On Tue, Jun 17, 2008 at 10:29 AM, Aristotle Pagaltzis <[EMAIL PROTECTED]> wrote: > > * Erling Wegger Linde <[EMAIL PROTECTED]> [2008-06-16 23:15]: >> Why not use RDF? Isn't RDF more general? > > Yes, which makes it more demanding to process properly. If Atom > was specified in terms of RDF, you can bet that the same thing > would happen as it did with RSS 1.0: aggregators would parse the > RDF as XML, instead of building a graph out of it, and would > therefore break if anyone actually used Atom as RDF, eg. changing > the precise XML serialisation or adding triples in unexpected > places.
But what about using RDF in the <content type="application/rdf+xml"> would this break the aggregators too? Regarding the atom:category(ies) I would like to state that a feed contains entries about for instance projects. Say: <feed> <category term="http://usefulinc.com/ns/doap#Project"; /> ... Would that mess up a client too? > > As I found out when I changed the namespace prefix bindings in my > feed, even just using Atom as actual XML will break a whole bunch > of naïve Atom parsers! > > I don't like this reality, but reality has a way of not caring > about whether I like it or not… > >> Could these categories be used anywhere outside of Atom? > > Not as it's written. > >> I'm not very familiar with the atom:category(ies) so I ask: can >> atom:categories be referred to / used in a meaningful way with >> RDF? > > An atom:category is basically a rudimentary shorthand for writing > an RDF triple (where the Entry is always the subject). > >> Why not try to use RDF when you can, not try to avoid it? > > See above. > > Note that there's nothing stopping anyone from GRDDLing Atom (or > otherwise creating a mapping), and there are community efforts > underway, eg. http://djpowell.net/blog/entries/Atom-RDF.html Just as you don't misunderstand me, I think it is great that Atom is kept simple, and as long as it is possible to use GRDDL or other approaches to map it to RDF I think it is great. However I think that the atom:category approach might not cooperate very we using GRDDL to transform Atom to RDF is greall with the future Semantic Web. This is just my intuition, and I think I need to see some more examples of atom:category schemes and how they are used before I conclude on this one. However, it should be possible to just use URIs to refer to which categories a Feed contains. Whether these URIs refers to an RDF document or the homepage of a company, a wikipedia site or whatever, it should be easy to deal with for a client. However, it would still make it possible for a client to look up that RDF document, or use a previously known ontology to find out that the category is actually a subClassOf another category/concept that it is already familiar with. I just don't understand why one are inventing the atom:category schemes? - Erling > > Regards, > -- > Aristotle Pagaltzis // <http://plasmasturm.org/> > > -- Med vennlig hilsen Erling Wegger Linde
