrency, maintaining consistency among copies, in the face of network latency and conflicting operations from independent team members, is a difficult task. Conflicts in software engineering analysis and design are often an indication of a need for further investigation, and are best resolved by discussion between stakeholders rather than by automatic and arbitrary actions. This project proposes a scheme to support the management and resolution of these conflicts, and demonstrates its correctness in a simulation environment. Belloni, Edgardo A. Marcos, Claudia A. http://sedici.unlp.edu.ar:80/handle/10915/184725 2025-09-20T04:10:39Z 2003-01-01T00:00:00Z

Objeto de conferencia Simposio Argentino de Ingeniería de Software (ASSE 2003) - JAIIO 32 (CABA, 1 al 5 de septiembre de 2003) In the last years, the software development based on mobile agents has received important attention because it introduces a new development paradigm for widely distributed and heterogeneous systems. Up to now, the usual approach for mobile-agent applications development has been focused on the implementation phase, mainly because of the vast lack of suitable modeling techniques or methodologies in the domain. In this sense, the development of powerful abstractions for the modeling of mobile agents becomes increasingly important. In this article, we present an approach to deal with the lack of appropriate concepts and notations in the standard Unified Modeling Language (UML) to capture relevant abstractions for the modeling of mobile-agent features. This approach is materialized by a coherent and comprehensive set of views and models, which extends UML contributing to the analysis and design phases of mobile-agent applications development.

2003-01-01T00:00:00Z In the last years, the software development based on mobile agents has received important attention because it introduces a new development paradigm for widely distributed and heterogeneous systems. Up to now, the usual approach for mobile-agent applications development has been focused on the implementation phase, mainly because of the vast lack of suitable modeling techniques or methodologies in the domain. In this sense, the development of powerful abstractions for the modeling of mobile agents becomes increasingly important. In this article, we present an approach to deal with the lack of appropriate concepts and notations in the standard Unified Modeling Language (UML) to capture relevant abstractions for the modeling of mobile-agent features. This approach is materialized by a coherent and comprehensive set of views and models, which extends UML contributing to the analysis and design phases of mobile-agent applications development. Pryor, Jane L. Marcos, Claudia A. http://sedici.unlp.edu.ar:80/handle/10915/184724 2025-09-20T04:10:39Z 2003-01-01T00:00:00Z

Objeto de conferencia Simposio Argentino de Ingeniería de Software (ASSE 2003) - JAIIO 32 (CABA, 1 al 5 de septiembre de 2003) Aspect-Oriented Software Development is an approach which supports the modularisation of crosscutting concerns through the development phases of an application. One of the limitations in the main approaches to AOSD is that they do not provide support for the definition and handling of conflicts that may arise between multiple competing aspects. This paper describes how conflicts are managed by a framework that we have developed for the construction of aspect-oriented applications. In the first place, we describe the main characteristics of this reflective multi-level framework and the tool that instantiates it. We then present our categorisation of different conflicts and how they are defined and handled by our environment for the development of aspect-oriented software.

2003-01-01T00:00:00Z Aspect-Oriented Software Development is an approach which supports the modularisation of crosscutting concerns through the development phases of an application. One of the limitations in the main approaches to AOSD is that they do not provide support for the definition and handling of conflicts that may arise between multiple competing aspects. This paper describes how conflicts are managed by a framework that we have developed for the construction of aspect-oriented applications. In the first place, we describe the main characteristics of this reflective multi-level framework and the tool that instantiates it. We then present our categorisation of different conflicts and how they are defined and handled by our environment for the development of aspect-oriented software. Colla, Pedro E. http://sedici.unlp.edu.ar:80/handle/10915/184717 2025-09-20T04:10:39Z 2003-01-01T00:00:00Z

Objeto de conferencia Simposio Argentino de Ingeniería de Software (ASSE 2003) - JAIIO 32 (CABA, 1 al 5 de septiembre de 2003) An alternate effort estimation methodology is proposed to be used at mature software engineering organizations where situations of insufficient historical data or activities resilient to prediction by statistical models exist, in this scenario experts become the most used and sometime only way to perform estimations As experts enter the scene subjectiveness come into play and consistency couldn’t longer be assessed when performing estimations across similar situations or technical contexts. This paper proposes the usage of pairwise comparison methodologies not only by its own merits but also to allow the subjective factors integrated in the experts’ effort estimation to be measured, an therefore assessed in terms of its consistency The benefits of the pairwise comparison methodologies in general, and the Saaty’s Analytic Hierarchical Process (AHP) methodologies often associated with complex data manipulation are briefly revisited and adaptations are made to it in order to get most benefits from their usage with simple requirements in terms of the hierarchical modeling of the problem and data manipulation. Criteria are discussed to be used as quality figures and validations of the proposed methodology, in particular the usage of the Consistency Ratio (CR)A preliminary experiment is discussed and evaluated in order to validate the scope of the work and to help identify areas that are candidate to receive further attention.

