ENSEMBLE

The project aligns well with the aspirations of the Semantic Web in that it is concerned with the specification and construction of a space of linked information in which data is continuously being enriched and extended by individuals and groups. At the same time it also conforms to the design principles and standards established by the World Wide Web Consortium, and will make use of a number of key technologies and specifications, notably Resource Definition Format (RDF), RDF Schema (RDFS) SPARQL and Web Ontology Language (OWL) as the basis for description of resources, concepts, terms, and relationships within the knowledge domains to be explored by the project.

The cases to be developed in the course of project activities in its various research settings represent significant challenges for developers of Semantic Web applications and implementations. Not only are the data to be included arbitrarily and unpredictably complex, they may be differently configured and combined by different users. A single case might include:

• Digital repository content
• Content harvested from publishers’ websites and libraries
• Datasets from domain-specific data sources
• Public domain information
• User generated content ranging from substantial documents and data sets either collated from different sources or generated by other software applications to annotations, comments and ratings

Cases might include both substantive content in formats including: structured and unstructured texts, images, video audio and data sets in generic and domain specific formats; and metadata in formats such as: OAI Dublin Core; BibTex, RIS or MARC records; RSS or Atom; VCard and VCal.

The SIMILE Project

The SIMILE project based at MIT Libraries and the Computer Science and Artificial Intelligence Laboratory at MIT (http://simile.mit.edu) and supported by the World Wide Web Consortium and Hewlett-Packard Labs has developed a suite of tools which are designed to support the development and provision of end-user services by drawing on the digital assets, schemas/vocabularies/ontologies, and metadata held in a wide range of individual, community and institutional stores. Given the variation in the nature, structure and content of these (as outlined above), the SIMILE toolkit is designed to provide interchange and conversion processes in order to support interoperability and the aggregation of disparate data and metadata. The use-cases which informed this development work were drawn predominantly from the libraries domain, so applications in varied pedagogical settings represents an opportunity to implement and evaluate the tools and gather new use cases to share across the user and developer communities.