What we need is for systems to start providing search capabilities for authority data, not just as it occurs in bib records, as an AAP, but as it appears in authority records, including variant terms, facet information, etc. If I search now on "central banks" in id.loc.gov's database of LCSH authorities, I get a result set of brief records, not a browse list of AAPs and VAPs. The brief records include the AAP as a main term, but also include VAPs as part of each authority's brief display, and could include other selected metadata elements from the full record. In the facet list there are options to limit the search by entity type, by source of term, etc. In this instance, authority data about an entity is treated in manner comparable to bib data about a resource for searching and navigation. So, a user's terms can lead in a search environment to discovery of the authority for the desired entity, which in turn could lead to accessing bib resources associated wtih that entity, all without recourse to a browse index.

I'm not claiming that id.loc.gov is a model for this sort of functionality, but it offers a better illustration of the way we need to frame our questions and propose new solutions that is more forward-looking than an approach which focuses on refining browse indexing. I love browse indexing and see lots of ongoing value in maintaining it. But given the current state and likely future of discovery, our focus should be on how to make authorities better targets and performers in environments based on search, not browse.



On Fri, Dec 21, 2018 at 2:04 PM Yang Wang <[log in to unmask]> wrote:

Yes, John, it was discussed on this forum at least earlier this year regarding “Time.” I was unable to determine in our Voyager catalog how many monographs/articles had been written on the magazine. Now, to complicate the situation further, some institutions that have started adding $0 to NAF-LCSH headings in their bibs run the risk of convoluting the two. I caught quite a few instances like this:


630 00 Time. $0 http://id.loc.gov/authorities/subjects/sh85135395


Is this the kind of sacrifice we are willing to make?


Adam, you are quite right! [Browsing is apparently a relic of the past, hidden away and not optimized for utility. Authority records apparently play no part.]. But where lies our responsibility to library users? Force them to do keyword searching?


Without authority file/controlled vocabulary, why is “Browse” even listed and given as a method of searching? If a patron only knows a variant form of an established name, s/he keys in the variant name but gets no help from the system—no results?  On the role of central banks in the world today, library users eagerly type in “Central banks” in subject browse box, hoping to get a list of central banks from different or some particular countries. But, alas, only a few “hits.” No help from the system to redirect users to “Banks and banking, central”? There are literally thousands, subdivided by countries and topical themes.


Enough of my “ranting.” I apologize for bringing up the subject again.


Best regards,




From: Program for Cooperative Cataloging [mailto:[log in to unmask]] On Behalf Of Hostage, John
Sent: Friday, December 21, 2018 2:10 PM
To: [log in to unmask]
Subject: Re: [PCCLIST] LCSH/NAF cross-references


Wasn't the question of whether access points in the name authority file should be checked for conflicts against LCSH discussed on this list some months ago?  As far as I can remember, no conclusion was reached; at least, no rule to prevent conflicts between the two files was identified.  I would guess that very many one-word subject headings would run the danger of conflicting with names or titles.



John Hostage

Senior Continuing Resources Cataloger

Harvard Library--Information and Technical Services

Langdell Hall 194

Harvard Law School Library

Cambridge, MA 02138

+(1)(617) 495-3974 (voice)

+(1)(617) 496-4409 (fax)
0000 0000 4028 0917


From: Program for Cooperative Cataloging <[log in to unmask]> on behalf of Yang Wang <[log in to unmask]>
Sent: Friday, December 21, 2018 13:27
To: [log in to unmask]
Subject: Re: [PCCLIST] LCSH/NAF cross-references


Yes, someone should have a long time ago (2006? When Voyager was merged with Ex-Libris?). Regardless, I think that catalogers should be making an effort to prevent this from happening in the future. I know we may have URIs to rely on in the future:


Time http://id.loc.gov/authorities/names/n94056291 = Time (magazine)

Time http://id.loc.gov/authorities/subjects/sh85135395 = Time (concept)


Machines can sort them out easily. But for library users, it difficult to differentiate which is which, when they brow subject headings. Certainly Time magazine can be used as a subject. My point is this: without seeing the leading MARC codes (130 vs 150, etc.), how can we expect library users to look for what they want?


I just did a subject browse search in the University of Washington Library online catalog. Under the heading Labor, I could not even see narrower terms. Is that normal? No cross-references of any kind.


Try the term “Epistemology,”  let’s see if your library users will be re-directed to the LCSH term: Knowledge, Theory of. Or, try “Augustine, Saint,” do they get NAF record: Augustine, ‡c Saint, Archbishop of Canterbury, ‡d -604?


BTW, PUL OPAC is in a similar situation. So, no offense to UW.







From: Program for Cooperative Cataloging [mailto:[log in to unmask]] On Behalf Of Adam L. Schiff
Sent: Friday, December 21, 2018 12:52 PM
To: [log in to unmask]
Subject: Re: [PCCLIST] LCSH/NAF cross-references


Complain to Ex Libris??

From: Program for Cooperative Cataloging <[log in to unmask]> on behalf of Yang Wang <[log in to unmask]>
Sent: Friday, December 21, 2018 6:10:32 AM
To: [log in to unmask]
Subject: LCSH/NAF cross-references


In our Voyager system, I noticed this problem a while ago. The oddity is not new, but I think it’s worth mentioning from time to time. An LC subject search under Labor in Voyager gives me this screen:



When a heading in NAF (be it a personal name, corporate body, geographic name, series/serial/uniform title heading, etc.) is an exact match of an LCSH term, it seems to me that adding a qualifier would suffice to avoid such embarrassing cross-references:   


430 #0 Labor (Instituto de Formação Social e do Trabalho (Portugal))

430 #0 Labor (Milan, Italy)

410 #0 LABOR (Universidade Federal do Ceará. Laboratório de Estudos do Trabalho e Qualificação  Profissional)


Maybe it’s just PUL’s local Voyager that has this problem? Any comments or suggestions?


Best regards,



Stephen Hearn, Metadata Strategist
Data Management & Access, University Libraries
University of Minnesota
170A Wilson Library (office)
160 Wilson Library (mail)
309 19th Avenue South
Minneapolis, MN 55455
Ph: 612-625-2328
Fx: 612-625-3428
ORCID:  0000-0002-3590-1242