ALA Midwinter Meeting 2020, Philadelphia
Report by Karen Peters (Library of Congress), Chair, Encoding Standards Subcommittee
MARC Advisory Committee Meetings (January 25 and 26)
Over the course of two meetings, the MARC Advisory Committee (MAC) considered two proposals and seven discussion papers. Two of the discussion papers deal with changes in MARC that will be needed to accommodate new RDA elements introduced as part of the RDA Toolkit Restructure and Redesign Project (“beta RDA”), while one of the proposals and several of the discussion papers have their genesis in issues related to the development of BIBFRAME and the process of converting or migrating data between BIBFRAME and MARC. Both proposals passed unanimously; the future of the discussion papers is unknown at present.
Prior to consideration of the 9 papers, brief mention was made of two fast-track proposals approved by the MARC Steering Group after the ALA Annual 2019 last June. MARC Proposal 2019-FT01 (Adding a Code for Audio Belt in Field 007/01 of the MARC 21 Bibliographic Format) was approved on July 9, 2019 and published on July 12 as part of an addendum to MARC 21 Update No. 28. MARC Proposal 2019-FT02 (Defining Subfield $g in Field 751 of the MARC 21 Bibliographic Format), which originally appeared as Discussion Paper 2019-DP04 during the MAC Meetings at ALA Annual 2019 and makes it possible to add qualifying information to Geographic Names used as added entries, was approved on October 18, 2019 and published as part of MARC 21 Update No. 29 on November 26.
MARC Proposal No. 2020-01: Defining a New Indicator Value for Human-generated Content in Field 883 of the MARC 21 formats
This proposal, from the German National Library, follows discussion papers presented in 2017 and 2018 dealing with the need to improve the coding of metadata provenance information in MARC. Note that the proposal is not an attempt to address the needs of beta RDA in this regard but is instead a short-term solution designed to meet the needs of the German National Library until such time as the requirements of beta RDA can be accommodated. After discussion, the proposal passed unanimously with minor modifications, including changing the new First Indicator 2 from “Created by a human cataloger” to “Not machine generated.”
MARC Proposal No. 2020-02: Adding Subfield $0 to Fields 310 and 321 in the MARC 21 Bibliographic Format
This proposal, from the Library of Congress’s Network Development and MARC Standards Office (NDMSO), follows a Discussion Paper presented at 2019 Annual that considered the possibility of coding URIs associated with controlled terms in MARC Fields 310 (Current Publication Frequency), 321 (Former Publication Frequency), and 521 (Target Audience Note). Field 521 was dropped from the proposal, as Field 385 (Audience Characteristics) is already available to accomplish that particular goal. After some discussion, the proposal was initially deemed not ready for a vote due to a misunderstanding of NDMSO’s wish to make Field 310 (Current Publication Frequency) repeatable. Subsequent discussion, however, clarified that repeatability was sought only as a way to accommodate links from different sources in the process of migrating MARC data into BIBFRAME, or of converting BIBFRAME data into MARC records. Once this intent was understood, the proposal passed unanimously.
MARC Discussion Paper No. 2020-DP01: Modernization or Replacement of Field 856 in the MARC 21 Formats
This discussion paper, from OCLC and the German National Library, attempts to deal with the need to update Field 856 (Electronic Location and Access) due to changes in both its use (e.g., decline in the use of a single bibliographic record for print and electronic versions of a resource) and in technology (e.g., decline in the use of Telnet) over time. Discussion indicated no clear consensus as to which of the 3 possible options presented—redefine/reuse currently “unused” 856 subfields, leave Field 856 alone but define a new Field 857 to be used for Open Access URIs, or make Field 856 obsolete and define a new Field 857 to be used in its place—might be preferable/least problematic. There are a lot of moving parts to this discussion paper: should the idea of modernization or replacement of Field 856 be pursued further it will likely return as 2 or possibly 3 separate papers.