2003-01-01T00:00:00Z An alternate effort estimation methodology is proposed to be used at mature software engineering organizations where situations of insufficient historical data or activities resilient to prediction by statistical models exist, in this scenario experts become the most used and sometime only way to perform estimations As experts enter the scene subjectiveness come into play and consistency couldn’t longer be assessed when performing estimations across similar situations or technical contexts. This paper proposes the usage of pairwise comparison methodologies not only by its own merits but also to allow the subjective factors integrated in the experts’ effort estimation to be measured, an therefore assessed in terms of its consistency The benefits of the pairwise comparison methodologies in general, and the Saaty’s Analytic Hierarchical Process (AHP) methodologies often associated with complex data manipulation are briefly revisited and adaptations are made to it in order to get most benefits from their usage with simple requirements in terms of the hierarchical modeling of the problem and data manipulation. Criteria are discussed to be used as quality figures and validations of the proposed methodology, in particular the usage of the Consistency Ratio (CR)A preliminary experiment is discussed and evaluated in order to validate the scope of the work and to help identify areas that are candidate to receive further attention. Slamkovic, Richard Fernandez, George McGovern, Jim http://sedici.unlp.edu.ar:80/handle/10915/184716 2025-09-20T04:10:40Z 2003-01-01T00:00:00Z

Objeto de conferencia Simposio Argentino de Ingeniería de Software (ASSE 2003) - JAIIO 32 (CABA, 1 al 5 de septiembre de 2003) A main problem for enterprise systems communication based on a client/server model is the likelihood of the end-points (client and server) being based on a heterogeneous infrastructure. In such cases, there is a need to convert session and transport information from the format pertaining to the source of the request to that of the target. The usual approach of implementing point-to-point connections that translate a source format to a target one is laborious, costly and inflexible since the bridges have to be handcrafted for each situation. To address this problem, in previous work we have described the architecture of a generic dynamic adaptor TUBE (The Ubiquitous Broker Environment), which intends to effect dynamic protocol-level systems integration by at run time bridging the gap between the disparate formats presented by a heterogeneous infrastructure. This paper discusses one of the central components of the system, the Dynamic Adaptive Marshaller (DAM), which is responsible for converting data between different formats at run-time. All marshalling is performed dynamically, so the DAM must "dynamically adapt” to encode or de-code the desired format, involving presentation, session and transport layer operations.

2003-01-01T00:00:00Z A main problem for enterprise systems communication based on a client/server model is the likelihood of the end-points (client and server) being based on a heterogeneous infrastructure. In such cases, there is a need to convert session and transport information from the format pertaining to the source of the request to that of the target. The usual approach of implementing point-to-point connections that translate a source format to a target one is laborious, costly and inflexible since the bridges have to be handcrafted for each situation. To address this problem, in previous work we have described the architecture of a generic dynamic adaptor TUBE (The Ubiquitous Broker Environment), which intends to effect dynamic protocol-level systems integration by at run time bridging the gap between the disparate formats presented by a heterogeneous infrastructure. This paper discusses one of the central components of the system, the Dynamic Adaptive Marshaller (DAM), which is responsible for converting data between different formats at run-time. All marshalling is performed dynamically, so the DAM must "dynamically adapt” to encode or de-code the desired format, involving presentation, session and transport layer operations. Caliusco, María Laura Galli, María Rosa Chiotti, Omar Juan Alfredo http://sedici.unlp.edu.ar:80/handle/10915/184715 2025-09-20T04:10:40Z 2003-01-01T00:00:00Z

Objeto de conferencia Simposio Argentino de Ingeniería de Software (ASSE 2003) - JAIIO 32 (CABA, 1 al 5 de septiembre de 2003) In business-to-business (B2B) relationships, enterprises share heterogeneous and distributed information that is the result of decision-making activities involved in inter-enterprise and intra-enterprise processes. The problem of bringing together heterogeneous and distributed computer systems is known as interoperability problem. Interoperability has to be provided on a technical and on an information level. The focus of this paper is to present the backgroung of information interoperability problem in B2B relationships and to identify the challenges in this area.

