Andreas Hartmann wrote: > Doug Chestnut wrote: >> Hi Andreas, >> My understanding of the dublin core terms (of type element-refinement) >> is that they are qualifiers which "enhance the precision" of the >> elements. >> http://dublincore.org/resources/faq/#dumbdown >> >> It seems that we are treating the terms as an extended set of >> elements. Perhaps the interface just needs to be changed so that >> the term can be selected on the dc element which it refines. >> >> Perhaps the conf below could reference terms like: >> <element name="abstract" refines="dc:description"/> >> <element name="accessRights" refines="dc:rights"/> >> <element name="alternative" refines="dc:title"/> >> [...] >> >> WDYT? > > Yes, that sounds reasonable. Can this concept be generalized for our > meta data handling? > > A little comment on the syntax: > refines="dc:description" depends on a "dc" prefix, which would have to > be globally unique and declared somewhere. I guess it is easier to use > the namespace URI instead, e.g.
in dc, all elements can have a cardinality > 1, which is something that is not currently covered by lenya either. if lenya were able to handle multiple occurrences of the same element name (which it should imho, since there may be co-authors etc), we could get rid of the "refines" stuff, and just say <element name="dc:contributor" qualifier="editor".../> <element name="dc:contributor" qualifier="author".../> ... mapping each qualified variant of a dc element to a lenya metadata element of cardinality one is only half the way... all in all, i'd love to see the expressive power of full xml for the lenya metadata, although i can see that it's awkward to implement. just dreaming... --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
