Date: 2001-08-20Present: Manon Théroux (chair and recorder), Steven Arakawa, Ellen Ellickson, David Faulds, Robert Killheffer, Beatrice Luh, Earl Roy, Dajin Sun.
Absent: Mickey Koth, Cécile Mandour, Audrey Novak, Rick Sarcia.
I. Brief MARS Status Report
--Basefile1 correction transactions still not loaded into production Orbis.
--Recon bib records (Nov-May) loaded into ORBDEV7; will QA this week
--Auth records (various files) loaded into ORBDEV7; will QA this week
--Basefle2 corrections transactions received - will be QAed after recon and auth records
--More detailed information is available on the committee's web page.
II. Web Page Status Report <http://www.library.yale.edu/cataloging/authorities/acac.htm>
--Committee minutes now available on committee web page
--Added new page on how to get into ORBDEV7 test region for new members
III. QA of Recon Records
--To make sure that OCLC made necessary corrections to our recon record headings
--To make sure that OCLC sent us the authority records that correspond to these headings
We will not concern ourselves with the quality of the descriptive fields.
--Each person should look at 100 random bib records in his/her key range (see below) by end of next Monday, Aug. 27th.
--Use the Excel spreadsheet (sent as an attachment) to record QA results; complete the form electronically if possible; anyone who is not familiar with Excel may complete it by hand (or this might be a good time to start learning!).
--Use two ORBDEV7 sessions and one ORBIS session
--Pull up recon bib records in one ORBDEV7 session
--Search all headings in the other ORBDEV7 session (looking for an auth record); use jx search because want to check that hdg use codes are set correctly in auth records
--If you find an auth record in ORBDEV7, check to make sure that the bib record hdg matches the auth record hdg; also check to make sure that the hdg use codes are set correctly (*see below for caveat).
--Keep in mind that search results in the current NOTIS release are ordered as follows: 1) hdg used as 6xx; 2) hdg used as 1xx, 7xx, 8xx; and 3) hdg used as 6xx with subdivisions. To find out if hdg has been used as a "name" access point, you may have to scroll through many screens of 6xx hdgs.
--Also keep in mind that for 6xx hdgs with subdivisions, an auth record for the entire string will not always exist. This is because many subdivisions are "free-floating". If you suspect that is the case with your hdg, search instead on the form of hdg that precedes the subdivision (e.g. Do not expect to find an auth record for "Artists--Biography" but do expect to find one for "Artists")
--If you don't find an auth record in ORBDEV7, look for one in ORBIS LTYL; OCLC may have already sent us the auth record as part of the "new auth" file that we loaded to ORBIS in June. If you find an auth record in ORBIS LTYL, determine whether it's an A, N, or Y range record; check to make sure that the recon record hdg matches the auth record hdg; also check to make sure that the hdg use codes are set correctly (*see below for caveat)
--If you don't find an auth record in either ORBDEV7 or ORBIS LTYL, look for one in ORBIS LTLC. If you find one, it probably means that OCLC *should* have sent it to us. Also check to see if the recon record hdg matches the auth record hdg. If it doesn't, it probably means that OCLC *should* have corrected it.
*Here are possible complications with the hdg use code part of the QA:
The auth records loaded into ORBDEV7 correspond not only to hdgs in our recon bib records but also to hdgs in our basefile bib records in ORBIS. The bib records in ORBDEV7 include our recon records, our basefile records, and most of our post-basefile records. So, if the hdg use codes in an auth record are not right, it might be because we are pulling up post-basefile bib records in our jx hdg search; OCLC hasn't yet processed these bib records, so we cannot expect the hdg use codes to correspond to them. It also might be because the basefile correction transactions haven't been loaded yet. We might be missing some bib records when we do a jx hdg search because they still contain variant (uncorrected) forms of the hdg. Although the hdg use code part of our QA cannot be perfect, it should be good enough. Just do the best you can to figure out what is going on.
Division of Records:
DEW9960 - DEZ2834 (Fixed and Keyed Records - 19,488)
all - Manon
DEZ2835-DFK1971 (Retro Records, pt. 1 - 77,589)
DEZ2835-DFC9999 - Steve
DFD0000-DFG9999 - Ellen
DFH0000-DFK1971 - David (no DFI)
DFK1975-DGJ5513 (Retro Records, pt. 2 - 187,573)
DFK1975-DFN9999 - Robert (no DFO)
DFP0000-DFS9999 - Mickey
DFT000-DFW9999 - Beatrice
DFX0000-DGA9999 - Cecile
DGB0000-DGE9999 - Earl
DGF0000-DGJ5513 - Dajin
Rick is on vacation this week and is excused from this QA; Audrey is always excused from QA--she coordinates our loads instead.
IV. Next Meeting
This meeting is a replacement for the Aug. 27 meeting, which is cancelled. Our next scheduled meeting is Sept. 24, but we may need to meet before then if Basefile2 records are ready for QA. We should keep our 10-12 time slot open on all Mondays, just in case, if possible.
Site URL: http://www.library.yale.edu/cataloging/authorities/acac.htm
Comments to: Manon Théroux <firstname.lastname@example.org>
© 2001 Yale University Library