- From: Henry Story <henry.story@bblfish.net>
- Date: Mon, 28 Aug 2006 17:38:46 +0200
- To: Jan Algermissen <algermissen1971@mac.com>
- Cc: Semantic Web <semantic-web@w3.org>, Davis Ian <iand@internetalchemy.org>
On 28 Aug 2006, at 17:05, Jan Algermissen wrote: > On Monday, August 28, 2006, at 04:59PM, Henry Story > <henry.story@bblfish.net> wrote: > >>> >>> Hmm.. I guess that is 'serialization', eh? >> >> No no. You can serialize an rdf graph into any number of different >> xml/rdf documents. When you crystalize it, you force a particular >> structure on the rdf serialization. > > Since this is a somewhat common task, it would be really useful to > have a declarative means for specifying the mapping. IIRC, [1] is a > nice approach. Good point. That would indeed be very useful. In my case BlogEd for example stores AtomOwl triples, and it would be really nice to be able to have a language like that to crystalize the graph into an atom feed. Henry > Jan > > [1] http://www.semanticplanet.com/2003/08/rdft/ > > > > > > o for example in the RSS 1.1 >> case you force the document to start with the <Channel> element, >> specify that it can contain <title>, <lint>, <items>..., and specify >> that <items> contain <item>s which contains <title>,<link>, ... >> elements. You do this using a Relax-NG schema. >> >> see the RSS1.1 spec for a good example of this: >> http://inamidst.com/rss1.1/ >> >> Henry
Received on Monday, 28 August 2006 15:39:10 UTC