MARC Discussion Paper No. 2020-DP02: Adding Subfield $0 to Fields 504 and 525 of the MARC 21 Bibliographic Format
This discussion paper from NDMSO, like MARC Proposal 2020-02, grows out of work being done to convert MARC records into BIBFRAME data, as well as the current evaluation of BIBFRAME to MARC conversion, with an eye towards the introduction of data entry efficiencies and improved data consistency. Here the intent is to investigate the possibility of coding URIs associated with controlled Supplementary Content terms in MARC Fields 504 (Bibliography, etc. Note) and 525 (Supplement Note). Discussion of the paper raised suggestions for alternative fields for this information (for example, in the 3XX range), as well as an expression of concern from the legal community regarding possible data loss in Field 525, which is heavily used by that community.
MARC Discussion Paper No. 2020-DP03: Defining New Subfields in Field 340 to Record Illustrative Content and Sound Content in the MARC 21 Bibliographic Format
This is another discussion paper from NDMSO that grows out of work being done in the realm of MARC/BIBFRAME conversion. In this case, Sound Content, a binary choice between sound and silent, is of particular relevance to moving image resources, projected graphics, and their electronic equivalents. On the other hand, the non-applicability of Sound Content to recorded sound resources appeared not well understood by a number of those present who suggested that, rather than Field 340 (Physical Medium), Field 344 (Sound Characteristics) might be the appropriate place for this information—a notion that had been considered and rejected by the writers of the discussion paper, who recognized that Field 344 “is highly oriented to sound recordings.” Under the circumstances, it has been strongly suggested that OLAC and MLA be consulted on any future work done on this issue.
MARC Discussion Paper No. 2020-DP04: Renaming Field 345 and Defining New Subfields for Aspect Ratio and Motion Technique in the MARC 21 Bibliographic Format
This discussion paper, also from NDMSO, is another intended to facilitate conversion between MARC and BIBFRAME. Addressing as it does certain features pertinent to moving image resources, it was strongly suggested that OLAC be consulted on any further development of the ideas in this paper.
MARC Discussion Paper No. 2020-DP05: Reinstatement of Field 241 in the MARC 21 Bibliographic Format
This, another discussion paper from NDMSO and a controversial one, has its genesis in the Library of Congress’s interest in experimenting with the use of vernacular scripts with minimal romanization in BIBFRAME and the resulting conversion of this data into MARC. The discussion paper’s authors propose that Field 241 (Transliterated title) be reinstated for use in carrying the Romanized title proper of a resource, in which case Field 245 rather than Field 880 would carry information in the vernacular script of the resource. According to the paper, “Transliteration is not useful to end users and it is both costly to produce and difficult to consistently implement for vocalized scripts.” Discussion of the paper, however, indicated wide disagreement with this assertion, arguing instead that not only is it useful to end users in cases where use of a resource may not be predicated on a user’s ability to read a vernacular script (as is generally the case for music resources), but also to library staff who deal with these resources and may not have the ability or support needed to work with vernacular scripts. The paper’s representatives clarified that that the Library of Congress simply wants a chance to experiment with the possibility of doing away with most Romanization and that the 880 model would still be available for use by libraries that wished to use it. As it was clear that most communities do not want to lose the ability to employ Romanization, any further work in the direction of limited Romanization will need to clarify how, for example, use of Field 241 will interact with use of 880 fields.
MARC Discussion Paper No. 2020-DP06: Defining a New Field for Manifestation Statements in the MARC 21 Bibliographic Format
MARC Discussion Paper No. 2020-DP07: Recording the Extension Plan for Bibliographic Works in the MARC 21 Bibliographic and Authority Formats
These two discussion papers, both from the MARC/RDA Working Group, are the first of what will be many proposals intended to accommodate new beta RDA elements. 2020-DP06 accommodates a whole range of new Manifestation Statement (“A statement appearing in a manifestation and deemed to be significant for users to understand how the manifestation represents itself”) elements, while 2020-DP07 is intended to accommodate the new Extension Plan (“A categorization reflecting an intention to extend the content of a work”) element. Many questions were raised during the discussion; suggestions made in connection with both papers should assist in the redrafting of both papers as future proposals.
