Catálogo de publicaciones - libros

In the paper [1] we propose a taxonomy for classifying COTS business applications, i.e. products that are used in the daily functioning of all types of organizations worldwide, such as ERP systems and document management tools. We propose the identification of characterization attributes to arrange the domains which these products belong to, and also we group these domains into categories. We define questions and answers as a means for browsing the taxonomy during COTS selection. We show the need of identifying and recording the relationships among the domains and propose the use of actor-oriented models for expressing these relationships as dependencies. Last, we explore the definition of quality models for the domains, to be used in COTS selection, focusing on their reusability and stepwise definition downwards the hierarchy.

Earned value management (EVM) has long been used by organizations to plan, monitor, and control the development and evolution of custom developed systems. EVM was developed for managing such projects, and assumes a waterfall development model. COTS-based systems (CBS), on the other hand, are formed and evolved through the selection and composition of pre-existing, off-the-shelf packages or components with potentially some number of custom components. Experience indicates that a spiral or iterative development process is a key to success with CBS. While EVM has been applied to CBS projects, the results have not been uniformly satisfying. This paper explores the fundamental challenges in using EVM with CBS, and proposes adaptations to some of the principals of EVM to render it more suitable for CBS development.

A business process is a collection of activities that are required to achieve a business goal and it is represented with an activity flow that specifies the orchestration needed to complete the goal. The definition of these processes allows business people to easily integrate the functionalities of the COTS in the company to support the business objectives. This activity flow can be implemented in two ways, using traditional methods or using a Business Process Definition Language (BPDL). Traditional methods encode the activity flow using state of the art programming languages such as Java, C#, etc. BPDLs describe the activity flow with a specific language that is directly interpreted by a BPDL engine. This paper analyses the use of BPDLs and traditional methods to develop solutions for services-based architectures. It presents a case study where the results obtained using a BPDL and a traditional method are compared.

Nowadays, we can take the European single market for granted. With old barriers gone, people, goods, services and money move around Europe as freely as within one country. However, single market is not possible just sweeping away the technical, regulatory, legal, bureaucratic, idiomatic, cultural and protectionist barriers, but it has been essential to work hardly to defend fair competition as a simple and efficient means of guaranteeing consumers a level of excellence in terms of the quality and price of products and services, and to grant consumer protection to improve the quality of life of all European citizens. In this paper, we analyse the impact of these areas EU legislation on the COTS industry, how the COTS single market is forming and what and mean and imply.

Organizations that are acquiring a COTS based system must adapt many of their acquisition process activities that are traditionally used for acquiring non-COTS based systems. Much of this adaptation becomes quite difficult within government environments where the process is often constrained by government rules and regulations. This paper provides an experience report on COTS based acquisition for a government agency during the early stages of the process. The impact on requirements engineering and the steps for developing the Request For Proposal (RFP) and evaluating the proposals are outlined. The parties involved in the acquisition process are identified, and their relationship within a project governance structure are discussed. The final discussion provides some guidance as to how the early stages of the acquisition process should be adapted to minimize risk through the project.

A vital issue in the correct use of commercial-off-the-shelf (COTS) components is the proper understanding of their functionality, quality attributes and ways of operation. Traditionally, COTS component vendors provide some of this information in accompanying documentation. However, the documentation is often informal and likely contains ambiguous and inconsistent statements. Even equipped with interface descriptions clearly defining the basic aspects of component use, such as operation signatures and operating platforms, this documentation does not provide a mathematically sound means for addressing the behavioural interoperability issues in component-based system design. In this paper, we propose a formal but user-friendly component specification approach which augments commercial IDLs with the capability of capturing component interoperability requirements. This approach uses unambiguous temporal operators to define sequencing and concurrency constraints between component operation invocations. Accordingly, it enables precise specifications of how a component provides its services and the correct way in which its services should be used.

COTS significantly complicates the IV&V process. The necessarily pessimistic culture of IV&V has a perspective on which COTS assessment attributes and techniques are relevant that differs greatly from developer’s typically optimistic, success-oriented perspective. There is no basis to assume that the COTS assessments made by developers will ultimately be consistent with IV&V COTS assessments. The result frequently results in a “lose-lose” situation where either large re-work costs are incurred to replace existing COTS with IV&V approved COTS, or higher risk and uncertainty must be tolerated (from the IV&V perspective) to continue with the COTS the developers chose. This work seeks to remedy this “culture clash” of COTS assessment perspectives by integrating IV&V and developers system level COTS assessments that provides a result that is both consistent and cost-effective.

The paper addresses the specific issue of interoperability in heterogeneous databases (DBs) and the possible use of COTS components that may alleviate the DB interoperability problem. A component-based software Architectural Style (AS) for interoperable DBs has been used, and an example of its application given, to identify which role the COTS components may play when populating the architecture. We discuss the characteristics of such COTS components and advocate that such COTS components should be developed with a specific component platform in mind, interoperate within a certain context, and adhere to constraints of our AS.

This paper proposes the adoption of a goal-based method called GBRAM for facilitating the process of building taxonomies of COTS components. Since GBRAM was defined in a different setting, the main result of the paper is to adapt it to this new context obtaining the GBTCM method. We show how the different activities and artifacts of GBRAM change, and we apply the proposal to obtain a taxonomy for requirements engineering oriented tools.

Organizational theorists have long touted as a market asset for reducing transaction costs. In some cases managers have learned to depend on social relationships of firms with whom they are familiar rather than judging products on merit. The trouble arises when trust is established with a firm, and superior products from other firms are not considered. The problem is exacerbated with software, as the product itself is intangible and often difficult to judge or understand. Smaller COTS software firms with superior products have had a difficult time entering the U.S. Government market. Government managers have traditionally turned to well-known contractors with whom they have had decade-old ties, rather than seeking newer and better COTS solutions that are lesser known. This paper examines some of the barriers to trusting lesser-known software products and suggests solutions to overcome such barriers.