SAC: ALA Midwinter Report 2020

Philadelphia, PA, January 25-27, 2020
Reported by Rebecca Belford (Oberlin College & Conservatory), Chair, MLA CMC Vocabularies Subcommittee

Reports from:

  • Faceted Subject Access Interest Group
  • Authority Control Interest Group
  • Technical Services Managers in Academic Libraries Interest Group (“Medium Heads”)
  • Heads of Cataloging Departments Interest Group
  • Subject Analysis Committee (SAC)
  • Subject Analysis Committee: Subcommittee on Faceted Vocabularies (SSFV)

Faceted Subject Access Interest Group

(ALCTS CaMMS) January 25, 2020

Three presentations followed a brief update from the FAST Policy and Outreach Committee (FPOC).

“FAST: Finding the Future of Subject Access at the British Library”

(Caroline Kent, British Library)
The British Library (BL) team (Alan Danskin, Caroline Kent, Janet Ashton) began its exploration of FAST (Faceted Application of Subject Terminology) as part of a review to identify cataloging efficiencies. FAST emerged as a means to get terms from an internationally recognized controlled vocabulary into the catalog where otherwise there are no current subject terms. Because at that time FAST was still a beta project at OCLC, the BL team investigated and determined that other institutions were already experimenting with using FAST before OCLC had committed to move the project to production.

While testing FAST application, they asked whether it would 1) not be detrimental, and 2) be an improvement. The team also considered if all current workflows could change so the only new headings added were FAST. Ultimately they decided that they could do more by using FAST in areas not covered by LCSH, but would continue to add LCSH for mainstream, English-language content. This exploration occurred simultaneously with the work of the FPOC and OCLC.

Currently, the British Library has imbedded application of FAST in workflows where either there was no provision for subject headings or where only local terms were applied. The goal was to make materials more easily and immediately discoverable by researchers. Implementation for grey literature was successful. Local terms were mapped to FAST, some of which mapped to exactly the same term, now in a controlled vocabulary. Additional FAST terms not previously used were added. They also successfully added FAST to the workflow for E-media (i.e., material received in electronic format through legal deposit or purchases), for which many records either had no subject access or contained only publisher terminology. They were able to map publisher terms and apply others directly. The BL took a different approach for purchased content in their Asia/Africa collections. Although much of the material is in non-Roman script, people tend to search in English because the search interface is in English, so adding English-language terms enhances discoverability. They maximize effectiveness by having the subject and language specialists with the most expertise select FAST terms from a dropdown list in a spreadsheet. That spreadsheet is then periodically mapped to MARC and loaded into the catalog. Archives and manuscripts are considering implementation. Finally, the BL implemented a policy to leave all FAST headings in incoming OCLC records, even those using LCSH.

Future plans include more training, fuller implementation and development, additional testing, and additional partnerships. Slides on ALA Connect: FAST – Finding the future of subject access.pptx

“FAST Headings for ETDs at Brown”

(Jeanette Norris, Brown University)
Jeanette presented on behalf of the Brown Digital Repository (BDR) team (Jeanette Norris and Joseph Rhoads).

Since Brown University moved to e-only dissertations (ETDs) in 2008, all indexing has been provided solely by the authors. Theses and accompanying metadata go directly to the BDR without any library intervention. With the authors focused primarily on meeting graduation requirements, metadata has not been a main concern. The Library’s BDR team decided to revamp the process to help the graduate school by ensuring an archival copy of record, and to improve discoverability by enhancing data in a way that is easy for authors.

The BDR team then built FAST headings into the ETD application, using the assignFAST API. Students begin typing a term, and suggestions appear for FAST and local terms previously used; students may also assign terms directly from FAST or create their own. The BDR team also implemented a retrospective project, using OpenRefine to correlate original keywords with FAST headings, with the goal of having at least one FAST term per ETD.

The team performed two analyses. In an examination of a large set of records in 2016, they found that over 85 contained FAST headings, including around 1700 unique FAST terms and 2000 unique uncontrolled terms. Of those uncontrolled terms, few had exact matches, there were some fuzzy matches, and most did not have a match. A smaller sample of 25 records indicated that 98% of the headings were ‘correct’; for example, students were correctly choosing “satellites” not “artificial satellites” for moons.

