Database Reference
In-Depth Information
mm:0012
http://mereabeers.co.me/sells
http://mereabeers.co.me/is_brewed_in
http://sws.geonames.
org/2640726
FIGURE 5.3
RDF Graph Demonstrating Blank Nodes.
But, what happens if we want to say something like, “The Isis Tavern sells beer
that's brewed in Mereashire”? We know that it is some sort of beer, but we can-
not just use the class Beer because it is certainly not true that all Beer is brewed in
Mereashire. All we can say is that an instance of a beer is brewed there. However,
since we do not know which one, we cannot give it an identifier. Instead, to model
this, we introduce what is known as a “blank node” or “bnode” for short. A blank
node is a node that acts as the subject or object of a triple but does not have its own
URI—it is anonymous. In a graph, it is easy to represent: We just do not bother
labeling the blank node, as in Figure 5.3.
But, if we write the same information in triples, we need to introduce a blank
node identifier, which is of the form “_:name.” For example, our beer blank node
could be “_:beer0.” This allows us to identify it locally within our dataset as there
could be a number of blank nodes in the same graph. Again, the second triple in
Table 5.4
is calling on the GeoNames dataset to supply a URI for “Mereashire.”
3
