Submitted by Geyciane Santos (geyciane_thamires@hotmail.com) on 2015-06-17T13:21:45Z No. of bitstreams: 1 Dissertação- Jeanne de Castro Trovão.pdf: 2079561 bytes, checksum: c951810ee47ee9da038b5fbcc6171763 (MD5) / Approved for entry into archive by Divisão de Documentação/BC Biblioteca Central (ddbc@ufam.edu.br) on 2015-06-17T20:22:49Z (GMT) No. of bitstreams: 1 Dissertação- Jeanne de Castro Trovão.pdf: 2079561 bytes, checksum: c951810ee47ee9da038b5fbcc6171763 (MD5) / Made available in DSpace on 2015-06-17T20:22:49Z (GMT). No. of bitstreams: 1 Dissertação- Jeanne de Castro Trovão.pdf: 2079561 bytes, checksum: c951810ee47ee9da038b5fbcc6171763 (MD5) Previous issue date: 2014-10-30 / FAPEAM - Fundação de Amparo à Pesquisa do Estado do Amazonas / Software companies are always concerned about improving the quality of their products and processes. By improving the quality of process, it also contributes to the quality of the product resulting from this process. Several strategies can applied to reach processes quality. One of them consists in the integration of support processes that assist the management of the main process and may be applied when required, contributing to the success and the quality of the main process and its products. When talking about software product quality, we can cite software testing activities, which are planned and executed in order to increase product quality through a process that also needs to be managed. Therefore, approaches that aim to contribute to the quality of the software testing process indirectly contribute to the quality of the software product. Thus, the strategy of integration of support processes that help to manage the testing process also contributes to this quality. However, this kind of approach is poorly developed in the technical literature, especially in the context of testing. Considering this necessity, this thesis presents the definition of two Support Processes instantiated to Software Testing Process, Testing Risk Management Process and Testing Human Resource Management Process, specified in order to contribute to the quality of management aspect of the testing process. These processes were defined based on several maturity models and norms/standards. In order to evaluate the proposed processes, a peer review was conducted with experts in the fields related to the specified processes and the results of this evaluation suggest that the proposed processes are appropriate and can be integrated into the testing process. / Empresas de software estão sempre preocupadas em melhorar a qualidade de seus produtos e processos. Ao melhorar a qualidade do processo, também se contribui para a melhoria do produto resultante desse processo. Diversas estratégias podem ser aplicadas visando à obtenção de qualidade de processos. Uma delas consiste na integração de processos de apoio, que são aqueles que auxiliam um processo principal como parte integrante, podendo ser empregado por ele quando necessário, contribuindo para o sucesso e a qualidade do processo principal e de seus produtos. Quando se fala em qualidade de produto de software, se remete às atividades de teste de software, que são planejadas e executadas a fim de aumentar a qualidade do produto através de um processo que também precisa ser gerenciado. Portanto, abordagens que visam contribuir para a qualidade do processo de testes de software contribuem indiretamente com a qualidade do produto de software. Assim, a estratégia de integração de processos de apoio que ajudam a gerenciar o processo de testes também contribui com essa qualidade. No entanto, esse tipo de abordagem é pouco desenvolvido na literatura técnica, principalmente no contexto de teste. Considerando essa necessidade, esta dissertação apresenta a definição de dois Processos de Apoio ao Processo de Testes, Processo de Gerência de Riscos de Teste e Processo de Gerência de Recursos Humanos de Teste, especificados e adequados com o objetivo de contribuir com a qualidade no aspecto gerencial do processo de testes de software. Esses processos foram definidos tendo-se como base diversos modelos de maturidade e normas técnicas. Para avaliar os processos propostos, uma revisão por pares foi conduzida com especialistas das áreas referentes aos processos especificados e os resultados dessa avaliação sugerem que os processos aqui definidos estão adequados e possibilitam a aplicação deles de forma integrada ao processo de testes.

