Dear Gary,

   Since the non-Latin 400 fields were taken from bibliographic  
records in WorldCat, would it be possible to do a reversed matching  
project? For example, run the non-Latin names against 880 fields that  
are linked to 100/600/700 in bibliographic records, then construct the  
citations for the new NARs from the matching records—245 $a/$c and 260  
$c (or Date 1 from 008). The 100 fields in the new authority records  
can be taken from the original NARs. Records created via this method  
should be marked, so that manual review can be done if necessary.

   You are correct in saying that 670 fields were not converted when  
the romanization was switched from Wade-Giles to Pinyin for Chinese  
names. As a result, the authorized headings often do not match the  
forms found in 670 fields. It might be advisable to take this  
opportunity to convert 670 fields to pinyin, if possible. (However,  
the mixture of Wade-Giles and Pinyin data in 670 fields right now  
might make it difficult, if not impossible, to do so.)

   Thank you very much.



   Sarah S. Elman
Head of Technical Services
C.V. Starr East Asian Library
Columbia University
307 Kent Hall
1140 Amsterdam Ave.
New York, NY 10027
Phone: (212) 854-2579
Fax: (212) 662-6286Quoting Gary L Strawn <[log in to unmask]>:

> Here's another interesting bit that may be of interest. By my count   
> there are 3848 undifferentiated name records that contain   
> non-vernacular 400 fields in them (and not a combination of   
> vernacular and non-vernacular).  Should it be found proper   
> automatically to tease undifferentiated records apart, in all but   
> 478 cases all of these non-vernacular 400 fields can be matched up   
> with the texts of 670 fields, and so (in theory) could be sorted  
> out  to the appropriate record(s).
> As an earlier report noted, almost none of the vernacular 400  
> fields  can be found in 670 fields because they were created from   
> information in bibliographic records and are not "justified" in   
> authority records. One suggestion for dealing with these might be  
> to  attempt to find bibliographic records containing vernacular  
> fields  whose 245s match titles in the 670s. Any such work would be   
> complicated by the switch in romanization schemes (which I think  
> was  not applied to authority 670 fields).  Other than that, I don't  
> see  much hope for any automatic handling of vernacular 400 fields   
> (either keep 'em all in each split-apart record, or discard 'em   
> all), though I would be happy to hear of other ideas.  There are by   
> my count 13,386 records at present with vernacular 400 fields in them.
> Gary L. Strawn, Authorities Librarian, etc.
> Northwestern University Library, 1970 Campus Drive, Evanston IL 60208-2300
> e-mail: [log in to unmask]   voice: 847/491-2788   fax: 847/491-8306
> Forsan et haec olim meminisse iuvabit.         BatchCat version: 2007.22.416