The purpose of this project is to implement an integrated piece of software consisting of a number of graphics programs that support mathematical computation and experimentation in dynamical systems.

O processo de desenvolvimento de sistemas de grande porte, envolvem particularidades que tornam a sua gestão mais complexa, quando comparada ao desenvolvimento de sistemas tradicionais. Estudos ilustram resultados favoráveis quando aplicado a filosofia ágil durante o processo de fabricação de um software, no entanto, desperdícios ainda são observados quando o contexto do desenvolvimento são aplicações de grande porte. Assim sendo, a pesquisa em questão procura a redução de desperdícios identificados durante o processo de desenvolvimento de sistemas de grande porte. Foram realizados estudos teóricos e práticos objetivando analisar as técnicas e ferramentas Lean para obtenção de resultados mais eficientes. O estudo apresenta e discute conceitos do Lean para desenvolvimento de software e a sua relação com as metodologias ágeis para desenvolvimento de sistemas de informação. Resultados apontam melhorias no processo de desenvolvimento de software além da redução de desperdícios com superprodução de funcionalidades e defeitos encontrados. A dissertação em questão será constituída de uma etapa de revisão da literatura, e duas próximas etapas com aplicações práticas de ferramentas e técnicas Lean para a redução de desperdício no processo de desenvolvimento de software de grande porte. / The development process of large systems involves particularities that make their management more complex when compared to the development of traditional systems. Studies illustrate favorable results when applied to agile philosophy during the manufacturing process of a software, however, wastes are still observed when the development context are large applications. Therefore, the research in question seeks the reduction of wastes identified during the development process of large systems. Theoretical and practical studies were carried out to analyze Lean techniques and tools to obtain more efficient results. The study presents and discusses Lean concepts for software development and its relationship with agile methodologies for the development of information systems. Results point to improvements in the software development process besides the reduction of waste with overproduction of functionalities and defects found. The dissertation in question consists of a literature review stage, and two next steps with practical applications of tools and Lean techniques to reduce waste in the large software development process.

South Africa has shown an increased interest and awareness of Open Source Software (OSS) in the past decade. One of the reasons for this was the support from the Shuttleworth Foundation for Open Source initiatives. Migrating to OSS is a difficult and time consuming activity that should not be underestimated by the migration team. Careful planning and roll-out procedures should be in place before one commence on this journey. Process reference models are used in different fields to capture the generic process flow of activities. For the OSS domain, no process reference models could be found for migration purposes. Therefore, this study has as aim the suggestion of an initial set of process reference models for an organisational OSS migration. These process reference models were identified by capturing the process models for a case study that entailed the migration of the CSIR software systems and desktops from proprietary to OSS. From this set of process models, the migration processes were identified and refined to a set of suggested process reference models for organisational OSS migrations. This set of process reference models are useful to determine the processes necessary for organisations considering migrating to OSS. The study is divided into four research questions, where the first focusses on use and value of process reference models and the second on what is already known about OSS migration processes. The third deals with key processes within an organisational open source migration (OOSM) and the last with process reference models for an OOSM. For the first research question, the use and value of process reference models and the usefulness of utilising process reference models is discussed as well as using process models as a modelling tool to identify and capture processes. For the second research question, a summary is provided of what we know about OSS migration processes and a description about what the researcher and others have learnt about OSS, OSS migrations, process reference models, the process and its structure. For the third research question, the key processes within an OOSM is discussed as well as all the processes that took place during the OSS migration project from basic administrative processes to complex processes, from the beginning of the project until its completion. Lastly, for the fourth research question, process reference models that are essential for an OOSM and possible generic migration process models bound to reoccur are identified by the researcher and validated using a focus group discussion. / M.Tech. (Information Technology)

Software project development is increasingly becoming geographically distributed especially in today’s global business environment. In order to build quality software faster andat a lower cost, companies in industrialised countries are turning to globally distributed software development projects. Emerging countries such as India and Israel are known to have large pools of highly trained software engineers at relatively low cost. However, recent trends have shown that new locations are emerging as developmental sites. Examples of those locations in the Middle East include Oman, Dubai, Abu Dhabi and Qatar. The main attraction of those countries is that they are close to the Arab market. However, distributed development also entails a number of risks and problems, for instance in terms of communication and knowledge transfer between sites, resulting in project failures. Being the project manager of a team that includes remote members has additional challenges compared to managing co-located team members. These projects face particular challenges that need careful managerial attention.This research aims to examine how globally distributed software development projects are managed in the Middle East. The main question to be answered was “How do companies in the Middle East successfully manage and organise software development in a globally distributed environment?”. We outline the factors that were perceived as important by project managers. We also look at how culture affects the operation of virtual teams in a globally distributed software development environment in the Middle East taking religion into consideration. To understand how these factors influence the success of globally distributed software development projects in the Middle East, we conducted a qualitative case study research involving three different companies operating in Dubai, Oman, Abu Dhabi, India and Pakistan. We used semi-structured interviews as the primary sources of information as these allow highly personalised data. This research outlines the following specific project management-related areas that need to be addressed to facilitate successful virtual team operation: Cross-cultural collaboration • Tools and technology used to facilitate GSD • An infrastructure that facilitates communication between virtual teams • Effective project management / Computing / M. Sc. (Computing)

