2026
BIBFRAME WORKSHOP IN EUROPE, Barcelona 2026
University of Barcelona
BIBFRAME Workshop in Europe
The aim of the BIBFRAME Workshop in Europe is to be a forum for sharing knowledge about the practice of, production with, and planning of BIBFRAME implementation.
We bring together people working in the transition from MARC to Linked Data using the BIBFRAME model and related tools.
The workshop areas are strongly focused on the practical implementation of BIBFRAME, not a theoretical Linked Data / Semantic Web event.
Programme
Tuesday, September 8
-
Works and Instances are the Model
The BIBFRAME model was first introduced in November 2012. That is when the notion that a Work – defined as the conceptual essence of something that may reference people or places or subjects - could relate to one or more Instances, which represent some material embodiment of the Work. BIBFRAME was developed, in 2012 as it continues to be now, to accommodate RDA cataloging but BIBFRAME was also designed, as it continues to be now, to be content model agnostic. It is not an accident that the BIBFRAME model, in 2012 like the one in 2026, does not contain classes for expressions or manifestations. Indeed, the simple complementary nature of Works and Instances remains the resilient core to BIBFRAME even while other components have been introduced to better handle the complexities of bibliographic description and RDA. This workshop session will explore how a model composed of two core entities, neither of which is hierarchically related to the other and both of which exist in a much larger graph of related entities, creates a basic foundation for description that allows flexibility when it comes to describing a variety of material, some of which may fit the RDA content model and some of which does not.
-
From Linked Data Cataloging to Linked Data Search
In 2018, the Swedish union catalog Libris made the leap from MARC to linked data, releasing its new linked data-based backend and cataloging tool. Until recently, however, the search service, through which most users come into contact with Libris, was still based on MARC. But in May 2026, the National Library of Sweden finally launched the new Libris linked data search service.
Libris is a service used daily by a wide range of users, from casual patrons to expert catalogers and API integrations. In this talk, we will discuss the challenges and possibilities of data modelling for a user-friendly search experience for a diverse audience when MARC is (almost) out of the picture.
When developing the new search interface, we wanted to shift away from presenting heavily interpreted MARC records toward exposing our BIBFRAME-based data in a form closer to its underlying structure. This shift has made the underlying data and data model more visible in the search interface, for better and for worse.
This exposes several longstanding challenges: material types described in expert terms unintuitive to many end users; cataloging rules and practices that have accumulated incrementally, generating inconsistencies and redundancies; untapped potential with regards to relationships between bibliographic entities; and standards that have not yet been adapted to accommodate emerging digital formats.
How can we unveil the power and usefulness of linked data to cataloguers and end users? How do we translate user feedback into the most meaningful improvements that leverage the possibilities of linked data?
Libris is now both a living union catalog and a practical testing ground for our BIBFRAME-based data model and for linked data search service design.
-
Bringing Linked Data into Daily Practice: JCricket as an Operational Entity Editor
The Share Family community has developed JCricket, the Linked Data Entity Editor designed for collaboratively curating linked data entities in a multi-institutional, shared space.
This presentation will show how JCricket is being further enhanced to streamline libraries’ operational workflows and real-world cataloguing practices: several features enable effective, granular cataloguing and Provenance-based data management, in addition to new functions designed to support cataloguers in validating data input and correctly applying BIBFRAME-based metadata.
Also, its ability to integrate with local library systems is a key factor of this cooperative cataloguing environment: data entered in libraries’ ILS/LSP is reflected in the central Cluster Knowledge Base underlying JCricket and vice-versa, thus allowing for a full roundtrip of data.
Since JCricket operates on a multi-Provenance dataset, its inherent complexity requires guidelines: a set of best practices is being elaborated within the Share Family community to consolidate shared cataloguing workflows and ensure consistent data integration with external systems.
Besides its standard application, the LOD Platform technology can expand metadata management and transformation processes to formats and types of resources other than MARC-based bibliographic descriptions, such as TEI - Text Encoding Initiative and others. This flexibility encompassing an extended range of use cases is an added value worth sharing with the whole linked data community.
-
Using BIBFRAME data at scale
Libraries around the world are at various stages in their transition to BIBFRAME. Some are learning the impact and benefits of leveraging BIBFRAME data; many are experimenting with BIBFRAME data in their cataloging and discovery workflows; and others are ready to fully productionize BIBFRAME data across their metadata management ecosystems.