Next steps, the team will be assessing the technical aspects including whether it makes sense to download a local copy of the full FAST file, identifying options for greater metadata quality control such as having catalogers spot check records, expanding vocabularies options to MeSH and AAT, and improving how FAST terms are used in discovery layer facets. Their Django ETD app is available on GitHub at Slides on ALA Connect: Norris-FASTetdsBrown.pptx

“Maximizing Discovery of Datasets in the Library Catalog”

(Rowena Griem, Tachtorn Meier, Yukari Sugiyama, all Yale University)
The Yale Library provides access to commercial and open access datasets to support research and the digital humanities and StatLab units. In April 2019, the library began a project to develop best practices for cataloging and discoverability. Cataloging challenges included the variety of content types (numeric, image, geospatial data, text) and file formats that a dataset may include. In MARC, there is no one way to code datasets so that they can be consistently and reliably described and discovered as datasets, distinct from other computer files or content types. This is an ongoing challenge and a cleanup project to address inconsistent data in the past. There are also few terms for datasets in controlled vocabularies.

The cataloging best practices the Yale team developed included MARC and vocabularies decisions. They assign all datasets “computer file” in 007/Leader, “dataset” in field 336, separate 347 fields for file type, encoding format, file size; a 520 that includes the text “dataset,” and MARC 786/787 as appropriate. The team made successful LCGFT proposals for Data sets, Biostatistics, and Medical statistics; successful LCSH proposals for Data mining–Statistical methods, Spatial data mining, Image data mining, and Text data mining; and their proposal for Text corpora is under consideration. The form subdivision “Data sets” was rejected from LCSH, in favor of Data sets + content GFT. The Yale best practices assign multi-level genres: all datasets get Data sets and a narrower term (Geospatial data, Images, Statistics, Text corpora).

The presenters shared their approach to record enhancement and discovery user interface recommendations. They identified approximately 10,000 records using known collections, keywords, form subdivisions, and genre headings; that number is after manual review to eliminate numerous false positives. In their discovery layer, 336 Computer dataset maps to Datasets in the facet. Using the term in the facet means that users will not miss results in keyword searches for “data sets” vs. “datasets.” Once Datasets is selected as a limit, narrower options appear (statistics, text corpora, census data, etc.). Any results that require specific logins offer a button to generate an automated email to the subject librarian to request access. Documentation for their project is available at and Slides on ALA Connect: Yale-FSAIG 2020 midwinter.pdf

Authority Control Interest Group

(ALCTS-LITA) January 26, 2020

Martha Rice Sanders (Innovative) moderated. Janis Young (LC) opened the session with updates on Library of Congress Projects (see report from the SAC meeting).

Anita Kazmierczak (Tulane University) gave a brief presentation, “Traditional versus Modern: Library of Congress Subject Headings,” as a discussion prompt. Generally, cataloging and classification are critical for discovery, but don’t serve their purpose if patrons can’t find what they are looking for. Problems arise when patrons don’t know how to best search for something by subject, subject terminology is outdated or incorrect, or when there is a lag in updating LCSH. Sample terms identified by the NACO-SACO group at Tulane that are problematic or illustrate biases include: Women in X vs. Men in X; Women illegal aliens vs. Male immigrants; Prostitution (default is female); Sexual ethics for women (only); Theology (Christian unless specifically qualified); Citizenship (default is citizenship by birth). National movements working toward change or alternatives include Radical Cataloging, various Equity, Diversity, and Inclusion initiatives, use of local alternatives in MARC 6xx$2 local or 69x, and FAST headings.

A lively floor discussion followed, despite a somewhat small group due to the conflict with the LC BIBFRAME Update Forum and LITA Top Technology Trends sessions. There were many comments, questions, respectful disagreements, and ideas, including:

  • It has become easier to expand and change LCSH since the SACO program began.
  • Proposed changes need to be carefully considered so they continue to be appropriate in the long term.
  • Explore using multiple source vocabularies to supplement LCSH, such as the Homosaurus.
  • The principle of literary warrant reflects bias and assumptions of the “default” in research and publication; maybe our conversation indicates we are searching for justification other than warrant.
  • LCSH retains the bias of binary in its lack of true umbrella terms (examples: no umbrella LCSH for “significant others” or “queer”).
  • Challenge of maintaining our files, and teasing apart headings where an unqualified plural currently refers to mixed genders or to males.
  • We pay for our collections and our discovery systems, and we want to make sure patrons can access our items by using current terminology in searches and without seeing themselves represented by offensive terms. There are quantifiable and unquantifiable costs.
  • Work to improve LCSH and the access it can provide can seem overwhelming. How can we build on what’s been done and create manageable blocks of work to effect change?
    As expected, the discussion raised more questions than it answered, but was a clear indication that this is a topic we are passionate about.

