|
|
Third International Workshop on Composition
Languages
In conjunction with |
The main focus of the workshop will be on software composition on an architectural level, and not on component-based systems in general. In particular, we would like to emphasize the important issues of (i) the design and implementation of higher-level models and languages for component-based software development, (ii) approaches that combine architectural description, component configuration, and component composition, (iii) paradigms for the specification of reusable software assets, (iv) expressing applications as compositions of software components, and (v) the derivation of working systems using composition languages and components. Furthermore, we would particularly like to encourage authors to submit position statements focusing on formal aspects of the issues mentioned above and case studies of using composition languages for real-world applications.
All submissions will be peer-reviewed by at least two members of the workshop paper selection committee. Based on the quality and originality, a selection of the best position statements will be presented at the workshop.
The workshop will be organized in several sessions. After an initial presentation session, where all participants can formulate one or more, possibly provocative, working hypotheses, we intend to split the workshop into task forces to foster the discussion a particular subject of common interest. At the end o the workshop the task forces will reunite and we will assemble the results and formulate future work, which we intent to present to the rest of the ECOOP community in the form of a poster at the conference.
A component-based software engineering approach mainly consists of two development steps: (i) the specification and implementation of components and (ii) the composition of components into composites or applications. Currently, there is considerable experience in component technology and many resources are spent for the first step, which resulted in the definition of component models and components such as CORBA, COM, JavaBeans, and more recently EJB and .NET. However, much less effort has been spent in investigating appropriate composition environments and languages, which allow application developers to express applications flexibly as compositions of components and, therefore, offer support for component-based software engineering.
Most available composition environments focus mainly on special application domains and offer at best rudimentary support for the integration of components that were built in a system other than the actual deployment environment. Furthermore, these systems do not enforce a clear separation of computational elements (i.e., components) and their relationships, which is needed to address the flexibility and maintainability of component-based systems. The reason for this situation is not only the lack of well-defined (or standardized) component interfaces, but the ad-hoc way the semantics of the underlying language models are defined.
In the recent past we observed a paradigm shift from component-centric development to model-centric and architecture-centric development. One of the recent developments in this area is the Model Driven Architecture™ (MDA) defined by OMG. MDA is considered to be the next step in solving software integration problems. MDA introduces a separation between application logic and infrastructure by encapsulating infrastructure specific aspects as far as possible in code generators. This separation allows for the architecture specification and software composition on a conceptual level and thus reduces architectural mismatches usually introduced by dependencies to infrastructure.
Model-centric and architecture centric development
Compositional reasoning
Aspect of Composition languages
The workshop will be organized in several sessions. After an initial presentation session, where all participants can formulate one or more, possibly provocative, working hypotheses, we intend to split the workshop into task forces to foster the discussion a particular subject of common interest. At the end o the workshop the task forces will reunite and we will assemble the results and formulate future work, which we indent to present to the rest of the ECOOP community in the form of a poster at the conference.
Authors are encouraged to address any aspects of the design and implementation of composition languages in their position statements. We solicit submissions on original research in the form of extended abstracts. Submissions should not exceed 8 pages (with a minimum 11pt font) and must have a cover page including the paper title, abstract, names and affiliations of authors, postal contact addresses, email addresses, and telephone numbers. In addition, we ask the authors to include a list of critical questions and/or some, perhaps provocative, statements at the end of their submission which will assist the organizers to define topics for discussion in advance. Submissions should be sent in an electronic format (PDF or Postscript) to Markus Lumpe and preferably prepared for letter or A4 sizes using Springer LNCS-style.
All selected submissions will be made available online prior to the workshop and be published by one of the affiliated organizations. Aspects of the best position statements as well as the workshop results will be discussed in a chapter of the ECOOP Workshop reader. The results of the workshop will also be presented to the rest of the ECOOP community in the form of a poster at the conference. We are investigating having a special issue of a journal for revisions of selected papers after the workshop.
For further information about the workshop, please refer to the workshop home page at http://www.cs.iastate.edu/~lumpe/WCL2003.