More Advanced Applications - User-Centered Data Management - page 55

Figure 4.5: Wikipedia page including info box on right.
PREFIX dbprop: <http://dbpedia.org/property/>
PREFIX db: <http://dbpedia.org/resource/>
SELECT ?who ?work ?genre
WHERE {
db:Tokyo_Mew_Mew dbprop:illustrator ?who .
?work dbpprop:author ?who .
OPTIONAL { ?work dbpprop:genre ?genre } .
}
Figure 4.6: SPARQL query of DBpedia RDF (from wikipedia.org/wiki/dbpedia ) .
Early HTML had been derived from SGML and so, for example, close tags were not required where
they could be inferred from the structure, for example, in “ < ul >< li > one < li > two < /ul > ,” a parser
knows that the first < li > tag encloses “one” as < li > elements cannot be nested, effectively yield-
ing “ < ul >< li > one < /li >< li > two < /li >< /ul > ,” not “ < ul >< li > one < li > two < /li >< /li >< /ul > .”
This is fine for a well-known structure such as HTML, but it means that new tags cannot easily be
added. In contrast, XML enforces strict nesting, so a parser does not need to know the kind of data
to be able to parse the hierarchical structure. A DTD can be associated with an XML document;
however, this can be used optionally for validation and is not essential for parsing.