***With apologies for cross-posting***


Please join the ALA Core Bibliographic Conceptual Models (BCM) Interest Group for its Core IG Week session. The BCM session will take place on Thursday, March 10, 2022, at 11am ET (8am PT, 10am CT, 16:00 UTC) and will consist of two presentations (see below for program details). 


The session is free and open to all, but registration is required. Register for this session at: https://ala-events.zoom.us/webinar/register/WN_7d2Ocd7KTTSlv-ANhFamSw 


Program details: 


Aggregator Ways to Account for Different Conceptual Models

Michael Phillips, OCLC

Work, Expression, Manifestation, and Item (WEMI) from the Functional Requirements for Bibliographic Records (FRBR) and now the Library Reference Model (LRM) have caused implementation issues for those creating bibliographic data, whether record-based, like MARC, or graph-based, using an RDF serialization. Specific to graph-based infrastructures many models have been proposed and enacted with three mainstream instantiations in Library of Congress’s BIBFRAME, Casalini Libri and @CULT’s Share-VDE, and OCLC’s Shared Entity Management Infrastructure (SEMI). With the goal of sharing and aggregating linked data, different models contain challenges for aggregators. 

This presentation will discuss the current conceptual models used by different library communities. Prior presentations for different events have compared LRM, BIBFRAME, Share-VDE, and SEMI. In March of 2021, OCLC presented its model and a comparison of all four. This presentation provides an update to these models and the evolution of aggregating data to fit them. 

A brief overview of the OCLC SEMI model will be followed by a look at the aggregation of data through several examples, comparing data points that make up the WEMI structure. The goal is to give the audience an understanding of working at scale from hundreds of sources which are primarily MARC record-based, and to discuss the potential for making better connections in the data by enriching MARC records prior to entity extraction with appropriate identifiers. These identifiers will help connect the dots between all the pieces of data, whether they are in MARC or in some other format.  


Modeling sequential relationships beyond MARC 21: Mapping preceding and succeeding entity relationships for Continuing Resources

Abigail Sparling, University of Alberta Library; Charlene Chou, New York University Libraries; Ian Bigelow, University of Alberta Library; Everett Allgood, New York University Libraries


Work on Share-VDE MARC to BIBFRAME conversion is an iterative process. Anticipating a forthcoming Share-VDE Sapientia BIBFRAME Knowledge Base update, the Sapientia Entity Identification Working Group (SEIWG) has been defining BIBFRAME entity creation and relationship management rules for the MARC 21 linking fields (i.e., 76X-78X). Refining these conversion and modeling processes will allow library catalogs to leverage linked data’s ability to clearly express relationships between entities and to encode these relationships more explicitly for machines, users, and catalogers.


This presentation describes the work of a serials-focused SEIWG subgroup to define the sequential relationship rules for the 780 (Preceding entry) and 785 (Succeeding entry) tags. Noting the complexity of many sequential relationships captured in MARC, the subgroup analyzed how this data might accurately be expressed in BIBFRAME and other linked data ontologies. Results highlight that while the majority of 780/785 relationships can be mapped to BIBFRAME without a loss of information, there are a number of cases where alternative ontologies or creative mapping solutions are required. 


With the goals of retaining the quality of relationship data captured in MARC, while also harnessing the ability to explicitly link and establish entity relationships in linked data, several key areas surfaced during this investigation. For example, the ability to : 

·       Capture relationships involving three or more titles ; 

·       Preserve sequential relationships when relevant data is encoded in tags outside of 780/785 ;

·       Harmonize MARC to BIBFRAME conversion specifications for 780/785 relationships so that converted data interoperates with relationships encoded natively in BIBFRAME


It is hoped that this presentation will inform further steps for the development of the BIBFRAME ontology and conversion processes for continuing resources as libraries work toward BIBFRAME implementation.


Friendly Regards,


Thomas Dousa (co-chair) | University of Chicago | [log in to unmask] 

Paloma Graciani (co-chair) | Harry Ransom Center | [log in to unmask] 

Nathan Putnam (co-vice-chair) | OCLC | [log in to unmask] 

Victoria Sokari (co-vice-chair) | Bayero University, Kano | [log in to unmask]