South African Computer Journal - Volume 2005, Issue 35, December 2005

This bumper edition of SACJ contains fifteen articles. They cover a wide spectrum of topics, illustrating SACJ's commitment to serve as a local platform for all relevant ICT-related research.

The most fundamental and far-reaching of changes we have seen in the field of software in the past few years, have not been technical, but rather changes in the normative assumptions surrounding the ownership of so called "intellectual property". A symptom of these changes has been an increase in the practice of patenting claims for software inventions. This paper provides a case study of one such patent which has been granted by the South African patent office. I argue that the patent itself is probably invalid on a number of grounds, not least of which being the exclusion of computer programs as patentable subject matter under the South African Patent Act. The paper outlines how a an under-resourced local patent office together with an under-developed national policy and legislation, is proving incapable of defending the national public interest in the face of determined international pressure and considerable private interests.

---

A brief overview is provided of the international "intellectual property" environment which forms the backdrop to this strange arrival in our local patent office. I go on to argue that this particular patent, if it were to be enforced, could have a number of negative consequences which far outweigh any value it may have to our local economy and well-being.

In this paper, the relative importance of trust and usability of E-commerce in South Africa is discussed. In order to investigate these issues, a heuristic evaluation and an empirical evaluation were conducted on a number of South African E-commerce sites. A comprehensive set of E-commerce design guidelines was compiled, which was used to conduct a heuristic evaluation of the selected E-commerce sites. The results of the heuristic evaluation indicated that a number of usability problems exist on the selected sites. The empirical evaluation results also identified a number of usability problems on the selected sites. A correlation analysis between the two sets of results was used to identify a number of E-commerce design guidelines that are of specific relevance to the South African E-commerce market. This paper proposes a trust model for E-commerce in South Africa, based on the guidelines that were identified as being significant for E-commerce in South Africa.

Alignment between business and information technology (IT) strategies has been a key concern for both IT and business managers for several decades. It is recognised that achieving alignment contributes immensely to ensuring that investments in IT result in improvements in organisational performance. Alignment is a concern, not only for commercial organisations, but also for tertiary-level educational institutions, many who have invested heavily in IT for teaching and educational purposes, as well as to improve efficiency and effectiveness of administrative processes. It is therefore necessary for research on alignment to be conducted in such institutions, as much of the literature assumes a commercial organisational context. In this paper, a case study is reported in which it was found that integration between business and information systems (IS) planning, rational-adaptation in IS planning, IT managerial resources, and IT implementation success are all factors likely to influence alignment. Their relative importance, however, is very much dependent on the organisational context, the timing of the study, and on whose perspective is being sought.

Measuring user satisfaction with information systems has attracted widespread research attention, given it is often used as an indicator of success. The Internet has allowed applications to be extended to customers of an organization, where interaction can take place through a web site, typically from home or office. The focus of attention with such applications is customer satisfaction. In this research, a 21-item, 7-factor instrument developed to measure customer satisfaction with web sites that market generic digital products and services was modified slightly, and then empirically tested and validated in the context of Internet banking specifically. A 19-item, 5-factor validated instrument emerged, the factors being Customer Support, Security, Ease of Use, Transactions and Payment, and Information Content and Innovation. The difference in number of factors as compared to the generic instrument was attributed to the unique nature of Internet banking web sites. These and other findings are discussed in the paper, and their implications examined.

The adoption of Information and Communication Technologies (ICTs) in developing countries often comes with the argument of how the technology will assist in bridging the 'digital divide'. This paper argues1 that often ICTs fail to realise their potential due to the overly technical focus of the information systems design process. A case study from South Africa illustrates how the design of an information system, which takes the social situation into account, can address exclusion. Using an interpretive participatory action research approach a community-based information system was designed, in one municipality in South Africa, to address the vulnerability of children. The revised information system is primarily a paper and oral based system, but links with the formal health facility information systems at higher levels. Thus, though ICTs were not directly used in the design of the information system due to the nature of the particular context, the need to connect with the higher levels of the computerised health information system was recognised. The case study illustrates the importance of designing information systems which are culturally and technologically appropriate and through this process attempts to address the exclusion of certain members of society from the health system.

