Home → Cataloging in NC Cardinal → Bibliographic Cataloging → Quarterly Catalog Update with Authorities
The Marcive updating is a quarterly process that starts on the 15th day of March, June, September, and December.
MOBIUS identifies new and updated bib records for cleanup. These records are sent to Marcive for comparison with Library of Congress authorities. Marcive will try to match those fields containing personal names, corporate names, and subjects (1XX, 6XX, and 7XX) with the contents of LoC authority files. If there is a match, they will send us a copy of the authority file. We then import the original bib record and the new authority file into the catalog so they can be linked.
As bib records finish processing, MOBIUS reimports them constantly, in the background. This ensures the process doesn’t interrupt normal library operations. The reimport overlays any changes catalogers may have made locally while the record was “away.”
The entire process can take a few weeks, but some records may be processed in just a few days. We will notify you as soon as we hear that the process is complete.
When the record is reimported/overlaid, we append a 902 field with the process date in the format MARCIVE YYYYMM. If the 902 field shows the current year-month, that bib record is ready for editing.
Whenever you create or edit a bib record, Cardinal will check it against our authority files. This occurs automatically in a nightly batch process.
All bib records that have 1XX, 6XX and 7XX contents that match authority records in our database will be linked to these authority records. Linked author, contributor, subject heading or title fields get a string like “$0 (CARDINAL)12345” appended. Whenever these authority records change, all linked records will stay up-to-date.
These currently fall through the cracks, because we are only sending Marcive records that we have changed in the last three months every three months.
When authors die, or series headings are created, or new subject headings are adopted, some bibs may linger with old/outdated information in these fields. We cannot get updates for records we do not send out. This means that any records that have not been changed in, for example, two years will not be updated by Marcive. These will have to be fixed manually.
Periodically check your catalog for old/outdated subject headings which might be offensive. (The best way to peruse headings is to the use the Browse the Catalog search function.)
An exception to this occurs when a heading includes a $0(CARDINAL) in its subfields. If a heading is on a record that has not been updated in several years but has the $0(CARDINAL) subfield, when a new authority record is uploaded into Evergreen, that particular heading will be updated, because it was already previously linked to a Cardinal authority record.
Why Are Some Headings Updated and Others Not?
Headings for names, corporations, and subjects must match the authority record exactly in order to be updated. If, for example, a subject heading is misspelled in the catalog, it will not be updated with the authority record, because the spelling does not match. Similarly, if the subject heading is singular (e.g. noncitizen) and the authority record plural (e.g. noncitizens), or vice versa, the subject heading and the authority record will not match. They have to match exactly character for character, however, punctuation such as a period or comma is ignored, so:
noncitizens. or noncitizens, would match noncitizens
Similarly, in the case of problematic subject headings that Marcive replaces with updated, Library of Congress-approved subject headings, if the problematic subject heading (e.g. illegal aliens) in the MARC record does not match exactly how it is transcribed in the authority record (e.g. it is misspelled or is singular), the subject heading will not be updated to the newer noncitizens subject heading.
Additionally, certain subfields will prevent subject headings and authority records from matching. These subfields include $v, $x, and $z.
So, for example:
=650 \0$aIllegal aliens would be updated to =650 \0$aNoncitizens
BUT
=650 \0$aIllegal aliens$vFiction would not be updated to =650 \0$aNoncitizens$vFiction
Other subfields, such as $d, do not prevent these updates from occurring.
For example:
=600 10$aWhite, Betty,$d1922- would be updated to =600 10$aWhite, Betty,$d1922-2021
This is another reason why it is important for catalogers to periodically check their catalog for old or outdated subject headings.
A good rule of thumb might be to check after the quarterly Marcive process has completed, but checking and making corrections/updates as catalogers are performing their regular, day-to-day duties is encouraged, as well. As noted above, the best way to perform searches for headings is to utilize the Browse the Catalog search function. Instructions for utilizing this function can be located here.
For example, browse for author "King, Stephen" and you will see that quite a few results may need altering to better reflect the proper authority heading, which is
$aKing, Stephen,$d1947-
Making changes to that field for one of the results that has no birth date (but you still determine to be by the horror writer) will make the record link overnight and then also get exported with the next Marcive batch.
Marcive would not have assigned any of these records lacking a date as by the author of It. They are all lacking an important field of information (birth date) and there are a dozen authorized Stephen Kings in the Marcive database.
Making manual changes will not update the authority records. However, any changes you save to the bib will mark them for export in the next Marcive process, so older changes can be caught in this way.
Occasionally when Mobius sends a file of MARC records to Marcive, some of the records will have errors that must be corrected. Depending on the error/s, Marcive will either go ahead and correct them or notify us of the errored records that they could not fix and provide examples. These errored records that Marcive cannot fix are deleted from the file that Marcive processes.
The most common causes of these errors are "junk" characters within the MARC records. However, these characters are not always visible, which is what makes them so troublesome. Such characters may include tabs, new line characters, and the like. Specifically, according to Marcive:
...Carriage Return/Line Feed characters or end of field characters that may be needed for the visual representation on the internet, but are not valid MARC, and that data is then in the record that way. This causes errors in the directory, structure, and the format of the record, throwing the indicators and tagging off.
These characters apparently do not cause any issues in our records as they exist in Evergreen, but they are not valid, and they typically occur as a result of copying and pasting data from the Internet directly into a MARC record (think copying a plot summary from an Amazon listing and pasting it into the =520 field of a MARC record when cataloging a book).
In order to try and prevent similar errors from occurring in the future, when catalogers do perform a copy and paste like this, it is recommended that they make sure to select the option to "Paste as Plain Text" (or use the shortcut Ctrl-Shift-V) instead of just "Paste." This will go a long way towards keeping the MARC records clean so they are not rejected by Marcive's process.
Alternatively, if you feel you may have made a mistake somewhere along the way and want to be on the safe side, you can copy the entire record and paste it as plain text using the following shortcuts:
Ctrl-A
Ctrl-C (or X)
Ctrl-Shift-V