Database Reference
In-Depth Information
•
Ontologies are independent of the implementing application and code,
increasing the visibility of the descriptions they provide of the data.
•
Developing OWL ontologies can be difficult, and methodologies exist to
aid the process.
•
OWL is based on first-order logic, and it can be difficult for nonlogicians to
fully understand all the logical consequences of its reasoning.
•
OWL classes are very different animals from OO classes; it is easy for the
novice with experience in OO design or programming to confuse the two.
Do so at your peril.
11.5 THE TECHNOLOGIES
First, let us reiterate Tim Berners-Lee's Linked Data principles:
1. Use URIs as names for things.
2. Use HTTP URIs so that people can look up those names.
3. When someone looks up a URI, provide useful information, using the stan-
dards (RDF, SPARQL).
4. Include links to other URIs so that they can discover more things.
To summarize, the Semantic Web relies on a stack of technologies, from URIs,
through RDF for data, SPARQL for querying, and OWL and RDFS for ontolo-
gies, to the more immature although fast-developing areas of Provenance (with the
Vocabulary of Interlinked Datasets, for example) and Trust. Although RDF can
be serialized in XML, it is often easier to read and write in Turtle format, which
exposes the triple structure more clearly. The learning point here, for those steeped
in XML technologies or those familiar with the tabular form used to represent data
in a GIS sitting on top of a relational database, is that RDF is more than just an
XML format or a table structure: It is a
graph data model
, which allows knowledge
to be structured much more flexibly, breaking away from both the relational and the
document structure.
A second take-home message for Linked Data technologies is that we must be
clear what exactly we are identifying with our URI: Is it a URI for a document
describing the thing or the URI of the thing (the “resource”) itself?
Third, we note that publishing Linked Data offers a more open and reusable alter-
native to the traditional REST (Representational State Transfer) API (Application
Programming Interface) as a means to access data. Whereas with a REST API the
structure of the data remains proprietary to each Web service, the RDF data model
is standard across all data providers, making it much easier to integrate data from
any Linked Data publisher.
Finally, as it has become apparent from our discussions in Chapter 8 onward,
many of the technologies relating to Linked Data, particularly those relating to link
discovery, authentication, and trust, as well as Linked Data user interfaces such as
browsers, are still very immature. They are often the subject of university research
projects and not at a stage where they can be reliably implemented in commercial,
high-volume workflows, although things are changing rapidly.
