USF MARCit! Clean-up and Load

From SharedBib

Jump to: navigation, search

USF MARCit! Clean-up and Load

Related to RT tickets #17613 (catalog queue) #17479 (report queue) and #16185 (SFX queue)

Proposed plan (waiting for USF feedback)

I. Plan for a full file replacement of the MARCit! records for USF in the Shared Bib database.

Instead of processing new, delete and changed records identified from SFX, a new file of all relevant records would be identified. The records would be loaded with guidelines being developed for other e-journal temporal records. This would result in a clean up-to-date file of USF e-journal holdings in the Shared bib database.

Question: do you see any problem with the full file replacement? If that isn't possible, then we need to look at even more details for how to proceed.

-- II. Remove all MARCit! records from the Shared Bib database.

These records fall into two categories:

A. Records which ONLY have USF holdings which would be deleted Note: need to ensure that these are ONLY e-holdings and not any print holdings

B. Records which have other institutional holdings attached A report has identified that there are approximately 11,500 records in this category.

USF e-holdings can be removed as well as any indication these were MARCit! records.

Explore the option of replacing these bibliographic records with records from OCLC. (Obviously needs careful coordination with the SUL libraries who's holdings are attached)

USF response---------------------

We can’t do these concurrently, though we can prepare for them concurrently. Once we do a full export, it should be overlaying on the SFX 035 in the bib record. If those 035 fields are not removed from the merged records before the export, new MARCit! records will overlay all of those merged records. Other comments:

I—We’re fine with doing a fresh export of MARCit! records and have created a MARCit! profile called “master” for this purpose.

IIA-- Yes, SFERS can go; SFTMP needs to stay. One problem with removing ALL MaRCit! records is that several of them have orders attached, and we need to keep those, though they can be overlaid with a fresh export. It’s fine to remove any that have just USF SFERS holdings and that do NOT have an order on them.

IIB—Yes. Rebel has supplied a list of global changes that would wipe out any MARCit! data from the bib records that have merged.

Alternatively, here’s the five steps that have been in my head:

1—Remove MARCit!/USF info from the shared records per Rebel’s list of requested global updates 2—Do a fresh MARCit! export to overlay the single-holdings records and add new bib records. 3—Get a list of the remaining bib records that have both an OCLC 035 and SFX 035. We’ll go through these to determine which ones need to be deleted. (assuming that there was no mass-delete before the fresh export) 4—Carol Ann will coordinate moving any LIB920 orders to the new MARCit! records. These might be on print bibs, or previously merged bibs. 5—Get a list of SFTMP records with 245|h [electronic resource] so that we can work on fixing the SFTMP records that were merged with MARCit! records. Or, if you are doing something at the state level to handle everyone’s print records, that would work too! ☺

Rebel's request:

After reviewing these records it would be my suggestion to run a global modification on these records to remove if found(Should not be case sensitive):

035 (SFX)*

655_7_a Electronic journals

655_7_h lcsh



951__1 10

951__a SFU*

951__5 FTS

949_a FTS

949_b SFERS

949_c JRNL

949_x MARCit

949_h (ONLINE)

It is my understanding that we do not have the ability to run these global changes, so if this could be complete by FCLA/FLVC staff that would be great. Please let us know if you agree with these suggestions and if this is something that can be completed by FCLA/FLVC on this set of records.

In addition to this we will also need to remove USF SFERS (not SFTMP) holding and item records associated with this set.

Carol Ann has also alerted us that the text within the 949_c JRNL may not always say “JRNL” but may also have AGGR and possibly JRNLG(?) so perhaps we just need to remove any text found within the 949_c. Better yet maybe we just need to delete all 949 fields with FTS in 949_a. Also, all 856 fields that have subfield 5 FTS can also be removed.

To confirm this list of records only contains Bib with two or more holdings and SFX 035? We have a plan to overlay/delete the remaining old records but need the SFX to be left on those records.

Personal tools