When it comes to widespread adoption and implementation, the need for scalable, consistent, and quality data becomes the cornerstone. And while the data formats and workflows evolve, the global infrastructure to support such an important transition already exists. For nearly 60 years, WorldCat has provided the foundation upon which the global library community creates, manages, and shares knowledge—evolving every step of the way as library collections and needs of the communities they serve change.
Building on OCLC’s history of collaboration with the library community on metadata management and linked data innovation, we’re excited to share the next phase of WorldCat’s evolution. In this presentation, we’ll share updates on WorldCat <> BIBFRAME interoperability, including the technical challenges and opportunities, workflows for managing data in both MARC and BIBFRAME, and how this data can be used in various BIBFRAME management services.
-
The System is Hybrid – The People Are Too: Rebuilding Australia’s National Union Catalogue with BIBFRAME
In 2025, Australia’s national union catalogue went live with a new BIBFRAME-based metadata engine at its heart. This once-in-a-generation infrastructure renewal project replaced the backbone of the Australian National Bibliographic Database (ANBD), a network that aggregates more than 41 million metadata records from over 900 partner organisations. While the project was initially conceived as a technical transformation to introduce BIBFRAME, the experience demonstrated that success depended as much on investing in people as in technology.
The implementation presented a unique challenge: introducing a BIBFRAME-based national infrastructure in a country where no large or medium-sized libraries had yet adopted BIBFRAME. Currently, metadata contributions continue to arrive primarily in MARC via automated workflows, while the new platform operates on a BIBFRAME foundation. Resource sharing, partner integrations, and the broader library ecosystem remain heavily MARC-dependent, requiring interoperability across standards and technologies. From the outset, the system needed to be genuinely hybrid, with the resulting architecture functioning as a transformation layer to support ongoing MARC-based operations while enabling a gradual transition toward a linked data future.
Building this environment required significant investment in internal capability. With limited local BIBFRAME expertise available, the National Library of Australia focused on developing its own workforce’s knowledge and technical skills, drawing heavily on international implementations and established standards while adapting the model to Australian requirements. The project also needed to preserve metadata elements valued within the Australian context, leading to the development of local extensions and hybrid workflows that combine BIBFRAME and selected MARC data elements.
However, the most significant challenge was not technical but organisational. While the National Library could invest in developing internal BIBFRAME capability, the broader Australian library sector was at a very different starting point. Feedback from partner organisations consistently highlighted limited awareness and understanding of BIBFRAME, particularly among staff whose day-to-day work remained firmly MARC-based. At a time of constrained budgets, workforce pressures, and limited opportunities for professional development, introducing a new metadata standard risked being perceived as an additional burden rather than an opportunity for innovation.
To support sector readiness, the National Library adopted a change-management approach that prioritised engagement, communication, and confidence-building alongside technical implementation. Working in partnership with ALIA and key cataloguing stakeholders, the Library delivered introductory “BIBFRAME Downunder” sessions, established advisory groups, hosted 26 national webinars, and developed a suite of self-paced learning resources. The emphasis was not on teaching the technical intricacies of BIBFRAME, but on helping libraries understand why the change mattered, what benefits it could deliver, and how a gradual transition could be managed within existing operational constraints.
This presentation explores the technical, organisational, and cultural lessons learned from implementing a BIBFRAME-based national union catalogue in a predominantly MARC-based ecosystem. Australia’s experience demonstrates that successful BIBFRAME adoption is not solely a metadata challenge but a sector-wide transformation that depends on supporting both systems and people through a hybrid future.
-
BIBFRAME and Open Metadata Sharing - Solving the Gordian Knot of Interoperability
The University of Alberta Library recently launched our Open Metadata Sharing Framework, and then promptly shifted gears to migrating to a new library services platform. UAL now needs to refocus on both our approach to BIBFRAME (BF) and open data. With so much work on BF, the importance of open metadata should not be understated. Indeed, since the formation of online catalogs, resource description standards and practice have been built on supporting interchange and open sharing of data.
Varied approaches to bibliographic metadata re-use and restrictions present challenges in MARC based workflows, creating something of a gordian knot like challenge for libraries to resolve, and it would be unfortunate to carry these challenges forward into linked data ecosystems. BF and linked data more broadly are very different from the MARC based record structure, both in terms of the approach to re-use in such an interconnected / “linked” ecosystem, but also because of the shift from a record based description to an entity-based one leveraging RDF and the need to account for metadata re-use policies at a more granular level.
This presentation will give an overview of existing challenges related to open metadata for libraries through the lens of the UAL Open Metadata Framework, but then extend the discussion to include challenges on the road ahead for BF as linked open data. Will any solutions to this gordian knot holding back BF as linked open data emerge?
