The Library of Congress >> Librarians, Archivists >> Standards
MARC Standards
MARC 21 HOME >> MAC

MAC Meeting Minutes
MARC Advisory Committee


Annual Meeting
Online Meeting - June 28-29, 2023


MARC Steering Group Members:

Sally H. McCallum               LC                Library of Congress
Hong Cui                        LAC               Library and Archives Canada 
Thurstan Young                  BL                British Library
Reinhold Heuvelmann             DNB               Deutsche Nationalbibliothek

MAC Chair and Secretary

Catherine Gerhart, Chair        UW                University of Washington
Everett Allgood, Secretary      NYU               New York University

MARC Advisory Committee Representatives and Liaisons:

Sherman Clarke                  VRA             Freelance art cataloger
Nick Curotto                    ARLIS/NA        Virginia Museum of Fine Arts
Ethan D'Ver                     MLA             The  Juilliard School
TJ Kao                          PCC             University of California, Davis
Yoko Kudo                       OLAC            University of California, Riverside
Susan M. Moore                  MAGIRT          University of Northern Iowa
John F. Myers                   CC:DA           Union College
Jackie Parascandola             RBMS            University of Texas at Austin
Kate Peck                       AALL            UC Berekeley, School of Law
Elizabeth Plantz                NLM             National Library of Medicine
Regina Reynolds                 LC/ISSN         Library of Congress
Ricardo Santos Muñoz            BNE             Biblioteca Nacional de España
Adam L. Schiff                  SAC             University of Washington Libraries
Jay Weitz                       OCLC            OCLC
John Zagas                      LC              Library of Congress

Other Attendees:

Allison Bailund                 San Diego State University
Bryan Baldus                    OCLC
Ardie Bausenbach                Library of Congress
Renate Behrens                  Deutsche Nationalbibliothek/RSC Chair
Rebecca Belford                 Oberlin College
Gaëlle Béquet                   ISSN International Centre
Juliya Borie                    University of Toronto
Jacqueline Brellenthin          Library of Congress
Cecilia Caride                  Yale University
Christopher Carr                Concordia University/CCM
Andrea Cawelti                  Harvard University
Charlene Chou                   New York University/PCC
Bonnie Dede                     University of Michigan
Corine Deliot                   British Library
Andrea Diedrich                 GBV Common Library  Network, Göttingen, Germany
Andrew Dunnett                  Library and Archives Canada
David Floyd                     Binghamton University
Kevin Ford                      Library of Congress
Paul Frank                      Library of Congress
Britannia Gammond               Northern Lights College
Kathy Glennan                   University of Maryland/RSC Past Chair
Juha Hakala                     National Library of Finland
Matthew Haugen                  Columbia University
Kirk Hess                       OCLC
Sarah Hovde                     University of Maryland
Mary Huismann                   St. Olaf College
Damian Iseminger                Library of Congress/RSC Technical Team Liaison Officer  
Audra Kackley                   St. Tammany Parish Library, Louisiana
Minna Kantanen                  National Library of Finland
Caroline Kent                   British Library
Gennie Kieffer                  East View Information Services
Kitty Kozisek                   Timberland Regional Library
Alex Kyrios                     OCLC
Elizabeth Lilker                National Library of Medicine
Maricelda Losoya-Rush           Bowling Green State University
Hayley Moreno                   OCLC
Charlene Morrison               OCLC
Cory Nimer                      Brigham Young University
Adrian Nolte                    Essen Public Library, Germany
Iris O'Brien                    British Library
Jean Pajerek                    Cornell University Law Library
Tatja Pusa                      National Library of Finland
Pat Riva                        Concordia University/CCM
Karen Ross                      Library of Congress/US ISSN Center
Melissa Rucker                  University of the Incarnate Word
Trina Soderquist                Library of Congress 
Peter Stephen                   Library and Archives Canada
Gary Strawn                     Northwestern University
Hermine Vermeij                 UCLA
Deanna White                    ISSN International Centre
Rebecca Wiederhold              Brigham Young University
Jenny Wright                    Bibliographic Data Services Ltd.
Jodi Williamschen               Library of Congress

[Note: anyone who attended and is not listed, please inform LC/Network 
Development and MARC Standards Office.]

Preliminaries

Cate Gerhart (University of Washington, Chair) began with an explication of the online meeting protocols and voting procedures.

Introduction of members

Cate Gerhart (University of Washington, Chair) performed a roll call and asked committee members to introduce themselves. 17 MAC voting members were present.

Approval of minutes from MAC January/February 2023 meetings
The minutes of the MAC Midwinter meeting, held online on January 31-February 2, 2023, were approved without correction.

Fast-track proposals
Two fast-track proposals were approved since the Midwinter meeting:

Business Meeting/Library of Congress report/ Other
None. [See below for discussion of $0 and $1 usage guidance.]

