Model and Schema Registry

Anvith Baddam, Prajakta Nivargi, Monte Gardner, R. M. Malyankar Department of Computer Science & Engineering Arizona State University, Tempe, AZ 85287-8809. E-mail: [email protected] URL: http://www.eas.asu.edu/~gcss/ Introduction With the proliferation of XML vocabularies and schemas, and the advent of semantic markup languages, registries for tags defined in such languages are becoming increasingly important. A General Accounting Office (GAO) report [11] mentions the proliferation of overlapping and incompatible vocabularies and structures as one of the most important problems in the adoption of XML in the Federal government. The same report discusses various XML registries (both planned and in various stages of implementation) as a component of the solution to this problem. One part of our project involves the development of MIML (Maritime Information Markup Language) [5,7], an XML vocabulary for the marine transportation system (MTS). Given the large number of stakeholders, many of whom already have their own information models (constructed independently over several years), creating and managing a new single information model or XML schema covering all the diverse sources of data would require a large investment of time and resources and be an extremely complex task for logistical and technical reasons. It was therefore decided to incorporate existing models into MIML wherever possible. The distributed nature of this effort requires integrating and managing different kinds of models and schemas. We present a schema management and registry system that can be used as a repository for XML schemas and which possesses certain other functionality useful for schema designers and application area programmers. The distinguishing features of this system, as compared to other registry and repository efforts, are its ability to contain information about information models (currently, in Protégé [3] format) as well as XML schemas, and the application of certain techniques based on ontology mapping research to the problem of detecting overlaps and conflict between XML schemas.