An ontology for software measurement

Abstract

Software measurement has evolved in such a way that it is no longer a marginal or atypical activity within the software development process and has become a key activity for software project managers. All successful software organizations use measurement as part of their day-to-day management and technical activities. Measurement provides organizations with the objective information they need to make informed decisions that positively impact their business and engineering performance [17]. As a matter of fact, CMMI (Capability Maturity Model Integration) includes software measurement as one of its requisites for reaching higher maturity levels and it helps organizations to institutionalize their measurement and analysis activities, rather than addressing measurement as a secondary function. Other initiatives such as ISO/IEC 15504 [11], SW-CMM (Capability Maturity Model for Software) and the ISO/IEC 90003:2004 standard [12] also consider measurement to be an important element in the management and quality of software. In all these initiatives measurement plays a fundamental role as a means for assessing and institutionalizing software process improvement programs. However, as with any relatively young discipline, software measurement has some problems. When we approach software measurement and compare diverse proposals or international standards, it becomes apparent that the terminology used is not always the same or the same term may refer to different concepts. Terms such as "metrics", "attribute", or "measure" need to have a single definition accepted by all the researchers and practitioners who work in software measurement. The most serious point is when inconsistencies appear between different measurement proposals or standards. Standards provide organizations with agreed and well-recognized practices and technologies, which assist them to interoperate and to work using engineering methods, reinforcing software engineering as an "engineering" discipline, instead of a "craft". Furthermore, the Internet is changing how business is done nowadays, promoting cooperation and interoperation among individual organizations, which need to compete in a global market and economy, and share information and resources. Standardization is one of the driving forces to achieve this interoperability, with the provision of agreed domain conventions, terminologies and practices. However, there is no single standard which embraces the whole area of software measurement in its totality, but rather there are diverse standards orientated towards specific areas such as the measurement process or function points. Without an overall reference framework managing these standards, inconsistencies arise in the measurement terminology. This issue has been recognized by ISO/IEC, which has created a work group for the harmonization of systems engineering standards within its Joint Technical Committee 1 (JTC1: "Information Technology", www.jtc1.org), and is trying to explicitly include in its directives the procedures which guarantee consistency and coherency among its standards. Furthermore, there has been an agreement in place since the year 2002 between the IEEE Computer Society and ISOJTC1-SC7 to harmonize their standards, which includes the terminology on measurement. In spite of these efforts, the problem of terminology harmonization still needs to be resolved in our opinion. The objective of this chapter is to present a coherent software measurement terminology which has been agreed upon by consensus, i.e., without contradictions or disparities in the definitions, and a terminology which is widely accepted. The terminology presented in this chapter has been obtained as a result of an exhaustive analysis of the concepts and terms used in the field of software measurement. First of all, similarities, discrepancies, shortcomings and weaknesses in the terminology used in the main standards and proposals have been identified, including ISO International Vocabulary of Basic and General Terms in Metrology (VIM) [13] in the comparison [5]. The result has been a software measurement ontology that provides a set of coherent concepts, with the relations between these concepts well defined, and which we hope helps to create a unified framework of software measurement terminology. This chapter is organized as follows. After this introduction, Sect. 6.2 guies a brief analysis of the current situation. Section 6.3 presents the Software Measurement Ontology proposal; the concepts of the ontology and relationships among them are presented in detail grouped according to the sub-ontology to which they belong. A running example based on a real case study is used to illustrate the ontology. Finally, Sect. 6.4 draws some conclusions, proposes some suggestions for harmonization, and identifies future research work. © 2006 Springer-Verlag Berlin Heidelberg.