RDA Forum: 2017 Midwinter Meeting

The 2017 Midwinter Meeting RDA Forum held on January 21 featured two speakers.

Kathy Glennan, ALA Representative to the RDA Steering Committee (RSC), spoke about the outcomes of the RSC’s meeting in November 2016, highlighting three major topics.

IFLA Library Reference Model

The RSC agreed to implement the IFLA Library Reference Model (LRM), currently awaiting final approval by the IFLA Committee on Standards. The draft version of this model went out for world-wide review last year under the title FRBR Library Reference Model. It consolidates and updates the three existing models, Functional Requirements for Bibliographic Records (FRBR), Functional Requirements for Authority Data, and Functional Requirements for Subject Authority Data. Like its predecessors, LRM is a high-level conceptual model, which uses an entity/relationship modeling framework. It emphasizes user tasks, rather than library operations. Unlike its predecessors, LRM is based on and fully compatible with the CIDOC Conceptual Reference Model (CRM), and with the object oriented version of FRBR (FRBRoo).

Implementing LRM in RDA entails modifying some existing entities, including Work, Expression, Manifestation, and Person. RDA will also need to accommodate new entities: Agent, Collective agent, Nomen, Place, and Time-span. Some of these changes will be minor; others will be significant. Of particular note is the change in the scope of LRM Person, which must be a real human being. This change is not popular with U.S. catalogers, but it is inherent in the underlying CRM. As a result, RDA will need to find new ways to capture data about fictitious, legendary, and non-human entities credited in statements of responsibility so that users can still find these resources.

Likewise, Place and Time-span exclude fictitious entities. Place encompasses both contemporary and historical places, on Earth or extra-terrestrial. The model recognizes that places can have changing boundaries over time. Time-span is a period of time with a measurable duration, no matter how short. It can be more or less precise, ranging anywhere from a geological era to a nanosecond.

Nomen is one of the more unusual entities in LRM. It is an association between an RDA entity and a character string that refers to it. The model notes that different entities can have the same nomen (e.g., multiple people with the name John Adams), and that the same entity can have different nomens (e.g., Mark Twain, Samuel Clemens, etc.).

LRM declares 37 attributes; these are representative, not exhaustive. Catalogers can record many attributes as a literal string or as a URI. Two newly-declared attributes are of particular interest in the model. The Representative expression attribute is a characteristic deemed essential in characterizing the Work, such as language. The Manifestation statement attribute helps users understand how a resource represents itself. It is normally transcribed from a Manifestation and includes strings like a Statement of Responsibility.

RDA Toolkit Restructure and Redesign (3R) Project - RSC’s Focus

Glennan spoke about the content changes anticipated in the upcoming RDA Toolkit Restructure and Redesign (3R) Project. Although some rewording revisions have taken place over the past year, the more significant changes require freezing the RDA Toolkit content for a year. Between April 2017 and April 2018 the RSC will not consider any community proposals for modifications of the instructions.

Along with incorporating the new and revised entities as part of implementing LRM, the 3R Project also allows the RSC to address long-standing problems, such as removing the instructions that say: “This instruction has been deleted as a revision to RDA.” The project also offers an opportunity to consider additional changes, including generalizing the instructions where possible; restructuring the layout; clarifying “transcribe” vs. “record” guidance; developing a new approach to relationship designators; and further developing guidelines for recording both pagination and foliation. Finally, the RSC anticipates fully implementing the “4-fold path” throughout RDA, which represents four different ways to capture data: unstructured description; structured description (such as authorized access points); identifiers; and URIs.

Although the RSC will not consider change proposals during the freeze, catalogers should still document problems and potential solutions for submission after April 2018. In addition, there will be some opportunities for community participation and feedback in relation to the 3R Project. The RSC will turn to its working groups and other experts for assistance and will seek input from RDA communities on some of the potential changes.

Community and Working Group proposals and discussion papers

Finally, Glennan summarized the outcomes of specific proposals and discussion papers considered at the 2016 RSC meeting. The 23 papers focused on internationalization, provenance, relationships, music resources, and rare resources. Of these, the RSC approved 15 either in full or in part, including the three ALA proposals (adding a controlled vocabulary for regional encoding, improving instructions for laws covering more than one jurisdiction, and providing greater flexibility in the construction of variant access points). The RSC referred the remaining documents to the 3R Project.

For more information, Glennan recommended consulting the RSC website and her recent report to CC:DA.

RDA Toolkit Restructure and Redesign (3R) Project – Publisher’s Focus

During the second half of the forum, James Hennelly, Director, RDA Toolkit, gave further details about the 3R Project. A driving goal is to have the Toolkit better meet the needs of its users and offer greater flexibility in the display of RDA instructions and supplementary resources, such as policy statements. To this end, ALA Publishing is working with the RSC on content changes, RDA developers to streamline the underlying data structure, and the 6-member 3R User Group to identify Toolkit improvements.

Over the summer, the RSC undertook efforts to better synchronize the content in the RDA Registry with the RDA Toolkit. Now updated definitions, for example, only need to be made in one place; this minimizes the need to make changes specifically for the Toolkit text. In addition, Hennelly will soon implement new translation software to support the growing adoption of RDA in non-English-speaking communities. He also demonstrated what the improved RDA Toolkit website will look like.

When the new version of the Toolkit rolls out in April 2018, it will include: a responsive design, which can be used on tablets; better accessibility, meeting the AA standard from the W3C; and more modular content, which will enable greater flexibility in displays, many of which will be user-driven. Hennelly identified the following areas for additional improvements: user-created content tools; shared ownership of documents; the administrative interface; eliminating the double login; and presenting the last-viewed page after a Toolkit time-out.

With the restructuring, the current instruction numbers will be replaced with a new numbering system that has yet to be developed. However, the current numbers will remain as search options, so catalogers can find existing instructions in their new locations.

The new Toolkit version will not have the following features: 1) the blue floating headers; 2) the print table of contents; 3) the print index; and 4) all search refinement metadata, except for the numerical label and the “core” designation.

Still under investigation are solutions to revision history, mapping, element display, and managing large tables and lists.

Hennelly closed by inviting attendees to participate in a more detailed discussion at the RDA Tech Forum, held on Monday January 23.

Reported by Kathy Glennan, University of Maryland