--On 26/11/99 9:17 AM +0100 Michael Jost wrote:
 
> I think this would work for (1[b]) (by specifying no or an own
> vocabulary).
> But it seems to me that it won't work with (2). I thought it was
> the aim of this qualifier discussion to come up with a *fixed* 
> sets of qualifiers for the DC elements. With option (2) you 
> would allow anyone to invent any new *qualifier* they want (at 
> least for agent roles).

Michael - yes this is correct. Andy raised this exact problem
in his last message (not in the archive yet):

  > No, not necessarily.  If your option 2 is my option 2 then, as I've
  > mentioned before, we should ask the DC Datamodel group for
  > clarification on how to select an element qualifier from a non-DC
  > controlled vocabulary.

and I have sent an email message to the DC DataModel WG asking
for clarification on this point.


Cheers... Renato
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
Dr Renato Iannella           ->  http://purl.net/net/renato
Principal Research Scientist ->  http://www.dstc.edu.au/


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