RE: [oic] first impression on conformance tests ? - <dc:creator> namespace abuse

["Dennis E. Hamilton" <dennis.hamilton@acm.org>]

Yes, that's right.  The RDF CURI is an attribute value and is distinct from
the <office:meta> <dc:creator> element.

I need to raise this at the ODF TC.  In the Ed.Note in 1.2 cd01-rev01, the
URL that is there is no longer appropriate.
<http://dublincore.org/documents/dcmi-terms/> needs to be used now.  And the
correct creator element needs to be referenced in that document, since there
is a newer namespace with a creator element.  Also, notice that
<office:annotation> <dc:creator> and <office:change-info> <dc:creator> have
no reason to be defined the same way as <office:meta> <dc:creator>.
Unfortunately, in ODF 1.1, definitions of <office:annotation> and
<office:change-info> cross-reference to the definition of <dc:creator> in
3.1.7, which contradicts what is said at those definitions.

One possible way out of this is to simply deprecate <dc:creator> in
<ofice:meta> (and those other places); add a <meta:last-modifier> to capture
the case that is intended in 3.3.2; then, introduce <dcterm:contributor> in
the annotation and change-info cases.

 - Dennis

-----Original Message-----
From: Hanssens Bart [mailto:Bart.Hanssens@fedict.be] 
http://lists.oasis-open.org/archives/oic/200903/msg00068.html
Sent: Tuesday, March 31, 2009 16:05
To: dennis.hamilton@acm.org; OIC TC List
Subject: RE: [oic] first impression on conformance tests ? - <dc:creator>
namespace abuse

http://lists.oasis-open.org/archives/oic/200903/msg00066.html
> With regard to the ODF 1.1 section 3.7.1 modification of the Dublin Core
1.1
> creator element, I note that the RDF CURI attribute value "dc:creator"
using
> the same namespace binding for the dc: prefix, and allowed in the ODF 1.2
> RDF Metadata, will always have the Dublin Core semantics and never have
the
> interpretation forced onto the ODF 1.1 <office:meta> <dc:creator> element
> except by coincidence.  

Mm, but that's only for the RDF part, right ?
Because 3.3.2.7 (non-RDF, starting 3.3.1) in 1.2cd01-rev01 still mentions
the
difference in semantics for plain old non-RDF meta data...

Bart