Technical Services Managers in Academic Libraries Interest Group (“Medium Heads”)

(ALCTS) January 25, 2020

The meeting of the interest group was dedicated entirely to discussion. Each of four full tables had an assigned topic; groups reported to the large group at the end.

“Doing More with Less: Technical Services Reorganizations and Merged Departments”

  • Retirements/vacancies through attrition, and new positions created through reorganizations; opportunities include retooling jobs, i.e., making job descriptions more vague for future flexibility
  • Planning needs include retraining and documenting institutional knowledge
  • Important to expand opportunities for longer term staff and opportunity to move up in rank
  • Labor/union issues are a consideration
  • Communication: Important with your staff and with higher-ups. Need to get buyin from staff in lower levels of the organization’s structure. Strategies for discussion: ‘try it and see where it goes’. Middle managers will need to admit that you don’t know either when staff ask you questions. Difficult conversations: what projects won’t get done?
  • “If you seized power for one day, what would you do?”
    • Unmerge positions
    • Reconsider a reorg, to reunite ERM and cataloging
    • Make staff think about busywork: what and why
    • Cloning
    • Redoing reorgs

“Using Task Management or Project Management Apps (e.g. Trello, Airtable, Jira) to Manage Ongoing Technical Services Workflows”

  • Great for e-resources; good for acquisitions; not useful for ongoing physical format cataloging or big backlog projects
  • Staff buyin is difficult
  • Lots of setup work
  • Helpful if a specific and obvious need
  • Can be useful as advocacy/analysis tool to track tasks and time
  • Excellent for handoffs among people who don’t necessarily work closely together

“Student Workers, and How They Are Helpful or Not”

  • Approaches varied widely among the group
  • Many types of students to consider: graduate, LIS, undergraduate, international, unionized students, work study/not work study. Different expectations for different types
  • Need to clearly establish what students are allowed to do if union shop (unionized students and/or staff)
  • Issues with retention and turnover
  • Rigorous training may be required; some places train for a full year
  • Are there concerns among staff that students will take over their jobs? Varied widely. It is a point of contention in some places; elsewhere students are ‘adopted’ by staff
  • Useful to create a learning experience for LIS grad students who might come back for a staff position after their MLS program

“Soft Skills and Technical Services”

  • What is “technical services”? This can be a fuzzy definition, many people have additional duties
  • Defining soft skills: those that are not hard skills like MARC, systems, RDA
  • Soft skills include: people skills, listening, play well with others, team building, managing change, critical thinking, flexibility, sense of humor, customer service (internal and external), initiative, time management
  • How to develop soft skills? Workshops, team building exercises, personality testing workshops

Heads of Cataloging Departments Interest Group

(ALCTS CaMMS) January 27, 2020

The three presenters were introduced by interest group co-chair David Van Kleeck.

“Exploring the Use of Wikidata: Highlights of a Sabbatical Project”

(Lori Robare, University of Oregon)
Lori gave rationale, details, and data modeling challenges of several Wikidata mini projects comprising her sabbatical research. Wikidata, part of the Wikimedia Foundation, is all structured data, editable and readable by both humans and machines. It is a single data store with interfaces and display available in multiple languages, housing items (entities, assigned unique Q numbers), statements (facts about entities, expressed as triples), and properties (assigned unique P numbers). The “pot of gold” at the end of an item is an identifier that works as a linking hub. The mini projects shared the goal of raising the profiles of people, organizations, and entities important to Oregon and the University of Oregon. Wikidata itself does not have robust documentation; most information is located in WikiProject. WikiProject is crowdsourced, meaning a self-selected group of contributors supply varying amounts of information and perspectives.
Each area of the projects raised challenges related to data modeling and best practices:

  • Women mayors in Oregon. Is it a best practice to create an item for a specific municipal position (e.g., mayor of Eugene); Lori initiated a query on project chat. (Answer: yes)
  • Oregon book award winners. Data modeling challenge because ‘‘Oregon book award’ actually consists of a set of separately named awards.
  • UO Libraries to items. Data modeling challenge of how to represent the relationship between branch libraries and the system. (Decision: “part of”)
  • Name changes. Decided to treat earlier names as aliases and include start and end dates.
  • “archives at” additions. Data modeling challenge: What level (university, library, specific location within library)?
  • University presidents. Had to decide whether University president (of) University as subproperty vs. President of UO. Project chat had few and conflicting recommendations.

