[Web4lib] Authors name standardization. Your opinion?

Diane I. Hillmann dih1 at cornell.edu
Tue Nov 14 18:54:02 EST 2006

>Mike Taylor wrote:
>>Oh.  That seems like a _bad_ thing to me in the case of author
>>identities.  I wonder whether you could suggest a couple of scenarios
>>that illustrate the kind of functionality that you imagine this extra
>>flexibility provides?

Karen replied:

>I was thinking of the LCCN = bib record, not authority. In terms of 
>authority... well, I don't know what kind of system we'd be using 
>other than the LC authorities, but in essence, I think we have to 
>assume that the records get distributed to many systems, and that 
>you may have a query that is specific to that system. That doesn't 
>mean that the author's identity changes, but different information 
>may be available from different sources.

Diane sez:

I think it's pretty clear that the days when 'name authority' meant 
just LC NAF are numbered. Other sources of name data are already 
available, with different functionality.  There are useful areas of 
information where we haven't traditionally supplied data where 
libraries may want to make an attempt to supply the data, or may want 
to consider other sources (affiliation comes to mind). It's 
interesting to note that some metadata formats (ONIX for instance) 
actually carry the biographical data along with the description of 
the book, because it's more convenient to do that for the online 

>  So I can imagine being in a local catalog, that has both bib and 
>authority records, and wanting to see the authority info linked from 
>an author on your screen. Assuming we have an identifier 
>(lccn:xxxxx), you might offer the user the local authority record 
>(which could have some added info if the library's community has a 
>particular interest in that author, say a professor at the 
>institution), the LC database record (neutral, but authoritative), 
>or the Wikipedia entry (which would be in a narrative form), if wp 
>is accessible through the same identifier string.

Yes, I think this is really where people want to go--to be able to 
use other than the traditional sources about authors and others 
associated with resources. I think the critical piece is 
identification, as Karen notes. Probably the most important part of 
identification in this realm is just what we're identifying. 
Traditional name authority records use the preferred name as the 
identifier--the record itself has an ID as well, but it identifies 
the record, not the person, so just blythely using the record 
identifier as if it were the identifier for the person isn't going to 
get us very far. In order to be able to use multiple sources of 
information about people, we're going to have to take a leap and 
start identifying people with something like a URI, and that's a 
pretty interesting problem, to say the least

>It seems that you are assuming that there is one central place for 
>author info. I guess I've given up on "one central place" for any 
>public info.


>I'll tell you what it means to me: distributed data across the web 
>(read: network) that is accessed and reused in many ways, often 
>using web services. Whatever we decide to call it (and I truly am 
>neutral on its naming), the era of separate databases, data records 
>staying put and unchanging, single "definitive" sources for 
>information -- that's gone. This "flexibility" seems too chaotic to 
>some folks, but having worked in the world of libraries for so many 
>years I have come to live with a rather large measure of ambiguity. 
>There are things that will defy the desire to create a perfect 
>solution; author identity is one of those.

I'm with you there, Karen.  If we're not willing to cope with a bit 
more chaos for the promise of a lot more functionality, we will go 
the way of the dinosaurs.


>Karen Coyle / Digital Library Consultant
>kcoyle at kcoyle.net http://www.kcoyle.net
>ph.: 510-540-7596
>fx.: 510-848-3913
>mo.: 510-435-8234

Diane I. Hillmann
Research Librarian
Cornell University Library
Email: dih1 at cornell.edu
Voice: (607) 387-9207

More information about the Web4lib mailing list