On Fri, Aug 27, 2010 at 8:01 AM, Ian Green1 <[email protected]> wrote: > - I'm not sure that these are really "two ways". More that the second is > an extension of the first. This is important since it means that consumers > that don't care or don't understand link properties can reasonably be > expected to ignore them, and that queries which don't talk about link > properties don't need to change when link properties are added or removed.
Good point. I did some rewriting and removed the "two ways" language. > - How about the RDF terms subject/object rather than source/target? Yes, that is better. Done. > - i suggest relaxing "there is an rdf:Description element" since such an > element need not exist - it could be given some rdf:type and thus be > abbreviated as part of the RDF/XML representation. I removed the sentence that you are referring to. It was not really helpful. Thanks for the feedback. - Dave
