The potential for using AI tools to increase the efficiency of library cataloguing has been increasingly recognised in recent years. Like many libraries, the University of Glasgow Library has a substantial backlog of non-English language material waiting to be appraised and catalogued, but few staff with the specialist language skills required to do this. However, recent advances in AI image analysis, translation, and transliteration have made it possible to begin to tackle this backlog without requiring specialist language skills.

This presentation will cover the CILIP Scotland Ian McCracken Innovator Award winning project which investigated the potential for using AI tools to help us copy catalogue non-English language books.

The project started by using Google Translate, OpenL, and ChatGPT to translate photos of title pages of Russian books, then used those translations to search for catalogue records on OCLC. This was not very successful, resulting in only a 52% success rate.

We then tried transliteration instead of translation, which is only supported by ChatGPT. Specific prompts were used to instruct ChatGPT to generate Library of Congress-compliant transliterations from images of our Russian material. These transliterations were very accurate, and we were able to use them to find records on OCLC with a 92% success rate. ChatGPT transliteration was also used to create a list of a donation of rarer Russian books, with a similar success rate. We then used this list to scarcity check the donation against our own and other catalogues, to inform decisions about retention or discard.

Our pilot project has demonstrated that AI tools have clear potential to help us catalogue a broader range of materials than was previously possible. While they cannot replace cataloguers’ deep subject knowledge and specialist skills, they can nevertheless provide valuable support for cataloguers without language skills to allow them to copy catalogue in languages they can’t read. AI tools must be used with caution and an awareness of their limitations, but they nevertheless have an increasingly important role to play in making our collections more accessible than ever before.

The conference presentation will expand on this background information, further detailing the evolution of the project, and include demonstration of prompting and results, before discussing further use and additional applications of the methodology.

This paper will describe a retrospective cataloguing project at Southampton Solent University to assess the feasibility of using generative AI to generate time periods describing the subject content of books for use in the 648 field in MARC catalogue records. The project aims to assess the accuracy of generative AI, with a view to possibly fully automating the addition of this field under certain conditions.

We will discuss:

• What value the 648 field offers?

This field is stipulated for use as part of FAST subject analysis, but it appears to be little used. It allows for simpler and more accurate description of the time period of the content.

• Practicalities of generating the 648 field with AI

We will describe our methodology and present our results, detailing the accuracy level of AI generated data and the strengths and weaknesses of this approach.

• Improving discovery with time period data

As well as discussing how the 648 field might be used in an ideal world, we will describe how we have implemented time period data in our discovery environment and the challenges of doing so. We will also report on the impact of this implementation.

Cataloguers and metadata professionals stand at the threshold of a new era, one where local cataloguing practices must not only be transparent and consistent but also machine actionable. This presentation introduces a metadata strategy that reimagines how cataloguing rules are authored, shared, and used—both by people and machines—laying foundational work for broader metadata interoperability and even the long-envisioned goal of Universal Bibliographic Control (UBC).

Traditionally, cataloguing guidelines have been published in formats suited only for human consumption: print manuals, PDF documents, or static HTML pages. KBR, the royal library of Belgium, has moved beyond these formats to develop a modular, machine-readable metadata ruleset that takes the form of a structured JSON file. This JSON file is now the single source of truth for our cataloguing rules, providing data to two complementary outputs:

1. A human-readable HTML interface that dynamically renders rule descriptions and logic using the JSON backend, and
2. A machine-readable interface that enables integration with generative AI tools and other metadata processing systems by sending the same JSON content as part of prompt engineering workflows.

This “triangle model” — JSON as the backbone, HTML as the human interface, and prompt ready data as the machine interface — represents a novel, scalable solution to metadata consistency and authority control in both traditional and AI-assisted cataloguing environments.

The implications are significant. If libraries and institutions adopt similar machine-readable frameworks for their local rules, the path to aligning and ultimately unifying metadata practices becomes more achievable. Our approach goes beyond standard creation or adoption — it promotes harmonization of logic, not just terminology. By enabling AI tools to respect local policies while operating at scale, we ensure that future metadata work maintains fidelity, transparency, and context.

