OCLC Expert Cataloging Community Sharing Session minutes, June 2018

Minutes of the OCLC Expert Cataloging Community Sharing Session
ALA Annual Conference
Friday, 2018 June 22
10:30 a.m.-12:00 noon
New Orleans, Louisiana

The ALA Annual 2018 Edition of Breaking Through: What’s New and Next from OCLC and the compilation of News from OCLC were distributed. The following items were highlighted:

• The next OCLC Americas Regional Council Conference will be held 2018 October 25-26 in Chicago. The theme is Change the Game, how to go beyond incremental progress to make transformative progress that will change libraries and communities.
• OCLC-MARC Update 2018: This update will implement the MARC Bibliographic and Holdings format changes announced in MARC 21 Updates No. 25 (December 2017) and No. 26 (April 2018), as well as MARC codes announced in nine Library of Congress Technical Notices (https://www.loc.gov/marc/marcginf.html#naa) issued since November 2017. The updates will be done in the second half of 2018 and will be widely announced. The corresponding authority updates will be implemented later in coordination with LC and NACO.
• OCLC Virtual AskQC Office Hours and Metadata Quality FAQ: OCLC Metadata Quality has a new Web page at https://oc.lc/mq, which links to the Virtual AskQC Office Hours page where all materials on the Office Hours are posted. The next Office Hours is scheduled for Wednesday, June 27 at 1:00 p.m. EDT. Then it will take a few months off while OCLC evaluates the results of a recent survey on the project to decide whether to continue with it. The initial impression is that those surveyed favored the sessions and they are likely to continue later in 2018. We also received many ideas for future topics to discuss.

Laura Ramsey (Section Manager, Quality Control) spoke about the OCLC Member Merge Project, which currently has eight participating institutions. The four libraries of the second cohort—Brigham Young University (UBY), University of Maryland (UMC), Western Washington University (XFF), and the University of California, Berkeley (CUY)—are now all independently merging book format records and will be trained to merge other formats. OCLC has already received about a dozen applications for a third cohort to begin later in 2018. We expect to select around seven. We are hoping to have an OCLC Community Center for the Member Merge Project by the time of the third cohort. If you are interested in participating in a future fourth cohort, please contact us at AskQC@oclc.org. During Fiscal Year 2017, participants in the project performed 4883 merges; during Fiscal Year 2018, 6347 merges were done as of the end of May 2018. Between July 2017 and the end of May 2018, members of the WorldCat Metadata Quality team performed 349,222 manual merges and corrected 136,413,412 bibliographic records. During the same period, member libraries modified 1,039,607 records.

Could a symbol such as “OCLCM” be used as a field 040 subfield $d designator for the project? Can a symbol be added to show the library doing the merge?

Because of the technicalities of the way merging happens, a symbol for the merging library could not be added until after the merge is complete. The merging institution’s symbol would be added, however, if they modify and replace the retained record.

How can we get the updated records after a merge?

In Collection Manager, you can sign up to be notified whenever there are updates to a record. This is the same functionality as for sending records. The matching points are determined by your Collection Manager settings. OCLC’s Jody Stroh (Product Manager, Metadata Services) adds: “Merges result in an update record being delivered. Because the change in OCN sometimes creates a bit of an issue for those libraries overlaying based on OCN, we deliver merges in a separate file, in case they need to deal with them in a different way. If they have ‘Reason for update’ configured in their MARC output, I believe it says ‘OCLC control number change.’”

John Myers (Union College) noted seeing some instances with JSTOR records that would snag due to post-merge records not matching. Their system wasn't set up to match OCLC numbers in field 019 with field 001, so libraries should be aware of that setup option.

There were no pre-submitted questions, so the floor was opened for further questions, answered by Bryan Baldus (Consulting Database Specialist, Metadata Quality); Robert Bremer (Senior Consulting Database Specialist, Metadata Quality); Marti Heyman (Executive Director, Metadata Strategy and Operations); Hayley Moreno (Database Specialist II, Metadata Quality); Charlene Morrison (Database Specialist II, metadata Quality); Sara Newell (Senior Product Analyst, Metadata Services); Rosanna O’Neil (Senior Library Services Consultant, Library Services for Americas); Laura Ramsey (Section Manager, Quality Control); Jay Weitz (Senior Consulting Database Specialist, Metadata Quality); and Cynthia Whitacre (Manager, Metadata Quality).

