As I said in a message to DC-uasge the other day, I was voting in the
spirit of distiunguishing well suppoorted qualifiers from noit well
supported qualifiers in the expectation that the DC-usage recommendations
would be brought to DC-AC where I would argue that we don't yet have the
proper frameworks to issue ANY qualifiers.
I understood this to be the process when Stu suggested to some members of
DC-AC that they might not want to participate in Dc-Usage since it would
involve a lot of nitty-gritty review (and implicitly or explicitly, I can't
remember, that DC-AC would get the final say.)
IF Dc-AC is not going to review the overall situation of DCMI with resperct
to issuing qualifiers before the are relerased, I will changhe my votes in
Dc-Usage to REJECT ALL qualifiers. Frankly, we aren't ready. Carl
articulated clearly why in his message this morning. 
David

At 09:12 AM 1/13/00 -0500, you wrote:
>It is the job of the DC-Usage Committee to identify the starting set of
>recommended DC Qualifiers based on the proposals brought forward in working
>groups and the subsequent discussion and balloting in this forum.
>Qualifiers which have attracted a clear consensus of 2/3 majority by
>February 11 will be included in that set.  The Directorate reserves the
>responsibility to change names of, and assign tokens to, qualifiers to
>promote clarity and consistency.
>
>The members of the Usage Committee are essentially the leadership of the
>DCMI, most of whom have been involved in this activity for a very long time.
>I have confidence that this group of theoreticians and practitioners has
>sufficient experience and insight to make reasonable choices that will
>reflect the needs and expectations of the community.  
>
>Carl's question reflects, I believe, a legitimate concern that there is
>implicit modeling embedded in the balloting that has not received sufficient
>scrutiny.  This can *only* be argued for one of the proposals in the set
>(Agents)however.  The argument has been made that we do not understand the
>underlying implications of the modeling implicit in this proposal, and these
>implications will somehow defeat us in the future.  I have seen no
>convincing evidence of this.  The current proposal will not obstruct
>deprecation of agents into a single agent element (probably our most
>important question of evolution) if we should choose this path.  In fact, it
>makes it easier.   What other deep structural problems are lurking?  I do
>not see them, and three years of data modeling efforts have not exposed
>them.
>
>In a previous message
><http://www.mailbase.ac.uk/lists/dc-usage/2000-01/0025.html> I gave a prose
>interpretation of what I believe the current agent qualifier proposal
>supports.  How does this model fail our objectives?  What does it leave out?
>How does it confound future evolution?
>
>If proponents of an alternate view are willing to articulate an alternative
>model for agents that accomodates the spirit of the recommendations of the
>Agents Working Group, I would support balloting such a proposal (either by
>modifying the existing proposal or balloting them in parallel).  What I am
>unwilling to do is to further delay completion of our publically announced
>objective (to ratify the initial version of the DC Qualifiers) in deference
>to an open-ended reverse engineering process on the part of the data
>modelers among us.
>
>I am reminded of the joke about economists... string them all end to end,
>and they wouldn't reach a conclusion.  We will reach a conclusion, and it
>will be announced as soon after our balloting is closed as possible.
>
>stu

David Bearman
President, Archives & Museum Informatics
2008 Murray Ave., Suite D
Pittsburgh, PA 15217 USA
+1-412-422-8530
fax +1-412-422-8594
[log in to unmask]
http://www.archimuse.com


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