Susanne Busse

Platform independent model transformation based on triple

Reuse is an important topic in software engineering as it promises advantages like faster time-to-market and cost reduction. Reuse of models on an abstract level is more beneficial than on the code level, because these models can be mapped into several technologies and can be adapted according to different requirements. Unfortunately, development tools only provide fixed mappings between abstract models described in a language such as UML and source code for a particular technology. These mappings are based on one-to-one relationships between elements of both levels. As a consequence, it is rarely possible to customize mappings according to specific user requirements.We aim to improve model reuse by providing a framework that generates customized mappings according to specified requirements. The framework is able to handle mappings aimed for several component technologies as it is based on an ADL. It is realized in Triple to represent components on different levels of abstraction and to perform the actual transformation. It uses feature models to describe mapping alternatives.

Analysis of compositional conflicts in component-based systems

Today, incompatibilities in component specifications make their composition hard to handle in practical terms. Incompatibilities can be classified into three conflict categories: type conflicts, behavioral conflicts, and property conflicts. This paper describes a framework for the identification of compositional conflicts in component-based systems that analyses conflicts of all three categories. Furthermore, the conflict analysis framework can be integrated into the software development process and handles component transformations between different abstraction levels.

Interoperable e-learning ontologies using model correspondences

Despite the fact that ontologies are a good idea for interoperation, existing e-Learning systems use different ontologies for describing their resources. Consequently, exchanging resources between systems as well as searching appropriate ones from several sources is still a problem. We propose the concept of correspondences in addition to metadata standards. Correspondences specify relationships between ontologies that can be used to bridge their heterogeneity. We show how we use them for building evolvable federated information systems on e-Learning resources in the World Wide Web. Beside this integration scenario, we also describe other interoperation scenarios where correspondences can be useful.

Schema Evolution in Federated Information Systems

Mediators—used in federated information systems—provide a homogeneous read-only access to a set of autonomous information sources. To achieve semantic integration of the heterogeneous data, correspondences between the autonomous schemas are specified manually by experts. Considering the continuous evolution of those typically long-living information systems, schema evolution is an important aspect. New concepts are necessary to adapt correspondences consistently to evolving schcmas. In this paper we propose a formalized schema evolution mechanism for federated information systems based on metamodeling and dynamic logic. We define useful transactions for schema evolution based on three elements: (1) a metamodel for ODMG schemas, (2) a metamodel for model correspondence assertions (MoCAs), and (3) a classification of evolution actions and their specification. For each evolution transaction, semantic preconditions are specified and the impact on the correspondence assertions are described. In this way, we contribute an important step towards a consistent evolution of federated information systems.

Software architectural modeling of the CORBA Object Transaction Service

The OMGs Object Transaction Service (OTS) is an important CORBAService that provides transaction processing facilities on top of object request broker technology. With the OTS, specific interfaces and interaction models are introduced that intrinsically impact the design of single components and of component configurations that are to participate in transactional computation. We argue to record impacts as introduced by the OTS as distinct abstractions of design, and propose a software architectural approach to software system representation. We present a pattern-like connector abstraction for one processing model of the OTS, and discuss its use for modeling CORBA applications that interface the OTS. Using our concepts of connector and component abstractions, a vocabulary of design can be established, and design rationales for introducing component features can be well recorded, supporting continuous development of ORB-based systems.

Connectors for modeling object relations in CORBA-based systems

We present connectors as part of a conceptual design framework for modeling CORBA distributed object architectures. We propose the concept of connectors for abstracting complex object dependencies, and introduce connectors for CORBA object relations based on OMGs event service specification and relationship service specification. With connectors, modeling of component integration and interaction is strongly supported.

OCLPrime - Environment and Language for Model Query, Views and Transformations

UML, MOF, and MDA currently do not provide a standardized means to describe manipulation of model-elements in algorithms or rules. In order to define specific UML-based methods in a product-independent and portable way, this capability is essential. This paper discusses design and implementation of the hybrid language framework Prime and its derivative language OCLPrime in the light of the OMG Query / Views / Transformations RFP. Prime allows and coordinates the reuse of different languages for validation, selection, and projection in the Transformation of models. Its design follows the Composite, Visitor, and Interpreter patterns and coordinates the sub-languages in transactions. OCLPrime is a reference language implementation in Prime employing OCL expressions to select parts of a source model and SQL DML to project these into a target model. Preand Post-validations are performed by a UML Profile Validator.