NOTE:
Full pre-meeting feedback commentary of the MARC proposals and discussion papers can be accessed on the MARC Listserv at: https://listserv.loc.gov/cgi-bin/wa?A1=ind2306&L=MARC

MARC PROPOSALS

 

PROPOSAL 2023-05: Attributes of Family in the MARC 21 Authority Format
URL: https://www.loc.gov/marc/mac/2023/2023-05.html
Source: PCC Standing Committee on Standards
Summary: This paper proposes new and revised fields and subfields for accommodating address of family, field of activity of family, occupation of family, and other attributes of families in various fields in the MARC 21 Authority Format.
Related Documents: 2023-DP04; 2009-01/1

Summary of pre-meeting comments:
There was general support for this proposal from Australia, Spain, LC and NLM.  There was also support from several other groups but with comments and suggestions. OCLC, RDA Steering Committee, and MLA were mainly concerned about 374 (Occupation) and whether it was appropriate for families. RDA-SC indicated it does not align with the LRM; OCLC would prefer "field of activity" rather than occupation for this kind of information; and MLA just found it odd to have occupation for family. Germany supported the proposal but feels they will continue to use the 5XX fields to make these kinds of relationships. Britain did not indicate support but commented that they have continuing concerns about the ethical issues surrounding adding personal information to authority records. PCC members were split on their support for this paper – those that did not support the proposal had various reasons: one preferred the 368 for this information, one disliked the use of the authority record as a mini-encyclopedia, and one felt this information is too dynamic to be of use in an authority record. Canada supported the proposal but did not like the word "other" in the title of 374$d.  They were also concerned that the word designation in the title is misleading since they did not think this would ever be used as a designation or qualifier for the family.

MAC Discussion:
Matthew Haugen (Columbia University) introduced the proposal. In response to some of the pre-meeting comments, Matthew said that the authors of this paper followed other MARC field and subfield name labels, definitions, etc. as precedents.

Thurstan Young (BL) appreciated that section 2.6 addressed some ethical concerns around the recording of this type of information for family names. He encouraged MAC to include some additional examples within the documentation to address ongoing ethical concerns with encoding these types of data attributes for family names, personal names, etc.

Adam Schiff (SAC) agreed with Thurstan Young (BL), and noted that some of these examples were consciously selected to illustrate this point: i.e., they were mostly not for living families and any examples of addresses had been omitted.

John Myers (CC:DA) suggested that MAC discuss some of the concerns raised around recording an Occupation for a Family Name NAR (i.e., Section 2.4).

Kate Peck (AALL), speaking from a Rare Book community perspective, indicated this information could often be helpful.

Adam Schiff (SAC) pointed out that the MARC 21 format is content-agnostic.

Everett Allgood (NYU, Secretary) raised this “Entity-through-Time” point before within Serials discussion. Family entities, like Serial entities, exist through time – often long expanses of time. Assigning one, or even a few "occupation" attributes to such an entity description appears illogical and incongruous. This is one reason that Subject attributes & Classification for Serial descriptions are typically assigned at a very broad, general level; so that they apply and persist over time. He preferred to see MAC recommend the "Field of Activity" attribute for Family Name NARS instead of "Occupation".

Adam Schiff (SAC) understood this concern. As a reminder though, he noted that just because a MARC field is available does not make it required. Many Family entity descriptions will likely not include these designations. But when it is known and useful for identification/disambiguation, the ability to record it will be valuable.

Jay Weitz (OCLC) asked Kate Peck (AALL) and the Rare Book community to address the value or preference here of using "Occupation" rather than "Field of Activity".

Kate Peck (AALL) and Jackie Parascandola (RBMS) spoke in favor of including the option to record "Occupation" for Family Name entity descriptions.

MAC Action:
Proposal approved as submitted.


PROPOSAL 2023-06:
Defining a New Field for Cluster ISSNs in the MARC 21 Bibliographic, Authority, and Holdings Formats
URL: https://www.loc.gov/marc/mac/2023/2023-06.html
Source: ISSN Review Group
Summary: This paper proposes the creation of a new MARC field 023 in the MARC 21 Bibliographic, Authority, and Holdings formats to store Cluster ISSNs. Cluster ISSNs, such as ISSN-L, identify a group of related resources as defined in ISO 3297:2020.
Related Documents: 2021-DP07; 2007-DP03; 2007-05; 2020-DP11; 2021-04

Summary of pre-meeting comments:
There was general support for this proposal from Australia, OCLC, Spain, MLA, LC, Canada, CC:DA and PCC. NLM supported this proposal but was concerned about the amount of lead time that will be needed to implement. Germany had general support, but wondered if the intent is to have a bibliographic or an authority record for the cluster as a whole. They were also unsure about whether the 76X-78X block needs a unique subfield for this number. The main concerns raised by Britain centered upon who the key players are that the paper mentions.  They were also worried about what happens if some cataloging communities choose not to implement this new field.  They would also like to see what this would look like to users without a graphical display. In addition, they questioned why the paper seeks the definition of both subfield $0 and $1 to record the ISSN as a URI.