This presentation will offer a walkthrough of the JSON schema and discuss technical considerations, and demonstrate how this system is already being used in conjunction with generative AI tools. It concludes with a call for collective action: imagine a future in which every institution’s rules are machine-readable and designed to interoperate. Why not aim not only for compatibility, but for shared cataloguing logic—a collaborative metadata ecosystem that supports both human judgment and machine efficiency?

This talk is especially relevant to metadata specialists, cataloguers, system librarians, and LIS students interested in the intersections of metadata governance, automation, and the evolving role of cataloguing professionals in AI-enhanced environments.

This session explores an ongoing research project at San Diego State University (SDSU) Library to integrate a natural language query (NLQ) to Boolean converter (https://library2.sdsu.edu/onesearch/ai/demo/) into library discovery in order to enhance metadata visibility and improve search transparency.

Vendor-produced AI search and discovery tools are being released at a rapid rate, yet most of these commercial systems search textual abstracts and article-level summaries, disregarding the underlying metadata that librarians carefully curate. In contrast, our project positions AI as a mediator between natural language and structured metadata. By converting user questions into advanced Boolean queries, the tool makes explicit the role of subject headings, synonyms, truncation, and facets in shaping retrieval. This approach not only improves search precision but also provides a learning opportunity for users by revealing how metadata influences results.

Unlike tools that cannot utilize structured metadata, our system applies the complex Boolean strings, developed from the researcher query using an AI prompt, directly to the library catalog, strengthening the value of librarian-produced metadata and reframing AI not as a replacement for discovery systems but as a bridge to access them more effectively. This work contributes to broader conversations about metadata ethics, user empowerment, and advanced search instruction.

Originally initiated in January 2024, the project is now moving into a second phase of research and design. Librarian researchers are working with pedagogy and metadata experts to build improved iterations of the tool that will address specific unique information needs. Initial testing with students and faculty has demonstrated both opportunities and challenges. These findings provide valuable insights into the future of metadata-driven search systems.

The conference paper will share the technical approach, user feedback, and lessons learned, alongside best practices for libraries considering similar tools. We will argue that applying AI to search metadata has the potential not only to improve discovery but also to give librarians new diagnostic methods for evaluating and strengthening their metadata. Ultimately, this project demonstrates a replicable and scalable pathway for libraries that want to harness AI while keeping metadata at the center of discovery.

How well can you articulate the mission and vision of your Metadata Team? Do you know what your strategy is to achieve that vision, and what your goals are? Do your colleagues understand what happens in the Metadata Team, and the centrality of metadata to the institution? This session will help you visualise the role of your Metadata Team as a central cog in the larger endeavour to transform discovery in the wider context of your library and parent institution. Mapping the library and institutional strategy to a mission and vision for the Metadata Team brings clarity, focus and increased value to day-to-day work. It also demonstrates how metadata contributes to strategic priorities, a particularly important skill in communication with senior managers and decision makers in your organisation. Using the model of a ‘Purpose Tree’, and in discussion with other conference attendees, this session will help you think through how to define the mission and vision of your team, what your strategy is to achieve that vision and what actions need to be taken to move towards that.

The implementation of the official RDA Toolkit, underpinned by the IFLA Library Reference Model (LRM), requires cataloguers to adopt new conceptual frameworks and workflows that differ significantly from AACR2 and legacy MARC-based practice. For many institutions, the greatest challenge is not technical but pedagogical: how to bridge the gap between abstract models and daily cataloguing tasks.

This paper explores training approaches that connect theory and practice by scaffolding learning from LRM’s entities and relationships to hands-on applications in an entity based cataloguing system. Strategies include integrated conceptual and practical modules, train-the-experts models to cascade expertise, and differentiated pathways tailored for both new professionals and experienced cataloguers.

Using BDS implementation of entity-based cataloguing and of the official RDA as a case study of a training programme, the paper demonstrates how cataloguers can become effective bridge-builders between legacy standards and future-oriented metadata, ensuring organisational readiness for linked data environments and sustaining professional expertise in a period of transition.