O objetivo deste trabalho foi definir um processo para a avaliação de componentes de software, que possa ser utilizado em organizações que utilizem componentes comerciais ou de software livre desenvolvidos por terceiros no desenvolvimento de sistemas de software. O Desenvolvimento Baseado em Componentes (CBD - Component-based Development) é um caminho para a diminuição de custos e prazos no desenvolvimento de sistemas de software. A adesão a essa prática pelas organizações incentivou o surgimento do mercado de comercialização de componentes de software e a multiplicação de projetos de componentes de código aberto, distribuídos livremente pela Internet. No entanto, o uso de componentes de software desenvolvidos por terceiros possui um risco associado, pois um componente escolhido pode não possuir a qualidade esperada, ou não apresentar o comportamento desejado dentro das condições reais de uso. Neste contexto, o objetivo do processo de avaliação de componentes é levantar dados referentes à qualidade de um componente e interpretar esses dados, seja para adquirir confiança em um componente, ou para a seleção de um componente em casos em que existe mais de uma opção disponível. Como peça central da avaliação está o modelo de qualidade, que define os atributos de qualidade de cada fator de qualidade, e as métricas para apoiar a avaliação. / The objective of this work is to define a process to software component evaluation to organizations which use third party, commercial off-the-shelf or open source components in the development of software products. The Component-based Development (CBD) is an approach to reduce costs and time-to-market of software projects. The need of components has become an opportunity to the growth of the software component market and to multiplication of projects of open source software components, which can be freely downloaded from Internet. However, the use of third-party components in a project can be risky. It is possible that a component does not present the expected behavior or does not have the needed quality for specific conditions associated to the software system in development. In this context, the objective of the software component evaluation process is to obtain trustful information from the quality of a component and to analyze this information, assessing the component use in a particular context or selecting one component among similar products. The main part of this process is the use of Quality Model which establishes the quality factors and metrics to support the evaluation.

The ubiquitous nature of software demands that software is released without faults. However, software developers inadvertently introduce faults into software during development. To remove the faults in software, one of the tasks developers perform is debugging. However, debugging is a difficult, tedious, and time-consuming process. Several semi-automated techniques have been developed to reduce the burden on the developer during debugging. These techniques consist of experimental, statistical, and program-structure based techniques. Most of the debugging techniques address the part of the debugging process that relates to finding the location of the fault, which is referred to as fault localization. The current fault-localization techniques have several limitations. Some of the limitations of the techniques include (1) problems with program semantics, (2) the requirement for automated oracles, which in practice are difficult if not impossible to develop, and (3) the lack of theoretical basis for addressing the fault-localization problem. The thesis of this dissertation is that statistical causal analysis combined with program analysis is a feasible and effective approach to finding the causes of software failures. The overall goal of this research is to significantly extend the state of the art in fault localization. To extend the state-of-the-art, a novel probabilistic model that combines program-analysis information with statistical information in a principled manner is developed. The model known as the probabilistic program dependence graph (PPDG) is applied to the fault-localization problem. The insights gained from applying the PPDG to fault localization fuels the development of a novel theoretical framework for fault localization based on established causal inference methodology. The development of the framework enables current statistical fault-localization metrics to be analyzed from a causal perspective. The analysis of the metrics show that the metrics are related to each other thereby allowing the unification of the metrics. Also, the analysis of metrics from a causal perspective reveal that the current statistical techniques do not find the causes of program failures instead the techniques find the program elements most associated with failures. However, the fault-localization problem is a causal problem and statistical association does not imply causation. Several empirical studies are conducted on several software subjects and the results (1) confirm our analytical results, (2) demonstrate the efficacy of our causal technique for fault localization. The results demonstrate the research in this dissertation significantly improves on the state-of-the-art in fault localization.

Software product lines provide the opportunity to improve productivity, quality and time-to-market of software-based systems by means of systematic reuse. So as to accomplish systematic software reuse, elicitation of commonality knowledge is to be upheld by the analysis and management of variability knowledge inherent in domain requirements. Considerable effort is devoted to the management of functional variability, often neglecting the impact of quality concerns originating from non-functional requirements. In this thesis, a hybrid approach concentrating on the modeling of quantitative as well as qualitative concerns on quality has been proposed. This approach basically aims to support the domain design process by modeling non-functional variability. It further aims to support application design process by providing trade-off selection ability among quality concerns to control functional features that belong to the same domain. This approach is implemented and evaluated on an example domain to reveal its benefits on non-functional variability.

