Call out that language information is preserved, but should not be depended upon due to differences in processing model.
--- a/microdata-rdf/index.html Mon Oct 31 15:10:14 2011 -0700
+++ b/microdata-rdf/index.html Mon Oct 31 15:16:12 2011 -0700
@@ -275,14 +275,13 @@
<p>Microdata also specifies that items are values are ordered, which is not typically the case for RDF
vocabularies. In fact, unless a property has an <code>rdfs:range</code> of <code>rdf:List</code>, or is
unspecified, it may not be appropriate to generate an <tref>RDF Collection</tref>.</p>
- <p>The Microdata JSON serialization does not retain datatype or language information that might be derived
- from the HTML DOM. The RDF Transformation does retain both datatype and language information when it
- is available.</p>
<p>This specification is an update to the original RDF transformation process in addition to
<tref>vocabulary</tref>-specific rules that affect the generation of property URIs and value serializations.
This is facilitated by a <tref>registry</tref> that associates URIs with specific rules based on matching
<aref>itemtype</aref> values against registered URI prefixes do determine a vocabulary and
vocabulary-specific processing rules.</p>
+ <p class="note">The Microdata JSON serialization does not retain datatype or language information that might
+ be derived from the HTML DOM. The RDF Transformation does retain language information when it is available.</p>
</section>
<section>