Database Reference
In-Depth Information
5
The Resource
Description Framework
5.1 INTRODUCTION
This chapter looks at the Resource Description Framework (RDF)-the standardized
way of encoding data for the Semantic Web. It has been said that, “RDF is for the
Semantic Web what HTML has been for the Web” (Yu, 2011, p. 24). Since RDF is
the primary building block of Linked Data and the Semantic Web, it is important to
understand how it works in terms of both its data model and its language and how to
interpret the meaning of its constructs, its semantics. In addition, in this chapter we
introduce some basic RDFS
1
(RDF Schema) ontologies that you are likely to come
across and may find useful to reuse when publishing your own data. Chapters 6 and 7
go into more detail about how to encode GI as Linked Data and how to publish,
query, and link it on the Linked Data Web, but for now we concentrate on preparing
the groundwork of your RDF understanding.
5.2 RDF: THE PURPOSE
So, what exactly is RDF trying to achieve? Its aim is to be both a “conceptual
framework” (a way of looking at things) and an actual language for describing Web
resources independently of any particular domain or subject area. The purpose of
this is to allow applications that exchange machine-understandable information on
the Web to
interoperate
more easily with each other. Since we are dealing with the
Web, this framework needs to be
scalable
, so that it still works with any amount of
data;
flexible,
, expressive enough to encode any sort of information we might need
to talk about; and yet
simple
, so that it is easy for anyone to read, write, or query it.
Let us illustrate all of this with an example. The fictional Web mapping com-
pany Merea Maps offers data about Topographic Objects via a Web service API
(Application Programming Interface). In the traditional Web scenario, a mobile user
wants to find out which pubs are nearby. Through an application on the user's phone,
an HTTP (Hypertext Transfer Protocol) request is sent to the Web service to retrieve
a list of Topographic Objects, categorized as “Pub,” within a certain radius of the
user's current location, specified as latitude and longitude. The list of pubs returned
might be something like that seen in
Table 5.1
.
Although the ID number may not be presented back to the user, as it is meaning-
less, the Web service will almost certainly internally store an identifier for each
Topographic Object. Our user decides on the Isis Tavern, visits it, and thinks it is
great. Now, if the user wants to publish a review of the pub, giving it five stars and
letting people know about its food and beer garden, the user will encounter some
65