-
From MARC to BIBFRAME in a Hybrid Metadata Ecosystem: The School of Salamanca Digital Library as an Enriched Linked Open Data Case using Marva Editor
The transition from MARC to BIBFRAME marks a shift towards an entity-based model aligned with Linked Open Data (LOD). As noted by the ThinkEPI Group, this process is hybrid and incremental.
This paper presents the School of Salamanca Digital Library, developed by the Fundación Ignacio Larramendi, as a case study of BIBFRAME implementation. It examines the conversion of MARC records into BIBFRAME using Marva Editor, with a focus on semantic enrichment through external vocabularies and links to Europeana.
BIBFRAME improves the linking of bibliographic and authority data through URIs and explicit relationships, enhancing interoperability and reuse. The study also explores the potential of AI-assisted workflows in metadata transformation. Results confirm the value of BIBFRAME for representing complex intellectual networks in a hybrid metadata environment.
-
From MARC to Linked Data: Institutional Strategies and Lessons Learned from a Large-Scale BIBFRAME Implementation in Latin America
The transition from MARC21 to Linked Data has become an operational challenge requiring institutional strategy, metadata governance, interoperability, and organizational transformation. This presentation shares the experience of the Universidad de Concepción (Chile) in implementing BIBFRAME at institutional scale, becoming one of the first academic libraries in Latin America to move from experimental testing to operational deployment.
The project included the assessment, normalization, and transformation of nearly two million bibliographic and authority records into BIBFRAME/RDF environments. Beyond technical migration, the initiative required redesigned cataloguing workflows, metadata transition policies, authority control strategies, and staff training focused on Linked Data and semantic interoperability.
A central component was the adoption and testing of the Linked Data Editor (LD Editor), based on MARVA and developed collaboratively within the FOLIO ecosystem. The implementation also incorporated RDF-based discovery through BiblioGraph and interoperability with external sources such as Wikidata and VIAF.
The presentation will discuss implementation strategies, challenges, and lessons learned, offering a practical framework for libraries beginning their transition toward sustainable Linked Data environments.
-
BCP 47 Language Tags in BIBFRAME: Increasing internationalization
The Library of Congress’s adoption and expanded use of BCP47 language and script codes to describe strings and labels is likely of interest to the participants of this annual European workshop. The Library of Congress began experimenting with recording language and script for individual labels in BIBFRAME in September 2024. This was a marked departure in bibliographic cataloging. Language codes have long been used in cataloging but generally in the service of recording the language in which the whole resource is presented, not to identify discrete strings or labels in the bibliographic description. The catalyst for the Library was the need for a Linked Data friendly solution to include multi-script access points in BIBFRAME-to-MARC converted records. This presentation will detail the problem the Library initially sought to solve, and describe why and how leveraging BCP47 codes were key components of the solution. It will also cover how the Library has expanded the use of language codes, both in BIBFRAME and introducing them to MARC. The adoption of BCP47 codes both in LC’s BIBFRAME and the wider bibliographic community (notably by MARC users) has been swift and a decisive step in the direction of greater internationalization of bibliographic description.
-
Linked Library Data transformation pipeline using local LLM to support feature vector methods in work clustering and authority record reconciliation
We present a novel approach to MARC 21 to BIBFRAME conversion utilizing the traditional algorithm based approach augmented by a local large language model tackling the challenges caused by irregularities in the source data and the clusterization of work-level entities. We describe a fully-local pipeline using a large language model to convert ~800,000 MARCXML records from the Finnish capital city area public libraries to ensure smooth transition to linked data.
Our solution addresses two failure modes of FRBR clustering and authority reconciliation where symbolic methods have difficulty to scale: (1) Work clustering across translations, editions, anthologies, and transliterations, where titles, languages, and uniform-title headings diverge but the underlying Work is the same; and (2) reconciliation against subject, authority, and value vocabularies where bilingual labels, fictional vs. real persons, same-name collisions, and cataloguing leftovers from ontology migrations confound rule-based linking.
A cascade architecture combines cheap heuristics, multilingual feature-vector embeddings under approximate nearest-neighbour search, and for the uncertain middle band, a structured-output local LLM judge emitting work-similarity decisions with confidence and rationale. The pipeline doubles as a corpus stress-test surfacing data-quality issues as actionable batches for cataloguer fix-up.
