I'm with Deborah Leslie--the practice of establishing undifferentiated
name authorities should be eliminated.
One of the greatest efficiencies offered by name authorities is the
possibility of linking an authority to its corresponding bib headings as
OCLC now does through its controlled headings feature. This should
enable WorldCat to automatically and reliably update a bib heading as
the authorized heading form changes. However, current LC rules make this
kind of linking unreliable, because they make the link between an LCCN
and a particular identity unstable. Consider this scenario: the name
"Smith, John" is established to represent person X. A second Smith, John
appears (person Y), and person X's authority is modified to represent
both as an undifferentiated authority. Then qualifying information is
found for person X's heading, and person X's identity is moved to a new
authority. The original authority now represents only person Y, and by
LC rules is recoded as being differentiated again. In effect, the
referent of the authority record's LCCN has become uncertain, and any
links based on the record are now prone to error. In general, LC
considers the re-use of an LCCN to be forbidden; but in the case of
undifferentiated personal name authorities, they essentially require
such re-use. Not a good thing.
Our systems are not good at helping us choose which authority is the
right authority for common names, and I agree with Ted that if all we
have to go by is an alphabetical list of candidate headings, then having
to check all the candidate headings is a task made more difficult by
scattering the candidates through the list based on multiple types of
qualifying data. But to me that's less of a problem than having the work
of identifying a resource with a person be undone by letting the meaning
of the authority become unstable. Systems should be able to make
choosing from a list of headings easier without having to open up each
record one by one to see the additional differentiating data contained
in 670s, etc.
In most cases, the reason undifferentiated authorities are used is NOT
because anyone believes the two or more persons cited are the same--it's
just because there's not enough information available to differentiate
their headings. Differentiating entities is the primary task that
authorities perform. Anytime a cataloger can make a judgment that two
persons are different entities, the cataloger should be able to
establish them with differentiated authorities. In my opinion, we'd have
fewer problems on the heading management side if we enabled universal
differentiation and required catalogers to make a best judgment effort
to establish only differentiated personal name authorities. We already
have routines for merging authorities when duplicates are discovered,
and the cases where two persons are genuinely established as one (i.e.,
NOT as an undifferentiated authority) are few and far between, and much
less of a problem on balance than is dealing with the undifferentiated
personal name authorities we have now.
Sorry, but this topic really pushes my buttons.
Stephen
Ted P Gemberling wrote:
> I don't understand the aversion to undifferentiated headings. They essentially provide a place for careful catalogers to record and share information about author identities, in hope of eventually creating differentiated headings for them. They basically mean, "we don't know enough about these people yet." If we differentiate headings right away, don't we risk doing it badly, and creating a mess that will be harder to clean up later? If there's a plethora of authors with |c qualifications related to subject or occupation, it seems it could be a big mess, with a lot of potential bib file maintenance if the differentiations are found to be incorrect.
>
> I would say the same could be true of headings qualified with LCCNs (for the authorities?), as Stephen mentioned. If the authorities were set up with uncertain information in the first place, they wouldn't differentiate the names very well, and it would be very difficult to sort things out if you had to look at multiple authority records and bibs instead of just one undifferentiated authority record.
>
> Just my two cents. I know I'm not saying anything new.
>
> I think Mary's examples we very good.
>
> Ted Gemberling
> UAB Lister Hill Library
> (205)934-2461
>
> -----Original Message-----
> From: Program for Cooperative Cataloging [mailto:[log in to unmask]] On Behalf Of Deborah J. Leslie
> Sent: Wednesday, June 02, 2010 9:36 AM
> To: [log in to unmask]
> Subject: Re: [PCCLIST] a NACO question
>
> I am in complete agreement with Richard, and pronounce anathema on
> undifferentiated name headings.
> _________________________
> Deborah J. Leslie, M.A., M.L.S.
> RBMS Chair 2009-2010 | Head of Cataloging, Folger Shakespeare Library
> 201 East Capitol St., S.E. | Washington, D.C. 20003 | 202.675-0369
> [log in to unmask] | http://www.folger.edu
>
>
>
> -----Original Message-----
> From: Program for Cooperative Cataloging [mailto:[log in to unmask]] On
> Behalf Of Moore, Richard
> Sent: Wednesday, 02 June, 2010 03:41
> To: [log in to unmask]
> Subject: Re: [PCCLIST] a NACO question
>
> I tend to agree with John on this. Identifying authors by means of their
> headings is useful, but differentiation and collocation are of primary
> importance in retrieval, in my book (so to speak). If the rules allowed,
> I'd qualify someone by their shoe size, rather than undifferentiate.
>
> Cheers
> Richard
>
> _________________________
> Richard Moore
> Authority Control Team Manager
> The British Library
>
> Tel.: +44 (0)1937 546806
> E-mail: [log in to unmask]
>
>
>
--
Stephen Hearn
Metadata Strategist, Technical Services Dept.
University Libraries, University of Minnesota
160 Wilson Library
309 19th Avenue South
Minneapolis, MN 55455
Ph: 612-625-2328 / Fax: 612-625-3428
|