added semantics section, fixed string literl desc
authorGuus Schreiber <guus.schreiber@vu.nl>
Tue, 19 Nov 2013 23:34:21 +0100
changeset 1357 41b583fca125
parent 1356 1e4f81f15b94
child 1358 501162b33a0e
added semantics section, fixed string literl desc
rdf-primer/index.html
--- a/rdf-primer/index.html	Mon Nov 18 11:05:06 2013 +0000
+++ b/rdf-primer/index.html	Tue Nov 19 23:34:21 2013 +0100
@@ -338,14 +338,10 @@
       tag</i>. For example "Léonard de Vinci" could 
       be associated with the "fr" language tag and "李奥纳多·达·文西"
       with the "zh" language tag.</p> 
-
-      <p>Literals may only appear in the object position of a triple.</p>
-
-     <p>The RDF Concepts document provides a
-     <a href="http://www.w3.org/TR/rdf11-concepts/#section-Datatypes">list
-     of datatypes</a>.  This includes many datatypes defined by XML
-     Schema, such as string, boolean, integer, decimal and date. </p>
-
+      
+     <p class="note">The RDF data model assigns the special datatype
+     <code>rdf:langString</code> to language-tagged literals.</p>
+      
      <p class="note">The 2004 version of RDF contained the notion of a
      "plain literal" with no datatype. This feature has been removed as the
      distinction between "plain" literals and literals with datatype
@@ -353,6 +349,13 @@
      writing literals without an explicit datatype and treat this
      as syntactic sugar for a <code>string</code> datatype.</p> 
 
+     <p>Literals may only appear in the object position of a triple.</p>
+
+     <p>The RDF Concepts document provides a
+     <a href="http://www.w3.org/TR/rdf11-concepts/#section-Datatypes">list
+     of datatypes</a>.  This includes many datatypes defined by XML
+     Schema, such as string, boolean, integer, decimal and date. </p>
+      
     </section>
 
     <section id="subsection-blank-node">
@@ -678,14 +681,11 @@
 string"</code> is equivalent to <code>"This is a
 string"^^xsd:string</code>. </p>
 
-<p>Strings can be language-tagged; the tag
-appears directly after the string, separated by a <code>@</code> symbol. In the
-case of language-tagged strings no datatype is specifed explictly.
-
-<p class="note">The RDF data model assigns the special datatype
-<code>rdf:langString</code> 
-to language-typed literals, but in the Turtle syntax this datatype is
-implicit.</p>
+<p>In case of language-tagged strings the tag
+appears directly after the string, separated by a <code>@</code>
+symbol, e.g. <code>"Leonardo da Vinci"@it</code>. In the
+case of language-tagged strings the datatype is not specifed explictly
+in Turtle.</p>
 
 <p>The above is by no means a full account of the Turtle syntax. For
 more details about the syntax of Turtle the reader is referred to the
@@ -822,10 +822,60 @@
 
     <h2>Semantics of RDF Graphs</h2>
 
-    <p>@@ example: derive that Alice is a person</p>
-    <p>@@ explain that one ITI can be a subject/object in one triple
-    and a predicate in another triple. </p>
-    <p>@@ etc </p>
+    <p class="note">Readers should view this section as optional.</p>
+   
+    <p>RDF is grounded in a formal semantics which is specified in the RDF
+    Semantics [[RDF11-MT]] document. This document takes a logical stance on RDF
+    graphs; it specifies truth-preserving conditions of RDF graphs as
+    well as valid derivations from RDF graphs. Such logical
+    consequences are called <a
+    href"http://www.w3.org/TR/2013/CR-rdf11-mt-20131105/#semantic-extensions-and-entailment-regimes">entailments</a>. For
+    example, consider the following two statements:</p>
+    <pre>
+    <code>ex:Bob foaf:knows ex:Alice .</code>
+    <code>rdfs:domain foaf:knows foaf:Person .</code>
+    </pre>
+    <p>The RDF Semantics document tell us that from this graph it is legal to
+    derive the following triple:</p>
+    <pre>
+    <code>ex:bob rdf:type foaf:Person .</code>
+   </pre>
+   
+   <p class="note"</p>RDF Semantics distinguishes
+   a number of different "entailment regimes". For example, there
+   are "simple" entailments that are true for every RDF graph,
+   as well as entailments based on datatypes. Technically speaking,
+   the entailment in the example 
+   above is not sanctioned by the basic RDF model, but stems from the
+   semantic extension of RDF with the RDF Schema vocabulary. For detailed
+   iinformation about entailment regimes and semantic extensions
+   please consult the RDF Semantics document [[RDF11-MT]]. </p> 
+
+   <p class="note">As we say in the earlier <a
+   href="#section-rurtle-example">Turtle example</a> we could have
+   used the shorthand <code>ex:bob a foaf:Person</a> for the example
+   above. </p>
+   
+   <p>The semantics of RDF also tell us that the triple:</p>
+    <pre>
+   <code>ex:Bob ex:Age "forty"^^xsd:integer . </code>
+   </pre>
+   <p>leads to a logical inconsistency, because the value does not
+    abide by the constrinats posed by the XML Schema datatype <a
+    href="http://www.w3.org/TR/xmlschema11-2/#integer">integer</a>.</p>
+
+   <p>@@ Tools </p>
+
+   <p>Unlike many other data
+   modelling languages, RDF Schema allows the same entity to be used both as
+   class and as proerty. Therefore, RDF semantics views the following graph
+   as valid:</p>
+   <pre>
+   <code>ex:Marriage rdf:type rdf:Property .</code>
+   <code>ex:Marriage rdf:type rdfs:Class . </code>
+   </pre>
+
+
 </section>
 
    <section id="section-data">