Print

Print


PCCLIST readers (c: Gary Smith, OCLC),

This follows Jay Shorten's question and Adam Schiff's response with regard to using VIAF as a source.  I record data from VIAF in field 670, and where appropriate use data such as a year of birth in an access point.  But the following is a cautionary tale received directly from another correspondent, forwarded with permission.

Thanks for your AUTOCAT post on Jan. 29 that included information about Gary Smith’s OCLC Cat post about “bad characters” and their behavior in authority records when loaded into a different system.  I had an authority record that drove me cuckoo for three years because every time we tweaked it and then loaded it, the 100 and the second 670 field would align with the right margin (in Sirsi/Dynix).  I did eventually get the 670 field to align correctly, but never could get the 100 field right (kept focusing on the diacritic).  I reported it to LC, but they couldn’t explain the behavior. ... The authority record in question:  ARN 8911153     Skjöldebrand, Anders Fredrik, |d 1757-1834

The post of mine to which she responded (Citation of ISNI, VIAF, etc. in authority records) can be found in the Autocat archives, and is available here
https://listserv.syr.edu/scripts/wa.exe?A2=ind1601E&L=AUTOCAT&P=R560&1=AUTOCAT&9=A&J=on&d=No+Match%3BMatch%3BMatches&z=4

The email from Gary Smith (Error message regarding "bad character") Tue, Dec 2, 2014 is available in the OCLC-Cat archives, http://listserv.oclc.org/scripts/wa.exe?A2=ind1412a&L=oclc-cat&T=0&F=P&X=79E4F9339A2D3A18B7&Y=ifairclo%40gmu.edu&P=3598

In the record cited (included below), you can identify the problem. Position the cursor on the ǂ (subfield delimiter) before the dates, then use your left-arrow key four times.  Notice how, on the second and fourth times, the cursor stays put.  It's actually "moving" across the hidden characters.

VIAF records are literally riddled with these characters.  Perhaps someone can explain the technical requirement for them in that database, if any.  It appears that in the NAR she cites (included below), the cataloger copied the data directly into both field 670 and 100.  If your system accepts the characters without notification, you'll never know about the problem.  But other folks will.  Furthermore, OCLC Connexion's Edit-- Marc-8 characters-- Verify function doesn't identify them.

Sincerely - Ian

Ian Fairclough
Cataloging and Metadata Services Librarian
George Mason University Libraries
[log in to unmask]

010  n 2011046463
040  DLC ǂb eng ǂc DLC
1001 Skjöldebrand, Anders Fredrik,‏ ‎ ǂd 1757-1834
670  Hermann von Unna, 1800: ǂb (name not given)
670  VIAF, July 11, 2011 ǂb (Skjöldebrand, Anders Fredrik,‏ ‎1757-1834‏)