Near the end of October, I added to all of our loaders for vendor
bibliographic records for current acquisitions some new features
related to the recent switch in LC's tracing policy. Specially, the
loaders now attempt to resolve 490 fields with first indicator zero
at the time the records arrive in the local database. 'Resolution'
in this case means: match the 490 field with an authority record; and
then, if the authority record says that the series should be traced,
either re-tag the series as 440, or as 490:1 with an 8XX, depending.
Although I intend to collect information over a longer period, I
thought you all might be interested in the results from the first two
weeks. The numbers are derived from records received from 5 or 6
domestic and foreign vendors (depending on how you count). I've
collapsed numbers for all of the vendors into one set of numbers for
simplicity in this brief note. (Some of the numbers for individual
vendors are very small, and wouldn't yet serve to make a useful distinction.)
1138 bibliographic records processed
256 440 fields present in records received
128 490:0 fields present in records received (see below)
49 490:1 fields present in records received
50 8XX fields present in records received
Of the 128 490:0 fields in the incoming records:
106 matched to an authority record: became a traced series
2 matched to an authority record: remains a not-traced series
20 problems reported for operator resolution (see below)
Of the 20 490:0 fields reported for operator resolution:
3 new 430 field added to existing authority record
3 new 643 field added to existing authority record (the
program requires at least an approximate match on publisher)
5 series changed to traced; new authority record created
3 series changed to traced, no new authority record created
(see below)
2 problem is not with heading, but only with form of
numbering (not resolved because item not in hand; does not affect
primary access)
2 deleted the 490 field (examples below)
1 series changed to traced to match authority record;
inserted $v code before the numbering
1 duplicate added volume to serial detected because of
series; provisional bib record eventually deleted
Here are the cases where I deleted the 490 field; these both occur in
'provisional' records from one vendor:
245 BLUE GUIDE : ROME
250 9TH ED.
490 BLUE GUIDE: ROME.9ED
245 BLUE GUIDE : GREECE:: THE MAINLAND
250 7TH ED
490 BLUE GUIDE: GREECE.7ED
Of the 3 series changed to traced without a new authority record: in
two cases, the program reported multiple matches (similar titles,
similar publishers) and it was a matter of picking the correct
one. In one case, it was not possible to figure out the correct
series without reference to the item, but I converted the series to
traced because the database already contained a handful of
bibliographic records with the same problem, and it seemed best to
put the new record with the others and let someone else deal with the
whole bunch at once.
At present, I'm not contributing the new authority records, because
they're created without item in hand, but that may change in the
future if we redesign the workflow. Same for the modifications to
existing records.
Although I don't think that separating out records by vendor is yet
meaningful because of the small numbers involved, there is one bit of
information I would like to tease out of the data for a closer
look. Some of the records from a particular vendor are provisional
or temporary records intended to be overlaid later by full records;
by definition, these provisional records have all series tagged as
490:0. Indeed, 115 of the 128 490:0 fields in this small sample
occur in such records; and the program was able automatically to
resolve 101 of these 115. These 490:0 fields are I believe similar
to the 490:0 fields we may expect to find in bibliographic records
created by institutions that follow the LC tracing decision: they are
likely to have been transcribed without reference to an authority
record. And yet the program doesn't have any more trouble with them
than with other 490:0 fields.
I suppose the conclusion to be drawn is that with appropriate
software support, deciding to continue to trace series isn't going to
make that much of an impact on staff. Five additional authority
records in two weeks' time isn't a crushing burden.
Gary L. Strawn, Authorities Librarian, etc.
Northwestern University, 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.
|