On Thu, 2 Mar 2000, Priscilla Caplan wrote:

> Eric Miller said:
> >> As such, the refinements identified by the LOC group for the CCP elements
> >> are listed in the ballot and all have the value of an 'DCMI Agent'.   The
> >> vocabularies for defining tha describing DCMI Agent is in a separate set.
> 
> Renato said:
> >Sorry Eric, but the Agent WG _did not_ recommend these qualifiers.
> >
> >There was no clear way on how to agree on which roles would be
> >recommended. So, the WG decided that that best way would be to
> >support a Role qualifier and recommend a number of vocabulaires
> >to choose from.
> 
> Again, I agree with Renato.  
> 
> Eric has made the argument that whether there is a Role qualifier with
> recommended vocabularies or whether there is a list of role-type
> subelements such as "illustrator" and "editor" is entirely a matter of
> syntax, not semantics.
> 
> If that is the case, then there is no reason not to present the
> recommendation as issued by the agent WG, as the ballot attempts to be
> syntax-neutral.  If as I suspect that is NOT the case, and there is some
> semantic difference, then the ballot should reflect what the agent WG
> recommended.  That is, either way, it should be on the ballot as a Role
> qualifier with recommended vocabularies.  

To be blunt... the Agents WG did not restrict itself to making
recommendations that fell within the two principles of element refiners
and value encoding schemes.  We have just had a vote that strongly
endorsed the notion that we should restrict the current vote to only those
two kinds of qualifiers.  Therefore, it seems clear to me that we can
*not* have a vote that conforms entirely to the recommendations of the
Agent WG.

In particular, a 'Role' qualifier (particularly a role qualifier that is
itself qualified with a scheme) does *not* seem to me to fit into the 2
qualifier principles.  By that I mean that, while we can encode

<meta name="DC.Contributor.Illustrator" content="Andy Powell">

in HTML, we cannot encode

<meta name="DC.Contributor.Role" content="...

in the same way.

I apologise for dropping into syntax, but this notion that we can somehow
discuss semantics completely independently from syntax excapes me...
ultimately it is very unhelpful because we simply have to be able to
encode this stuff! :-(

Andy
--
Distributed Systems and Services
UKOLN, University of Bath, Bath, BA2 7AY, UK       [log in to unmask]
www.ukoln.ac.uk/ukoln/staff/a.powell             Voice: +44 1225 323933
Resource Discovery Network - www.rdn.ac.uk         Fax: +44 1225 826838




%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%