Screenshots for projects demonstrated some of the front-end functionality that can result. This includes a timeline feature, shared queries (e.g., timeline of Little Women and the actresses who played Jo), entity graphs located near the Q number.

In summary, positive outcomes were: being able to put structured data on the web, Wikidata is easy to edit, it was open and collaborative, and there is no requirement for completeness of data. Challenges included the learning curve, locating documentation, struggling with the consensus model where answers depend on who happens to respond, and the variation in data quality with the most minimal resulting from mass imports. There is a now a Wikidata working group at the University of Oregon that will continue monitoring Wikidata activity and exploring possible future activities.

Q: Have you experimented with batch extraction from university information sources? A: Not yet, but would like the group to try. The only test was with Wikidata’s QuickStatements and TABernacle. It makes sense to work toward batching.

Q: Regarding granularity, how is Wikidata using the hierarchy? Does it produce results from a broader level when you search for a narrower one? A: The query aspect wasn’t part of this project, but it is a great question. So far it has not seemed easy or intuitive. For the example of mayors, there isn’t an easy way to get a full list. “Mayor of Eugene” is a subclass of “mayor of a place in Oregon,” so to get a list of mayors of all places, each place must be searched separately.

Q: Is there a Wikidata a role for libraries? A: It is a logical place to start. Faculty suspicious of metrics can generate some skepticism: if their Wikidata entry is created but not maintained, it looks like they aren’t publishing or have not published recently. The UO working group is focusing on collections to start.

“Investigating the Identifier Ecosystem of the University of Colorado Boulder Faculty”

(Chris Long, UC Boulder)

Chris detailed two UC Boulder (UCB) projects exploring linking identifiers. The first was a local Wikidata project based on UCB’s digital collection of women poets of the Romantic period, resulting in exposure of the collection on the web and information about the poets themselves. In this project, they added Wikidata entries, created NACO and ISNI records if not already present, then added the NACO, VIAF, and ISNI records to Wikidata. UC Boulder is also midway through a pilot PCC URI project, a “road test” of the conceptual work. One phase of the project was adding identifiers. The group at UCB limited the scope to current UCB faculty that have NACO records with UCB affiliation, 179 records. The PCC lifted the moratorium of MARC 024 just for this pilot, in which they added MARC 024 fields with VIAF, ISNI, ORCID, and Wikidata identifiers. They found VIAF identifiers available for 100% (by definition), ISNI/ORCID 85%, due to promotion efforts by the library, around a quarter had all three and a small number had none.

They then looked at the presence of identifiers within other records—the interconnectedness of the ecosystem. The majority of Wikidata records had NACO, VIAF, and ISNI identifiers; fewer than half had ORCIDs. However, Wikidata itself is not heavily used by faculty. In VIAF, a large percentage had ISNIs but almost none had ORCIDs. Almost no ORCIDs have ISNIs, perhaps due to the unmediated input of ORCID information (directly claimed by faulty). NACO records almost entirely lacked identifiers, perhaps due to the moratorium on use of MARC 024 to add identifiers; in the future, there may be much potential in addition of URIs in MARC.

Chris admitted to bringing “lots of questions and absolutely no answers.” These include where best to direct our energy, asking ‘if not us, who’; if we are capable but don’t have the capacity for these projects (CUB has 19,000 faculty); do we limit work to our domain; how much they can do and still get the cataloging done; considering the barriers and labor-intensive work for NACO and ISNI; and considering the low barriers but ‘nascent at best’ best practices in Wikidata. In short, what should we be doing?

“Alphabet Soup: Balancing the (Sometimes) Competing Demands in Shared Collections”

(Ben Walker, University of Florida)
The University of Florida (UF) participates in many shared collections: regional collections, Google, HathiTrust shared print, FDLP (Federal Depository Library Program), FLARE (Florida Academic Repository), Rosemont shared print alliance, and more. Cataloging for UF is separate from that for shared collections, but there is a lot of crossover. Desired outcomes of shared print include compatible data, benefits to stakeholders, and efficiencies. However, the actual outcomes did not necessarily match. They discovered inefficient/nonexistent data crosswalks with systems that do not don’t easily talk to each other; unrealized benefits, such as reconciling a plan to move duplicates to storage with the need for people and space; and static budgets, with a dwindling catalog department as staff was shifted to shared collections and no money for additional staff.