O desenvolvimento de software entre equipes geograficamente distribuídas tem adquirido relevância devido a interesses das organizações, relacionados a custos de projetos, busca de profissionais multidisciplinares e especializados ou visando ampliar a cobertura de seus clientes, atendendo-os de forma mais globalizada. Porém, apesar dos benefícios existentes, há algumas dificuldades oriundas da distribuição de equipes que são inerentes da dispersão temporal e geográfica. Estas dificuldades incluem as diferenças culturais, linguísticas, temporais entre outras, que comprometem a qualidade da comunicação e consequentemente interferem na compreensão mútua entre as equipes que desenvolvem um mesmo projeto de software. O objetivo deste trabalho é apresentar uma melhoria de um processo de desenvolvimento distribuído de software (DDS), baseado em Scrum, capaz de minimizar as interferências e perdas de informações, descentralização de conhecimentos em pessoas específicas e melhorar no fluxo de comunicação e disseminação das informações. Um caso real foi abordado e para o desenvolvimento deste trabalho o método de pesquisa adotado foi a pesquisa-ação. Neste cenário real haviam problemas originados da comunicação ineficiente entre seus atores e exigiu que o desenvolvimento desta pesquisa tivesse várias iterações e retroalimentações até atingir a maturidade. A melhoria da comunicação foi alcançada e é a principal contribuição deste trabalho. / Software development between geographically distributed teams has acquired relevance due to interests of organizations, project-related costs, looking for multidisciplinary and specialized professionals or aiming to expand the coverage of its customers, serving them in a globalized form. However, despite the existing benefits, there are some difficulties originated from the distribution of teams that are inherent in the temporal and geographic dispersion. These difficulties include the cultural, linguistic, temporal among others, which affect the quality of communication and consequently interfere with the mutual comprehension between the teams which develop the same software project. Thus, the aim of this work is to present an improvement in the process of distributed software development (DSD) based on Scrum, capable to minimize interference and information losses, decentralization of knowledge in specific people and improve the communication flow and dissemination of information. A real case was discussed and as for the development of this work the action research was the approach employed in this study. In this scenario, problems originated from inefficient communication between their actors and demanded that the development of this research present several iterations and feedbacks in order to reach its maturity. Improved communication has been achieved and is the main contribution of this work.

O desenvolvimento de software entre equipes geograficamente distribuídas tem adquirido relevância devido a interesses das organizações, relacionados a custos de projetos, busca de profissionais multidisciplinares e especializados ou visando ampliar a cobertura de seus clientes, atendendo-os de forma mais globalizada. Porém, apesar dos benefícios existentes, há algumas dificuldades oriundas da distribuição de equipes que são inerentes da dispersão temporal e geográfica. Estas dificuldades incluem as diferenças culturais, linguísticas, temporais entre outras, que comprometem a qualidade da comunicação e consequentemente interferem na compreensão mútua entre as equipes que desenvolvem um mesmo projeto de software. O objetivo deste trabalho é apresentar uma melhoria de um processo de desenvolvimento distribuído de software (DDS), baseado em Scrum, capaz de minimizar as interferências e perdas de informações, descentralização de conhecimentos em pessoas específicas e melhorar no fluxo de comunicação e disseminação das informações. Um caso real foi abordado e para o desenvolvimento deste trabalho o método de pesquisa adotado foi a pesquisa-ação. Neste cenário real haviam problemas originados da comunicação ineficiente entre seus atores e exigiu que o desenvolvimento desta pesquisa tivesse várias iterações e retroalimentações até atingir a maturidade. A melhoria da comunicação foi alcançada e é a principal contribuição deste trabalho. / Software development between geographically distributed teams has acquired relevance due to interests of organizations, project-related costs, looking for multidisciplinary and specialized professionals or aiming to expand the coverage of its customers, serving them in a globalized form. However, despite the existing benefits, there are some difficulties originated from the distribution of teams that are inherent in the temporal and geographic dispersion. These difficulties include the cultural, linguistic, temporal among others, which affect the quality of communication and consequently interfere with the mutual comprehension between the teams which develop the same software project. Thus, the aim of this work is to present an improvement in the process of distributed software development (DSD) based on Scrum, capable to minimize interference and information losses, decentralization of knowledge in specific people and improve the communication flow and dissemination of information. A real case was discussed and as for the development of this work the action research was the approach employed in this study. In this scenario, problems originated from inefficient communication between their actors and demanded that the development of this research present several iterations and feedbacks in order to reach its maturity. Improved communication has been achieved and is the main contribution of this work.

This project created a program that takes predicate calculus formulas and creates a visual Semantic Tableaux truth tree, thereby proving or disproving a conclusion. Formal methods used in developing and verifying software and hardware are mathematically based techniques for describing and reasoning about system properties. Such formal methods provide frameworks within which people specify, develop, and verify systems in a systematic, rather than ad hoc, manner. Formal methods include the more specific activities of program specification, program verification and hardware verification.

The purpose of this thesis is to provide the California State University, San Bernardino, Department of Computer Science with an analysis and recommended solution to improving the software development process.

Call number: LD2668 .T4 1986 C68 / Master of Science

Computing / (M.Sc. (Information Systems))