Dolors Costal

Development of service-oriented architectures using model-driven development

ContextModel-Driven Development (MDD) and Service-Oriented Architecture (SOA) are two challenging research areas in software engineering. MDD is about improving software development whilst SOA is a service-based conceptual development style, therefore investigating the available proposals in the literature to use MDD when developing SOA may be insightful. However, no studies have been found with this purpose. ObjectiveThis work aims at assessing the state of the art in MDD for SOA systems. It mainly focuses on: what are the characteristics of MDD approaches that support SOA; what types of SOA are supported; how do they handle non-functional requirements. MethodWe conducted a mapping study following a rigorous protocol. We identified the representative set of venues that should be included in the study. We applied a search string over the set of selected venues. As result, 129 papers were selected and analysed (both frequency analysis and correlation analysis) with respect to the defined classification criteria derived from the research questions. Threats to validity were identified and mitigated whenever possible. ResultsThe analysis allows us to answer the research questions. We highlight: (1) predominance of papers from Europe and written by researchers only; (2) predominance of top-down transformation in software development activities; (3) inexistence of consolidated methods; (4) significant percentage of works without tool support; (5) SOA systems and service compositions more targeted than single services and SOA enterprise systems; (6) limited use of metamodels; (7) very limited use of NFRs; and (8) limited application in real cases. ConclusionThis mapping study does not just provide the state of the art in the topic, but also identifies several issues that deserve investigation in the future, for instance the need of methods for activities other than software development (e.g., migration) or the need of conducting more real case studies.

Formal semantics and ontological analysis for understanding subsetting, specialization and redefinition of associations in UML

The definition of the exact meaning of conceptual modeling constructs is considered a relevant issue since it contributes to their effective and appropriate use by conceptual modelers. This paper studies three related constructs that enhance the expressiveness of the UML language about associations and which still lack a complete and comprehensive study, namely, association subsetting, specialization and redefinition. It formalizes their semantics, analyses them from an ontological perspective and compares them. The semantic formalization is based on mapping the studied constructs to a basic UML layer which have a previous formal definition in the literature. Furthermore, the ontological analysis developed here is based on a formal theory of relations which is part of the Unified Foundational Ontology (UFO).

Improving the definition of general constraints in UML

An important aspect in the specification of conceptual schemas is the definition of general constraints that cannot be expressed by the predefined constructs provided by conceptual modeling languages. This is generally achieved by using general-purpose languages like OCL. In this paper we propose a new approach that facilitates the definition of such general constraints in UML. More precisely, we define a profile that extends the set of predefined UML constraints by adding certain types of constraints that are commonly used in conceptual schemas. We also show how our proposal facilitates reasoning about the constraints and their automatic code generation, study the application of our ideas to the specification of two real-life applications, and present a prototype tool implementation.

On the use of association redefinition in UML class diagrams

Association redefinition is a new concept in UML 2.0 that makes it possible to impose additional constraints on some instances of associations. In this paper, we describe how to use association redefinition to declare additional referential integrity and cardinality constraints for associations. We also analyze the interactions between taxonomic constraints and association redefinitions and their impact on the satisfaction of taxonomic constraints. Finally, we establish several conditions that are necessary to guarantee well-formed association redefinitions.

Facilitating the definition of general constraints in UML

One important aspect in the specification of conceptual schemas is the definition of general constraints that cannot be expressed by the predefined constructs provided by conceptual modeling languages. In general this is done by means of general-purpose languages, like OCL. In this paper we propose a new approach to facilitate the definition of such general constraints in UML. More precisely, we define a profile that extends the set of UML predefined constraints with some types of constraints that are used very frequently in conceptual schemas. We also study the application of our ideas to the specification of two real-life applications and we show how results in constraint-related problems may be easily incorporated to our proposal.