Wednesday, September 9
-
Aggregates in BIBFRAME: Two Approaches
From this year on, Finnish libraries apply Official RDA cataloguing rules. The Official RDA includes the concept of aggregate as a new approach to resources like collective works or multipart monographs.
RDA offers several ways of recording aggregate resources. In this presentation we’ve chosen the following options: 1) recording just an aggregate manifestation and at least one aggregated resource and 2) recording work, expression and manifestation of an aggregating resource. We focus on multipart monographs modelled as a bf:Series. In general, this approach is also applicable to other types of collection aggregates, but they may require some additional modelling.
A light approach is to create the aggregate manifestation as a BIBFRAME Instance. The only required addition to BIBFRAME is the bf:Aggregate as a subclass of bf:Instance. The aggregate manifestation includes the title of the series, relations to each of the individual bf:Works in the series, media and carrier types representing the whole series, as well as other typical properties of bf:Instance.
As a more thorough approach, we built an alternative model utilizing the Finnish BIBFRAME adaptation BFFI, where we have separated the bf:Work into bffi:Work and bffi:Expression. This allowed us to build also the aggregating expressions and works allowing us to explicitly record information such as genreform, language and intended audience of the aggregate series. The result is richer but requires more additions to BIBFRAME.
-
Mapping from RDA to BIBFRAME
The RDA Steering Committee (RSC) is the managing body for the RDA cataloging standard. In March of 2025, the RSC reached a protocol agreement with the Share Family / Share-VDE Advisory Council. As part of the RSC’s continuing work to develop RDA as a dynamic and responsive standard that can be used by various communities, including the BIBFRAME community, part of the RSC’s action plan for 2026 and 2027 is to develop a mapping from RDA to BIBFRAME. This presentation will give a brief update on news from the RSC over the past year about the development of RDA. The presentation will then discuss the RDA-to-BIBFRAME mapping project established as a collaboration between the RSC Technical Working Group and Share-VDE. It will give an update on the project, as well as an overview of the technical issues involved in such a mapping. Issues to be discussed include:-the differences between RDA and BIBFRAME, particularly in the structure of their classes and properties-a description of the different RDA element sets, and which ones are appropriate for a mapping to BIBFRAME-the syntactical issues in representing a chain of BIBFRAME classes and properties in a mapping-the different RDA recording methods, and how these relate to mapping to BIBFRAME-the similarities and differences between mapping RDA to BIBFRAME and previous mapping projects (RDA to MARC21, RDA to LRM, etc). The presentation will provide the audience with an update of the current state of the mapping project, as well as an understanding of the technical issues involved in the project.
-
Modelling Shared Print Actions
Research libraries have recognised that the long-term collective management of physical collections requires national coordination and shared infrastructure. In Sweden, this is set to be realised through Shared Print Sweden. This presentation outlines how Shared Print Actions (SPA) and MARC field 583 can be modelled as linked data, especially focusing on flexibility, scalability and item/holding entities. While the 583 notes in MARC have been enhanced with guidelines and a controlled vocabulary, its structure reflects a compromise where fundamentally different types of actions are expressed within a single framework. This creates challenges when translating the SPA model into a coherent linked data representation.
-
Mapping KCR5 as content rules to BIBFRAME
Cataloging rules as content standards and encoding formats based on the LRM conceptual model should be closely interrelated to ensure effective representation and organization of resources. Numerous studies have examined the mapping among these standards, including mappings from conceptual models to cataloguing rules, from cataloging rules to encoding formats, and from conceptual models to encoding formats. Examples include studies on mappings between LRM and RDA, RDA and BIBFRAME, as well as between LRM and BIBFRAME. Last year, the Korean Cataloging Committee developed KCR5 (previous KCR4) as content rules based on the original RDA. Accordingly, this study aims to evaluate the suitability of KCR5 for the exchange and sharing of bibliographic data within the BIBFRAME frameworks by mapping KCR5 to BIBFRAME.