Becky Culbertson (University of California, San Diego) opened not with a question, but with a plug for the National Academies Press Project, which is digitizing and making available online—free, searchable, and in full text—thousands of the Academies’ reports and other publications from 1863 onward—through Google Book Search. A search in Connexion for this publisher retrieves 30,954 records. Becky and others have created metadata for this collection, which is both current and growing. This open access collection is in Collection Manager. It was recommended that this could certainly be promoted on the Knowledge Base forum of the OCLC Community Center.

There was a recommendation that in Collection Manager, there was a need to download multiple files at once. Jody Stroh responds: “They can certainly do that via SFTP, but that isn’t currently something we’ve talked about for My Files.” Sara Newell adds: “This means that libraries would need to download a group not from the UI but by FTP.”

What is happening with FAST? Some members are having trouble getting FAST headings added to their records. Are the FAST headings not being generated correctly and in a timely manner?

Marti Heyman responded that OCLC is currently working on a road map to move FAST from research to production. It was established as a research experiment and so resources have not been entirely adequate. OCLC would like it to evolve into a cooperative project, like Dewey's EPC, but more lightweight, something like a FAST editorial policy committee processing candidate terms. Another attendee was curious to hear how this institution uses FAST genre, names, and topics. The institution stated that any subject heading they use in their collection gets FAST headings, which are included in their knowledgebase.

Richard Sapon-White (Oregon State University) wondered why he is seeing so many LC records with uncontrolled headings. Another said he catalogs from CIP and always controls all the headings, but they never show up in OCLC as controlled.

When records are batch loaded into OCLC from LC or from other member libraries none of the headings show up as controlled initially; it takes a while for the control to be implemented, but it will be there eventually. OCLC will check to make sure there is no problem with the program. Those who see the uncontrolled headings are encouraged to control them. If the delay is more than a few days, please report those to bibchange@oclc.org to determine what is causing the delay.

Rebecca Henning (Amherst College) asked if records for rare materials are being merged. This is of concern because there are so many seemingly insignificant details on these records that really do make a difference and may result in incorrect merges if ignored.

Regarding manual merging, OCLC is very cautious about merging records for any rare materials because they are aware of the problems. Emphasizing that caution is definitely part of the Member Merge Project training. As far as automated Duplicate Detection and Resolution (DDR) is concerned, OCLC currently exempts from online processing all bibliographic records that are coded in field 040 subfield $e for any of the following description conventions for rare and archival materials: amim, amremm, appm, bdrb, cgcrb, cco, dacs, dcgpm, dcrb, dcrmb, dcrmc, dcrmg, dcrmm, dcrmmss, dcrms, dmbsb, enol, estc, gihc, iosr, ohcm, rad, rna, vd16, vd17. These DDR exemptions apply only to the original resources. The complete MARC Description Convention Source Codes list can be found on the Library of Congress Web site at http://www.loc.gov/standards/sourcelist/descriptive-conventions.html. OCLC urges members of the cooperative to code records for rare and archival materials using the applicable descriptive convention you’ve followed. In addition, records for resources with dates of publication/production earlier than 1801, cartographic materials with dates earlier than 1901, and all resources that can be identified as photographs (Material Types "pht" for photograph and/or "pic" for picture) are exempt from automated online DDR processing.

There are many records in the authority file with a 667 note that says the record cannot be used in RDA until it has been reviewed and possibly updated. What should I do if I need to use one of these headings?

Robert Bremer explained that these notes were intended to alert NACO participants of the need to review the name to make sure it is compliant with RDA. If the name is compliant the NACO participant can remove the note; if the name is not compliant, they can change it to become RDA compliant. If you are not a NACO participant, use the heading as is and control it so that it will be automatically updated when the authority record is.

How can we be alerted to these updated authority records?

Attendees suggested going to such vendors as MARCIVE who can send you the updated authority records. LC also has a form that tells you what has been changed. Collection Manager can also send you a file of controlled and updated authority headings.

It seems that, especially at the start of a semester, it takes a long time to get YBP records into Collection Manager. Do you know anything about this issue?

We would need more detail and some specific examples because this could be a KB or WCP issue. Please get back with more information to Sara Newell (newells@oclc.org) or Jody Stroh (strohj@oclc.org).

Jim Soe Nyun (University of California, San Diego) recently did a large number of batch updates and would like to send all the updated records to OCLC but is afraid that some of the updates may not be quite as good as they should be. Should they be sent anyway?

