I’ve referred to linked open data and the semantic web in previous blog posts with a brief overview of these terms. I’d like to attempt describing linked open data in the context of the semantic web in more detail for a better understanding. Chapter 11 in our textbook, Metadata for Digital Collections, and the Linked Data website provide useful information.
According to the Linked Open Data website, linked data is about connecting related data on the Web. Wikipedia defines linked open data as “a best practice for exposing, sharing and connecting pieces of data, information, and knowledge on the Semantic Web using URIs and RDF.” Steven Miller in our textbook defines linked open data as “a method of exposing, sharing, and connecting data via dereferenceable URIs on the Web.” While the terms linked data and semantic web can be used interchangeably, the textbook treats these separately explaining that linked open data is a subtopic of the wider semantic web.
Currently the Web is an environment of linked documents via hyperlinks. Searching on the Web involves entering keywords for search engines to match within documents. Linked open data proposes linking data instead of documents via the RDF structured format and URIs. This is similar to how databases are searched: querying for metadata in structured statements about a resource. This searching method is based on metadata with a level of semantic meaning. When searching on the Web for links to documents, the links do not have a level of semantic meaning. So, the idea of the Semantic Web, and the purpose of linked data and RDF, is to pull semantically meaningful metadata out of the database “silos” of digital collections and make available on the Web.
Metadata for digital collections existing in database records and XML records needs to be converted into RDF format and exposed to RDF software, similar to the process of OAI metadata harvesting. This would make digital resources residing in closed-off repositories accessible in the open web environment or in a linked data controlled environment.
Miller states that the Dublin Core Metadata Initiative is moving away from the DC metadata element set and toward core vocabulary and metadata structures for linked data. The new library cataloging standard RDA has also been adopted primarily because of its compatibility with linked data and the semantic web. As commented on in a previous blog post, controlled vocabularies from Getty and the Library of Congress are being made available as linked open data.
This application of metadata as Linked Open Data and RDF in the semantic web environment has great potential but is not yet fully realized.
More detailed description of RDF in next blog post.
Metadata for Digital Collections, by Steven J. Miller