Specific barriers included the complexity of getting data out, including working with spreadsheets and version control; “bad data,” or, ”mistakes of the fathers”; complex communication needed for setup problems introduced by automatic processes (batch upload for Hathi trust chopped off digits from OCLC number); and, again, limited resources. There is no single system that can reduce all of these barriers.

Shared collections are important for the future of libraries, and we must ask what our core mission is. What gets abandoned when a new initiative is added? How do we support complex processes with communication, software, grants, and non-library partners? How can we support initiatives if we don’t have the human resources?

Subject Analysis Committee (SAC)

(ALCTS CaMMS) Sunday, January 26 and Monday, January 27, 2020 (Brian Stearns, Chair)

Liaison reports

Written liaison reports are available from ALA Connect in SAC’s community library (“ALCTS CaMMS Subject Analysis Commmunity [OPEN]”) [sic] and linked below.

Library of Congress Policy, Training, and Cooperative Programs Division

(Janis Young) (See: SAC20-MW/1.3). The full LC report prepared for ALA Midwinter is available at

Highlights/questions (includes notes from ACIG session, SSFV meeting, PCC SACO liaison report, and new business):

  • Staff changes: Retirement of Lucas Graves, who did many of the bib and global changes for name changes.
  • RFI for next-generation ILS is now closed to responses.
  • Jump in number of new LC Subject Headings in FY19 (7,259 from 2,221 in FY18) is due to the multiples project.
  • Authorized access point and corresponding use in LCSH changed from Kiev (Ukraine) to Kyïv (Ukraine), which applies the ALA/LC Romanization table for Ukrainian rather than Russian. Classification updates are also complete.
  • Moratorium on LCDGT proposals is still in place; end date not determined.
  • Bibliographic Record Control Numbers in MARC 670 in Authority Records: “In order to support linked data, the bibliographic record control number will be contained in subfield $w beginning in spring 2020. The subfield $w will appear as the last element in the field. The control number itself will be preceded by the MARC code for the agency to which the control number applies, enclosed in parentheses, e.g., $w (DLC)2015300502.” This applies to LCSH, LCGFT, LCMPT, and LCDGT. This is only required of SACO libraries if they are CIP participants. Please do not make change proposals solely to recode, but when general revisions proposals are made, the proposal should be formatted according to the new policy. Each 670 should have only one $w even if multiple control numbers are available. NAF: “stay tuned.” Records with $w will begin distribution in April 2020. Corresponding updates will be made to the Subject Headings, Genre/Form, and Demographic Group Term manuals.
  • Q: Can the SACO proposal form be updated to have a validation feature to require $w?
  • A: Intentional decision not to require validation on templates in order to prevent submitters’ giving up due to validation errors.
  • ClassWeb 4 released August 2019; runs on Mozilla Firefox, Google Chrome, Microsoft Edge, and Safari.
  • LC project to revise access points for first-level administrative subdivisions of France and Norway (merges of regions) and update systematically. Old names will remain valid for use to apply to unmerged time period.
  • Multiples cancellation project (e.g., “[Topic]–Religious aspects–Buddhism, [Christianity, etc.]”): Created 2470 individual authority records, canceled 294, redesignated 75 validation records as full ARs. Non-SACO libraries can submit a proposal when needed. Records will resume distribution in March or April 2020 pending fixes to ClassWeb backend functionality.
  • Q: How fully is LCGFT implemented by LC catalogers? A: Yes for music, moving images, cartographic resources, cookbooks.
  • Q: Draft GFT manual status? A: No firm timeline.
  • Q: How current are PDFs of term lists? A: Last update was April. Should be updated approximately February 2020.

Committee on Cataloging: Description & Access (CC:DA)

(Robert Maxwell) (See: SAC20-MW/1.4 Report from CCDA)
3-R Task Force commented on the RDA unconstrained elements and a proposal to develop a set of user-friendly labels that could be used for public display (particularly for relationship designators). The Task Force helped CC:DA prepare a joint change proposal from the Art Libraries Society of North America (ARLIS/NA) and CC:DA to add an RDA element for curators who play a role at the work level (e.g. a curator who organizes an exhibition that publishes and exhibition catalog). The proposal was forwarded to NARDAC. Although not approved, this process helped NARDAC gain insight into how future North American RDA proposals will be handled.

