另外，参考： bulding the semantic web on XML ; Yin/Yang web.  google...
最本质的区别如楼上所述：XML的语义是隐含在文档的标记和结构中的，不能表达机器
可理解的语义。

我最近写了一篇文章，就是扩展Yin/Yang Web的，有兴趣看的话，留个email. 希望您能给我一些comments.

1） XML有XML Schema约束它的语法结构，而RDF的语法虽然是XML格式，但并没有XML Schema来约束它的结构，而RDF Schema是解释性的 ，即解释RDF 中的词汇的意义。

2） RDF可以看成一种Web知识表示语言，或者说一个逻辑语言，有形式化的语法，模型论语义，证明论，还有完全性，可靠性定理，总之，是个完备的形式化系统。 而XML只是一种用于
描述语法的语言，即其他语言的语法都可以是基于XML。

3）RDF文件可以被合并，聚集，而XML文件的合并和聚集，有谁知道吗？

4）RDF的数据模型是 无序的边带有标识的有向图，XML的数据模型是有序的节点带有标识的树

。。。。。
欢迎大家见仁见智，多多补充。

   It's best to tend to think of RDF as what it is: another language, just as SVG is, rather than viewing it simply as XML surrounded in rdf:RDF tags.

    With XML, you expect to get a tree with its elements in a predictable order. If an expected element is missing, then it tends to render your whole XML document invalid with respect to the schema you are using. If an element you don't expect to be there is present, then again your document becomes invalid. With RDF, you have no particular expectations; if the particular strand of spaghetti you're looking for isn't there, the rest of the ball remains. If there are new strands you don't expect, you won't even notice their presence if you don't go looking for them.

   The big advantage of being failure-friendly is the enhancement of interoperability. Consider what happens if you add a new property to your description of my weblog and it becomes like this:
<rdf:Description rdf:about="http://usefulinc.com/edd/blog">
   <dc:title>Behind the times</dc:title>
   <example:coolness>10</example:coolness>
</rdf:Description>
My "find the title" processor can still deal with your description just fine, as all it cares about is the dc:title property. All RDF processors are automatically future compatible, and all RDF descriptions are automatically backwards compatible. This is a huge benefit over traditional XML processing.

参考：
【1】 Edd Dumbill Sticking with it -- RDF http://usefulinc.com/edd/blog/contents/2003/08/08-whyrdf/read
【2】 Missing isn't broken: data validation and freedom on the Semantic Web http://rdfweb.org/mt/foaflog/archives/000047.html

Justification for doing this as RDF rather XML, in one line?

there's no algorithm for merging 2 random XML docs (but there is for RDF/XML)/

http://rdfweb.org/pipermail/rdfweb-dev/2003-July/011457.html

A story about RDF and XML

1.XML Documents represent the XML Infoset; RDF graphs represent what those Infosets are trying to tell us about objects, their inter-relationships and properties.

2.RDF gives us a model for namespace mixing and data merging:

There is no algorithm for merging two XML Infosets, to enable us to pool knowledge acquired from diverse sources. The RDF information model, by constrast, was designed with data aggregation (rather than structured documents) in mind. Merging RDF data is trivial: add the triples extracted from two RDF/XML documents, and store them in a new one.

3.RDF views of the Infoset are explicit about the information we can throw away:

Transforming Infosets into their RDF graph allows us to throw away irrelevant information, such as the aspects of the Infoset concerned with preserving a representation of document ordering. When we define transformations from an XML Infoset into RDF, we show XML processors which parts of the Infoset can be discarded without losing the essence of the message encoded in that XML.

说RSS是RDF的子集，是强调RDF可用于描述很多知识，而RSS是其中一种；
说RDF是RSS的子集，是强调RSS可以用多种格式来写，而RDF只是其中一个可选的格式。

有我上传的电子版，请不吝指教。