Catálogo de publicaciones - libros

A number of structural measures have been suggested to support the assessment and prediction of software quality attributes. The aim of our study is to investigate how class-level measures of structural properties can be used to assess the maintainability of a software product as a whole. We survey, structure and discuss current practices on this topic, and apply alternative strategies on four functionally equivalent systems that were constructed as part of a multi-case study. In the absence of historical data needed to build statistically based prediction models, we apply elements of judgment in the assessment. We show how triangulation of alternative strategies as well as sensitivity analysis may increase the confidence in assessments that contain elements of judgment. This paper contributes to more systematic practices in the application of structural measures. Further research is needed to evaluate and improve the accuracy and precision of judgment-based strategies.

The management and deployment of reuse-driven and architecture-centric requirements engineering processes have become common in many organizations adopting Enterprise Resource Planning solutions. Yet, little is known about the variety of reusability aspects in ERP projects at the level of requirements. Neither, we know enough how exactly ERP adopters benefit from reuse as part of the requirements engineering process. This paper sheds some light into these questions and suggests a practical approach to applied ERP requirements reuse measurement by incorporating reuse metrics planning as part of the implementation of metrics on an ERP project. Relevant process integration challenges are resolved in the context of SAP R/3 implementation projects in which the author participated while being employed at the second largest telecommunication company in Canada.

This paper presents a software process definition language that is targeted towards the development of software and systems using Model Driven Engineering methods. The dynamics of a process model are based on Petri Nets. This allows us to use a process definition model to plan and track the execution of actual projects. This new language can be integrated with existing approaches for software process modeling such as Software Process Engineering Metamodel.

A well-organized defect management process is one of the success factors for implementing software projects in time and in budget. The defect management process includes defect prevention, defect discovery and resolution, defect causal analysis, and the process improvement. However, establishing an organization-wide defect management process is a complicated task. The main research question in this paper is what kind of difficulties organizations have regarding the defect management process. Our findings show that problems are related to defect resolution reports, limited project resources for fixing defects, and challenges in creating a test environment. Results are based on our observations from four case organizations. The main contribution of this study is to help organizations to identify and avoid typical problems with defect management.

The creation of a proper work breakdown structure (WBS) is essential in performing successful project effort estimation and project management. The use of WBS is required on the level 1 of CMMI. There is, however, no standard WBS available. In this paper, the results of a pilot project in which new activities were introduced into the TietoEnator’s WBS are reported. The activities were non-construction activities which are necessary but not directly related to the actual software construction. The study shows that the success of the introduction of such activities very much depends on the naming of the activities and how they are introduced to the employees. Additionally, it turned out that the pre-thought set of non-construction activities included activities that should not have been in the set at all as individual activities.

The use of formal models such as Role Activity Diagrams (RADs) for analysing a process often hide what really happens during that process. In this paper, we build on previous research on informal aspects of the prototyping process and look at the key concerns that prototypers had during the prototyping process. We contrasted those concerns with an analysis of whether documented practice during prototyping was likely to exacerbate or lessen those concerns. The basis of our analysis was a set of interviews with prototypers all of whom were part of a team actively producing evolvable prototypes in an industrial setting. Grounded Theory was used to extract the relevant data (concerns and mitigating practice) from the interview text. Interestingly, only a small number of the concerns of prototypers seemed to be supported by any supportive action, suggesting that there are factors that contribute to project success or failure beyond the control of the prototyping team. However, time and cost pressure seemed to figure largest in our analysis of prototyper concerns. The research highlights the problems that prototypers face and the benefits that an informal analysis can have on our understanding of the process. It also complements our understanding of the formal analysis of process using techniques such as RADs and the human factors therein.

Effective usage of a general purpose modeling language in software engineering poses a need for language customization – adaptation of the language for a specific purpose. In the context of the Unified Modeling Language (UML) the customization could be done using two mechanisms: developing profiles and extending the metamodel of UML. This paper presents an industrial case study on the choice between metamodel extensions and profiles as well as the influence of the choice on the quality of products based on the extensions. The results consist of a set of nine prioritized industrial criteria which complement six theoretical criteria previously identified in the literature. The theoretical criteria are focused on the differences between the extension mechanisms of UML while the industrial criteria are focused on development of products based on these extensions. The case study reveals that there are considerable differences in effort required to develop comparable products using each mechanism and that the quality (measured as correctness of a product) is different for these comparable products by an order of magnitude.

In this paper we present our preliminary findings from a survey conducted during 2005 of Australian Multimedia Application Developers. Our objective was to understand what development processes and techniques are used and how these relate to practices cited in the literature. We were also interested in what impact the presence of multimedia content has on the process, as well as the differing skill sets it requires in relation to “traditional” software development. In our findings we report on the process models used and the process tasks most often performed, as a first step to determining what is considered best practice in the industry. We found that developers appear to have a much keener sense of their processes than previous studies have suggested.

In order to survive in a strong competition software houses need to design high-quality software. To achieve this some companies try to certify their software development processes in accordance with well-known industrial standards. Through a case study we investigated what characterizes the use of a quality system among developers and project managers in a large software company that has successfully achieved an ISO 9001:2000 certification. We found that certification not always indicates that the company successfully uses the practices in accordance with quality standards. This caused serious problems, such as projects that follow outdated practices, project managers faking quality documentation before audits, resources wasted by producing documents no one needs, problems created for new employees since they cannot find descriptions of the processes people are working in accordance with, and an expensive system no one uses.

In this paper we present findings from our empirical study of software process improvement (SPI) implementation. We aim to provide SPI practitioners with insight into designing appropriate SPI implementation initiatives in order to achieve better results. Thirty-four interviews were conducted with Australian practitioners. Three SPI implementation issues were investigated: reasons for embarking on SPI initiatives, SPI benefits to the management, and factors that play a positive role in SPI implementation. We have found that most common reasons for embarking on SPI initiatives are to: improve the quality of software developed, reduce software development cost, and increase productivity. Our results show that 71% of the practitioners said that SPI initiatives provided clear benefits to the management. We have also found that most frequently cited SPI implementation factors are: SPI awareness, defined SPI implementation methodology, experienced staff, staff time and resources, senior management commitment and training. Our aim of conducting this study is to provide a SPI implementation framework for the design of effective SPI implementation initiatives.