Art Libraries Society of North America (ARLIS/NA)

(Sherman Clarke) (See: SAC20-MW/1.6 Report from Art Libraries Society of North America)

  • Waiting for the release of the new version of RDA to resume work on the update of Cataloging Exhibition Publications.
  • Catalogers are discussing how “black and white” should be treated in catalog records, i.e., description, subject heading, genre/form terminology. For example, the subject heading Color photography goes back decades; Black-and-white photography was only established in 1999 (previously assumed the “default”).

Library and Archives Canada Published Heritage Branch

(Annie Wolfe) (See: SAC20-MW/2.1: Library and Archives Canada Published Heritage Branch)

  • Canadian Subject Headings (CSH) are available only in OCLC WMS Record Manager or by SFTP upon request to LAC. Q: We lament the loss of being able to search CSH on the web. Can LAC post an annual PDF? A: Currently, access is only SFTP. They receive this request quite a lot and are thinking about an alternative.
  • CSH names of indigenous peoples: LAC is developing replacements using the term “Indigenous” for occurrences of the word “Indian” or “aboriginal” in preferred headings, collaborating with First Nations, Inuit and the Métis Nation communities.

MARC Advisory Committee

(Adam Schiff) Updates from MW meeting:

  • DP02: Discussion indicated use of 3xx instead of 504/525 for controlled vocabulary of bibliographic content instead of using a note field. A coded value subfield will be needed.
  • DP03: Did not favor illustrations/sound in 340. Discussion: explore new 3xx fields for content-related characteristics; reconsider color as expression-level; maybe Fixed Fields can map to BIBFRAME and then back to MARC; maybe eliminate the 008.
  • DP04: Field 345 and aspect ratio/motion pictures. Requested to return as a proposal after revising in collaboration with OLAC, which did not coauthor the DP.

Dewey Classification Editorial Policy Committee

(Deborah Rose-Lefmann) (See: SAC20-MW/2.4 Report from Dewey Classification EPC)

  • Podcasting: requested discussion paper on internet media generally
  • Looking forward to paper/development of Graphic Design

Library of Congress Dewey Program

(Caroline Saccucci) (See: IFLA Working Group on Automated Subject Analysis)
IFLA will be conducting interviews regarding automatic processing of subjects/assignments. Please contact Caroline if you are willing to be interviewed. The end result will be a white paper to be shared on IFLA’s website.

PCC Subject Authority Cooperative Program

FYI from Adam Schiff: PCC Standing Committee on Standards (SCS) has started discussion about the possibility of adding subfield for URI in $0 for permalink or Hub. Janis Young suggested that any SCS proposals $0 for subject authority records include the relevant staff at LC.

See written reports…

Working Groups and Subcommittees

SAC Subcommittee on Faceted Vocabularies

(Casey Mullin) (See: SAC20-MW/1.8 Report of SAC Subcommittee on Faceted Vocabularies)

  • Report from MW meeting Jan 25, updates from task groups
    • Chronological: first draft of Best Practices; likely related MAC papers
    • Genre/Form: mappings from form subdivisions and MARC fixed fields; writing up what would be required context for retrospective projects
    • Geographic: working on retrospective project based on subdivisions esp music and literature potentially generating 370 and 386, discussing what is in scope and the types of records more appropriate to run on
    • Demographic: still working on literature analysis. Mapping to LCSH; PCC SCS has been working on mapping of demonyms to place names. If/when the group issues BPs, may focus more on intellectual/ethical process than mechanics.
  • Headed to SAC for attention:
    • Chronological BPs and G/F retrospective package
    • SSFV applications/reappointments
  • Q: Would SAC plan related programming? A: Sounds like a good idea!
SAC “Illegal aliens” Working Group

(Violet Fox) (See: SAC20-MW/1.9 Draft Report of the SAC Working Group on Alternatives to LCSH Illegal aliens)

  • Violet and others gave a related presentation, “Implementing Local Alternatives to the “Illegal aliens,” at the Catalog Management IG on January 25.
  • Work focused on the survey, and the survey report will be available in the next month. In addition to the formal report, the group would like to offer a clearinghouse of local approaches. ALCTS Board will look into whether ALA can create a web space that the group can easily edit to house this information.
Future of the Research and Presentation Working Group

Midwinter 2020, there was no presentation due to lack of WG chair. Originally the WG was formed to track relevant research throughout the year and use that to generate a bibliography and ideas for presentations. Presentations at SAC meetings have been successful in bringing people to the meetings, and timing is most successful when it is at the beginning of the second meeting with a reliable start time. Publicity through listservs has been successful in drawing a crowd. Future focus will be on a presentation at Annual.

