Large flexible software systems tend to incur “bloat”, here defined as the runtime overhead induced by the accumulation of excess functionality and objects. Removing bloat is hard as these overheads are a side-effect of the same trends that have fuelled software growth. Even defining and measuring bloat is non-trivial, as software doesn’t come with built-in labels that indicate which portions of computation are necessary for a given application and which lead to bloat. Much progress has been made in novel analysis tools that aid (human experts in) the process of finding bloat by highlighting signs of excessive activity and data flow. However, there has been very little research focus on understanding the connection between sources of bloat and its system level implications. In particular, excess resource usage due to bloat could be a significant source of power-performance inefficiencies, but the relation between bloat and energy efficient design remains unexplored. In order to systematically devise effective mechanisms for reaping power-performance benefits through bloat mitigation, we require a deeper insight into exactly when excess features can originate bloat, when the resource overheads of bloat are most pronounced and when bloat matters for power performance. This dissertation explores the problem of software bloat and its energy efficiency implications from multiple perspectives to develop a better understanding of these connections. First, we establish the need for a whole systems perspective in assessing potential energy efficiency benefits of bloat reduction, based on a systematic empirical and analytical study that highlights a curious interplay between bloat, energy proportionality and system bottlenecks. Second, we present a novel static analysis algorithm to perform an automated code transformation for object reuse that mitigates bloat involving the generation of excess temporary objects within loops. Third, we introduce the idea of concern augmented program analysis (CAPA), to identify sources of bloat due to excess features; the technique uses externally supplied information about program concerns and their properties as an abstraction of underlying intent. Fourth, as an early diagnostic aid, we use a statistical topic model to automatically discover latent concerns from source code statements and then correlate these latent concerns with resource usage properties that vary at statement granularity. The statistical model has a built-in sensitivity to the context of individual statements so that it can discover even diffused concerns without any apriori concern information. Together, our findings show that presence of excess features, in itself, may not lead to (runtime) bloat, unless these features have some structural interaction with essential features. Further, the overheads due to such structural interactions, in turn, may not cause substantial bloat in the resource consumption of a long running (server) application unless incurred repeatedly during program execution. Finally, even such bloated resource usage has a pronounced impact on power-performance only if it affects a system bottleneck or a hardware resource that has a high degree of energy proportionality and consumes a high fraction of power compared to the other system resources. We conclude that energy wastage due to bloat need not be an inevitable consequence of over-provisioning flexibility. Instead, the extent to which excess features result in runtime bloat and poor power-performance is determined by certain characteristics of the program structure and of the underlying hardware system --these represent potential control points that could be exercised to develop principled design approaches for mitigating bloat without sacrificing flexibility or productivity.

The purpose of this study was to explore nurse-patient encounters from the perspective of the Home Healthcare Registered Nurse. A qualitative descriptive design was used to collect data from a purposive sample of 20 home healthcare registered nurses from Connecticut, Massachusetts, and Rhode Island currently or previously employed as a home healthcare nurse. Four themes and one interconnecting theme emerged from the data: Objective Language; Navigating the Unknown; Mitigating Risk; Looking for Reciprocality in the Encounter; and the interconnecting theme of Acknowledging Not All Nurse-Patient Encounters Go Well. One goal of the study was to propose an empirically informed definition of what constituted a difficult encounter. An important early finding was that the terms difficult patient and difficult encounter were not generally used by study participants. HHC RNs voiced a preference for objective and nonjudgmental language to communicate outcomes of nurse-patient encounters. Three types of HHC RN-patient interactions emerged from the data, with constructive encounters the norm and non-constructive or destructive encounters less frequent. A constructive encounter is when two or more human beings, the nurse on the one side, and the patient, caregiver, or both on the other, interact to achieve a mutually agreed upon outcome. A nonconstructive encounter is when one or more human beings obstruct efforts to achieve at least one positive outcome. A destructive encounter is when one or more human beings direct anger at or physically aggress toward another human being. Strategies to promote reciprocality are routinely employed during HHC RN-patient encounters, but HHC RNs who miss cues that a strategy is ineffective or failed may be at risk in the home. Study data lend support to key concepts, assumptions, and propositions of Travelbee’s (1971) Human-to-Human Relationship Model. Study results provide a foundation for further research to increase the understanding, recognition, and development of empirically derived responses to non-constructive or destructive encounters such that HHC RNs are safe and best able to meet patients’ healthcare needs.