OCLC Linked Data Roundtable: Stories from the Front (January 25)
A video is available on YouTube: https://www.youtube.com/watch?v=5u2q6hUEW-c )
Nathan Putnam (OCLC) opened the roundtable with remarks on OCLC’s new Entity Management Infrastructure project, about which he spoke more fully at the LC BIBFRAME Update Forum; please see that report for details. His introduction was followed by three presentations on Linked Data projects:
In the first, Beyond Penn’s Treaty: A Digital Collections Linked Data Project, Sarah Horowitz (Haverford College) discussed the work involved in digitizing and building a web site for Penn’s Treaty, a collection of Quaker primary source documents. As it turned out, the developers found that digitization of the collection and the value added by the application of linked data went hand in hand. Note that this particular project has been largely dependent on the participation of undergraduate student workers from Haverford and Swarthmore Colleges.
Collection website: https://pennstreaty.haverford.edu (note: consideration is being given to rebuilding the site over the summer).
In the second presentation, Honor Moody (Harvard University) discussed the LD4 Wikidata Affinity Group. Note that Harvard is one of the cohorts in Linked Data for Production 2 (LD4P2), funded by the latest in a series of Mellon grants supporting the building of a pathway that will enable the cataloging community to transition to the use of linked data for description of library resources. The LD4 Wikidata Affinity Group is concerned with investigating how library metadata can be integrated with Wikidata to help improve the library discovery environment, and reciprocally, how libraries can help fill gaps in Wikidata. Along with past activity and future plans, Ms. Moody discussed ways in which others can participate in this work.
Affinity group site: https://wiki.lyrasis.org/display/LD4P2/LD4-Wikidata+Affinity+Group
In the final presentation, LC: a BIBFRAME Update, Kevin Ford (Library of Congress) outlined recent developments in BIBFRAME. Since ALA Annual, development in MARC to BIBFRAME conversion has been relatively minor, primarily bug fixes; at the same time, changes to the BIBFRAME editor have been made to improve cataloger experience. The BIBFRAME Pilot at LC has expanded and now includes 107 cataloger participants. Monthly meetings are held for the participants, during which project updates are communicated, changes to the editor or database are demonstrated, and participant feedback is elicited. Other topics touched on in the presentation, especially BIBFRAME to MARC conversion and BIBFRAME hubs, were discussed more fully during the LC BIBFRAME Update Forum and the Bibliographic Conceptual Models Interest Group meeting respectively; please see those reports for details.
ALCTS CaMMS Catalog Management Interest Group Meeting (January 25)
The Interest Group meeting consisted of three presentations. The first, Cataloger’s Judgment: Bringing Ethical Considerations Into a Fluid Space, by Paromita Biswas (University of California, Los Angeles) and Ann Hallyburton (Western Carolina University), probed the possible use of cataloger’s judgment (the “fluid space” of the title) as a counterbalance against the limitations of certain LCSH Subject Headings available for use. Examples of judgment given included the use of more precise subject headings, such as “Gujaratis (Indic people)” in place of “East Indians,” which carries colonial implications; replacing culturally-loaded terms like “Female impersonators” or “Cross-dressers” with more neutral terms like “Performance artists” and “Drag shows”; and the use of cross-referenced (4XX) terms that are arguably less objectionable than the established terms to which they are attached, for example using “Enslaved persons” in place of “Slaves.” Catalogers are advised of the need to keep abreast to changes to and improvements in terminology, as well as to consider making changes or additions to existing records to incorporate the use of more culturally sensitive headings. The speakers also pointed out that the LCSH pages for individual terms at id.loc.gov include a “Suggest Alternative Terminology” link: see, for example: http://id.loc.gov/authorities/subjects/sh85123347.html) (the link is on the right-hand side of the page under the “Subject of Works” box).
The second presentation, Implementing Local Alternatives to the “Illegal aliens” LCSH, involved members of the CaMMS Subject Analysis Committee’s Working Group on Alternatives to LCSH “Illegal aliens”—Violet Fox, Tina Gross (St. Cloud State University), Cate Kellett (Yale Law School), and Karl Pettitt (University of Denver)—who presented the results of a survey they conducted about issues connected with replacement/non-replacement of the subject heading “Illegal aliens” after a proposed change to “Undocumented immigrants” was rendered impossible by the passage of the Stopping Partisan Policy at the Library of Congress Act. Over 35 libraries, library systems, and library consortia responded to the survey; for the most part, those who did report making changes either (1) retained the SH “Illegal aliens” but added additional subject headings; (2) replaced the SH “Illegal aliens” with a different subject heading or headings; or (3) retained the SH “Illegal aliens” in the system but displayed an alternate subject heading. The changes made did not generally result in complaints from the public: although the institutions surveyed reported a few negative reactions, most noted no reaction at all. Overall, the institutions themselves viewed the changes in a positive light: those who made specific comments indicated a wish that they had made these changes sooner. In conclusion, while the Working Group found it impossible to make prescriptive recommendations, the members felt that doing anything by way of change was better than doing nothing at all.
