|START Conference Manager|
In order to enable the provision and sharing of environmental data across domains and organizations, a great deal of work has been done in the last years on the definition of standardized data models. This work has taken place both within the Open Geospatial Consortium, as well as in various working groups set up by the European Commission for the implementation of the INSPIRE initiative. As evolution, change and growth have proven to be some of the only constants in the environmental domain, the data models defined for the provision of data on the environment must be able to change and grow together with their subject matter. The INSPIRE data specifications have already been extended for European Air Quality Reporting; at present Member States are working on extensions to support their national requirements. Unfortunately, UML currently does not provide mechanisms to allow for such extension of data models in a well-structured and harmonized manner. In this paper, we first analyze the requirements for such continuous evolution and growth of environmental data models, and show that through the provision of reusable URI-Properties this process could enabled. We then analyze various options for the implementation of reusable URI-Properties within UML, as well as their encoding within XML Schema. The options analyzed for this work were: • Data Types • Interfaces • Stereotypes • MOF level adjustment of UML Based on this analysis, we have come to the conclusion that reusable URI-Properties can best be implemented via the use of UML stereotypes. This approach is also well suited for transformation to XML Schema in accordance with the requirements stemming from ISO and INSPIRE, and in a manner that allows for easy identification of the URI-Properties included in the extended data models utilizing the URI-Properties. Finally, we provide an overview of the organizational requirements for the large-scale implementation of the reusable URI-Property concept within a large-scale data modelling community. In addition we provide an outlook on how the implementation of stereotype based URI-Properties would help pave the way for serialization of INSPIRE data to semantic technologies such as RDF and OWL.
Topic Area: [2.9] Challenges and approaches to standardization of data and interoperability of systems. Abstract Type: Oral Presentation
Comments: Data Modeling, UML, URI-Properties, Extension
START Conference Manager (V2.61.0 - Rev. 4195)