Use of (Brief) Order Records in NZ/IZ

See Best Practices for Decision Pages and Tags for groups

Legend: not started IN PROGRESS STALLED decided

Recommendation

The Acquisitions Operations Subteam recommends that all order records (especially brief records) be ingested to the IZ, and only linked to the NZ after being updated with a full record. This includes records containing OCLC #s. We are discussing primarily GOBI in this data analysis, but recognize that other vendors provide data, sometimes including OCLC #s, in brief records supplied in mrc, spreadsheet, or other formats. For campuses that are manually downloading OCLC records to import for materials received, it has been the experience of larger campuses like UCLA that keeping them in the IZ is still the safer option so catalogers can make adjustments like switching between set and stand-alone record without having to make adjustments in the NZ.

Impact

Reasoning

GOBI identified several ways in which their data was “dirty” and could create complications if order records were loaded directly into the NZ for later Update Inventory loads. Several campuses are already using regular “Link to NZ” jobs for materials loaded into the IZ to ensure proper cataloging before linking to the NZ. Given the difficulties involved in correcting errors in the NZ, especially for cases where entire records are erroneously overlaid or items of multiple titles incorrectly wind up on a single record, it was determined that the routine use of the “Link to NZ” jobs would be less trouble than tracking down and untangling mismatch and overlay problems.

Background

The Phase 4 decision RMFG (Go-live and beyond) OCLC numbers in UC Network Zone bibliographic records references “on order” brief bib records may eventually be added as the workflows listed [at] AEFG (Go-Live) Vendor bib records/overlay rules and Import Profiles - AEFG w/ RMFG (MVP 003). GOBI’s API option for firm order materials provided an excellent test case for such workflows, as the API brief records include OCLC numbers for both physical and electronic materials, making them a seemingly ideal sample.

In advance of testing loading these records into the NZ, staff at UCI met with the Resource Management Operations Subteam and GOBI staff to discuss possible problems and how to rectify them. GOBI’s staff strongly advised against this practice as they know many titles will change OCLC # during the Gold Cataloging process, and are aware of a statistically significant number of items in their database which may have faulty brief record data - including titles listing the wrong OCLC #. We were initially hoping the change of OCLC# would be due to records being merged, and those OCLC#s being relegated to the 019 field, but the existence of the second case is almost impossible to overcome. Data for a batch of UCI’s API orders can be found at https://docs.google.com/spreadsheets/d/1xxt1OwT_hc-LHv9InWBWuF2RRoGg1hHc/edit?usp=sharing&ouid=114323008871213565779&rtpof=true&sd=true but may not include examples of all the errors GOBI tells us are known to exist.

Options Considered [remove if not needed]

Dependencies

Questions to consider

Action Log