In the final presentation, Strategies and Tools for Transforming Unstructured Metadata into Structured Linked Data, Darnelle Melvin (University of Nevada, Las Vegas) discussed the process by which the digital collections in UNLV Special Collections are being migrated out of CONTENTdm and into Islandora 8 in support of open access, improvements in data operability through use of linked data, and digital preservation. An important feature of the process is the use of workflows built around spreadsheets designed for ease of use by staff (such as curators) who typically do not have metadata expertise.
ALCTS Metadata Interest Group Meeting (January 26)
The Interest Group meeting consisted of a presentation follow by a business meeting. In the presentation, Bringing Everyone to the Table: Collaborative Ontology Development, Melanie Wacker (Columbia University) discussed her work on the LD4P 1 ArtFrame project. ArtFrame, one of a number of domain-specific BIBRAME ontology extensions, was the result of a collaborations between the Art Libraries Society of North America, the Library of Congress, and a number of museums and libraries, including the Columbia University, which owns a sizeable art collection. While originally designed to enhance descriptions of two- and three-dimensional artworks, the participants ultimately decided to develop a single ontology for Art and Rare Materials; they are now expanding into the Archives domain as well.
During the Business Meeting that followed, Timothy Mendenhall (Columbia University), the IG’s Liaison to CC:DA, announced that the beta RDA Toolkit will switch over to become “the” RDA in December 2020, but noted that the date on which the beta text becomes official will be somewhat later. Task Groups are currently working to develop application profiles. Moving forward, RDA will be updated quarterly, a timeline that may make it more difficult to process/digest revision proposals; the details of doing so are still to be worked out.
LC BIBFRAME Update Forum (January 26)
Slides are available here: http://www.loc.gov/bibframe/news/bibframe-update-mw2020.html
In her introduction, Sally McCallum (Library of Congress) outlined some recent BIBFRAME activities, including the creation of BIBFRAME descriptions by Stanford University-led Mellon grant cohorts and the completion of the Sinopia editor, the involvement of the Program for Cooperative Cataloging, OCLC’s new Mellon grant-funded Entity Management Infrastructure project, and the 2019 European BIBFRAME Workshop in Stockholm that featured the National Library of Sweden’s working BIBFRAME system. She then outlined the Library of Congress’s goals for 2020: to move its BIBFRAME system into the cloud and stabilize it, probably reconverting all files with improved algorithms when this is done; to stabilize model issues involving title authorities, non-Latin scripts, and BIBFRAME’s relationship to MARC (including MARC to BIBFRAME and BIBFRAME to MARC conversions); to redevelop the BIBFRAME editor; to move the rest of the LC Classifications schedules into id.loc.gov; to refine the name authority module; and to review and improve BIBFRAME documentation.
The introduction was followed by four presentations. In the first, BIBFRAME Implementation at the University of Alberta Library, Abigail Sparling (University of Alberta) provided an update on the activity of the Canadian Cohort of LD4P2. The Cohort’s work has been focused on the accomplishment of four broad goals: continued enhancement of conversion, reconciliation, and enrichment process for MARC to BIBFRAME; exploration of new forms of authority control through the enrichment of MARC and BIBFRAME data with URIs; conversion of the monograph cataloging team’s operations to working in a BIBFRAME environment; and utilizing the cohort’s activity to continue building and strengthening the BIBFRAME community.
