Dear Colleagues,
	
In a teleconference Wednesday, the Policy Advisory Committee of DC agreed
that it was essential before we meet in Washington to articulate the
arguments which have been advanced on meta2 both for and against the
"agents" proposal. I was asked to take on this task and hope I'm up to it.

With apologies in advance to anyone whose position I have misrepresented or
overlooked or whose name I have misspelled (and for any other sins real and
imagined), I have tried this below.  With apologies to all, since the
categorization severely over-simplifies everyones more nuanced and complex
views.

The Proposal, in its over-simplified essence:
1) To conflate the DC elements Creator, Publisher and Contributor into one
element
2) To adopt the terms in the USMARC Relator list as a scheme for Role in
qualified DC

Rationale:
1) Metadata creators will not consistently chose an appropriate element in
which to record information about those involved in the production through
distribution of information resources, hence users will need to look in all
three locations.
2) The range of roles is such that the creator and publisher are
insufficient and contributor too vague, so a list of roles will need to be
adopted for qualified DC and the USMARC list is a good, place to begin and
it is maintained/extensible.


I see two fundamental arguments advanced for and against the proposal:

1) TIMING
	
Pro: Now, before NISO CEN standardization process starts and before much
implementation
[Pricilla Caplan, Martin Dillon, Rebecca Guenther, Diane Hillman]

Anti: This is destabilizing and discredits DC and/or there is already too
much implementation
[Leif Andresen, Ann Apps, Simon Cox, Tony Gill, Rachel Heery, Thomas
Hofman, John Knight, Carl Lagoze, Jenny Thornely, John Tipler]

2) USAGE

Pro: Creator, Publisher, Contributor not widely understood or consistently
applied and/or the nature of roles is changing sufficiently to
create/enhance confusion
[David Bearman, Martin Dillon, Diane Hilman, Mary Larsgaard]

Anti: These are distinctive (enough) and understood (enough) to be used
consistently
[Tony Gill, Peter Graham, Robin Wendler…with a history lesson by Roger
Brisson]


In addition, I see several subsidiary arguments. The first would be germane
if the proposal is essentially endorsed:

3) NAME
If we are to have one element, it should be called something other than
"Agent"
[Liz Parrot]
It would be best to 'retreat' to contributor since it is a name we already
use.
[Dan Brinkley, Alex Satrapa, Stu Weibel]

The next four would be germane if the proposal is essentially rejected:

4) PUBLISHER IS DIFFERENT
Intellectual responsibility is distinct from dissemination, e.g. maybe we
need 14 elements.
P Burnhill,Rebecca Guenther

5) BETTER GUIDELINES
The elements are confusing but better instructions can overcome the problem
[Simon Cox, Jacque Russon, Roland Schwaenzl, Jim Weinheimer]

6) LEAVE IT TO INTERFACE
The user could be confused, but its best to leave the solution to interface
design which may conflate them or not depending on the community they are
trying to serve
[Simon Cox, Rachel Heery, Thomas Hofman, Nigel Ward]

7) WAIT FOR MORE IMPLEMENTATION EXPERIENCE
We don't know if this radical step is necessary yet
[Nancy Morgan, Frank Roos]

I look forward to a vigorous discussion of these issues and to our arriving
at a common position that will strengthen DC and the DC process. We will
take up the question in a plenary discussion on Tuesday and in a break-out
to follow. New positions are, of course, welcome and I will be monitoring
meta2 so that if these are articulated prior to our discussions at the
meeting they will be introduced.

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