Anglican diocesan libraries in Ireland have long played a quiet but significant role in preserving theological knowledge, institutional memory, and ecclesiastical identity. Often located within cathedrals and shaped by the bibliographic passions of reforming bishops and scholarly deans, these collections represent an important but under-explored aspect of Ireland’s Anglican and library collection heritage. Yet many remain uncatalogued, poorly resourced, or functionally severed from the communities they were created to serve. The absence of robust metadata, authority control, and discovery mechanisms has left them vulnerable to neglect and, in many cases, invisible to both researchers and the wider public.

This paper explores the evolving role of metadata, cataloguing, and cross-sectoral collaboration in revitalising these overlooked collections, using the Diocese of Cashel, Ferns and Ossory as a case study. Once home to four significant diocesan libraries - Christ Church Cathedral in Waterford City, St Canice’s Cathedral in Kilkenny City, the Bolton Library in Cashel, Co. Tipperary, and the Cotton Library in Lismore, Co. Waterford - the diocese offers a compelling lens through which to examine how cataloguing decisions have affected the survival, visibility, and function of these collections over time.

Drawing on both historical and contemporary examples, the paper traces the transition of these libraries from working clerical resources to heritage archives, often relocated to academic settings. Where professional cataloguing has taken place, such as at Maynooth University or the University of Limerick, collections have found new life as research resources, though often at the cost of dislocation from their original ecclesiastical context. By contrast, Christ Church’s collection remains materially intact but largely uncatalogued and seldom accessed, its intellectual value obscured by a lack of metadata infrastructure.

The Cotton Library is the only one still in situ, a rare survival within its founding cathedral. Though modest in scale, its holdings are rich in ecclesiastical, theological, and antiquarian content. Its recent acquisition of a first-edition, seventeenth-century publication of the Bidel Bible, the first translation of the New Testament into Irish, signals renewed potential for relevance. Yet the Cotton Library’s metadata footprint remains minimal, its online presence scarcely visible, and access tightly constrained. In an era where discoverability hinges on metadata, this library risks becoming the best-kept secret in Irish ecclesiastical heritage.

This paper considers how such collections might support theological education, public programming, and community history, if more effectively integrated into digital catalogues, discovery platforms, and collaborative networks. It concludes with practical recommendations for partnerships between cathedral chapters, public libraries, academic institutions, and heritage professionals.

By making ‘hidden’ collections visible and accessible, these fragile yet culturally rich repositories can be reactivated, not only for ecclesiastical scholars of the past but for the wider communities and cultural imaginations of the present and future.

Metadata librarians and cataloguers collaborate all the time. We share advice, good practice, documentation, and metadata itself where we can. This interactive session will get us to do all of this in real time, now.

Drawing Room is a gallery and exhibition space with a specialist library that is open to the public. Drawing Room’s librarian is committed to having the collection professionally catalogued and has recently overseen a Mellon funded project to start this process. While it raises further money to complete the work the organisation is keen to keep the cataloguing going in the meantime, if possible. Using digital surrogates, we will do our best to contribute our enthusiasm and expertise to this task.

On the one hand then, the session is about trying to come together to help a library on its cataloguing journey. As a small specialist collection, it is not unreasonable to think that, one day, it could be completely catalogued to a consistently high standard and the metadata shared via the National Bibliographic Knowledgebase. This is tantalising enough of a prospect for many of us with vast swathes of legacy metadata and uncatalogued materials.

On the other hand, however, this collaborative cataloguing exercise is also an experiment in adapting to a shared brief and collaborating quickly, efficiently, and professionally, in real time. If we can come together to organise ourselves around a goal like this, it is possible that we have more power to shape our sector and the future of the cataloguing profession than we realise.

Let’s get cataloguing!

This paper will introduce and give detail on the UK Print Book Collection (UK PBC), a national initiative designed to support long-term management, retention and access to print monographs across UK libraries. The UK PBC enables libraries to make informed decisions about print retention and foster shared stewardship of a national monographic collection. This paper will cover the origins of the project, explore the findings of the pilot phases, and explain how libraries can participate and benefit from the initiative.