In the second presentation, PCC Roles and Files, Jennifer Baxmeyer (Princeton University) reported on the involvement of the Program for Cooperative Cataloging with linked data projects, specifically in LD4P2. PCC’s role in LD4P2 is to create and maintain PCC-approved application profiles, and to develop and provide ongoing training for linked data, BIBFRAME, and the Sinopia editor. PCC has discovered that this work is taking much longer than anticipated—for example, application profiles still have yet to be created (PCC is now considering reinstating the BIBFRAME Task Group to work on these), while the PCC Standing Committee on Training’s Sinopia Training Task Group is now focused on creating training for catalogers new to Sinopia, with drafts of the first couple of modules completed at this time. Besides these roles, PCC is planning to collaborate with OCLC on their Entity Management Infrastructure project, and is currently considering the creation of a freely-available PCC data pool that will contain a BIBFRAME representation of all PCC cataloging past, present, and future—including both MARC cataloging for conversion to BIBFRAME, and cataloging created in BIBFRAME.
In the third presentation, Nathan Putnam (OCLC) gave an OCLC update on its CONTENTdm Linked Data pilot and its Entity Management Infrastructure project. The CONTENTdm Linked Data pilot aims to work with libraries to improve their users’ ability to discover, evaluate, and use digital resources in CONTENTdm repositories. In Phase 1, which ended in December 2019, OCLC worked with the Huntington Library, Art Museum and Botanical Gardens; the Minnesota Digital Library; and the Cleveland Public Library, focusing on mapping textual metadata and including the development of an image annotator. Currently, in Phase 2, the University of Miami and Temple University have joined the pilot, is concentrating on the creation of tools for managing metadata in Wikibase. The last phase of the pilot, which will be directed towards the use of Wikibase entities to drive end-user discovery, is expected to begin in mid-2020 and run through August. For more information: oc.lc/cdm-linked-data-pilot . OCLC’s Entity Management Infrastructure project (announcement: oc.lc/mellon-grant), funded by a Mellon Grant that OCLC will match, is a 24-month project that looks to begin building the infrastructure necessary to provide the reliable and persistent identifiers and metadata necessary to the success of linked data, focusing on a production infrastructure for Work and Person entities. The project will draw upon the work of several previous projects, including the CONTENTdm Linked Data Project that was just described, and is intended to provide support for multiple descriptive and encoding standards. It is intended as a collaborative project with the library community, and includes a project advisory board, the LD4P community, the PCC, and the “Big Heads” (ALCTS Directors of Technical Services at Large Research Institutions) among its partners. OCLC also welcomes participation by others (email email@example.com if you are interested). Project updates will be available in the coming months, and at ALA Annual in Chicago in June. The update ended with a word on cross-partner collaboration: in November, the PCC Policy Committee agreed to share its data with LD4P for use in its Sinopia infrastructure. As a result, OCLC will provide the PCC MARC records currently in WorldCat to LD4P in exchange for PCC BIBFRAME records created by LD4P participants, which OCLC will convert back to MARC and add to WorldCat. After completion of the Entity Management Infrastructure project, OCLC anticipates being able to add BIBFRAME records directly to the new entity infrastructure.
In the last presentation, Library of Congress BIBFRAME to MARC data conversion, Jodi Williamschen (Library of Congress) spoke about efforts to convert BIBFRAME descriptions into MARC records, a project that has going on for about a year. Testing of these records is part of the project, and so far 22 of the converted records have been successfully loaded. These records differ from records created directly in MARC in a number of ways:
- During MARC to BIBFRAME conversions, many URIs are added to the BIBFRAME descriptions; the intent is to preserve these URIs in the converted MARC records.
- Converted MARC records do not contain Field 007: during the MARC to BIBFRAME conversion, Field 007 bytes are assigned to BIBFRAME properties or classes and given URIs; in BIBFRAME to MARC conversions, these elements are placed in the appropriate 3XX field and subfield.
- To avoid duplication of information, Field 008 in converted MARC records is more general, including only date entered, date and place of publication, and language.
- All genre form terms in the converted MARC records appear in Field 655 (only).
- There are no 880 fields in the converted MARC records, as data is only encoded in the vernacular script; Field 241 has been revived in order to experiment with its use in carrying Romanized titles when needed.
- Data consistency measures have been introduced into the converted MARC records, such as the placing of Field 020 qualifiers in $q, and the use of Field 264 in place of Field 260 for publication information.
- Less punctuation appears in the converted MARC records.
The conversion is not yet perfect; adjustments still need to be made, but there is a definite focus on reducing duplication/redundancy in the converted records.