The death penalty has been an issue of debate for decades and it is of great relevance at present. Different reasons have emerged that make recourse to the death penalty appear necessary, such as, that it serves as a deterrent, it meets the need for retribution and that public opinion demands its imposition. Conversely, more convincing arguments have been raised for its abolition, amongst which is the argument that it is a violation of human rights. Africa is seen as one of the “death penalty regions” in the world, as most African states still retain the death penalty despite the growing international human rights standards and trends towards its abolition. Further, the African Charter on Human and Peoples’ Rights makes no mention of the death penalty. The death penalty in Africa is therefore an issue that one has to be particularly concerned about. During the 36th Ordinary Session (2004) of the African Commission on Human and Peoples’ Rights, for the first time, the death penalty was one of the issues discussed by the Commission. Commissioner Chirwa initiated debate about the abolition of the death penalty in Africa, urging the Commission to take a clear position on the subject. In view of this and the international human rights developments and trends on the death penalty, discourses on the abolition of the death penalty in Africa are much needed. Accordingly, this study examines the death penalty in Africa from a human rights perspective. It seeks to determine why African states retain the death penalty, the ways in which the current operation of the death penalty in African states conflicts with human rights, what causes obstructions to its abolition in Africa, and whether it is appropriate for African states to join the international trend for the abolition of the death penalty. The current status and operation of the death penalty in Africa is first examined. The historical background to the death penalty in Africa from a traditional and western perspective is also discussed. Subsequently, the main arguments advanced by Africans (including African leaders, writers, priests and government officials) for the retention of the death penalty in Africa are evaluated. The study goes further to examine the death penalty in African states in the light of the right to life, the prohibition of cruel inhuman and degrading treatment and fair trial rights at both the international and national levels. After examining the death penalty in African states, the study arrives at the conclusion that it is appropriate for African states to join the international trend for the abolition of the death penalty, considering that the death penalty in Africa conflicts with human rights, the justifications for its retention are fundamentally flawed, and that alternatives to the death penalty in Africa exist. A number of recommendations are then made, which are geared towards the abolition of the death penalty in Africa. / Thesis (LLD)--University of Pretoria, 2006. / Centre for Human Rights / unrestricted

Stakeholder involvement in strategic planning: a strategy to mitigate the effects of HIV and AIDS on secondary education in Botswana. This study explores the involvement of stakeholders in strategic planning to mitigate the effect of HIV and AIDS in secondary education in Botswana. The prevalence of HIV and AIDS-related illness and deaths is high in Botswana and affects both teachers and learners. Education provision has been affected through increased mortality and morbidity and increased absenteeism which affect education-related personnel and the demand for education has been reduced due to growing numbers of orphaned and vulnerable children as a consequence of parent/guardian mortality and morbidity related to HIV and AIDS. The problem was investigated by means of a literature review and an empirical inquiry which combined quantitative and qualitative data collection. Based on the findings, in addition to medical and other interventions, an education management approach is required to mitigate the effects of HIV and AIDS on secondary education in Botswana. The strategies of coping, caring and preventing have been effective in this regard. Some internal stakeholders of the Department of Secondary Education (DSE) are not meaningfully involved in strategic planning. Similarly, most of the selected external stakeholders were not involved in the DSE HIV and AIDS strategic plan. Both external and internal stakeholders should be involved at all stages of planning. Furthermore, inducement-contribution exchanges and teacher credibility should be considered in a strategic plan. To improve the current DSE strategic plan, a stakeholder involvement model to involve internal and external stakeholders was designed. Based on this model and the above findings, recommendations for practice and suggestions for future research are made. / Educational Studies / D.Ed.(Educational Management)

Stakeholder involvement in strategic planning: a strategy to mitigate the effects of HIV and AIDS on secondary education in Botswana. This study explores the involvement of stakeholders in strategic planning to mitigate the effect of HIV and AIDS in secondary education in Botswana. The prevalence of HIV and AIDS-related illness and deaths is high in Botswana and affects both teachers and learners. Education provision has been affected through increased mortality and morbidity and increased absenteeism which affect education-related personnel and the demand for education has been reduced due to growing numbers of orphaned and vulnerable children as a consequence of parent/guardian mortality and morbidity related to HIV and AIDS. The problem was investigated by means of a literature review and an empirical inquiry which combined quantitative and qualitative data collection. Based on the findings, in addition to medical and other interventions, an education management approach is required to mitigate the effects of HIV and AIDS on secondary education in Botswana. The strategies of coping, caring and preventing have been effective in this regard. Some internal stakeholders of the Department of Secondary Education (DSE) are not meaningfully involved in strategic planning. Similarly, most of the selected external stakeholders were not involved in the DSE HIV and AIDS strategic plan. Both external and internal stakeholders should be involved at all stages of planning. Furthermore, inducement-contribution exchanges and teacher credibility should be considered in a strategic plan. To improve the current DSE strategic plan, a stakeholder involvement model to involve internal and external stakeholders was designed. Based on this model and the above findings, recommendations for practice and suggestions for future research are made. / Educational Studies / D.Ed.(Educational Management)