Page History

Michelle Mascaro

At some point in the past catalogers were less able/confident in updating OCLC master records. Using Alma, Connexion and vlookup in Excel, I'm finding those records that have enhanced 505 notes locally but not in OCLC, and then assigning them to catalogers in order to enhance the master record. We've so far done this with Special Collections as well as music scores and the place where we have DVDs. Future project will be to look at 6xx fields and 520, but I'm not sure yet how I'll do the 6xx fields.

Example resulting spreadsheet. In this sheet, I compared the 245 fields, which meant I normalized the text by removing spaces and most punctuation in order to account for subfield coding. This is not necessary in 505 fields). A brief explanation of my process:

1. Identify subset of records (in this case, a location).
2. Pull out OCLC numbers (can be done multiple ways).
3. Export set of bib records from Alma.
4. Put OCLC numbers into Connexion, then export the bib records from Connexion.
5. Run both files through Marcedit, and export specific fields as CSV (you will need a match point-- OCLC number--and the field you're comparing such as the 505).
6. Open both of those CSV files in Excel, then use vlookup to put the two lists in the same sheet (Column 1=OCLC number, column 2=Alma 505, column 3=OCLC 505)
7. From there... compare the Alma and OCLC columns (are they equal? true/false; character count-- if there's a big difference, have a cataloger examine; etc-- depends on the field and how similar you want them)

Josh Hutchinson

We are trying to move local data from the 500 field to the 590 field, starting with Special Collections material but also including other locations, if we have time before migration.

Annotations providing information about contributors (e.g., editors, engravers, illustrators, etc.)

Josh Fiala

I've discovered a lot of binding notes in particular in the 500 field that will need to be moved to 590s.

We've also added local data in non-local 7xx fields regarding printers and publishers over the years. I'm not sure how we'll identify which ones were only added locally and protect them.

There is an ongoing project to clean-up legacy practices that cause confusion. In the past it was common practice to add multiple 590s for one item or have one 590 for multiple items.

The goal is to have one 590 per item or set and to add a prefix for each 590 denoting the location and copy number when necessary for identification purposes (Special Research Collections copy 1: ...) Unfortunately, it largely requires manual intervention so it's a slow process.

We have other local data that will need to be identified and moved, but we're in a holding pattern on those clean-up projects until there is hopefully some consensus on what other UCs will do.

If they were to flip to 590s then there would be concern that we're just adding to our current 590 confusion listed above. It will be even harder to pull things apart later without having to physically look at the items.

(Note: covers general and special collections) We pulled records with these fields from our old Millennium catalog, acknowledging that some may slip through the cracks if they were added in Alma (but it is not our current practice to use anything other than 590). Assessed contents and divided into categories by action: delete, keep, move, review/cleanup. Some fields require manual cleanup (e.g. identifying LCSH terms in place of the local subject heading) and others can be deleted in bulk (e.g. local headings for dissertations that came from UCB in our WRCA records). Most of the existing 59x are being consolidated to 590.

Other fields moving to 59x: 501 (With note) to 591, 506 (Restrictions on access) to 598, 586 (Awards note – education collection only) to 596

We still need to conduct a similar analysis for 79x.

It has been our practice to add cover artists and other specialized illustrators in our Eaton Collection. Sometimes this information is also in a 500.

We're still reviewing and evaluating how to retain this data.

UCR has around 9,100 bib records with non-English 040 $b. We initiated a project to replace them with the correct English version of the OCLC records.

• After manual replacements began, we realized that some records had already been updated to 040 $b eng in OCLC. We are batch importing/overlaying all of our non-English records to eliminate the ones that have been converted to English records already.
• For the remaining records, we will be experimenting with MatchMarc for all records that have ISBN or LCCN to see if it can identify decent English replacements in batch.
• Remaining records without ISBN or LCCN will be updated manually