Linked Data

Although the concept has many different definitions, most of them agree that Linked Data is an way to publish structure data in such a way that rather than being a collection of separate data silos, the data is interlinked and one data set can reference data items in another data set.

What is Linked Data ?

The most common implementations of the Linked Data concept are based on standard Web technologies such as the HTTP protocol for retrieving, creating and updating data; and URIs to address and identify data items. The format of the data exchanged is more variable; depending on the application it might be a standard format such as RDF/XML or topic maps or it may be a proprietary data model serialized in XML or JSON.

The four basic principles of linked data outlined by Tim-Berners Lee in his Design Issues: Linked Data document are:

Linked Data provides a means for people, and more importantly applications, to incrementally discover information by starting with one data set; browsing or querying it and then following the links in the results that are returned. Some of those links may lead the consumer to other data sets which help to expand the amount of information available.

Why Publish Linked Data ?

Making your structured information available as Linked Data enables access to that data in a way that is completely neutral about how that data is consumed and used - it opens up the data to being analyzed, integrated and combined with other data in ways that you as the original publisher of the data might never have considered. Within the firewall of an organization, Linked Data can be used to open up data silos and to make integration tasks between data stores much easier. On the Internet publishing quality Linked Data can be used as a way to generate more traffic; make product offerings easier to find; or to establish a leadership position by being the place to go to in order to get information about a particular subject area. In addition as mobile and tablet devices with constant internet connections become more widespread, Linked Data provides an approach which enables a new breed of data-centric applications to be developed and distributed with their data sets continually updated and extended on the Web.

Web3 and Linked Data

Web3 adheres to the principles of Linked Data. Every topic map and topic in Web3 has its own unique HTTL URL which, when retrieved can provide human-readable information in HTML format; or machine-readable information in a variety of formats including RDF/XML and XML Topic Maps (XTM) standard syntaxes. Where a topic references other topics or associations, a full URL to the other topic or association is provided allowing the consumer to generate new requests to browse around the data. Web3 also supports the SPARQL query language for consumers to generate their own queries against the data-store. In fact we think that with the intuitive schema-driven editor Web3 is the easiest way to create and publish manually curated Linked Data sets as well as being a perfect way to manage and publish larger, automatically created data-stores.