I strongly disagree with the way the relations element has been
handled the last few months.
Firstly, the issue of the DC.Relation.Type and DC.Relation.Identifier
dichotomy has been raised by Andy Powell
(<URL:http://www.roads.lut.ac.uk/lists/meta2/1998/02/0030.html>) and
myself (<URL:http://www.roads.lut.ac.uk/lists/meta2/1998/02/0038.html>)
We never got any responses on our views.
Secondly, there is differences between the qualified DC RFC as posted
on this (<URL:http://www.roads.lut.ac.uk/lists/meta2/1998/02/0002.html>) list
and the document
<URL:http://purl.oclc.org/metadata/dublin_core/wsubelementdraft.html>
Why is there? Isn't the RFC going to be the authoritive document? Or
is the RFC going to be revised?
Thirdly, the DC.Relation.Type and DC.Relation.Identifier dichotomy
makes implicit assumption of a syntax which is of another kind than
for all other proposed subelements. It is my conviction that the we
need a small set of core sub-elements, and they should all be possible
to use also in the simplest encodings, and its usage should be
coherent.
The Date, Coverage and The-Other-Subelements working groups all propose a
substructure which is coherent and adhere to the Canberra qualifier
philosophy. Neither of which requires a grouping mechanism (as is
available in RDF). All three proposals are compatible with the older HTML
v2-3.2 and HTM 4.0 syntaxes. The embedding of relation metadata of the
kind that is required in
(<URL:http://www.roads.lut.ac.uk/lists/meta2/1998/03/0003.html>) will be
significantly harder to parse for harvesting robots, and will break
existing Dublin Core implementations.
The HTML meta-tagging is a dirty hack, a kludge. But we don't have to make
it even worse, do we? Therefor I urge that we return to the semantics
in the original proposal
(<URL:http://www.roads.lut.ac.uk/lists/meta2/1997/12/0046.html>).
However, In connection with the earlier discussion on Relation, I proposed
that we should start talking of DC metadata in _compact_ and _extended_
forms (see
<URL:http://www.roads.lut.ac.uk/lists/meta2/1998/02/0038.html>). In that
vocabulary I would say that the relations working group's proposal now is
in an extended form, whereas their original posting was using the compact
form.
I think that it is important that we are aware of existence of these two
forms and we do need both, but we should never mix the two.
Good afternoon
Sigfrid
|