Clarify that lexicographical order uses code point ordering.

For #416.

Author: gkellogg

common/terms.html

@@ -1,5 +1,5 @@
 <section><h3>Terms imported from Other Specifications</h3>
-<p>Terms imported from [[[ECMASCRIPT]]] [[ECMASCRIPT]], [[[RFC8259]]] [[RFC8259]], [[[INFRA]]] [[INFRA]], and [[[WEBIDL]]] [[WEBIDL]]</p>
+<p>Terms imported from [[[ECMASCRIPT]]] [[ECMASCRIPT]], [[[RFC8259]]] [[RFC8259]], [[[INFRA]]] [[INFRA]], [[[XPATH-FUNCTIONS]]] [[XPATH-FUNCTIONS]], and [[[WEBIDL]]] [[WEBIDL]]</p>
 <dl class="termlist" data-sort>
   <dt><dfn data-cite="INFRA#list" class="preserve">array</dfn></dt><dd>
     In the JSON serialization,
@@ -55,6 +55,15 @@
     is a sequence of zero or more Unicode (UTF-8) characters,
     wrapped in double quotes, using backslash escapes (if necessary).
     A character is represented as a single character string.</dd>
+  <dt><dfn data-cite="XPATH-FUNCTIONS#codepoint-collation" data-lt="code point order|code point ordered|unicode codepoint collation">Unicode code point order</dfn></dt>
+  <dd>This refers to determining the order of two Unicode strings (`A` and `B`),
+    using <a>Unicode Codepoint Collation</a>,
+    as defined in [[[XPATH-FUNCTIONS]]] [[XPATH-FUNCTIONS]],
+    which defines a
+    <a href="https://en.wikipedia.org/wiki/Total_order">total ordering</a>
+    of <a>strings</a> comparing <a data-cite="INFRA#code-point">code points</a>.
+    Note that for UTF-8 encoded strings, comparing the byte sequences gives the same result as <a>code point order</a>.
+  </dd>
 </dl>
 
 <p>Terms imported from [[[RFC3987]]] [[RFC3987]]</p>

index.html

@@ -3526,13 +3526,13 @@ <h3>Using the Document Base for the Default Vocabulary</h3>
     would still be in effect when traversing into another <a>node object</a>.</p>
 
   <p>When expanding, each value of <code>@type</code> is considered
-    (ordering them lexicographically) where that value is also a <a>term</a> in
+    (in <a>code point order</a>) where that value is also a <a>term</a> in
     the <a>active context</a> having its own <a>type-scoped context</a>.
     If so, that the <a>scoped context</a> is applied to the <a>active context</a>.</p>
 
   <p class="note">The values of <code>@type</code> are unordered, so if multiple
     types are listed, the order that <a>type-scoped contexts</a> are applied is based on
-    lexicographical ordering.</p>
+    a>code point ordering</a>.</p>
 
   <p>For example, consider the following semantically equivalent examples.
     The first example, shows how properties and types can define their own
@@ -12877,7 +12877,7 @@ <h3>Serializing/Deserializing RDF</h3>
         and furthermore comply with the following constraints:
         <ul>
           <li>It MUST NOT contain any unnecessary whitespace,</li>
-          <li>Keys in objects MUST be ordered lexicographically,</li>
+          <li>Keys in objects MUST be in <a>code point order</a>,</li>
           <li>Native Numeric values MUST be serialized according to
             <a data-cite="?ECMASCRIPT#sec-tostring-applied-to-the-number-type">Section 7.1.12.1</a> of [[?ECMASCRIPT]],</li>
           <li>Strings MUST be serialized with Unicode codepoints from <code>U+0000</code> through <code>U+001F</code>
@@ -13877,6 +13877,7 @@ <h2>Changes since Recommendation of 16 July 2020</h2>
     <li>Fixed typo in <a href="#alternate-document-location" class="sectionRef"></a> where "HTTP document" was intended to be "HTML document".</li>
     <li>Allow further structured extensions of `application/ld+json` by using
       `+ld+json` as a structured media extension. See <a href="#structured-extension-ld-json"></a>.</li>
+    <li>Use <a>code point order</a> as a better definition for the more vague "lexicographical order".</li>
   </ul>
 </section>