On Thu, 2 Mar 2000, Priscilla Caplan wrote:
> HOWEVER, I agreed to Agent Core, and tried to say this over and over again,
> ONLY on the condition that basic pieces of information needed by most
> projects to describe CCP-type agents be included as qualifiers in DCMES, or
> at least be considered for inclusion on the ballot.
I broadly agree with this, though I would much prefer to see the word
'identify' in place of 'describe' and we clearly need to rethink how the
qualifiers can be included given that we seem to agree that the current
vote can be based only on the two qualifier principles.
> To me, these include Role (and not the way it ended up on the ballot, I
> will send a separate note on that), Affiliation (I agree arguments can be
> made either way here, but it is not right to pre-emptorily reject this),
> name, and agent type (which I would recast as type of name).
We need to see enough information encoded in the value of a Creator,
Contributor or Publisher element to 'identify' the agent. This
information will form part of a more general description of the agent -
the nature of that description will be the subject of a reformed Agent
Core (or Authority) WG. The danger with pushing the whole thing out to a
Sepatarate Agent Core is that we lose the focus on 'identifying' the agent
- and, I presume, we'll have a vote that doesn't include many qualifiers
for the CCP elements!
I'd still prefer to include a DCAGENT encoding scheme (a scheme that
encodes the name, affiliation and type semantics defined by the WG - see
previous message) in the current vote *if* someone has time to write it
and *if* the process allows us to add something new at this stage.
If not (and I suspect that is the case), then I think that we should push
this stuff out to a separate Agent Core WG with a specific remit to
consider agent 'identification' (rather than 'description') and a
corresponding encoding scheme in the first instance.
We cannot vote on 'affiliation' or 'agent type' as things stand, because
they are neither element refiners nor value encoding schemes.
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
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|