Over the course of the past year, the OCLC Research Library Partnership (RLP) Metadata Managers Focus Group has delved into challenges related to staffing, succession planning, and thinking about metadata changes. Metadata Managers planning group member Chloe Misorski (Ingalls Library and Museum Archives, Cleveland Museum of Art) highlighted that these challenges are particularly pronounced for staff managing specialized collections and archives, who often operate with smaller workforces and budgets than larger, general collections.
To gain a deeper understanding of how these collection managers are navigating emerging next-generation metadata environments, we invited RLP cataloging and metadata colleagues in museum libraries, independent research libraries, art libraries and special collections or archives within larger campus networks. These are the metadata and cataloging colleagues of those who may be participating in broader discussions led by my RLP colleague Chela Weber as part of the Archives and Special Collections Leadership Roundtable.
We asked our participants to reflect on three prompts:
- How are you preparing for next-generation metadata and linked data?
- How are you developing new metadata workflows?
- What other factors are impacting your current metadata operations and planning?
Our discussions revealed a sense of apprehension stemming from the ongoing tension between the unique needs required of managing special collections and archives and the limited resources available, particularly in areas of cataloging and metadata management.
Backlogs
Several participants expressed that their most pressing challenge is addressing backlogs that predate the pandemic and have grown in the subsequent years due to limited cataloging resources. Some participants noted that they were only now back to previous staffing levels and that their primary focus was on tackling these backlogs using existing practices rather than looking ahead to next-generation metadata practices.
finding time to look to the future is difficult when keeping up with the present is so challenging
Workforce
Many of our participants are still dealing with a wave of retirements and organizational changes. These impacts are widespread but particularly severe in areas requiring specialized knowledge for cataloging unique materials. Staff reductions and retirements have exacerbated backlogs, and new hires often only slow their growth rather than reduce them.
Larger organizations have consolidated cataloging units, combining general and special cataloging staff. This has increased the workload for managers, who must re-imagine unit operations and provide growth opportunities for staff. Cross-training generalists to work with special collections is one strategy, but it brings challenges in balancing expertise levels across teams.
Managers often secure term-limited positions for specific projects, but training these catalogers is time-consuming, and they frequently leave when their terms end. This lack of continuity burdens the development of good documentation for cataloging practices. Recruiting for term-limited positions is also difficult due to the unique needs of special libraries, leaving descriptions stagnant after funding or staff departures.
Workflows
“Good enough” workflows: In the face of too much work and too few resources, many are working to define “good enough” workflows and standards for cataloging. Where, in the past, records for special materials may have been held back until they were “perfect,” resource limitations mean thinking about how to move records of sufficient quality forward.
Identifying efficiencies: Others are taking a hard look at their workflows and identifying opportunities to find efficiencies. Staff at Princeton University Special Collections shared how they were finding efficiencies by developing a “MARC Factory” that takes spreadsheets provided by book dealers and converts them to MARC records “good enough” to bring into other cataloging workflows. This is also an interesting example of what we’ve been calling “social interoperability” because staff in cataloging, acquisitions, and book dealers participated in crafting a workflow that worked for each of them.
Systems
Adding to existing challenges, several libraries are in the midst of system migrations. Even the smoothest implementations can cause additional disruptions, exacerbating issues related to staff shortages, backlogs, and reorganizations. Staff often need to freeze work, learn the intricacies of the new system, and rethink previous workflows.
Some libraries that invested in bespoke or open-source systems (OSS) to handle their special materials are finding it difficult to maintain them in the face of reduced resources. These systems are frequently built on technology stacks that are continually changing and, therefore, need close attention to maintain them to prevent security breaches. Maintaining bespoke systems may require gaining buy-in from leadership who are competing for limited resources, even if they share the same goals. Consequently, libraries seek commercial-off-the-shelf (COTS) solutions requiring less maintenance. At the same time, several participants mentioned how Yale University’s LUX platform (a multi-system open-source integration) is providing leadership in this area, even if their own institutions cannot build or sustain a similar platform with internal resources.
Abandoning a home-grown solution may come with other costs—if metadata in the source system is highly heterogeneous and not standards-based, it may need to be cleaned up to be migrated into new systems with less tolerance for creative descriptions. Even when moving high-quality metadata, work is required to ensure that the migration happens smoothly in ways that activate beneficial new features in the target system.
Next-generation metadata & artificial intelligence
Despite the challenges of staffing, backlogs, and reparative metadata work, many participants noted that they continue to pay attention to developments around linked data by attending webinars, creating test accounts, and exploring new tools.
As the market for library linked data tools is still emerging, many are taking a wait-and-see approach. When it is challenging to maintain existing systems and services, discussion participants find it difficult to consider extending workflows and financial obligations. Others continue to use home-grown solutions, especially for managing entities for digital and cultural collections that are not dependent on MARC-based workflows. For example, the University of Nevada, Las Vegas, has developed a portal that allows entities identified with multiple URIs to connect across LC/NAF, local data, and external resources.
For several participants, artificial intelligence seems like a distant solution, especially for special materials. There is skepticism about whether existing chatbot services can produce good descriptions. Despite an interest in “good enough” records, AI-generated may not be worth developing the workflows needed to validate and remediate issues—especially for descriptions of special materials. Concerns are amplified for institutions with large archival collections of analog items. While the Library of Congress’s experiment on the use of AI for cataloging ebook backlogs is promising, it doesn’t overcome the hurdles faced by staff in archives and special collections.
From the view of our participants, AI is drawing attention away from the day-to-day realities and complexities of cataloging workflows. While it may prove useful in the future, current applications of AI still need the guidance and expertise of catalogers who are knowledgeable about special materials.
Next steps
Multiple participants noted the value they’d found in Total Cost of Stewardship: Responsible Collection Building in Archives and Special Collections. Chela and I discussed how we might bring this resource to the attention of Metadata Managers and what it might help us to do based on the challenges reported in this session. We are currently planning a future round-robin as a follow-up.
The Metadata Managers Focus Group will also take a closer look at emerging next-generation metadata workflows, starting with Activating URIs in linky MARC in January 2025 (see the RLP Events calendar for dates and times). These sessions will be tied to follow-up conversations with OCLC colleagues who are building the future of cataloging. We hope to explore emerging use cases for how OCLC is bridging existing expertise and workflows that meets libraries where they are today – whether that’s in new editing environments or through a suite of APIs that enable creation and curation of linked data entities and descriptive relationships.