This paper explores how the National University of Singapore (NUS) Libraries have transformed access to Southeast Asian cultural heritage through a robust digitisation and open access programme. Central to this initiative is the creation, management, and discovery of high-quality metadata, supporting the long-term preservation and global visibility of unique collections.

Drawing on large-scale projects such as the Digital Gems portal—the flagship digital repository of NUS Libraries, which provides open access to a curated selection of rare, unique, and fragile materials that showcase the cultural and historical heritage of Singapore and Southeast Asia, enhance understanding of the region, and potentially serve as data sources for digital humanities research—this session details the workflows and standards used for cataloguing, classifying, and indexing rare materials, including manuscripts, private papers, newspapers, and biodiversity records. The adoption of professional metadata standards and best practices ensures discoverability and interoperability across global repositories and knowledge bases, while also contributing to the inclusivity and accessibility of Southeast Asia’s diverse historical narratives.

The presentation highlights the integration of digital collections into library discovery systems, as well as the National Library Board of Singapore’s OneSearch platform, facilitating open access and supporting research, teaching, cross-disciplinary scholarship, and public enrichment. Special attention is given to data management strategies, including digital preservation infrastructure, authority control, and the use of controlled vocabularies to enhance search engine visibility.

Collaboration and cooperation underpin the success of the NUS Libraries programme. The talk will showcase partnerships with NUS academic departments, international institutions, and the wider library and archives community, including initiatives such as the Transforming University Libraries Leadership & Innovation Programme (TULLIP). These collaboration have enabled shared expertise, joint digitisation projects, and resource sharing, thereby setting a model for the wider GLAM sector.

This case study will provide attendees with practical insights into:

• Metadata creation and management for digitised heritage collections
• Integrating digital collections with discovery systems
• Building sustainable collaborations within and beyond the academic library
• Addressing challenges in inclusive metadata, authority control, and digital preservation

NUS Libraries’ experience demonstrates how academic libraries can lead the way in creating accessible, sustainable, and discoverable digital heritage, offering a replicable model for institutions worldwide.

When an authority list has got out of hand, can we make it work for library users and the staff who support them?

The Health Libraries Midlands consortium (HeLM) was formed in 2022 from a merger of smaller consortia and individual libraries. Its 48 NHS libraries share a Koha librarymanagement system and serve NHS staff in hospitals and communities around the region. Users need highly specialist information, accessed quickly with the minimum time spent locating materials. NHS libraries are small and delivery-focussed, with few specialist cataloguers and a high turnover of stock.

Authority control on subject headings made sense when the consortium formed, but has been hard to maintain. Many records use unauthorised headings, and manyheadings do not come from a recognised source. When adding authorised headingstakes staff extra time, and users are not seeing search benefits, should we abandon authority control; or find a way to prune it back to functionality?

Recent developments in the Wessex classification scheme gave us an impetus: itsheadings are regularly updated to better reflect modern UK healthcare. Wessex forms the basis of our classmarks and subject authority list and we want to put it to use.

This paper will discuss and reflect on our approach: improving the workflow for assigning headings, including experimenting with AI agents; fixing thousands of recordswith non-preferred headings; and making authority control sustainable in a distributed consortium.

LSE is taking its first steps into providing computations access to collections and bibliographic data. The first project on this journey was turning the metadata held for LSE print theses into a publishable dataset to complement the Wikidata work we have and are doing on our electronic theses. This talk looks at the goals of the project, lessons learnt, the hurdles encountered and navigating around them.

The library is about to enter a partnership with OCLC, working with them on their Meridian product to deliver linked data entities for two unique and distinctive collections at the University of Southampton. The transition towards a linked data environment in the UK has been slow to get off the ground. We approached OCLC with a view to being early adopters of the product, believing that the experience and resulting feedback could benefit not only the University of Southampton but the wider metadata community. This presentation will explore the process of setting up the project, and the benefits and pitfalls of working within a linked data environment. It will look at the impact of linked data on discoverability of resources and seek to explore benefits for users.