2003-01-01T00:00:00Z In business-to-business (B2B) relationships, enterprises share heterogeneous and distributed information that is the result of decision-making activities involved in inter-enterprise and intra-enterprise processes. The problem of bringing together heterogeneous and distributed computer systems is known as interoperability problem. Interoperability has to be provided on a technical and on an information level. The focus of this paper is to present the backgroung of information interoperability problem in B2B relationships and to identify the challenges in this area. Ruiz, Francisco Genero, Marcela García, Félix Piattini, Mario Calero, Coral http://sedici.unlp.edu.ar:80/handle/10915/184714 2025-09-20T04:10:41Z 2003-01-01T00:00:00Z

Objeto de conferencia Simposio Argentino de Ingeniería de Software (ASSE 2003) - JAIIO 32 (CABA, 1 al 5 de septiembre de 2003) Software measurement, as a young discipline, suffers from the symptoms of any unconsolidated discipline. Unfortunately, there does not yet exist a consensus about the terminology used in this field. Therefore, the first step towards software measurement become a well established engineering discipline is to get a common agreement among software measurement researchers and practitioners about the concepts used in software measurement and the relationships between them. To reach this goal, it is well-known the usage of ontologies, due to an ontology represents a certain view of an application domain in which the concepts that live in this domain are defined in an unambiguous and explicit way. So that, the aim of this is to present a preliminary, practical and semi-formal ontology for software measurement.

2003-01-01T00:00:00Z Software measurement, as a young discipline, suffers from the symptoms of any unconsolidated discipline. Unfortunately, there does not yet exist a consensus about the terminology used in this field. Therefore, the first step towards software measurement become a well established engineering discipline is to get a common agreement among software measurement researchers and practitioners about the concepts used in software measurement and the relationships between them. To reach this goal, it is well-known the usage of ontologies, due to an ontology represents a certain view of an application domain in which the concepts that live in this domain are defined in an unambiguous and explicit way. So that, the aim of this is to present a preliminary, practical and semi-formal ontology for software measurement. Sánchez, Marisa A. Felder, Miguel A. http://sedici.unlp.edu.ar:80/handle/10915/184713 2025-09-20T04:10:41Z 2003-01-01T00:00:00Z

Objeto de conferencia Simposio Argentino de Ingeniería de Software (ASSE 2003) - JAIIO 32 (CABA, 1 al 5 de septiembre de 2003) For the case of the software we use today, the information outside the specification is increasing. For quality assurance purposes, it would be desirable to integrate the analysis of software and hardware components with testing. In this paper, we propose a fault-based approach for generating test cases, thus overcoming the limitations of specification-based approaches that derive from the intrinsic incompleteness of the specification, and from the focus of specifications on correct behaviors, rather than potential faults. We use Fault Tree Analysis to determine how certain undesirable states can occur in a system. The results of the analysis expressed in terms of Duration Calculus are integrated with statechart based specifications. As a result, we obtain a testing model that provides a representation of the way the system behavior can be compromised by failures or abnormal conditions or interactions. In this way we can automatically derive fault-based test cases from this model.

2003-01-01T00:00:00Z For the case of the software we use today, the information outside the specification is increasing. For quality assurance purposes, it would be desirable to integrate the analysis of software and hardware components with testing. In this paper, we propose a fault-based approach for generating test cases, thus overcoming the limitations of specification-based approaches that derive from the intrinsic incompleteness of the specification, and from the focus of specifications on correct behaviors, rather than potential faults. We use Fault Tree Analysis to determine how certain undesirable states can occur in a system. The results of the analysis expressed in terms of Duration Calculus are integrated with statechart based specifications. As a result, we obtain a testing model that provides a representation of the way the system behavior can be compromised by failures or abnormal conditions or interactions. In this way we can automatically derive fault-based test cases from this model. De Castro Bertagnolli, Silvia Blanck Lisbôa, Maria Lúcia http://sedici.unlp.edu.ar:80/handle/10915/184670 2025-09-20T04:10:41Z 2003-01-01T00:00:00Z

Objeto de conferencia Simposio Argentino de Ingeniería de Software (ASSE 2003) - JAIIO 32 (CABA, 1 al 5 de septiembre de 2003) Many software projects have failed because they are based on a poor set of quality attributes. Moreover, the growing use of these requirements in the software has made researchers focus their attention on methodologies and processes to deal with it. We propose a well-defined model intended to guide the developer through the software development lifecycle. Its goal is to determine a systematic way to elicit and model non-functional requirements, and establish a practice manner to improve the software quality. The FRIDA model is founded on the aspect-oriented software development. In this paradigm, the aspect is a complementary manner to describe the non-functional features related to any application. In our approach, the crosscutting nature of the requirements is considered during all software lifecycle. Furthermore, each NFR is represented with one or more aspects and FRs are restricted to classes hierarchies

2003-01-01T00:00:00Z Many software projects have failed because they are based on a poor set of quality attributes. Moreover, the growing use of these requirements in the software has made researchers focus their attention on methodologies and processes to deal with it. We propose a well-defined model intended to guide the developer through the software development lifecycle. Its goal is to determine a systematic way to elicit and model non-functional requirements, and establish a practice manner to improve the software quality. The FRIDA model is founded on the aspect-oriented software development. In this paradigm, the aspect is a complementary manner to describe the non-functional features related to any application. In our approach, the crosscutting nature of the requirements is considered during all software lifecycle. Furthermore, each NFR is represented with one or more aspects and FRs are restricted to classes hierarchies Díaz Pace, Jorge Andrés Campo, Marcelo http://sedici.unlp.edu.ar:80/handle/10915/184669 2025-09-20T04:10:42Z 2003-01-01T00:00:00Z