Οι εφαρμογές συναίσθησης (awareness applications) αποτελούν συστήματα ενίσχυσης της αντίληψης της παρουσίας (presence) και εντάσσονται στην ευρύτερη κατηγορία των συστημάτων υποστήριξης της επικοινωνίας με τη διαμεσολάβηση υπολογιστών (computer mediated communication systems). Μια πληθώρα τέτοιων εφαρμογών – μικρών και εξειδικευμένων ακόμη – έχει αναπτυχθεί τα τελευταία χρόνια (π.χ. MSN, Facebook), ενώ αναμένεται η ευρύτερη διάδοσή τους ως αποτέλεσμα της ανάπτυξης των κινητών επικοινωνιών. Η παρούσα διπλωματική εργασία βασίζεται στην πλατφόρμα ανάπτυξης εφαρμογών συναίσθησης ASTRA. Το συγκριτικό πλεονέκτημα της πλατφόρμας απέναντι στις υπάρχουσες εφαρμογές είναι ότι είναι γενικευμένη και συνεπώς μπορεί να οδηγήσει στην ανάπτυξη πλήθους εφαρμογών με σχετικά εύκολο τρόπο, δίνοντας τεράστια ώθηση στην ευρύτερη ανάπτυξη του πεδίου. Η εργασία αποσκοπεί στην δημιουργία ενός πλαισίου αξιολόγησης (evaluation framework) εφαρμογών συναίσθησης, το οποίο θα βασίζεται κυρίως στο πρότυπο ISO 9126. Για το σκοπό αυτό θα αναλυθεί το εν λόγω πρότυπο και θα εντοπιστούν οι ιδιαίτερες διαστάσεις των συστημάτων συναίσθησης (τα οποία θα αναλυθούν ως εφαρμογές διαδικτύου). Στη συνέχεια θα ομαδοποιηθούν τα χαρακτηριστικά τους και θα ενταχθούν στα τέσσερα στοιχεία εξωτερικής ποιότητας που προβλέπει το πρότυπο. Ως πεδίο εφαρμογής του πλαισίου θα χρησιμοποιηθεί η πλατφόρμα ASTRA, η οποία θα αξιολογηθεί με βάση αυτό. Για το σκοπό της εργασίας θα αναπτυχθούν και ορισμένες εφαρμογές συναίσθησης με τη χρήση της πλατφόρμας. / This thesis presents a quality evaluation framework for awareness systems. These are Computer Mediated Communication (CMC) systems, which support people’s needs to communicate and enhance their sense of presence. In general, quality is defined by the extent to which the software meets user needs and requirements. Our framework is based on the ISO9126 standard, which includes two quality dimensions, external and internal, and defines a set of quality criteria for each. We have chosen to adopt the end-user standpoint in evaluating awareness systems, thus we model only the four criteria groups of external quality, that is, functionality, usability, reliability and efficiency. Firstly, we present a brief introduction to software quality and the ISO9126 standard. Then, we present presence and awareness systems, and discuss their emergence and the needs they serve. We continue with a broad survey of presence and awareness systems and social software. For each of the surveyed systems (the list includes Facebook, MSN, Twitter, YouTube, LinkedIn et al), we present an extensive list of features, which support user/community management and application management. A comparative table summarizes the features of the systems surveyed. Based on this discussion, we develop the proposed evaluation framework, by listing detailed criteria for each of the four groups of external quality, according to ISO9126. For each group, we list two tables of criteria, one for community management evaluation and one for application evaluation. The criteria are sub-grouped into presentation, navigation, escalation, content, applications, communication, awareness, and connectedness. Each table is followed by a brief explanation of the criteria it contains and the range of values they assume. Then we present in detail ASTRA, the first pervasive CMC system that was designed and developed in order to enhance presence without requiring users to use a computer to join. ASTRA applies a pervasive computing interface on top of a social computing system, which supports community and awareness application management. The focus-nimbus model is used to describe the awareness framework and a novel connectedness theory is validated with the system. Finally, we evaluate and classify the presence and awareness systems, including ASTRA, according to the proposed evaluation framework.

The development of high quality software the first time, greatly depends upon the ability to judge the potential quality of the software early in the life cycle. The Software Engineering Research Center design metrics research team at Ball State University has developed a metrics approach for analyzing software designs. Given a design, these metrics highlight stress points and determine overall design quality.The purpose of this study is to analyze multiple software releases of the Advanced Field Artillery Tactical Data System (AFATDS) using design metrics. The focus is on examining the transformations of design metrics at each of three releases of AFATDS to determine the relationship of design metrics to the complexity and quality of a maturing system. The software selected as a test case for this research is the Human Interface code from Concept Evaluation Phase releases 2, 3, and 4 of AFATDS. To automate the metric collection process, a metric tool called the Design Metric Analyzer was developed.Further analysis of design metrics data indicated that the standard deviation and mean for the metric was higher for release 2, relatively lower for release 3, and again higher for release 4. Interpreting this means that there was a decrease in complexity and an improvement in the quality of the software from release 2 to release 3 and an increase in complexity in release 4. Dialog with project personnel regarding design metrics confirmed most of these observations. / Department of Computer Science

