Skip to main content
  • Subscribe by RSS
  • Ex Libris Knowledge Center

    Question about Authority records duplicate fields

    • Article Type: General
    • Product: Aleph
    • Product Version: 15.2

    We have catalogers who have insisted on using parallel fields in AUT records. They have done this since implementation for adding non-Roman script fields to the records.

    For example, an AUT record will exist with a Latin term. The cataloger will add a non-Roman script term, but instead of adding it in a 400 field which is standard practice, they will add it in another 100 field. We've managed to escape major repercussions except for having many ambiguous headings, but then last week, we had a situation where many 100 terms in bib records were flipped to the Hebrew term. Of course, this Partner freaked out, and I've been trying to convince them that they should just catalog according to standards and put the vernacular script into a 400.

    I have told them that using parallel fields in bib records is fine, that we can run the redo 880 fix for records we export from the system, etc. However, using two 100 fields (one for Latin, one for non-Roman script) in an AUT record renders headings ambiguous and confuses the system...and isn't standard practice.

    I've told them that putting parallel 100 fields in AUT records triggers actions, whereas using them in bib records does not. There are some AUT records with two 100 fields that don't cause the terms in the bibs to flip to Hebrew, and others that do. So I can't discern a pattern, however, it's screwy and I'm sure the system is confused. What is standard Aleph practice?

    It seems to me that there are three options:

    1. They can ontinue to use two 100 fields in AUT records and risk unwarranted flipping - lessen these chances by always using parallel fields in the bib record for terms linked to the respective AUT records with two 100 fields.

    2. They can put only one authorized term in a 100 and put the vernacular term in a 400, as is standard LC practice

    3. They can create two AUT records - one for the Latin term (to go in 100) with the vernacular term in 500, and one for the vernacular term (to go in 100) with the Latin term in 500 field. Link the Latin terms in the bibs to the Latin term AUT record, and the vernacular terms in the bibs to the vernacular term AUT record."

    What does EXL recommend to customers? Can you clarify that our system would not be set up to have duplicate 100 fields in AUT records? We get an error message when we save an AUT record with two 100 fields - this has been disabled for our bib records but not for our AUT records, which I assume means that EXL did not intend for us to use duplicate 100 fields in AUT records.

    This is high priority because of a flipping problem that I will report in a separate SI. We just need to know what you recommend doing within the Aleph context asap so we can tell our catalogers and get this sorted out.

    We strongly advise agains the use of paired 1xx fields in the authority record. The 1xx represents the single established form of a heading. To have two established forms represents a conflict in logic and can lead to unpredictable results. This can cause data corruption if you have coded the authority record to automatically update bibliographic headings.

    There are several ways the non-Roman form can be represented. It can be in a separate authority record altogether. This appears to be the most common practice among ALEPH libraries that have a single authority database with both Roman and non-Roman forms.

    It can be coded in a 5xx field, which is the most common practice among US ALEPH libraries. It allows the non-Roman form to be present without risking unanticipated heading correction in Bib records.

    The MARC21 standard itself calls for use of the 880 field for representing these alternate forms in Authority records.

    Finally, there is the possibility of using the 4xx field. This has generally been avoided among US ALEPH libraries. The 4xx represents the non-preferred form of the heading, and this doesn't actually represent the nature of the non-Roman form. However, beginning in June 2008, OCLC began distributing Authority records from the NACO project with the non-Roman form in a 4xx. This was deemed inappropriate by several ALEPH libraries, who approached OCLC and asked that this non-standard use of the 4xx be somehow flagged so that ALEPH could treat the records a a separate case. OCLC declined to do so. Consequently, Ex Libris, with input from the participating ALEPH libraries, modified the ALEPH software to insure that unexpected heading changes wouldn't occur from these NACO records. The rep_change is included in the following versions of ALEPH (and higher):

    V.16 - rpc #2166
    V.18 - rpc #1729
    V.19 - rpc #332

    Here are the details of the rep_change:

    Description: A new parameter has been implemented for the fix_doc_ref_1 program. This program is responsible for the update of the bibliographic record from the authority linked record. The new parameter enables Aleph customers that desire to "turn-off" the automatic authorities update mechanism in cases where the bibliographic heading is a non-Latin heading and the authority heading derives from a Latin authorized heading (or the opposite). In other words, the link to the authority record is created but if there is a difference in the scripts (Latin vs. non-Latin), then the update of the bibliographic record does not take place. This enhancement has been performed due to an announcement made by The Library of Congress on the addition of non-Latin script (such as Arabic, Cyrillic and Hebrew) to the LC/NACO name authority records distributed. These references will be under the 4XX reference fields. The Romanized form will continue to be the authorized heading (authority record 1XX field). Distribution of the new records is planned for mid-July.

    • Article last edited: 10/8/2013