We propose an approach (GQM/MEDEA) for defining measures of product attributes in software engineering. The approach is driven by the experimental goals of measurement, expressed via the GQM paradigm, and a set of empirical hypotheses. To make

Abstract. In software engineering it is widely acknowledged that the usage of metrics at the initial phases of the object oriented software life cycle can help designers to make better decisions and to predict external quality attributes, such as maintainability. Following this idea we have carried out three controlled experiments to ascertain if any correlation exists between the structural complexity and the size of UML class diagrams and their maintainability. We used 8 metrics for measuring the structural complexity of class diagrams due to the usage of UML relationships, and 3 metrics to measure their size. With the aim of determining which of these metrics are really relevant to be used as class diagrams maintainability indicators, we present in this work a study based on Principal Component Analysis. The obtained results show that the metrics related to associations, aggregations, generalizations and dependencies, are the most relevant whilst those related to size seem to be redundant.

evaluation of software processes is nowadays a very important issue due to the growing interest of software companies in the improvement of the productivity and quality of delivered products. Software measurement plays a fundamental role here. Given the great diversity of entities which are candidates for measurement in the software process improvement context (process models, projects, resources, products) this measurement must be performed in a consistent and integrated way. This will facilitate the making of decisions in process improvement. In this paper, a proposal for the integrated management of the software measurement is presented. The goal is to provide companies with a generic and flexible environment for software measurement which facilitates and establishes the basis for a common and effective measurement process and which is not restricted to only one kind of software entity or to a single quality or evaluation model. In order to achieve this, the proposal adopts the Model Driven Engineering philosophy and provides: a metamodel for the definition of software measurement models; a flexible method to measure any kind of software entity represented by its corresponding metamodel and GenMETRIC, which is the software tool that supports the framework.

Abstract. Due to the different nature of the available dynamic interactions between components afforded by some middleware infrastructure, distributed information systems (DIS) behave differently from traditional centralized ones. This results in a different view of their quality attributes, which require specifically customized measures for accurate estimation. In previous work we proposed and theoretically validated design measures purposely defined for DIS. In this paper we investigate the relevance of the newly proposed measures to estimate one of the quality attributes of interest for DIS. To this end, we have applied the proposed measures to a proof-of-concept DIS in the context of an Australian university, for the ultimate purpose of efficiency estimation. The research concludes that most of the proposed measures are indeed correlated to efficiency and are suitable to be used as part of estimation models. 1.

ABSTRACT. The quality of an object oriented information systems (OOIS) depends greatly on the decisions taken at the initial phases of their development. In a typical object oriented information systems development a class diagram is first built. Class diagrams lay the foundation for all later design work. So, their quality heavily affects on the product that will be ultimately implemented. Even though the appearance of the Unified Modelling Language (UML) as a standard of modelling OOIS have provided a great contribution towards building quality OOIS, it is not enough. Early availability of metrics is a key factor in the successful management of OOIS development. The goal of this work is to propose a set of metrics in order to assess the complexity of UML class diagrams. We also put the proposed metrics under empirical validation in order to provide empirical support to their practical significance and usefulness.

Abstract. The open source software ecosystem comprises more than a hundred thousand applications of varying quality. Individuals and organizations wishing to use open source software packages have scarce objective data to evaluate their quality. However, open source development projects by definition allow anybody to read, and therefore evaluate their source code. In addition, most projects also publish process-related artefacts, such as bug databases, mailing lists, and configuration management system logs. The software quality observatory is a platform that uses these product and process data sources to automatically evaluate the quality of open source projects. A plugin-based service-oriented architecture allows the mixing and matching of metrics extraction suites, source code repositories, and transformation filters. The resulting platform is aimed at it consultants and managers, the open source community, and researchers. 1

Abstract—The metrics formulated to date for characterizing the modularization quality of object-oriented software have considered module and class to be synonymous concepts. But, a typical class in object-oriented programming exists at too low a level of granularity in large object-oriented software consisting of millions of lines of code. A typical module (sometimes referred to as a superpackage) in a large object-oriented software system will typically consist of a large number of classes. Even when the access discipline encoded in each class makes for “clean ” class-level partitioning of the code, the intermodule dependencies created by associational, inheritance-based, and method invocations may still make it difficult to maintain and extend the software. The goal of this paper is to provide a set of metrics that characterizes large object-oriented software systems with regard to such dependencies. Our metrics characterize the quality of modularization with respect to the APIs of the modules, on the one hand, and, on the other, with respect to such object-oriented intermodule dependencies as caused by inheritance, associational relationships, state access violations, fragile base-class design, etc. Using a two-pronged approach, we validate the metrics by applying them to popular opensource software systems.

Abstract. Navigability is a main concern in the design of Web applications. In order to assess such navigability a number of measures has been proposed. From them, measures defined on conceptual models are specially relevant, as it is well known that high quality conceptual models are critical to the success of the deployed system. However, measurement methods associated to such measures, as well as the design modifications that need to be performed on the models in order to improve their values, are usually tightly coupled with particular Web Engineering approaches. This fact compromises their effectiveness and their propagation capacity to different environments and/or methodologies. Our aim in this paper is to illustrate how navigability measures can be captured in a general manner, by instantiation of a measuring metamodel that is based on the Software Measurement Ontology proposed by García et al. In this way, not only is it possible to define a reusable set of relevant measures for a given family of applications, but also such measures can be queried in the context of MDA transformation rules. These rules capture both the measure decision criteria and the design modifications that should take place if the measure value for a given navigational model does not match such criteria. 1

Class cohesion is an important object-oriented software quality attribute. It indicates how much the members in a class are related. Assessing the class cohesion and improving the class quality accordingly during the object-oriented design phase allows for cheaper management of the later phases. In this paper, the notion of distance between pairs of methods and pairs of attribute types in a class is introduced and used as a basis for introducing a novel class cohesion metric. The metric considers the methodmethod, attribute-attribute, and attribute-method direct interactions. It is shown that the metric gives more sensitive values than other well-known design-based class cohesion metrics.

This study focuses on students' performance in a distance-learning computer-based C course offered by EPGY (Education Program for Gifted Youth) at Stanford University. Its main goal is testing the feasibility of automated evaluation. Three automated methods consisting of a criterion set and weighted scoring methods were developed empirically, then applied to four hundred programs of ten types of problems and thoroughly analyzed. For most of the assignments, the more significant criteria for evaluation were related to execution time features, (i.e.