Objeto de conferencia Simposio Argentino de Ingeniería de Software (ASSE 2003) - JAIIO 32 (CABA, 1 al 5 de septiembre de 2003) Architecture-based design consists essentially of making certain decisions whose results will affect various quality attributes of the architecture being developed. However, the elaboration of these designs is still a technically difficult problem, because it naturally involves a complex process of searching, evaluation and composition of design alternatives. Although some handbooks documenting proven design practices exist, this knowledge lives mostly in the developer’s mind. In order to alleviate this problem, this work focuses on the exploration of design alternatives and the transformation of architectural models using predefined design operations. More specifically, we propose a planning-based approach called DesignBots to assist designers in the generation and configuration of design alternatives, which should follow from principled decisions made by the designer. As expected contributions of this approach, we aim at reduce the inherent complexity of analyzing architectural solutions and provide a better context to evaluate design decisions.

2003-01-01T00:00:00Z Architecture-based design consists essentially of making certain decisions whose results will affect various quality attributes of the architecture being developed. However, the elaboration of these designs is still a technically difficult problem, because it naturally involves a complex process of searching, evaluation and composition of design alternatives. Although some handbooks documenting proven design practices exist, this knowledge lives mostly in the developer’s mind. In order to alleviate this problem, this work focuses on the exploration of design alternatives and the transformation of architectural models using predefined design operations. More specifically, we propose a planning-based approach called DesignBots to assist designers in the generation and configuration of design alternatives, which should follow from principled decisions made by the designer. As expected contributions of this approach, we aim at reduce the inherent complexity of analyzing architectural solutions and provide a better context to evaluate design decisions. Villarreal, Pablo David Martínez, Virginia Povolo, Josefina Stuchi, Lorena Chiotti, Omar Juan Alfredo http://sedici.unlp.edu.ar:80/handle/10915/184668 2025-09-20T04:10:42Z 2003-01-01T00:00:00Z

Objeto de conferencia Simposio Argentino de Ingeniería de Software (ASSE 2003) - JAIIO 32 (CABA, 1 al 5 de septiembre de 2003) During the last years, agent oriented software engineering has received a growing interest. Several methodologies for analysis and design of multi-agent systems (MASs) have been proposed. However, most of these methodologies lack of a complete development process. At the same time, several development platforms have been developed. The development platforms allow the rapid development of MASs, and they can be divided in frameworks and development environments. A well-defined development process for MASs should involve all the stages of the MAS life cycle. Therefore, a method to derive the development platform concepts from the models of the analysis and design methodology is required. Towards this direction, in this work we present a method to derive the Zeus development environment concepts from the concepts of the Gaia agent-based analysis and design methodology. This method can be used in the implementation stage with the purpose of defining a complete development process for MASs

2003-01-01T00:00:00Z During the last years, agent oriented software engineering has received a growing interest. Several methodologies for analysis and design of multi-agent systems (MASs) have been proposed. However, most of these methodologies lack of a complete development process. At the same time, several development platforms have been developed. The development platforms allow the rapid development of MASs, and they can be divided in frameworks and development environments. A well-defined development process for MASs should involve all the stages of the MAS life cycle. Therefore, a method to derive the development platform concepts from the models of the analysis and design methodology is required. Towards this direction, in this work we present a method to derive the Zeus development environment concepts from the concepts of the Gaia agent-based analysis and design methodology. This method can be used in the implementation stage with the purpose of defining a complete development process for MASs Grigorjev, Fabio Lascano, Natacha Staude, Julio L. http://sedici.unlp.edu.ar:80/handle/10915/184667 2025-09-20T04:10:43Z 2003-01-01T00:00:00Z

Objeto de conferencia Simposio Argentino de Ingeniería de Software (ASSE 2003) - JAIIO 32 (CABA, 1 al 5 de septiembre de 2003) Error seeding is a technique, which can be used to evaluate the amount of residual errors after the system software test phase. This method helps to decide how to release the software depending on results that fault-seeding technique can provide. The advantages and disadvantages of the method are discussed, and various strategies for seeding errors are presented. The lessons learnt from experiences performed at Motorola Global Software Group Argentina are summarized, and suggestions for further research are offered.

2003-01-01T00:00:00Z Error seeding is a technique, which can be used to evaluate the amount of residual errors after the system software test phase. This method helps to decide how to release the software depending on results that fault-seeding technique can provide. The advantages and disadvantages of the method are discussed, and various strategies for seeding errors are presented. The lessons learnt from experiences performed at Motorola Global Software Group Argentina are summarized, and suggestions for further research are offered.