Adoption of OSS components

Open Source Software (OSS) has become a strategic asset for a number of reasons, such as short time-to-market software delivery, reduced development and maintenance costs, and its customization capabilities. Therefore, organizations are increasingly becoming OSS adopters, either as a result of a strategic decision or because it is almost unavoidable nowadays, given the fact that most commercial software also relies at some extent in OSS infrastructure. The way in which organizations adopt OSS affects and shapes their businesses. Therefore, knowing the impact of different OSS adoption strategies in the context of an organization may help improving the processes undertaken inside this organization and ultimately pave the road to strategic moves. In this paper, we propose to model OSS adoption strategies using a goal-oriented notation, in which different actors state their objectives and dependencies on each other. These models describe the consequences of adopting one such strategy or another: which are the strategic and operational goals that are supported, which are the resources that emerge, etc. The models rely on an OSS ontology, built upon a systematic literature review, which comprises the activities and resources that characterize these strategies. Different OSS adoption strategy models arrange these ontology elements in diverse ways. In order to assess which is the OSS adoption strategy that better fits the organization needs, the notion of model coverage is introduced, which allows to measure the degree of concordance among every strategy with the model of the organization by comparing the respective models. The approach is illustrated with an example of application in a big telecommunications company.

QuESo a quality model for open source software ecosystems

Open source software has witnessed an exponential growth in the last two decades and it is playing an increasingly important role in many companies and organizations leading to the formation of open source software ecosystems. In this paper we present a quality model that will allow the evaluation of those ecosystems in terms of their relevant quality characteristics such as health or activeness. To design this quality model we started by analysing the quality measures found during the execution of a systematic literature review on open source software ecosystems and, then, we classified and reorganized the set of measures in order to build a solid quality model.

Integrating the goal and business process perspectives in information system analysis

There are several motivations to promote investment and scientific effort in the integration of intentional and operational perspectives: organisational reengineering, continuous improvement of business processes, alignment among complementary analysis perspectives, information traceability, etc. In this paper we propose the integration of two modelling languages that support the creation of goal and business process models: the i* goal-oriented modelling method and Communication Analysis, a communication-oriented business process modelling method. We describe the methodological integration of the two modelling methods with the aim of fulfilling several criteria: i) to rely on appropriate theories; ii) to provide abstract and concrete syntaxes; iii) to provide scenarios of application; and iv) to develop tool support. We provide guidelines for using the two modelling methods in a top-down analysis scenario. We also present an illustrative case that demonstrates the feasibility of the approach.

Understanding Redundancy in UML Models for Object-Oriented Analysis

A phenomenon that frequently appears when designers define analysis specifications is that of redundancy between models. A correct and deep understanding of this phenomenon is necessary to help the task of the designer. In this paper, we study the problem of redundancy in UML Models for Object-Oriented Analysis. In this context, we identify different kinds of redundancies that may arise. We evaluate the impact of redundancy in specifications from the point of view of their desirable properties. We also propose how to obtain a canonical analysis model, which does not include any of the identified redundancies, and we sketch the possibility of having redundant views of some aspects of the canonical model.

Open source software ecosystems: A Systematic mapping

Abstract Context: Open source software (OSS) and software ecosystems (SECOs) are two consolidated research areas in software engineering. OSS influences the way organizations develop, acquire, use and commercialize software. SECOs have emerged as a paradigm to understand dynamics and heterogeneity in collaborative software development. For this reason, SECOs appear as a valid instrument to analyze OSS systems. However, there are few studies that blend both topics together. Objective: The purpose of this study is to evaluate the current state of the art in OSS ecosystems (OSSECOs) research, specifically: (a) what the most relevant definitions related to OSSECOs are; (b) what the particularities of this type of SECO are; and (c) how the knowledge about OSSECO is represented. Method: We conducted a systematic mapping following recommended practices. We applied automatic and manual searches on different sources and used a rigorous method to elicit the keywords from the research questions and selection criteria to retrieve the final papers. As a result, 82 papers were selected and evaluated. Threats to validity were identified and mitigated whenever possible. Results: The analysis allowed us to answer the research questions. Most notably, we did the following: (a) identified 64 terms related to the OSSECO and arranged them into a taxonomy; (b) built a genealogical tree to understand the genesis of the OSSECO term from related definitions; (c) analyzed the available definitions of SECO in the context of OSS; and (d) classified the existing modelling and analysis techniques of OSSECOs. Conclusion: As a summary of the systematic mapping, we conclude that existing research on several topics related to OSSECOs is still scarce (e.g., modelling and analysis techniques, quality models, standard definitions, etc.). This situation calls for further investigation efforts on how organizations and OSS communities actually understand OSSECOs.