Publishing and Using Cultural Heritage Linked Data on the Semantic Web
San Rafael, Calif. Morgan & Claypool Publishers, 2012, 161 pages
Series: Synthesis lectures on the semantic web, theory and technology
Publishing Cultural Heritage (CH) collections and other content on the Web has become one of the most successful application domains of Semantic Web and Linked Data technologies. After a period of technical research and prototype development, boosted by theW3C Semantic Web Activity kickoff in 2001 and the Linked (Open) Data movement later on, major national and international CH institutions and collaboration networks have now started to publish their data using Linked Data principles and Semantic Web technologies.
This work is highly interdisciplinary, involving domain expertise of museum curators, librarians, archivists,and researchers of cultural heritage, as well as technical expertise of computer scientists and Web designers. Applying a new technology in the rapidly evolving Web environment is challenging not only for non-technical personnel in CH institutions, but also for computer scientists themselves.
This book aims at fostering the application of Linked Data and Semantic Web technologies in the CH domain by providing an overview of this fascinating application domain of semantic computing.
My own work in this field started in 2001 after the W3C Semantic Web Activity launch by establishing the Semantic Computing Research Group (SeCo) focusing on this field. We first developed a semantic photograph search and recommender system for a university museum, followed by semantic portal prototypes for publishing heterogeneous collections of different kinds, including artifacts in cultural history museums, historical events, folklore, maps, fiction literature, and natural history museum data. This book reflects experiences gained during this work.
From the very beginning in 2002, after developing our first ontologies and transforming the first collection databases into RDF, it became clear that the possibility of reusing existing data, metadata models, and ontologies, and linking it all together in an interoperable way, will be a central benefit of Semantic Web applications.
W3C recommendations, such as RDF(S), SKOS, SPARQL, and OWL are the corner stones for facilitating cross-domain, domain-independent interoperability, but this is not enough. We also need domain-dependent metadata-models and domain ontologies based on the generic semantic principles, as well as domain specific datasets. From a practical viewpoint, we also need ontology services so that the shared resources can be published and used in legacy and other application systems in a cost-efficient way. In short, a Semantic Web content infrastructure needs to be built in a similar vein as railroad, telephone, and other communication networks were created during earlier technological breakthroughs.
Creating a Semantic Web infrastructure, as well as content for it, requires collaboration between content providers. Co-operation is needed not only for sharing data through joint portals such as Europeana, but also for developing shared metadata models and ontologies used in representing the contents in an interoperable way.
Publishing CH content is becoming a game of cross-domain networking where the traditional boundaries of memory organizations based on content types are breaking down. From a user's viewpoint, the focus is on data, knowledge, and experience, be it based on a book in a library, an artifact in a cultural history museum, a story in an archive, a painting in an art gallery, a photograph taken by a fellow citizen, or a piece of music on a record.
During these years my faith in Semantic Web and Linked Data has become strong even if there are great challenges ahead, too. This is a truly promising way for providing richer content to users through more intelligent and usable interfaces, and at the same time for facilitating memory organization with better tools for collaborative, open content publishing on the Semantic Web.