On Mon, 30 Sep 2002, Ann M Wrightson wrote:
> I think that the aspect of what I was saying here that has been missed, is
> that the dc:creator may be the entity (legally or otherwise) responsible for
> the creation of the resource, even though the content was all actually
> generated by (other) contributors.
> The UK e-Government Metadata Standard (which is intended to be an extension
> of dc) takes this view.
Ann,
apologies. Yes, I had missed what you were getting at here.
FWIW, I am not convinced that you are using dc:creator in the most
appropriate way here... but I agree that *if* your interpretation of the
semantics of dc:creator is correct, then dc:creator cannot be said to be
an element refinement of dc:contributor.
Let me try and explain why I think your interpretation of dc:creator is
wrong...
Clearly, the semantics of dc:creator are defined using words and it is
possible to interpret those words in different ways. My recollection is
that dc:creator was called 'author' in the earliest iterations of DC.
The name was changed, not because new semantics were required, but because
it was recognised that not all resources are text based - so, for example,
it didn't make sense to refer to the creator of a photograph as an
'author'.
If the name of the element was author, would you still feel happy with a
government dept claiming to be the 'author' of a document that it had
commissioned but that was actually written by someone else?
From another perspective... Imagine a cross-search of various databases
looking for documents written by me. (Surely this is what DC is really
about!?) A dc:creator search is going to find most stuff, but not
documents where I've been sub-contracted by the UK GVMT to write a
document (because in that case the government department is listed as the
creator). That doesn't feel right to me. Interoperability has been
compromised because of the way that you've chosen to use dc:creator IMHO.
Of course, if I know that you've chosen to use dc:contributor to list the
author, then I might phrase my search in terms of both dc:creator and
dc:contributor, or I might phrase my search in terms of dc:contributor
only (but assume that dc:creator is generally used as a sub-property of
dc:contributor and therefore allow for matches against dc:creator as
well).
But that rather brings us back to the distinction between the two
elements... :-)
Andy.
> -----Original Message-----
> From: DCMI Architecture Group [mailto:[log in to unmask]]On
> Behalf Of Andy Powell
> Sent: 27 September 2002 15:47
> To: [log in to unmask]
> Subject: Re: multiply affiliated refinements
>
>
> On Fri, 27 Sep 2002, Ann M Wrightson wrote:
>
> > For example, in designing the metadata for UK Govt XML schemas, we had a
> > problem deciding what to do about a subcontractor who had actually created
> > the content, vs the commissioning owner who was ultimately responsible.
> The
> > decision made was to make the commissioning owner the creator, and use
> > dc:contributor for the subcontractor, with the role played reflected in
> the
> > metadata value. See the following extracts from the v1.0 document:
> >
> > Examples
> > For an XML schema developed using the GovTalk process, through the
> > Government Schema Group:
> > CONTRIBUTOR: Government Schema Group
> > For an XML schema developed by a subcontractor for a Government
> > organization:
> > CONTRIBUTOR: developed by alphaXML Ltd http://www.alphaxml.com
> >
> > Where a schema is developed by a subcontractor or external consultant for
> a
> > Government organization, then the Government organization is considered to
> > be the creator, and the subcontractor a contributor (see above)
> > Where the schema is developed using the UK GovTalkT collaborative process,
> > then the Creator is UK GovTalk, and substantial contributors may be
> > recognized as described above.
> > Where the schema is developed within a Government organization, then the
> > name of the organization (or some relevant subunit) appears.
>
> Perhaps I don't understand what you are suggesting here? Nothing I've
> said would prevent you from using dc:creator and dc:contributor in the way
> you describe above. I'm simply observing that anyone that you choose to
> list as being a dc:creator is 'by definition' also a dc:contributor (based
> on the DCMI element definitions).
>
> Also note that if you are suggesting that the value of dc:contributor
> should contain both the name of the contributor and the role they played,
> e.g.
>
> <dc:contributor>
> developed by alphaXML Ltd
> http://www.alphaxml.com
> </dc:contributor>
>
> as you appear to indicate above, then that seems to me to be a little
> questionable. dc:contributor is defined as 'An entity responsible for
> making contributions to the content of the resource.'. That definition
> does not include space for an indication of what role the entity played.
> Apologies if I have mis-interpretted your message above?
>
> Andy
> --
> Distributed Systems, UKOLN, University of Bath, Bath, BA2 7AY, UK
> http://www.ukoln.ac.uk/ukoln/staff/a.powell +44 1225 383933
> Resource Discovery Network http://www.rdn.ac.uk/
>
Andy
--
Distributed Systems, UKOLN, University of Bath, Bath, BA2 7AY, UK
http://www.ukoln.ac.uk/ukoln/staff/a.powell +44 1225 383933
Resource Discovery Network http://www.rdn.ac.uk/
|