MAC Discussion:
Deanna White (ISSN International Centre) introduced the proposal. Responding to some pre-meeting comments, Deanna addressed the amount of lead time concern – per usual, this would be at least 6 months. But the ISSN IC would clearly want to work with each of the distribution nodes, OCLC, et al., to coordinate this properly. A move from recording cluster identifiers in field 022 to 023 could occur quickly, although there may be a need for the community to maintain its use of 022 $l and $m for now. As for choosing whether subfield $0 or $1 would be more appropriate to carry a URI, this remains a complex issue. It is currently unclear as to what community guidance would be. The current linked data application profile for ISSN will be reviewed to ensure the correct approach is being taken.

Hong Cui (LAC) supported the paper as written and was pleased to hear that the ISSN IC is initiating a database cleanup project. She wondered if that may have implications of database maintenance within the CONSER database. If so, she wondered whether the ISSN Centre will be working with OCLC and the various national ISSN bodies for distribution, etc.

Deanna White (ISSN International Centre) confirmed this was the case.

Regina Reynolds (LC/ISSN) noted that the 023 field does provide a MARC location for a “title-level” serial identifier across all serial title changes at the Manifestation level. She would like to have confirmation from OCLC that they continue to believe that it will not be difficult for them to move current 022 subfield $l (ISSN-L) and subfield $m (canceled ISSN-L) into the new 023 MARC field once it is available.

Jay Weitz (OCLC) responded that OCLC still believes this will be a relatively straightforward process in terms of moving data to the new 023 field and then later possibly deprecating 022 subfield $l and subfield $m.

Thurstan Young (BL) asked for additional information regarding which vendors, nodes, etc. were contacted and consulted and specifically what they were asked.

Deanna White (ISSN International Centre) responded taht a briefing paper was sent to 20 contacts, including Clarivate, Ithaka, Clarabelle, OCLC and RDA. The ISSN IC also considered responses and conducted some follow-up interviews. No concerns were raised with regard to deprecation of recording cluster identifiers in field 022 and their move to field 023.

John Myers (CC:DA) noted that deprecating field 022 $l and $m would just mean their removal from the MARC 21 format; it would not automatically lead to their removal from local databases.

Thurstan Young (BL) wondered about confusion among users who may search for one ISSN and then retrieve numerous Bib description "hits" in response to ISSN-L, ISSN-H and other ISSN cluster identifiers.

Deanna White (ISSN International Centre) responded that the ISSN Centres are looking carefully at this issue from the perspective of graphical display. The intention is that searching for an ISSN-H in the ISSN portal would display all the records which share that identifier.

John Myers (CC:DA) noted that the ISSN-H should not result in any confusion arising; in his view it was a feature rather than a bug. Serial cataloging is dependent upon a daisy chain of links being established using 780 and 785 fields. If there is a break in this link, then the ISSN-H should provide the means of resolving this problem.

Everett Allgood (NYU, Secretary) pointed out that the search and retrieval concern Thurstan Young (BL) raised is also an indexing and database issue applicable within MARC and library retrieval systems/mechanisms extending far beyond the current paper.

Gaelle Béquet (ISSN International Centre) responded, regarding the user confusion issue, that the ISSN Data Review Group has decided that prospectively Cluster ISSNs will be unique ISSN identifiers, as opposed to the earlier practice of "re-using" the earliest-assigned ISSN to represent both a Manifestation-specific ISSN and the ISSN-L for a Continuing Resource title. This was welcome news for many.

Thurstan Young (BL), addressing the  MARC 21 application of subfields $0 and $1 (which impacts this and other papers under discussion at present), said that there may be a need for a larger conversation than that which MAC can undertake. Advice from linked data experts with more technical knowledge could help avoid a situation arising in which the coding of URIs is inconsistent and previous usage has to be unpacked at a later date.

Regina Reynolds (LC/ISSN) responded that the use of $0 and $1 is open to a community decision on which is the appropriate one to use for the purposes of recording URIs.

Everett Allgood (NYU, Secretary) mentioned that, in the pre-meeting comments, there were suggestions to consider adding these Cluster ISSN subfields within the MARC 760-788 Bibliographic Linking fields. For the Serials cataloging community, this might prove troublesome because of the long, granular practice of using MARC 76X-78X linking fields in order to relate Manifestation-to-Manifestation bibliographic descriptions. Including group-level or Cluster ISSN identifiers within these fields may prove problematic for Serialists. He could not speak though for current or legacy practices within other User Communities.

