The Web of Data is a decentralized paradigm, where each data producer is the source of authority of his own structured data. By structured data, we refer to the Linked Data principle. We want to encourage each laboratory involved in Semantic Web research to cook their own semantic data using Croquette virtual machine, so that the sum of all croquette instances over the world would constitute a Semantic Web dog food. Such a distributed corpus should match the idea of a real Web of Data playground, in which the different issues that the Semantic Web is facing today such as provenance, consistency, scalability, performance, temporal evolution, etc. could be experiment on a worldwide scale. Croquette only takes as input the list of bibtex files (filesystem or URLs) that you are already used to managed in order to fill the triplestore.
|dc||http://purl.org/dc/elements/1.1/||Basic metadata on a document (creator, title, ...)|
|foaf||http://xmlns.com/foaf/0.1/||People (authors) and organisation (their university) in the corpus, using foaf:Person and foaf:Organization|
|swrc||http://swrc.ontoware.org/ontology#||Bibtex publication metadata traduction into Semantic Web (category, abstract, pages, ...)|
|geo||http://www.w3.org/2003/01/geo/wgs84_pos#||Geolocalisation of conference venues and organisation work places.|
|ical||http://www.w3.org/2002/12/cal/ical#||Time-related description of conferences and workshop.|
|swc||http://data.semanticweb.org/ns/swc/ontology#||Conference event metadata such as a given talk at a conference.|