Software requirements, and how they are constructed, shared and translated across software organizations, express uncertainties that software developers need to address through appropriate structuring of the process and the organization at large. To gain new insights into this important phenomenon, we rely on theory of work design and the travelling metaphor to undertake an in-depth qualitative inquiry into recurrent development of packaged software for the utility industry. Using the particular context of software provider GridCo, we examine how requirements are constructed, shared, and translated as they travel across vertical and horizontal boundaries. In revealing insights into these practices, we contribute to theory by conceptualizing how requirements travel, not just locally, but across organizations and time, thereby uncovering new knowledge about the responses to requirement uncertainty in development of packaged software. We also contribute to theory by providing narrative accounts of in situ requirements processes and by revealing practical consequences of organization structure on managing uncertainty.

The current generation of conventional software productivity tools is likely to achieve at most a factor of two reduction in life-cycle costs by the early 1990s. With projected order of magnitude increases in system complexity and size, a far greater improvement (factor of 10 or higher) is needed. Significant cost reductions and qualitative improvements for many kinds of applications can be demonstrated by means of a knowledge-based integrated tool environment that both adheres to the software development standards of the software development organization and promotes rapid development of high quality knowledge-based systems and their integration within highly specialized application environmentsThis investigation has led to the construction of the (K)nowledge (I)ntegration (T)ool, an operational testbed and architectural framework for the rapid development of highly extensible artificial intelligence systems and environments that both support the conventional life-cycle paradigm and facilitate the evolution of a knowledge-based life-cycle paradigm. A knowledge-based system is a programming system characterized by the ease with which objects, the relationships between them, and higher-level concepts composed of such objects and relationships, are manipulated and presented graphically and textually. The KIT consists of: knowledge-based integrated tool environments, integrated assemblages of knowledge-based systems that possess a man-machine interface that adjusts to the needs of individual users by means of user-profile and application-specific information; and a knowledge-based based system development environment, a knowledge-based system that supports the construction and maintenance of software systems, and acts as a mechanism to improve the reliability of the software development process. This thesis describes the synthesis of these system types in the KIT.Following the KIT's successful prototyping and demonstration, it is being scaled up and incrementally developed to provide life-cycle automation capabilities for a roboticized factory of a major aerospace company. In this thesis the historical and theoretical foundations, capabilities, current and planned uses of the KIT are described.Key Words: Artificial Intelligence, Knowledge Base, Life-cycle Automation, Knowledge-Based Environment, Knowledge-Based Systems, Knowledge-Based Project Management, Knowledge-Based Configuration Management, Knowledge-Based System Development, Knowledge-Based Software Engineering. / Department of Computer Science

Information Systems (IS’s) have provided organisations with huge efficiency gains and benefits over the years; however an outstanding problem that is yet to be successfully tackled is that of the troublesome maintenance phase. Consuming vast resources and thwarting business progression in a competitive global market place, system maintenance has been recognised as one of the key areas where IS is failing organisations. Organisations are too often faced with the dilemma of either replacement or the continual upkeep of an unwieldy system. The ability for IS’s to be able to adapt to exogenous influences is even more acute today than at any time in the past. This is due to IS’s namely, Web Information Systems (WIS’s) increasingly and continually having to accommodate the needs of organisations to interconnect with a plethora of additional systems as well as supporting evolving business models. The richness of the interconnectivity, functionalities and services WIS’s now offer are shaping social, cultural and economic behaviour on a truly global scale, making the maintenance of such systems and evermore pertinent issue. The growth and proliferation of WIS’s shows no sign of abating which leads to the conclusion that what some have termed as the ‘maintenance iceberg’ should not be ignored. The quandary that commercial organisations face is typically driven by two key aspects; firstly, systems are built on the cultural premise of using fixed requirements, with not enough thought or attention being paid to systems abilities to deviate from these requirements. Secondly, systems do not generally cope well with adapting to unpredictable change arising from outside of the organisations environment. Over the recent past, different paradigms, approaches and methods have attempted to make software development more predictable, controllable and adaptable, however, the benefits of such measures in relation to the maintenance dilemma have been limited. The concept of flexible systems that are able to cope with such change in an efficient manner is currently an objective that few can claim to have realised successfully. The primary focus of the thesis was to examine WIS post-development change in order to empirically substantiate and understand the nature of the maintenance phase. This was done with the intention to determine exactly ‘where’ and ‘how’ flexibility could be targeted to address these changes. This study uses an emergent analytical approach to identify and catalogue the nature of change occurring within WIS maintenance. However, the research framework design underwent a significant revision as the initial results indicated that a greater emphasis and refocus was required to achieve the research objective. To study WIS’s in an appropriate and detailed context, a single case study was conducted in a web development software house. In total the case study approach was used to collect empirical evidence from four projects that investigated post-development change requests in order to identify areas of the system susceptible to change. The maintenance phases of three WIS projects were considered in-depth, resulting in the collection of over four hundred change requests. The fourth project served as a validation case. The results are presented and the findings are used to identify key trends and characteristics that depict WIS maintenance change. The analytical information derived from the change requests is consolidated and shown diagrammatically for the key areas of change using profile models developed in this thesis. Based on the results, the thesis concludes and contributes to the ongoing debate that there is a discernable difference when considering WIS maintenance change compared to that of traditional IS maintenance. The detailed characteristics displayed in the profile models are then used to map specific flexibility criteria that ultimately are required to facilitate change. This is achieved using the Flexibility Matrix of Change (FMoC) tool which was developed within the remit of this research. This tool is a qualitative measurement scheme that aligns WIS maintenance changes to a reciprocal flexibility attribute. Thus, the wider aim of this thesis is to also expand the awareness of flexibility and its importance as a key component of the WIS lifecycle.

