Meeting of Technical Services/System Representatives
September 12, 2011
10 am – 12:15 pm
Sara Amato (via phone), Venice Bayrd (Maine InfoNet), Sarah Campbell (Portland Public), Karl Fattig (Bowdoin), Sharon Fitzgerald (UMO), Valerie Glenn (MSCS Program Manager), Clem Guthro (Colby), James Jackson-Sanborn (Maine InfoNet), Toni Katz (Colby), Judie Leighton (Bangor Public), Lanny Lumbert (USM), Mary Macul (Bowdoin), Peggy O’kane (Maine State Library), Deb Rollins (UMO), Mary Saunders (Maine State Library), Sharon Saunders (Bates).
The primary topic of discussion for the meeting was an OCLC reclamation project. The rationale for undertaking this project is to facilitate the use of a collection analysis system, as well as clean up MaineCat (shared catalog).
Sarah Campbell of Portland Public Library gave the group some background information on the reclamation project that they undertook in the fall of 2009.
- PPL cleaned up their system before sending records to OCLC. (They spent 3 aggressive weeks working on clean-up; 33,000 records were removed from their catalog during this process.)
- Made the decision to delete any records where status of missing, lost, billed, was set 18 months ago or later
- For others, used III’s batch check-in function to set things to available, then suppressed those records using a different suppression code (to separate from more current records)
- Sent around 240K records in batches of 60K (this was done in one long evening)
- III advises that batches should be limited to ~50K
- Generally, OCLC accepts batches of up to ~100K files
- A test batch (of XX records) was kept in order to track progress
- PPL did NOT send:
- records without an OCLC number
- bib records that did not have an item record (this was a way to identify e-resource records, which were not sent)
- Serial records were sent, but were set to “deflect” so that those items weren’t available for ILL
- Once records were sent, it took about 10 days for OCLC to turn it around, clean up OCLC numbers, and apply holdings to that batch (first step of scan/delete process)
- It took around 5 days to delete (holdings were deleted for items dated before the export date)
- PPL decided to re-load records into their local system – had to determine what local fields weren’t to be overlaid; had to wait for III to help them re: load profile
For Maine InfoNet: what are the matchkey points for MaineCat? Is overlaying of the records important?
Update from III:
The record matching is first on the o tagged MARC 001 with a subsequent check of the first three words of the title and if that fails an I tag matching of either MARC 020 or 022 with a subsequent check of the first three words of the title. Because of a bug in release 2006 which is fixed in current software there is an incorrect match because of its o001 and i020 matches. If the local record already exists at central then this matching does not occur. We decontribute and recontribute the records in question to fix these issues in older software.
Re: URSUS: What happens if records are re-loaded? How will this impact libraries not involved in the reclamation project?
- How are records with multiple formats handled? (ie, print record with microform, electronic holdings attached)
- If a library decides to contribute a subset of records to the reclamation project, how are other records handled in the scan/delete process? (exclusion list)
- How are existing LHRs handled during the reclamation? (Bates, USM and UMO use LHRs) ; would the Order Checklist for LHRUS Batchload come into play here? http://www.oclc.org/us/en/support/documentation/batchprocessing/using/checklistfororderingLHRUS.pdf
- What OCLC collection records are available? http://www.oclc.org/worldcatsets/sets/default.htm
- Can records be deflected for ILL purposes if there are no LHRs?
A subgroup, comprised of Sara Amato (contract Systems Librarian), Sarah Campbell (PPL), Karl Fattig (Bowdoin, CBB), Sharon Fitzgerald (UMO; URSUS), Toni Katz (Colby, CBB), Deb Rollins (UMO, Project Team), and Sharon Saunders (Bates, CBB) was formed to discuss and determine answers to the following questions for the project:
- What are books, ebooks, journals, ejournals, from a cataloging perspective?
- How should multi-format records be handled?
- If partner libraries are considering re-loading records from OCLC, what fields need to be protected?
- It has been decided that only records for owned electronic resources should be included; how will those records be identified?
- Determine the common answers for the project to certain questions on the OCLC Order Checklist for Bibliographic Batchload (http://www.oclc.org/us/en/support/documentation/batchprocessing/using/checklistfororderingBib.pdf)
It was suggested that clear guidelines for cleanup before the reclamation be provided; it was also suggested that workflows for this project be established.
Timeline: if we choose to subscribe to WorldCat Collection Analysis, their quarterly snapshot of WorldCat will be in January. We’d like for the majority of the reclamation projects to be finished by then.
- James will review matchkey points for MaineCat and send to attendees in order to determine whether or not records will need to be re-loaded (and if yes, how much of the record)
- Valerie will contact OCLC Batchload Services with the specific questions listed above, and will send answers to attendees
- Valerie will schedule a meeting of Collection Development representatives and will get their feedback on:
- materials that will be covered in the collection analysis
- capabilities we want from a collection analysis system
- These decisions will be communicated to Tech Services/Systems folks, as well as the Project Team