Discovering useful knowledge is a problem that remains elusive, even for publicly solved games such as Checkers. This paper presents a formal language F that represents knowledge that can be used by two-player board game agents. The language assumes minimal domain knowledge and as basic symbols it employs information that can be obtained directly from the observation of a game position. The language operators can be used to construct and manipulate knowledge within the context of an evaluation function.

With the advent of the Internet came the promise of global information access. In keeping with this promise, Digital Libraries (DLs) began to emerge across the world as a method of providing structured information to their users. These DLs are often created using proprietary monolithic software that is usually difficult to customise and extend. The Open Digital Library (ODL) project was created to demonstrate that DLs can be built as a network of components instead of as monolithic systems. Although the ODL approach has largely been embraced by the DL community, it is not without a few shortcomings. This paper introduces a graphical user interface and its associated framework for creating DLs from distributed components, consequently addressing a number of the limitations of ODL-like systems, as well as presenting a novel and generic approach for creating component-based systems. This system was subject to a user-based evaluation to confirm its utility and provide insights into possible extensions.

A new model of computation based on axiomatic logic is explored. The significant difference from other models is how reasoning about variables is expressed. The computation is described only by defining variables in terms of other variables. The variables have the mathematical property that each variable has a unique reference. The number of variables are not limited. Generalized relationships allow define classes of infinitely many variables. The benefits are: the simplicity of semantics of the axiomatic logic that simplifies the formal reasoning about a program; the model only captures the data dependencies resulting in an inherently parallel model of computation, and the expression of a program maps directly onto the model of computation. The unanswered question is the feasibility of an efficient hardware implementation. The model is described in the form of a tutorial that develops the model from simple natural deduction using axiomatic reasoning to do the computation. A simple example is used to illustrate how a program can be expressed in this model and how a result is inferred for a given input.

In response to the South African Computer Journal's editorial of June 2003, this paper reviews some of the philosophical issues underpinning academic research in Information Systems (IS). This is a reflective or theoretical paper, which considers a number of research paradigms and traditions and explores the relevance of Critical Realism as an appropriate philosophical base for IS Research. This paper does not intended to provide a definitive description of these research traditions. Rather it is hoped that the paper will stimulate a discussion about the use of these different paradigms, and in so doing respond to the call for IS researchers to be more aware of the research methodological options available to them. The paper concludes by suggesting that Critical Realism adds to the bouquet of philosophical options available to the IS researcher, and that it is an appropriate approach that transcends the limitations of both positivism and interpretivism.

Inadequate requirements specification remains to be one of the predominant causes of software development project failure today. This is mainly due to the lack of suitable processes, techniques and automated tool support available for specifying and analysing system requirements. In this paper we suggest a way to improve the approach to requirements specification that is the most popular at the moment - use case modelling. Despite their popularity, use case models are not adequate for creating comprehensive and precise requirements specifications. We amend the traditional use case metamodel such that more structured models with a precise meaning can be built. Further, we define several analysis schemes for these structured use case models that assist in discovering inconsistencies and other errors in the models. These analysis schemes are automated in a tool that we developed called the Structured Use case Model Analyser (SUM Analyser). The SUM Analyser provides an accessible interface that allows the user to construct use case models, configure and execute several analysis options and view the produced results. The existing NuSMV model checker is used to perform the actual verification tasks for the analysis. To facilitate this, the SUM Analyser transforms use case models to NuSMV programs and also interprets the produced results so that they can be understood by the user.

This paper describes a survey of 497 final year scholars at schools in the Western Cape, South Africa. They were asked about their inclination to study information systems(IS) or computer science(CS), their knowledge of the areas, and their opinions of salaries and job opportunities. Unlike much past research, significantly more interest came from those previously 'deprived' in the IT area. Those from schools previously disadvantaged by the apartheid school system and without access to computer or Internet access at home were more inclined to study CS or IS. These categories also expressed most optimism about IT job prospects and starting salaries, as did female scholars. Perceptions of the activities of IS students were far less accurate than of CS students. Males were more predisposed to study IS than females, and black students and those without school computer access favoured CS. Possible reasons for this are discussed.