This case study examined the self-directed and team-based learning activities of a software quality assurance organization in central Indiana. The skills required to assure a high level of software quality evolve rapidly and software quality professionals must embrace ongoing technology and process changes. The thirty focus group participants performed a variety of quality assurance tasks including configuration management, research, automated test development, test planning and execution, and team leadership. The case study was based on semi-structured interviews of four focus groups of software quality professionals, and explored the learning styles, preferences, and activities deployed to learn new technologies and solve complex software problems. Software products are becoming increasingly pervasive in our culture. The study of continuing education for the software quality profession is important due to our increased reliance on this profession to meet customer expectations for high-quality software products. The proliferation of software products in our culture has also increased the demand for software quality professionals. Those professionals who have access to continuing professional education to improve and maintain skills have the opportunity to meet customer expectations. There is no mandated certification or licensing for this profession therefore professionals are left to chart their own course of learning. This study sought to understand how these software quality professionals meet their continuing professional educational needs. As well, the study identified key resources required to support such continuing professional education both within the workplace and off the job. Future study of the role of critical self-reflection in establishing learning objectives could enhance our understanding of how software quality professionals identify and plan their learning activities. Further investigation of the value of computer programming and logic knowledge to the software quality professional would benefit our understanding of baseline skill requirements for the various roles performed in the profession. There are also opportunities to engage in future action research projects on co-location of teams, mentoring, and job rotation strategies, as employees were found to learn effectively from peers. / Department of Educational Studies

While software subcontracting is a common business transaction in the information technology field, metrics specifically aimed at effectiveness of software subcontracting arrangements are not commonly available. This thesis makes a review of the literature and derives such metrics from fields of software quality, COTS acquisition and IS success. A case study is performed on software subcontracting projects of a Turkish defense contractor, and the project metrics are compared according to their success. The results suggest that metrics regarding the requirements are good indicators for subcontracting success and that larger projects may enjoy more success compared to smaller ones.

The main objective of this thesis is to identify how the software repository features influence software quality during software evolution. To do that the mining software repository area was used. This field analyzes the rich data from software repositories to extract interesting and actionable information about software systems, projects and software engineering. The ability to measure code quality and analyze the impact of software repository features on software quality allows us to better understand project history, project quality state, development processes and conduct future project analysis. Existing work in the area of software quality describes software quality analysis without a connection to the software repository features. Thus they lose important information that can be used for preventing bugs, decision-making and optimizing development processes. To conduct the analysis specific tool was developed, which cover quality measurement and repository features extraction. During the research general procedure of the software quality analysis was defined, described and applied in practice. It was found that there is no most influential repository feature and the correlation between software quality and software repository features exist, but it is too small to make a real influence.

A complexidade dos dispositivos embarcados propõe novos desafios para o desenvolvimento de software embarcado, além das tradicionais restrições físicas. Então, a avaliação da qualidade do software embarcado e seu impacto nessas propriedades tradicionais torna-se mais importante. Conceitos como reúso abstração, coesão, acoplamento, entre outros atributos de software têm sido usados como métricas de qualidade no domínio da engenharia de software. No entanto, elas não têm sido usadas no domínio do software embarcado. No desenvolvimento de sistemas embarcados outro conjunto de ferramentas é usado para estimar as propriedades físicas, tais como: consumo de energia, ocupação de memória e desempenho. Essas ferramentas geralmente envolvem custosos processos de síntese e simulação. Nos complexos dispositivos embarcados atuais deve-se confiar em ferramentas que possam ajudar na exploração do espaço de projeto ainda nos níveis mais altos de abstração, identificando a solução que representa a melhor estratégia de projeto em termos da qualidade de software, enquanto, simultaneamente, atenda aos requisitos físicos. Neste trabalho é apresentada uma análise da correlação entre métricas de qualidade de software, que podem ser extraídas antes do sistema ser sintetizado, e as métricas físicas do software embarcado. Usando uma rede neural nós investigamos o uso dessas correlações para predizer o impacto que uma determinada modificação no software trará às métricas físicas do mesmo software. Esta estimativa pode ser usada para guiar decisões em direção a melhoria das propriedades físicas dos sistemas embarcados, além de manter um equilíbrio em relação às métricas de software. / The complexity of embedded devices poses new challenges to embedded software development in addition to the traditional physical requirements. Therefore, the evaluation of the quality of embedded software and its impact on these traditional properties becomes increasingly relevant. Concepts such as reuse, abstraction, cohesion, coupling, and other software attributes have been used as quality metrics in the software engineering domain. However, they have not been used in the embedded software domain. In embedded systems development, another set of tools is used to estimate physical properties such as power consumption, memory footprint, and performance. These tools usually require costly synthesis-and-simulation design cycles. In current complex embedded devices, one must rely on tools that can help design space exploration at the highest possible level, identifying a solution that represents the best design strategy in terms of software quality, while simultaneously meeting physical requirements. We present an analysis of the cross-correlation between software quality metrics, which can be extracted before the final system is synthesized, and physical metrics for embedded software. Using a neural network, we investigate the use of these cross-correlations to predict the impact that a given modification on the software solution will have on embedded software physical metrics. This estimation can be used to guide design decisions towards improving physical properties of embedded systems, while maintaining an adequate trade-off regarding software quality.