New business and discussion

New liaisons
  • Welcome to LAC as of this meeting
  • OLAC will send a liaison to Annual 2020
The phrase “foreign speakers” in relation to languages in LCSH
  • Comes up in discussion of whether “–Textbooks for foreign speakers” for materials covering textbooks for a language in which the reader is not fluent does not reflect that many people learning a language are not foreigners. “Foreign” in other topical subdivisions is rare, mostly “[Subject] students, foreign–[Place].
  • Alternatives? “Second language learners” (LAC)? “Non-native speakers”? “Non-fluent speakers”? Any alternative would need to be a substantial improvement; change would necessitate massive bib file maintenance.
  • SAC discussed a working group for a discussion paper outline that rationale for a proposed change (if an alternative suitable for a change proposal is identified).
Subject headings of the form “[Topic]–Fiction” as variants of “[Topic] stories” in LCSH
  • Discussion at SAC and SSFV meetings about challenges when $v Drama or $v Fiction in a SH reference changes the heading to a different meaning or scope. For instance, Lawyers $v Drama (applicable for films, plays, operas) is a reference that controls to SH Legal drama (limited to plays); or Law $v Fiction controls to Legal stories
  • Change proposals become particularly complex when a heading is needed to be available for use with $x History and criticism
  • Related challenge for place names, used as settings vs. fiction about that place
Response to Code of Ethics for Cataloguers

Form a working group to respond when there is a report available for comment? Sounds good.

Midwinter and virtual meetings
  • [After the Midwinter meeting, ALCTS announced there will be no business or programming meetings at Midwinter 2021.]
  • Periodic virtual meetings were discussed; the committee agreed to hold a test meeting before Annual 2020. A guest asked whether virtual meetings would be open, if notes would be taken, and if recordings will be made available. Meetings will be open and notes will be taken. However, several people felt that distributing recordings of meetings would inhibit free discussion.
Future of SAC

Much of the meeting time was dedicated to discussing how to keep SAC viable and how the committee will engage in activities in addition to communication among liaisons and the committee. There was also discussion about the role of working groups related to vocabularies that ultimately pointed to having SAC itself take on much of the work.

  • SAC will begin a pilot project between now and Annual to comment on LC tentative Subject and Classification lists, including a trial virtual meeting. This work will be undertaken by SAC members; Lisa Robinson volunteered to coordinate. Comments may be on any aspect, including identifying needed references, technical aspects, scope notes, hierarchies…any constructive comments. Janis confirmed that LC welcomes feedback from SAC and/or individual members taking turns commenting, and will forward notices of tentative lists to the SAC list.
  • A corresponding pilot will begin for two 2020 Dewey proposals in advance of the September 2020 Dewey meeting.
  • Candy Riley volunteered to compile an internal list of the subject/language expertise/interest of SAC members to assist in identifying contacts for particular subjects beyond what might be obvious from job titles.
  • Adam Schiff, coordinator for SSFV chronological group, will suggest a virtual meeting for discussion of the group’s best practices document.
  • Brian encouraged liaisons to let their respective organizations know that SAC is actively seeking projects.
  • ALA Annual 2020 plans: Daniel Joudrey volunteered to conduct a one-hour training session on aboutness as the SAC presentation.

Subject Analysis Committee: Subcommittee on Faceted Vocabularies (SSFV)

(ALCTS CaMMS) January 25, 2020 (Casey Mullin, Chair)

See also SSFV report to SAC (SAC20-MW/1.8 Report of SAC Subcommittee on Faceted Vocabularies). The SSFV employs the task group model. Each group gave a report as a discussion prompt.