-
Mapping Iranian Music Information Elements to the BIBFRAME Model: Identifying Gaps and Proposing Model Extensions
The transition from traditional cataloging formats such as MARC to Linked Data using the BIBFRAME model offers unprecedented opportunities for richer representation of cultural resources. However, Western metadata models often face significant limitations when describing cultural resources from non-Western contexts with unique conceptual structures. Iranian music, with its “radif”-based, “dastgah”-oriented, and “gusheh”-driven tradition, stands as a prominent example of this challenge; its foundational concepts—including “dastgah” (modal system), “gusheh” (melodic unit within a “dastgah”), “radif” (repertoire/canon), “maqam” (melodic mode), and the complex relationships among performers, composers, and instruments—cannot be fully represented within current metadata frameworks.This study was designed and executed in two phases to bridge this gap. In the first phase, fundamental information elements and conceptual structures governing this musical tradition were identified and classified through text analysis and interviews with Iranian music specialists. These elements include structural features (such as “dastgah” and “gusheh”), performance aspects (such as solo performance and ensemble performance), traditional musical instruments, and inter-versional relationships among different performances of a single work. In the second phase, these elements were compared against the classes and properties of the BIBFRAME model using qualitative mapping.The mapping results reveal that some Iranian music information elements either lack direct equivalents in BIBFRAME or require a combination of multiple classes and properties at a level of complexity that renders practical application difficult.Based on these findings, a set of proposed classes and properties for extending the BIBFRAME model has been developed. These proposals maintain compatibility with the core BIBFRAME structure and Linked Data principles while enabling more accurate description of Iranian music resources. These extensions are applicable not only to the cataloging of Iranian music materials but may also serve as a model for adapting BIBFRAME to other non-Western musical traditions.This presentation, while outlining the two-phase research methodology, focuses on the practical aspects of adapting BIBFRAME to a non-Western musical tradition. The findings demonstrate that fulfilling BIBFRAME's promise of representing cultural diversity requires continuous model revision and extension based on contextual needs. Situated at the intersection of modern bibliographic standards and musical cultural heritage, this research can serve as a foundation for developing BIBFRAME application profiles in the domain of world music.
-
The Alma Linked Open Data Editor: Advancing BIBFRAME Workflows in Production
This presentation introduces the Alma Linked Open Data (LOD) Editor and showcases recent product advancements supporting BIBFRAME-based cataloging in production environments. The session highlights how the LOD Editor enables creation, editing, and management of linked data within Alma, while supporting hybrid MARC and BIBFRAME workflows and gradual adoption. A brief live demo illustrates key capabilities, including entity-based editing, workflow integration, and interoperability considerations. The presentation concludes with insights from ongoing collaboration with the library community and outlines how current and upcoming enhancements are shaping practical BIBFRAME implementation.
-
From Records to Relationships: transforming Public Library Cataloguing in Flanders
Two years ago, we announced our decision to transition the public library network in Flanders from a centralized cataloguing system based on MARC 21 to a model grounded in entity-relationship principles. Today, that transition has become a reality, and we’d love to share our experiences with the community.In this presentation, we will reflect on our journey: how we approached the data conversion process, which strategic and technical choices were made? We will also discuss how cataloguers have adapted to this new way of working and consider a broader question: has MARC really died (in Flanders)?Beyond reflecting on the past, we aim to look ahead. What would we do differently with the knowledge we have now? Which successes can already be highlighted? And what challenges remain for the coming years?A key concern is ensuring compatibility with ongoing international developments. Are our data modelling choices aligned with emerging standards, or do they diverge? Public libraries often have different clustering needs than academic institutions, which has led us, at times, to deviate intentionally from frameworks such as RDA or BIBFRAME. We welcome a dialogue on these decisions and their implications for the future of bibliographic data.
-
Moving Marva beyond monographs
In 2024, catalogers at the Library of Congress began using Marva, the Library’s BIBFRAME editor, to catalog monographs and by spring 2025 all the catalogers were in production with BIBFRAME. After much consultation and development, Marva has evolved into a well-rounded tool for cataloging monographs. In 2026, the Library began working with catalogers of other types of material to adapt Marva for their needs. This presentation will share some challenges faced with the addition of maps, music scores, sound recordings, and videorecordings to Marva.
-
From the Shadows: Building Archival Entity Workflows Toward BIBFRAME
As libraries and cultural heritage institutions build toward BIBFRAME and linked data environments, archival and special collections present a distinct set of challenges that differ from traditional bibliographic workflows.
This presentation focuses on the development of an early-stage workflow for entity work using the OCLC Meridian editor at the University of New Mexico, focusing on regional archival and Latin American/Iberian materials as part of participation in the PCC Entity Management Cooperative (EMCO) project. The work centers on identifying, normalizing, and reconciling entities across materials with significant descriptive variation, while establishing consistent patterns that can support broader linked data implementation.To support this process, the project also draws on experimentation with APIs and OpenRefine to parse unstructured ingest data and align it with established vocabularies such as LCSH. These methods highlight both the technical and interpretive challenges involved in preparing archival data for use in BIBFRAME and linked data environments.