Yes, a not-quite-perfect update is certainly preferable to none at all. No process is perfect.

Becky Culbertson wondered about records from the National Library of Scotland (NLE) containing 506 fields saying to go to the main reading room.

Robert Bremer has asked NLE about these fields but has yet to hear back.

Is it preferable to upgrade a non-English Language of Cataloging vendor record or to create a new English-language record?

If such descriptive fields as 3XX and 5XX are actually in English, that means the record is incorrectly coded in field 040 subfield $b. In that case, correct the Language of Cataloging and use the record. Otherwise, and only in the absence of another English Language of Cataloging record, create a new English record.

John DeSantis (Dartmouth College) says that a certain PCC institution has been adding 588 fields with a statement like “This record is the property of this library” plus copyright information. Is this a correct use of field 588 and an acceptable practice?

No, such a field should not be in the master record.

Could we have an update on Bibliographic Formats and Standards (BFAS)?

The complete overhaul of BFAS is still in progress. The date of the last update appears on each page that has been touched. The updating of Chapter 4, “When to Input a New Record,” was completed and made available in March. The updating of Chapter 5, “Quality Assurance,” was completed and made available in May. We are actively working on Chapter 3, “Special Cataloging Guidelines,” to which we’ve added new sections. Everything else in the chapter is being updated, although some parts may be moving elsewhere in BFAS. Each of the initial chapters is a single “page” and our content management system doesn’t allow us to update only parts of a single page. That means we can’t activate any chapter until it is completely ready, which slows down our ability to make revised sections public. As we’ve been working on BFAS, we were given the added responsibility of integrating information and fields for Local Holdings Records (LHRs) and Local Bibliographic Data (LBDs). Those are not within the realm of expertise of any of us working on BFAS, so that’s been a challenge. On the other hand, our BFAS revision meetings are more fun and stimulating than you might imagine. Please do continue to report to us (AskQC@oclc.org) any errors or incorrect examples you find, and we’ll try to correct them as quickly as possible. But realize that sometimes corrections may not be so quick because of concurrent work that may be in progress on a particular page.

Can more fields be made available in LBDs?

Sara Newell noted that we have a list of fields that have already been requested, and we continue to take suggestions. Charlene Morrison said that OCLC staff is currently working on an internal proposal for the addition of LBD fields.

How do we handle local and other nonstandard fields, especially subject headings that are not standard LC subject headings or FAST headings? Do we retain them all, delete them all, retain some but not others? If we keep them, do we display and index them?

Don’t change them on the master record, but locally, feel free to do whatever meets your institution’s needs. Keep, index, and display all, some, or none; keep them but don’t index or display; keep only those with Second Indicator 2—whatever makes the most sense for your institution. If you are part of a consortium you will want to consider the needs of the entire consortium and consult with other participants. Do the other members want everyone in the consortium to keep and index and display all or some of the nonstandard headings? There are various macros and tools such as MarcEdit that you can use to deal with unwanted fields. Various local systems also have ways to deal with such fields.

Caitlin Angelone (cangelone@collegeofphysicians.org), Reference Librarian at the Mütter Museum and the Historical Medical Library of the College of Physicians of Philadelphia (https://cpp.ent.sirsi.net/client/en_US/library, OCLC symbol: PPC), reported that they recently merged the records of the museum into the library’s catalog and loaded the records into WorldCat. There are no catalogers on the library’s staff, so they are aware that there must be a lot of errors in these records. As non-catalogers, they are not even sure that they can recognize patterns of errors that would be immediately apparent to a cataloger, so she asks for help from the OCLC Expert Community. If we can look at their records and let her know what errors we find, she would be most grateful for any feedback.

Has an end date been set for Connexion?

No, there is currently not an end-of-life date for Connexion, and we will give you plenty of warning once such a date is determined. In the meantime, Record Manager is available for those who want to become familiar with it and to help OCLC make it as functional as possible. Sara Newell says that we are looking at extending the Record Manager advisory group so that we can get more input from members of the cooperative on how to improve the tool. Please let Sara know if you are interested.

Respectfully submitted by
Doris Seely
University of Minnesota
2018 June 28

With contributions and edits from Bryan Baldus, Hayley Moreno, Sara Newell, Jody Stroh, Cynthia Whitacre, and Jay Weitz.

OCLC
2018 July 30