Linked data as data model

It has been longer than 15 years since the famous article by Tim Berners Lee in which there is a talk for the first time about the web of data. This article was the germ of the semantic web initiative, an effort to evolve the way in which information was represented with the goal of moving from text (unstructured content) to data (structured content), facilitating the operations and more complex computer processes.

As a result of this initiative the linked data paradigm was born, whose most basic conceptual proposal refers to the building a web of data, where information has an enough level of representation to allow the construction of relationships and links, which enable the navigation among datasets from different sources, including different providers.

Thus, the linked data, by itself, is nothing more than a set of good practices for publishing and connecting structured data on the web. A proposal that is articulated through a set of standard W3C technologies:

• URIs, a generic mechanism to identify entities and concepts that work in an analogous way to proper names in natural language. The main advantage with other identifiers is that they are global (on the web) and unique (unambiguously refer and identify the same resource).
• The HTTP protocol, a simple and universal mechanism that allows to recover the resource that identifies a URI. In this way, when we "solve" a URI it is as if we were calling the resource through its own name, being able also to accede to its description (its "meaning") constructed through another resource, that is to say, other URIs.
• RDF data model. This technology, cornerstone of the technological stack of the semantic web, is a framework for describing resources (entities and objects on the web) that relies on the naming and access mechanisms mentioned above (URIs and HTTP respectively).

Thus, RDF is nothing more than a syntax for the triplet expression: "subject - verb / relation / link - object", which allow to describe resources in a simple and natural way, in addition to an open resource modelling, without the necessity of schemas or vocabularies, which facilitate the integration of heterogeneous datasets.

The original purpose for which RDF was designed is data exchange. RDF was not conceived with the aim of replacing any technology, but complementing and providing added value to the existing ones, allowing to raise the level of representation of the data published on the web and facilitating the communication between systems and agents; regardless of how they are built and the underlying technology.