Rather than presenting a fully implemented system, this session focuses on the development of a cohesive, iterative workflow prior to broader implementation. By sharing lessons learned from this early-stage work, the presentation offers a practical perspective on how institutions can approach entity management and workflow design in ways that support reuse, consistency, and future BIBFRAME integration.
-
Filling the Gap: Metadata Maker as a Practical Tool for BIBFRAME Data Creation
As implementation of BIBFRAME, the emerging metadata standard, continues to advance, there remains a notable lack of simple tools for creating BIBFRAME data. Metadata Maker, developed in 2015 and updated continuously with new templates and data output options, now includes a current version of BIBFRAME output for Monographs, Monographs-LD (Linked Data), and Electronic Book templates. To support BIBFRAME interoperability, Metadata Maker produces two types of BIBFRAME output: one for standard BIBFRAME and one for Alma, one of the Library Service Platforms. As MARC becomes one of many metadata formats rather than the sole standard, direct BIBFRAME data generation from source information is increasingly critical.This presentation will share the mappings used for producing BIBFRAME in Metadata Maker and demonstrate how the tool benefits users without prior knowledge of BIBFRAME or library metadata, making it especially valuable for libraries managing materials in special languages or subject areas where specialist cataloging staff may not be available. The session will also examine the metadata profiles of available modules and discuss the challenges of designing transformation workflows tailored to different resource types while maintaining alignment with BIBFRAME. Metadata Maker's decade-long evolution offers a compelling example of how an open-access, open-source metadata tool can remain relevant amid ongoing shifts in metadata standards and information landscape.
-
Bridging the Skills Gap: Capacity Building and Training Frameworks for Effective Transition from MARC to BIBFRAME in Diverse Library Environments
Bridging the Skills Gap: Capacity Building and Training Frameworks for Effective Transition from MARC to BIBFRAME in Diverse Library EnvironmentsAbstractThe shift from Machine-Readable Cataloguing (MARC) to Bibliographic Framework Initiative (BIBFRAME) represents a significant transformation in how libraries manage metadata. While MARC has been the dominant system for over 50 years, its design is rooted in outdated technologies and does not align with the needs of modern web-based environments. BIBFRAME, introduced by the Library of Congress, aims to modernize cataloguing practices to support linked data and the semantic web. This transition, however, presents unique challenges, especially for libraries in developing countries such as Nigeria. The limited technological infrastructure, inadequate personnel training, and budget constraints are just some of the obstacles faced by these institutions. As such, a structured, tiered approach to capacity building is necessary to ensure that libraries can effectively navigate the transition to BIBFRAME without being left behind in this global shift.The core distinction between MARC and BIBFRAME lies in their underlying data models. MARC is based on a flat, record-oriented structure, which is difficult to adapt to the modern, interconnected world of the web. In contrast, BIBFRAME uses a more flexible, entity-relationship model that makes metadata more accessible, reusable, and interoperable across platforms. The BIBFRAME model breaks down bibliographic data into core entities such as Work, Instance, Item, Agent, and Subject. This structure allows for more dynamic, web-friendly cataloguing that can support the growth of linked data technologies. While the advantages of BIBFRAME are clear, libraries in developing regions lack the necessary skills and infrastructure to make this transition. Many cataloguers are unfamiliar with linked data concepts such as RDF (Resource Description Framework) and OWL (Web Ontology Language), and the libraries themselves often lack BIBFRAME-compatible systems. These issues are compounded by the absence of institutional policies, budget allocations, and curricula that incorporate modern cataloguing techniques.To address these challenges, this paper proposes a tiered capacity-building framework designed to support libraries at different stages of readiness. The framework consists of three levels: foundation, intermediate, and advanced. At the foundation level, the training is aimed at all library staff and Library and Information Science (LIS) students. It focuses on introducing core concepts of linked data and the differences between MARC and BIBFRAME. Basic tools, such as MARCEdit and the BIBFRAME Editor, are also introduced at this stage. The intermediate level targets practicing cataloguers and systems librarians, offering hands-on experience with RDF, OWL, authority file integration, and cataloguing within a BIBFRAME environment. This level also includes training on linked data publishing, SPARQL queries, and migration workflows. At the advanced level, senior librarians and LIS educators are trained on policy development, change management, and institutional transition planning. This level also involves training-of-trainers (ToT) methodologies to ensure that knowledge can be disseminated widely within the institution.The implementation of this framework is structured into four phases. The first phase, lasting three months, focuses on needs