Normally I just lurk here, but have been observing this conversation as I think it's bringing up some important concepts about the web and the role of HTML.
For example: <article id="interesting-article"> Interesting article <article id="self-contained-article"> <a href="http://www.example.org/interesting">Interesting article also here</a> </article> </article> While #self-contained-article is self-contained in that it gives a user agent everything it needs, and given an internet connection, transfers it to the user, it's not self-contained in the traditional, dictionary sense. Because its content is simply not there in this document. Additionally, this example points out that in its dependence on #interesing-article, #self-contained-article *should* retain that information if it's to be semantically rich enough. While #interesting-article in fact is an element of the document obtained at its URL, #self-contained-article exhibits two relationships, one which is specified by the link it contains, and another, unspecified, by the URL of the document it lives in AND the #interesting-article in that doc which contains #self-contained-article. Just some food for thought... Aaron -- Aaron Miller Co-founder, BookGlutton/ReadSocial Senior Developer, Netgalley.com