[e] (0) Make the note more pedantically correct, since there are pars…

…e error differences, it turns out.

Fixing http://www.w3.org/Bugs/Public/show_bug.cgi?id=8699

git-svn-id: http://svn.whatwg.org/webapps@4673 340c8d12-0b0e-0410-8428-c7bf67bfef74

complete.html

@@ -78052,12 +78052,13 @@ <h3 id=parsing-html-fragments><span class=secno>12.4 </span>Parsing HTML fragmen
        <dd>Leave the tokenizer in the <a href=#data-state>data state</a>.</dd>
 
       </dl><p class=note>For performance reasons, an implementation that
-      uses the actual state machine described in this specification
-      directly could use the PLAINTEXT state instead of the RAWTEXT
-      and script data states where those are mentioned in the list
-      above. They are equivalent, since there is no <a href=#appropriate-end-tag-token>appropriate
-      end tag token</a> in the fragment case, but involve far fewer
-      state transitions.</p>
+      does not report errors and that uses the actual state machine
+      described in this specification directly could use the PLAINTEXT
+      state instead of the RAWTEXT and script data states where those
+      are mentioned in the list above. Except for rules regarding
+      parse errors, they are equivalent, since there is no
+      <a href=#appropriate-end-tag-token>appropriate end tag token</a> in the fragment case, yet
+      they involve far fewer state transitions.</p>
 
      </li>
 

index

@@ -72347,12 +72347,13 @@ document.body.appendChild(text);
        <dd>Leave the tokenizer in the <a href=#data-state>data state</a>.</dd>
 
       </dl><p class=note>For performance reasons, an implementation that
-      uses the actual state machine described in this specification
-      directly could use the PLAINTEXT state instead of the RAWTEXT
-      and script data states where those are mentioned in the list
-      above. They are equivalent, since there is no <a href=#appropriate-end-tag-token>appropriate
-      end tag token</a> in the fragment case, but involve far fewer
-      state transitions.</p>
+      does not report errors and that uses the actual state machine
+      described in this specification directly could use the PLAINTEXT
+      state instead of the RAWTEXT and script data states where those
+      are mentioned in the list above. Except for rules regarding
+      parse errors, they are equivalent, since there is no
+      <a href=#appropriate-end-tag-token>appropriate end tag token</a> in the fragment case, yet
+      they involve far fewer state transitions.</p>
 
      </li>
 

source

@@ -89074,12 +89074,13 @@ document.body.appendChild(text);
       </dl>
 
       <p class="note">For performance reasons, an implementation that
-      uses the actual state machine described in this specification
-      directly could use the PLAINTEXT state instead of the RAWTEXT
-      and script data states where those are mentioned in the list
-      above. They are equivalent, since there is no <span>appropriate
-      end tag token</span> in the fragment case, but involve far fewer
-      state transitions.</p>
+      does not report errors and that uses the actual state machine
+      described in this specification directly could use the PLAINTEXT
+      state instead of the RAWTEXT and script data states where those
+      are mentioned in the list above. Except for rules regarding
+      parse errors, they are equivalent, since there is no
+      <span>appropriate end tag token</span> in the fragment case, yet
+      they involve far fewer state transitions.</p>
 
      </li>