assessment, planning, and resource mobilisation. This phase includes skills audits, stakeholder mapping, and securing buy-in from key institutions. The second phase, spanning months four to six, focuses on curriculum development, including the design of modular training content and pilot testing. The third phase, from months seven to twelve, involves the delivery of training through workshops, online courses, hands-on labs, and peer learning networks. The final phase, starting from month thirteen, includes monitoring, competency assessment, and certification, along with a focus on sustainability through continued advocacy, policy development, and scale-up efforts.The paper draws lessons from global BIBFRAME transitions, highlighting the experiences of leading libraries such as the Library of Congress (USA), the British Library (UK), and NACSIS-CAT (Japan). These examples demonstrate the importance of strong institutional leadership, domain-specific piloting, and cultural adaptation. For example, the Library of Congress has been at the forefront of the BIBFRAME initiative, investing in research and development and establishing a national linked data service. The British Library's pilot projects, particularly for music and rare materials, illustrate how specific domains can benefit from early experimentation before full-scale adoption. In Japan, NACSIS-CAT developed national linked data cataloguing standards that were adapted to local language requirements. These examples highlight the importance of adapting global frameworks to local contexts, which is especially relevant for non-English-speaking environments like Nigeria.In conclusion, the transition from MARC to BIBFRAME is not just a technical upgrade; it represents a fundamental shift in how libraries approach metadata and knowledge sharing. The proposed tiered training framework provides a clear, phased approach for libraries to adopt BIBFRAME, even in resource-constrained environments. By addressing the skills gap at multiple levels, this framework aims to ensure that libraries in Nigeria and other African countries are not left behind in this global transition. However, for the transition to succeed, it is crucial that institutions, professional bodies, and government agencies collaborate to provide the necessary training, infrastructure, and policy support. BIBFRAME readiness should therefore be seen as a strategic priority for the future of metadata-driven library services in Africa.Keywords: MARC, BIBFRAME, linked data, cataloguing, metadata transition, capacity building, Nigerian libraries, semantic web, library education, digital transformation.
-
¿BIBFRAME en español? From Pilot Workshop to Open Training Infrastructure for Linked Data Adoption
BIBFRAME adoption is not simply a change in cataloging practice or an exchange of one standard for another: it represents a new sociotechnical ecosystem. For professionals whose working language is not English, this challenge is compounded by a lack of educational resources and documentation, reflecting a structural gap with real consequences for who can participate in shaping the future of bibliographic metadata.
This presentation reports on a multiyear initiative that grew out of a 2024 pilot at the Universidad Autónoma de San Luis Potosí (UASLP), Mexico. Led by Yale Library metadata professionals, this initiative seeks to build open, reusable Spanish-language infrastructure for BIBFRAME learning and adoption. The work has three facets: (1) an open, modular Spanish-language curriculum covering linked open data principles and the BIBFRAME model; (2) Spanish translations of BIBFRAME 3.0.1, BIBFRAME LC Extension 3.0.0, and MADS/RDF 1.5.0, with a bilingual glossary; and (3) a train-the-trainers model, anchored in a three-day Yale workshop in June 2026 and to be applied at UASLP during a one-week online continuing education course in September.
BIBFRAME is a community-driven, international standard, and its global relevance depends on participation reaching beyond any single linguistic or geographic context. Accessible Spanish-language infrastructure contributes to future adoption, widening the community of practitioners who can shape the standard.
-
Transitioning from the legacy constraints of MARC21 to the dynamic possibilities of BIBFRAME presents unique hurdles for smaller institutions. At Brandeis University, our journey toward linked data has focused on incremental integration, from pilot-testing various editors to fostering department-wide engagement. This presentation examines the practical challenges of a small technical services department, specifically the 'dual-track' burden of maintaining production-level MARC21 workflows while simultaneously developing BIBFRAME competencies. We will discuss strategies for cultivating staff interest, the role of community-driven support through groups such as the PCC Sinopia Affinity Group and LD4, and how moving toward BIBFRAME promises to revolutionize discoverability for smaller academic collections. We explore the reality of the learning curve, the assessment of various linked data editors, and the specific management strategies used to balance existing workloads with pilot projects.
Bridging the Gap: Implementing BIBFRAME in Small-Staff Environments
Proceedings
Pre-Conference
A Pre-Conference about the Catalan and Spanish BIBFRAME context is planned for 7 September 2026 afternoon. This session is linked to the local GLAM-Wiki Gender 2026. See the pre-conference program here. The pre-conference sessions will be held in Spanish and Catalan languages.