Chronological Group

(Casey Mullin, lead) Since ALA Annual 2019, their draft of best practices for use of 048 and 388 fields is approaching completion. Issues of interest include:

  • Faceted approach under discussion does not replace LCSH or FAST.
  • System agnostic recommendations are difficult to issue in practice. Details on use in Primo, see “Getting More out of MARC with Primo: Strategies for Display, Search and Faceting” by Kelley McGrath and Lesley Lowery in Code4Lib Journal Issue 41, 2018-08-09.
  • MARC 045 may be useful for retrospective addition of 046
  • The group is looking for an example of a serial where the date of publication is known to be the same as date of creation
  • Challenge: Field 046 definitions for a motion picture complete in one year and released significantly later. So far the group has identified two subfields $k as the best option
  • Challenge: Recorded music created long before release. Currently the group prefers subfields $and $p even if they are identical. Challenge of assign decades similar to LCSH subdivisions: what do you do with ‘late 40s/early 50s’ etc. For record albums, do you record dates of composition of individual songs, the release date, or both? [VS chair’s note: Many areas of this document use music as examples, thus the discussion is of high importance to the music community. It is hoped that MLA will have an opportunity to review and comment on the draft after it is released by SAC. In the meantime, the VS chair can provide comments as an individual SSFV member.]

For the next 6 months: Polish best practices and disseminate for comment (with SAC approval), and issue suite of related MARC proposals.

Genre/Form Group (Casey Mullin)

(Casey Mullin, lead) Since ALA Annual 2019

  • LCSH form subdivisions mapping spreadsheet first pass complete. The group began with a list of form subdivisions, then mapped to: LCGFT; Audience; CreatorDemographic; Free-floating (SHM instruction sheet number); Pattern (SHM instruction sheet number); [group] notes; fields populated by subdivision record (180, 185, 480, 485, 580, 585, 680, LCCN). They are also noting whether the form subdivisions can be subdivided geographically and including FAST headings. The next pass will address unmapped headings, and they will need to address how to identify the non-free-floating form subdivisions.
  • MARC fixed field code mapping spreadsheet, mapping fixed field combinations to LCGFT and identifying those as candidates for automated mapping. Codes are often used to indicate supplementary material that does not meet the 20% rule for subject headings (and therefore possibly may not warrant a LCGFT mapping).

For the next 6 months: Disseminate spreadsheets and accompanying white paper describing considerations around large-scale retrospective implementation efforts.

Geographic Group

(Lisa Cavalear, lead) Since ALA Annual 2019, the group has been mapping 6xx $z to 370 and 386, beginning with countries. Their work raised several questions, particularly what subject headings would work with an algorithm that generates 370/386, and what level of association/subfield to use to house the generated terms. A participant asked about the use of MARC 751, used by German libraries, vs. 370, and there are pros and cons. Table discussion indicated at least one library is applying terms at multiple levels in the hierarchy (e.g., city, state, country) so users can click on either in a faceted discovery layer.
For the next 6 months: Consider specifications that can be shared publicly.

Demographic Group

(Netanel Ganin, lead) Since ALA Annual 2019:

  • Environmental scan of literature relevant to demographic metadata
  • Mapping of LCDGT terms to LCSH headings, now going through working on close matches or terms without a match (primarily demonyms)
  • Search for other demographic vocabularies, finding that most are not available for machine export

For the next 6 months:

  • Consider Best Practices as a consideration of process rather than a focus on mechanics. This might cover when is mapping always acceptable, when it is not, and the many areas in between where due diligence is required.
  • Follow work of Cataloging Ethics Steering Committee. Subgroups have submitted reports and full report expected to be available for comment in advance of Annual 2020.
  • Work with Geographic group to consider assignment of demographic terms based on geographic headings.
  • Address when, if ever, applying multiple terms from different levels of a hierarchy is acceptable
  • Potentially relevant FYI: The PCC/ PTCP has mapped from demonyms to the name authority file; these are relationship mappings, not equivalence mappings.

New business/updates

  • ARLIS: Getty iconography vocabulary has been added to the MARC source list
  • East Asian: LC has again rejected the proposal for Manga; the term is currently available in AAT. They have a proposal under consideration for K-Pop.
  • The SSFV Chair will be attending the PCC Standing Committee on Standards meeting at MW.
  • Netanel asked if people use the hierarchies compiled on his website? Yes, they do.
  • If SSFV reaches the point of retrospective application, will we coordinate with the PCC? Yes, Committee on Applications.
  • Status of art SACO? An instruction sheet is in progress, and they want to complete it before opening proposals. There is a concern that proposals coming in may be so narrow they will be out of scope for GFT.
  • Structural issues with LCDGT includes a challenge of the nationalities hierarchy–what is the distinction between residency and demonym? For example, if someone lives in Place X during Y time, does that mean it is appropriate to assign them a demonym of that place? LC Classification has the same challenge for the literary author schedule, e.g., how to classify an American living in Paris or English-language literature from France.