This is a pilot project to assess the viability and benefits of using OCLC's Meridian system for adding linked data entities to bibliographic records. Our focus is to enhance metadata quality, improve discovery, and explore future-facing alternatives to traditional name authority control.

The UoS Metadata & Discovery team (including core cataloguing staff, Special Collections, and Winchester School of Art), with appropriate training and support from OCLC, will engage with the Meridian product to enrich metadata for two unique and distinctive UoS collections:

• Parkes Collection – Holocaust Memoirs, Special Collections, Hartley Library (focused on Jewish/non-Jewish relations and anti-Semitism research)
• Visionaire Collection, Winchester School of Art (Avant-garde art/fashion publication with varied formats and creators)

The aims of the project are:

• Improved user discovery: richer, contextual metadata to support better search results and serendipitous exploration.
• Staff development: practical training in linked data systems, building capability for future cataloguing practices.
• Collection visibility: enhanced records may increase research usage and external engagement with unique UoS holdings.
• An evidence-based assessment of linked data as a viable alternative to LC name authorities.

In a research-producing institution it’s likely that the organisational strategy speaks to increasing research visibility, reach and impact. It’s also likely that metadata or research services teams play an important, if sometimes unsung, role in contributing to that by creating or managing the metadata which ensures that each output is discoverable in an institutional repository, and on into wider aggregator services.

But how could a metadata team go the extra mile to support the wider institutional strategy, in the process making the team itself more visible as well?

This paper will demonstrate how the Metadata team at LSE Library have developed their role in supporting LSE’s ambitious Research for the World Strategy. It will outline the creation of new metadata sources to strengthen institutional infrastructure, researchers and research outputs in a complex and crowded research landscape, before exploring the place of Wikipedia in disseminating research to cultivate wider understanding of research, support open science and potentially increase public engagement. Finally it will explore the development of a Wikimedian in Residence role within LSE’s Metadata team, and the scope to support greater reach and impact of institutional scholarship and collections.

This presentation will detail the University of Leeds Libraries' initiative to systematically broaden its coverage of Open Access (OA) books, a key component of a larger, institution wide project to promote Open Educational Resources (OER). Beginning in Spring 2025, the Metadata and Discovery team developed and implemented a proactive workflow to identify, select, and ingest high-quality OA titles into the library catalogue.

The primary objectives of this project were multifaceted:

• To provide library staff and faculty with comprehensive access to a wider range of OA resources, directly supporting the OER initiative's pedagogical aims.
• To enhance the data wrangling skills of the Metadata and Discovery team by engaging in new processes for large-scale data acquisition, transformation, and management.
• To integrate the Metadata and Discovery Team more deeply into the broader OER ecosystem.

We will discuss the project’s foundational context, and the challenges and successes encountered along the way. We will reflect on the key lessons learned, particularly regarding building a workflow that can be sustained as ‘business as usual’.

Open access books promise unrestricted access to knowledge, but unreliable, inconsistent, and siloed metadata that cannot be easily shared and re-used due to licensing restrictions prevents discovery and library integration at scale. Libraries and discovery platforms still struggle to ingest and expose OA book content properly because metadata is fragmented, incomplete, or incompatible with existing library workflows.

Emerging efforts such as Thoth’s metadata work, and activities happening in the COMET & Barcelona Declaration communities of practice, etc. are making progress – but more concerted efforts are needed in the books space so that publishers, platforms, and libraries can move from pilot projects to scalable practice. And as an extension to this approach, the idea of a collaborative, shared open metadata cataloguing commons comes into view.

This panel brings together metadata experts with different backgrounds to diagnose these systemic metadata failures, from ingest pipelines to adoption of PIDs and controlled vocabularies in the books space. Panelists will also seek to discuss benefits of, and potential barriers to, establishing such a shared open metadata commons – while also proposing practical, community-led solutions that make OA books findable, and re-usable within as well as outside of library systems.