A Pre-Conference about the Catalan and Spanish BIBFRAME context is planned for 7 September 2026 afternoon. This session is linked to the local GLAM-Wiki Gender 2026. See the pre-conference program here. The pre-conference sessions will be held in Spanish and Catalan languages.
Registration
The BFWE meeting and the pre-conference will be held in a hybrid format and are free to attend, but registration is required. Please register here. In-person registration will be closed on August 30th.
All BFWE sessions will be recorded, and the recordings together with the presentation slides will be made available on the conference website a few days after the workshop.
The BFWE meeting and the pre-conference will be held in a hybrid format and are free to attend, but registration is required. Please register here. In-person registration will be closed on August 30th.
All BFWE sessions will be recorded, and the recordings together with the presentation slides will be made available on the conference website a few days after the workshop.
Organiser Group
Andreas Andersson, National Library of Sweden
Michele Casalini, Casalini Libri and Share Family
Núria Ferran Ferrer, University of Barcelona (Host 2026)
Matias Frosterus, National Library of Finland (Spokesperson)
Ignasi Labastida i Juan, University of Barcelona (Host 2026)
Nancy Lorimer, Stanford University Libraries and LD4P
Hannes Lowagie, Royal Library of Belgium
Sally H. McCallum, Library of Congress - NDMSO
Bjørge Vestli, National Library of Norway
Andreas Andersson, National Library of Sweden
Michele Casalini, Casalini Libri and Share Family
Núria Ferran Ferrer, University of Barcelona (Host 2026)
Matias Frosterus, National Library of Finland (Spokesperson)
Ignasi Labastida i Juan, University of Barcelona (Host 2026)
Nancy Lorimer, Stanford University Libraries and LD4P
Hannes Lowagie, Royal Library of Belgium
Sally H. McCallum, Library of Congress - NDMSO
Bjørge Vestli, National Library of Norway
Practical information
Location
The conference will take place at the
Faculty of Information and Audiovisual Media, Universitat de Barcelona, Carrer de Melcior de Palau, 140, 08014 Barcelona, Spain.
The Faculty of Information and Audiovisual Media is located within a short walk from major public transport stops. Many key areas of the city are easily accessible on foot. The nearest metro station is
Plaça de Sants (L1, L5), just a few minutes' walk from the venue.
Recommended Hotels
The following hotels are located within walking distance of the conference venue at the University of Barcelona and are recommended for participants:
- Abba Sants Hotel: Carrer Numància, 32, Sants-Montjuïc, 08029 Barcelona, Spain (5 minutes’ walk from the venue).
- Barceló Sants Hotel: Plaça dels Països Catalans, s/n, Sants-Montjuïc, 08014 Barcelona, Spain (8 minutes’ walk from the venue).
- Via Sants Hotel Barcelona, Tapestry Collection by Hilton: Carrer de Nicaragua, 5–7, L’Eixample, 08029 Barcelona, Spain (10 minutes’ walk from the venue).
- AC Hotel Sants: Passeig de Sant Antoni, 36–40, Sants-Montjuïc, 08014 Barcelona, Spain (10 minutes’ walk from the venue).
Location
The conference will take place at the Faculty of Information and Audiovisual Media, Universitat de Barcelona, Carrer de Melcior de Palau, 140, 08014 Barcelona, Spain.
The Faculty of Information and Audiovisual Media is located within a short walk from major public transport stops. Many key areas of the city are easily accessible on foot. The nearest metro station is Plaça de Sants (L1, L5), just a few minutes' walk from the venue.
Recommended Hotels
The following hotels are located within walking distance of the conference venue at the University of Barcelona and are recommended for participants:
- Abba Sants Hotel: Carrer Numància, 32, Sants-Montjuïc, 08029 Barcelona, Spain (5 minutes’ walk from the venue).
- Barceló Sants Hotel: Plaça dels Països Catalans, s/n, Sants-Montjuïc, 08014 Barcelona, Spain (8 minutes’ walk from the venue).
- Via Sants Hotel Barcelona, Tapestry Collection by Hilton: Carrer de Nicaragua, 5–7, L’Eixample, 08029 Barcelona, Spain (10 minutes’ walk from the venue).
- AC Hotel Sants: Passeig de Sant Antoni, 36–40, Sants-Montjuïc, 08014 Barcelona, Spain (10 minutes’ walk from the venue).
Contacts
For further information please contact us at conference@bfwe.eu.
For further information please contact us at conference@bfwe.eu.