Overlay matching should evaluate all repeatable subfields (e.g., every $a in the 019), not just the first occurrence
When a load table's overlay match point targets a repeatable field such as the 019 (OCLC number cross-reference), the load process only compares the first subfield occurrence against Sierra's index. Although Sierra correctly indexes every $a in the 019 individually, the overlay-match routine stops after the first $a. Any additional OCLC numbers stored in the second, third, or subsequent $a subfields are ignored when determining whether an incoming record should overlay an existing one.
As a result, records that should match an existing bib overlay as new duplicates instead, because the matching value lives in a $a other than the first.
Use case:
OCLC regularly merges master records and moves superseded control numbers into the 019 field. A single existing bib may legitimately carry several OCLC numbers — one current number in the 035/001 and one or more prior numbers in repeating 019 $a subfields. When we receive an incoming record whose control number matches any of those values, we want it to overlay the existing record rather than create a duplicate. Today that only works if the match happens to fall on the first 019 $a.
Libraries running lean cataloging operations — including those that have lost cataloger positions and rely more heavily on automated overlay logic — are especially exposed to the duplicate records this creates.
Requested behavior:
When an overlay match point (@ov_tag) references a repeatable field, the load process should iterate through all subfield occurrences of that field and test each against the corresponding Sierra index. A match on any occurrence should trigger overlay, consistent with how Sierra already indexes each occurrence.
Impact if implemented:
Fewer duplicate bib records, less manual dedup/cleanup, and overlay behavior that matches the indexing Sierra already performs — a meaningful gain for libraries with reduced cataloging staff.
-
Kimberly Allen
commented
As noted, many of us are short staffed these days. Depending on the collection, part of our load file prep routine can require significant time searching for possible duplicates, and there are always the ones that sneak through and need post load work. This would be a time saver and contribute to a cleaner database for us and patrons.
-
Gudrun McCollum
commented
This significant enhancement will streamline record management by eliminating the need for labor-intensive manual comparison of the 001 and 019 fields, thereby preventing duplicate record creation.
-
Julia Bullock
commented
This enhancement would save hundreds of hours of staff time when managing e-records. Currently, our cataloger must do extensive manual cleanup when loading e-records, because the system only compares the first instance of the 019 $a, instead of comparing all the data.