Authority Control Advisory Committee

Authority Control Advisory Committee Minutes

Date: 2007-01-16

Present: Ellen Ellickson, Tachtorn (Wheat) Meier, Jenn Nolte, Rick Sarcia, Michael Scott, Eileen Smith, Karen Spicher, Manon Théroux (chair and recorder).

Absent: Jerry Anne Dickel.

1. Membership

Karen, Eileen, and Ellen have renewed for another term. Eric has rotated off of the committee. Wheat has joined the committee. She is a catalog librarian in the History and Social Science Team in the SML Catalog Department.

2. Authority Control Issues Relating to the Voyager 6.1 Upgrade

NEW FEATURE: Subject subdivisions now display as they should (preceded by double dashes) in results of "Subject" searches in the OPAC and "Staff Subject Heading" and "Staff Subject Subdivision" searches in the cataloging client. Unfortunately, the double dashes do NOT display in the following subject searches: "Special Collections Subject" (OPAC and client); "Subject: Genre" (client); "Subjects" (client); and "Names" (client). Endeavor Knowledge Base incident number: #128664.

NEW BUG: In the authority validation box, the "Search" button does not work properly if the heading contains diacritics. The diacritics are translated into nonsense characters in the search string and the search thus does not bring you to the correct place in the index. Endeavor Knowledge Base incident number: #123366.

OLD BUG: The authority duplicate detection window still does not properly display headings with diacritics. Sometimes the diacritics display as nonsense characters, sometimes they actually cause the heading to appear truncated. The display problem does NOT affect the merge action, however. Endeavor Knowledge Base incident number: #97895.

The committee may make these the topics of authority control "tips" on YULCAT-L.

3. ACAC Testing of MARS Series Processing Fixes

Manon summarized the committee's testing last fall of fixes being proposed by MARS to their series heading processing routines. The biggest fix involved series with initial articles (e.g. MARS has always changed 440s to 490/8xx combinations when the only difference is an initial article; the new fixes would allow the series to remain as 440s). MARS was able to use our test results to make further enhancements to their fixes. Manon will ask whether the fixes have been implemented yet in production processing.

4. Non-Filing Indicator Clean-Up Project

Manon reported on a MARS clean-up project completed in December 2006. A small subset of Orbis bibliographic records needed corrections to their non-filing indicators because of the change in non-filing indicator practice implemented a couple of years ago in preparation for the move to a Unicode environment. In words where the first filing letter has a diacritic, the diacritic used to precede letter and thus was counted; in the Unicode environment, the diacritic follows letter and thus is not counted. Manon and Steven Arakawa identified the tags that used non-filing indicators. Yue Jin in ILTS identified the records needing clean-up by writing a program that looked for records with these tags that had the non-filing indicator(s) equal to something other than "0" and the first "filing characters" equal to a diacritic character. Records in English and JACKPHY languages were excluded because English does not use diacritics, CJK languages do not have initial articles, and all APHY records were recently processed by MARS when we did RLG snapshots (and thus would have been corrected by MARS at that time).

Before Unicode conversion:
245 15 Los [acute]ultimos
After conversion (before clean-up):
245 15 Los u[acute]ltimos
After clean-up:
245 14 Los u[acute]ltimos

The program identified 3113 records that needed clean-up. These records were extracted and sent for processing by MARS. Staff working in the cataloging client were asked to observe a 1-day mini record freeze on the records until the reload of the processed records was complete. Afterwards, Jenn sent Manon a report on the nonfiling indicators in the 245 field (before and after views). Manon identified 352 records that needed additional manual clean-up. The manual fixes largely broke down into four categories, as follows:

a). Records that had a diacritic where there should have been an apostrophe. There were a lot of these, many being recon records but many also created by staff who had incorrectly used a diacritic in place of an apostrophe (the problem records mostly dated from the early 1990s). Example (before and after manual fix):
245 13 La[acute]rt ...
245 12 L'art ...

b). Records that had nonfiling indicators that should have been "0" because there was no initial article. These seemed to result when staff incorrectly identified a word as an initial article. Example (before and after manual fix):
245 14 Il e[acute]tait une fois ...
245 10 Il e[acute]tait une fois ...

c). Records that had the title page in a language other than the primary language of the text (MARS uses the language code in the fixed field when attempting to determine whether or not the initial word is an article and the fixed field corresponds to the language of the text). Example (before and after manual fix):
245 15 The S[diacritic]rauta ...
245 14 The S[diacritic]rauta ...
[title page in English, fixed field code and text in Sanskrit]

d). Records without an initial article but with an initial title letter starting with a diacritic. Manon will report this to MARS. Example (before and after manual fix):
245 11 U[umlaut]bersinnlich
245 10 U[umlaut]bersinnlich

Jenn is currently working on a similar report for the 440 field.

5. Authority Control Workflow for C&T Staff

Manon announced that this document will be updated to remove references to Fastcat. Because of the LC decision to stop tracing series, the Fastcat workflow has changed. Fastcat staff are no longer required to verify every series and change to 490-1/8xx if needed. Instead, they are simply changing the 490-0 to 440 using a macro (and setting the non-filing indicators manually). We are relying on MARS to correct the 440 series as needed (although we know there are some categories that MARS will not be able to fix, most notably series with qualifiers).

6. Series Workshop

Manon reminded the committee that a couple of years ago they had almost completed preparations for a hands-on series workshop for cataloging assistants before the project got derailed for various reasons. It would be good to try to resurrect the project. Upcoming vacations by several of the committee members involved means February might not be a good time to revisit this but perhaps March.

Site URL:
Comments to: Manon Théroux  <>

© 2007 Yale University Library

Top of Page