The purpose of this article is to expose issues and concerns about the possibilities and limitations of Computer Assisted Qualitative Data Software (CAQDAS). Recent literature elucidates a rapid development in this field. Such developments may impact negatively on student learning; particularly those who wish become skilled in Qualitative Research (QR) and more specifically in Qualitative Data Analysis (QDA) with the use of advanced software. The software developers are speedily improving programs for qualitative research, adding to the complexity, comprehensiveness and methodological rigour of QDA. In turn, the demand for sophisticated computer software, certainly internationally, has increased greatly, that software developers had to respond to this need. That is why both students and experienced researchers will have to take cognisance of these rapid improvements in the field of QDA. To this end, I unravel several misconceptions of CAQDAS in QDA and shed light on intended purposes and processes, with special reference to postgraduate supervision and propose how CAQDAS may add methodological rigour to QDA. Finally, I put forward my concluding reflections and some affinities of CAQDAS.

Web-based learning (WBL) can be defined as any form of learning that is partly or fully based on material, applications or communication that is delivered over the World Wide Web (WWW). WBL tools provide integrated environments of various technologies to support diverse educators' and learners' needs via the Internet. One of the biggest challenges facing WBL environments is the organization, communication and presentation of information in a logical and usable manner via the WWW. A good information architecture design can increase the usability of a website. Information architecture is concerned with the design of organization, labeling, navigation and searching systems to help users find and manage information. Several Higher Education (HE) institutions in South Africa have developed their own WBL tools some of these, however, exhibit usability problems with respect to information architecture (IA). The goal of this paper is to investigate and discuss the issues involved with the IA of WBL tools developed in South Africa.

The Established Strategy for drawing up a Z specification document focuses on a more abstract activity of specification, namely, how to combine schemas but is largely silent about how to construct a schema. Schema construction may benefit from the application of certain heuristics for establishing its content. While formal specification can be seen as a subsection of software engineering and design heuristics in this area are firmly in place, corresponding principles and strategies for constructing a formal specification have been relatively rare. In this paper we examine a number of formal specifications written in Z as well as some design principles from software engineering and areas of general design. On the strength of these, we propose a preliminary set of heuristics for the construction of a formal specification and show how these may be embedded in an enhanced strategy for specification work. We illustrate how one such enhancement, namely the use of primitives, allows a specifier to discharge an important proof obligation arising from a formal specification, where otherwise a proof is not easily arrived at.

This paper presents an account of information systems evaluation, in which it is argued that the problems of evaluation as described in the literature only seem to exist because we accept and depend on the dualism or assumed split between subject and object. The resultant subjective / objective continuum in IS evaluation points to one of two stereotypes of IS evaluation and the manager engaged in this process: the objective / rational manager utilising objective / rational methods versus the subjective / political manager engaged in political manoeuvring, utilising objective / rational methods only as ritual or symbolism. Neither of these opposing stereotypes is satisfactory. Instead, this paper proposes a dialectic view of information systems evaluation, in terms of which, rather than being a decision maker, the manager is in-the-world, evaluating systems in order to get the job done, on the basis of her thrownness in-the-world.

---

In the organisation, managers evaluate systems through a process of organisational learning as encultured knowing, in terms of which a narrative, situated, pragmatic knowledge is most useful: Evaluation happens in the course of conversation. Such conversation is, however, not always skilful because the organisation is not just a collection of individuals but also a network of power relations. Conversations are never held outside of power: Systems evaluations as conversations cannot take place outside of a regime of truth. Therefore, the means to achieve a genuinely skilful conversation is to ensure that it is both improvisatory and deconstructive, by applying the suggested principles for information systems evaluation based on these ideas.