This research explores the concepts of defensive programming as currently defined in the literature. Then these concepts are extended and more explicitly defined. The relationship between defensive programming, as presented in this research, and current programming practices is discussed and several benefits are observed. Defensive programming appears to benefit the entire software life cycle. Four identifiable phases of the software development process are defined, and the relationship between these four phases and defensive programming is shown. In this research, defensive programming is defined as writing programs in such a way that during execution the program itself produces communication allowing the programmer and the user to observe its dynamic states accurately and critically. To accomplish this end, the use of defensive programming snap shots is presented as a software development tool.

Globally distributed software teams are widespread throughout industry. But finding reliable methods that can properly assess a team's activities is a real challenge. Methods such as surveys and manual coding of activities are too time consuming and are often unreliable. Recent advances in information retrieval and linguistics, however, suggest that automated and/or semi-automated text classification algorithms could be an effective way of finding differences in the communication patterns among individuals and groups. Communication among group members is frequent and generates a significant amount of data. Thus having a web-based tool that can automatically analyze the communication patterns among global software teams could lead to a better understanding of group performance. The goal of this thesis, therefore, is to compare automatic and semi-automatic measures of communication and evaluate their effectiveness in classifying different types of group activities that occur within a global software development project. In order to achieve this goal, we developed a web-based component that can be used to help clean and classify communication activities. The component was then used to compare different automated text classification techniques on various group activities to determine their effectiveness in correctly classifying data from a global software development team project.

This thesis will address problems of cross-platform software development for mobile devices. The author will analyze existing frameworks and tools; he will point out their major strengths and weaknesses. Based on this analysis, a pilot cross-platform application will be designed and implemented using the most versatile framework. Particular attention will be paid to the methodology of cross-platform development.

Call number: LD2668 .T4 1986 P68 / Master of Science / Computing and Information Sciences

This paper analyses and develops models for programmer participation and effort estimation in open source software projects. This has not yet been a centre of research, although any results would be of high importance for assessing the efficiency of this model and for various decision-makers. In this paper, a case study is used for hypotheses generation regarding manpower function and effort modeling, then a large data set retrieved from a project repository is used to test these hypotheses. The main results are that Norden-Rayleigh-based approaches need to be complemented to account for the addition of new features during the lifecycle to be usable in this context, and that programmer-participation based effort models show significantly less effort than those based on output metrics like lines-of-code. (author's abstract) / Series: Working Papers on Information Systems, Information Business and Operations

Background. Agile Software Development (ASD) promises agility and flexibility in dealing with uncertainty by prioritizing interaction between people supported by informal communication and knowledge sharing. The lack of practices to manage the knowledge as a resource might jeopardize the application of knowledge in the production of goods and service. The utilization of Knowledge Management (KM) strategies can significantly support achieving and sustaining competitive advantage and brings several benefits to software development. However, how to manage knowledge in ASD is still not well understood or investigated.  Objectives. The main objective of this thesis is to contribute to the software engineering field by providing a different perspective on directions that KM can take to improve knowledge-based resource (KBR) management in ASD. The detailed objectives are: (i) Understand the current ASD environment regarding KM; (ii) Identify KBRs in ASD and its implications for KM; and (iii) Provide an initial set of variables to evaluate knowledge criticality of knowledge items in ASD.  Method. We used a mixed-methods approach to address the objective of this thesis. The methods selected to conduct the studies include systematic literature review, grounded theory, and improvement case study. The data collection comprised a literature review, semi-structured interviews, and practitioners’ feedback through static validation.  Results. From our SLR we observed that that KM strategies in ASD promote mainly knowledge transfer through practices that stimulate social interaction to share tacit knowledge in the project layer, increasing the risk of losing knowledge by keeping the knowledge localized inside a few individual’s minds. When it comes to coordination, practitioners utilize KBRs in their routines, through social collaboration within teams’ environment/settings. However, this process is nonsystematic, which brings inefficiency to KBR utilization resulting in knowledge loss. It can generate negative implications to the course of the software development, including meaningless searches in databases, frustration because of recurrent problems, and unawareness of knowledge sources. To support decision making related to knowledge retention, we have developed an initial version of the method to evaluate the criticality (KCEM) of a knowledge item, which is divided into two categories, relevance, and scarcity.  Conclusion. The current results of this thesis are of particular interest. However, we recognize that the work is unfinished. As a complement to this thesis, we have planned our long-term objective, which is to contribute to creating scalable KM solutions for companies adopting ASD.We divide this long-term objective into three studies: Carry out a complementary study to apply KCEM in different companies; explore efficient ways of storing codified knowledge in combination with the KCEM, and investigate how to define metrics to evaluate the outcomes of KM practices. / S.E.R.T.E.R.T. Research Profile

A dissertation submitted to the Faculty of Science, University of the Witwatersrand, Johannesburg, for the degree Master of Science / The recording of seismic signals in remote areas requires a portable, low power recording system that can be left in the field for a few weeks at a time. Three components of ground motion are generally measured, and some form of event recording, rather than continuous recording should be available. [Abbreviated abstract. Open document to view full version] / AC2017