Examining the Quality of Evolution Frameworks and Metamodeling Paradigms of Information Systems Development Methodologies

Information systems development methodologies and associated CASE tools have been considered cornerstones for building quality into an information system. The construction and evaluation of methodologies are usually carried out by evaluation frameworks and metamodels, both considered as meta-methodologies. This chapter investigates and reviews representative metamodels and evaluation frameworks for assessing the capability of methodologies to contribute to high-quality outcomes. It presents a summary of their quality features, strengths, and weaknesses. The chapter ultimately leads to a comparison and discussion of the functional and formal quality properties that traditional meta-methodologies and method evaluation paradigms offer. The discussion emphasizes the limitations of methods and metamethods used to model and evaluate software quality properties, such as computability and implementability, testing, dynamic semantics capture, and people’s involvement. This analysis, along with the comparison of the philosophy, assumptions, and quality perceptions of different process methods used in information systems development, provides the basis for recommendations about the need for future research in this area.

Developer-Driven Quality

Much attention has been given to Software Process Improvements (SPIs) based on the premise that system development outcomes are largely determined by the capabilities of the software development process. The content of this chapter presents the results of a set of research projects investigating why SPIs have not been diffused and utilized in the software engineering community as expected (Fayad et al., 1996; Fichman & Kemerer, 1997; Luqi & Goguen, 1997; Pfleeger & Hatton, 1997). We show that a software developer’s perceived control over the use of an SPIimpacts its diffusion success. Additionally, we show that a software developer’s perceptions of enhanced software quality and increased individual productivity achieved through SPI use impact the successful diffusion of the SPI. Results of these research efforts support the compilation of a clear set of management guidelines to ensure the effective use of SPIs in software development organizations.

Process-Centered Contributions to Information Systems Quality

The growing attendance at seminars and conferences dedicated to quality programs attests to the increasing recognition of the continued importance of quality. Unfortunately, in many organizations, this intensified quality focus has not been effectively applied to information systems — a surprising outcome given the many demonstrations of a direct relationship between information systems delivery process and information systems quality and success. In this chapter, we analyze process-centered contributions and solutions to the increasing challenges of producing high-quality systems. We provide a balanced overview of evidence that has emerged from practical, real-world experiences and empirical research studies, an overview that incorporates the positions of both proponents and opponents of process-centricity. We then provide an assessment of the contexts in which software process improvements and quality- enhancing initiatives can thrive.

Making Real Progress with the Requirements Defects Problem

Requirements defects remain a significant problem in the development of all software intensive systems including information systems. Progress with this fundamental problem is possible once we recognize that individual functional requirements represent fragments of behavior, while a design that satisfies a set of functional requirements represents integrated behavior. This perspective admits the prospect of constructing a design out of its requirements. A formal representation for individual functional requirements, called behavior trees makes this possible. Behavior trees of individual functional requirements may be composed, one at a time, to create an integrated design behavior tree (DBT). Different classes of defects are detected at each stage of the development process. Defects may be found at the translation to behavior trees, and then at the integration of behavior trees and when individual component behavior trees are projected from the DBT. Other defects may be found by inspection and model-checking of the DBT.

Quality Metrics and Bayesian Analysis

This chapter presents an empirical assessment of the quality of the process of building software systems in light of the recent emergence of agile development methodologies, which were designed to to help with the development of higher quality information systems under given conditions. The purpose of this research was to assess one of these agile approaches, eXtreme Programming (XP), through a comparison with a traditional (design-driven) software construction process. With this aim we observed and measured the work of several student groups using different approaches to produce software for commercial companies during a semester. The data collected were analyzed following the Bayesian approach. Our results indicate that that XP could cope with small/medium size projects of software construction delivering a measurable improvement in the quality of the system as judged by the business clients.

Improving Quality through the Use of Agile Methods in Systems Development

We introduce, define, and elaborate on agile development methods and how quality information systems are created via the values and practices of people using agile approaches. Key differences among agile methods, the SDLC, and other development methodologies are covered and suggestions for improving quality in IS through agile methods are given. We recommend adopting the principles of agile methods, encouraging more education about the values of agile approaches, including more types of people in the development of agile systems; adopting agile methods for a variety of organizational cultures; and renaming agile methods to signify the value system inherent in the approach.

Software Quality and Open Source Process

This chapter introduces the open source software development process from a software quality perspective. It uses the attributes of software quality in a formal model and attempts to map them onto the principles of the open source process. Many stages of the open source process appear to have an ad-hoc approach. Although open source is not considered to be a formal methodology for software development, it has resulted in the development of very high quality software, both in the consumer and in the enterprise space. In this chapter, we hope to understand the open source process itself, and apply it to other methodologies in order to achieve better software quality. Additionally, this chapter will help in understanding the “Wild West” nature of open source and what it may hold for us in the future.

Academic Storytelling

This chapter introduces the concept of rich picture stories or scenarios to establish a collaborative effort in the design of new information systems. Systems designers need contextual information about their users in order to design and provide information systems that will function effectively and efficiently within those contexts. Story-telling or scenarios allow developers and consumers alike to create that all-important “meeting place” from which comes a collaborative effort in the design of new systems. One of the primary issues inhibiting the further use of storytelling or scenarios in information systems is the issue of validation, which is explored. It is hoped that this chapter will provide insights into the opportunities available to information systems users and designers to improve the quality of information systems through the use of valid user-driven scenarios.

User Participation in the Quality Assurance of Requirements Specifications

Improper specification of systems requirements has thwarted many splendid efforts to deliver high-quality information systems. Scholars have linked this problem to, between others, poor communication among systems developers and users at this stage of systems development. Some believe that specifying requirements is the most important and the most difficult activity in systems development. However, limitations in human information processing capabilities and the inadequacy of the structures available for communicating specifications and obtaining feedback and validation help to exacerbate the difficulty. This chapter presents an overview of both longstanding and newer requirements specification models and evaluates their capability to advance user participation in this process and incorporate stated quality attributes. It also reports on preliminary evaluations of animated system engineering (ASE), the author’s preferred (newer) technique, which indicate that it has the capability to improve the specification effectiveness.

The Different Views of Software Quality

This chapter examines the different definitions of quality and compares the different models and frameworks for software quality evaluation. It will look at both historical and current literature. The chapter will give special attention to recent research on the Software Evaluation Framework, a framework for software evaluation, which gives the rationale for the choice of characteristics used in software quality evaluation, supplies the underpinning explanation for the multiple views of quality, and describes the areas of motivation behind software quality evaluation. The framework has its theoretical foundations on value-chain models, found in the disciplines of cognitive psychology and consumer research, and introduces the use of cognitive structures as a means of describing the many definitions of quality. The author hopes that this chapter will give researchers and practitioners a better understanding of the different views of software quality, why there are differences, and how to represent these differences.