Deanna White (ISSN International Centre) and others agreed. Part of the mechanism they use to generate and identify Cluster ISSNs is through the Manifestation-specific ISSN assignments catalogers encode within the 76X-78X MARC fields.

Regina Reynolds (LC/ISSN), in light of the comments from Thurstan Young (BL) and others regarding subfields $0 and $1 within this paper, asked that the proposal be approved as written with the understanding that the use of $0 vs. $1 will be determined by the community at a later date.

MAC Action:
Proposal approved as submitted, with the $0 and $1 usage guidelines to be determined at a later date.


MARC DISCUSSION PAPERS

 

DISCUSSION PAPER 2023-DP05: Refinement of Accessibility Field 341 in the MARC 21 Bibliographic Format
URL: https://www.loc.gov/marc/mac/2023/2023-dp05.html
Source: Canadian Committee on Metadata Exchange (CCM)
Summary: This paper suggests adjustments to the Field Definition and Scope of MARC Bibliographic field 341 (Accessibility Content) and to the description of 341 subfield $a (Content access mode).
Related Documents: 2018-03; 2017-11; 2017-DP03

Summary of pre-meeting comments:
All groups agreed that a case was made for the need for these refinements. Germany and MLA did not indicate which option they preferred.  Option 1 was chosen by LC and PCC. Option 2 was preferred by NLM, Australia, Spain, OCLC, OLAC, and CC:DA (by a slight margin).  Britain was split in their choice of options and wondered if the two options could somehow be combined. Spain and NLM did not see the need for the sentence in option 2 beginning "use terms from the same vocabulary…"  They would remove it.  OCLC also did not like this sentence and asked for it to be edited. LC indicated that they preferred option 1 because it was the least disruptive. CC:DA suggested adding the $0/$1 and also they would look at the option in terms of what would work best in BIBFRAME.

MAC Discussion:
Christopher Carr (Concordia University/CCM) introduced the discussion paper, along with mention of the connection to Schema.org vocabulary and implications. In response to some of the pre-meeting comments and questions, he noted that CCM and PCC SCS mutually agreed that CCM would pursue the changes in this paper, while PCC would pursue the subfield $0/$1 separately.

Thurstan Young (BL) suggested that, pursuing the U.K.'s suggestion to combine proposed options 1 and 2, they might consider enhancing Option 1 with the last sentence of Option 2.

John Myers (CC:DA) pointed out that MAC members had already expressed resistance to the last sentence of Option 2.

Thurstan Young (BL) acknowledged concerns of the wider community on the last sentence of Option 2.

Christopher Carr (CCM) said that the sentence would change depending on the option chosen by MAC.

Kevin Ford (LC) mentioned the backwards compatibility problem(s) if MAC suddenly redefines the use of this field and the applicability of subfield $2 when there are existing occurrences of the field encoded following the earlier definition(s) and field instructions. This may result in a certain disconnect between legacy use of the field and prospective uses.

Jay Weitz (OCLC) pointed out that a quick look at the OCLC WorldCat database reveals that the use of this 341 field has been limited (maybe thousands of bibliographic record occurrences, perhaps fewer). Historically, the MARC community has not had a controlled vocabulary for this field; now it does (i.e., Schema.org, etc.). Going forward, catalogers are more likely to use that or another vocabulary and therefore wish to cite this usage in subfield $2. This may represent a break with the past and with legacy data, but it will be understandable.

Sally McCallum (LC) and others observed Adam Schiff’s (SAC) description of repeating MARC fields when necessary to use terms from multiple vocabularies. Additionally though, MARC does not and has not made certain subfields mandatory – such as suggesting here that going forward subfield $2 might be required for 341 field occurrences. By not requiring subfield $2 for each occurrence MARC allows for the use of uncontrolled terms here and elsewhere throughout the formats.

Ethan D’Ver (Juilliard) pointed to previous MAC discussion and development for the 382 (Medium of Performance) field. This may prove helpful.

Christopher Carr (CCM) asked if MAC could move forward with the paper by providing the subfield $2, but not making it mandatory, and not suggesting/requiring use of a specific controlled vocabulary. After all, other additional vocabularies may be developed in the future.

A straw poll of attendees demonstrated a slight preference for Option 2 over Option 1 with 13 in favour of the former and 10 in favour of the latter.

Thurstan Young (BL) noted that, as currently written, there is no requirement of using terminology from a controlled vocabulary; if a cataloger/library does use controlled language terms, then they all need to come from the same vocabulary identified in subfield $2.

Sally McCallum (LC) mentioned that this is the case with many MARC fields, especially others in the 34X range of fields.

Ethan D’Ver (Juilliard) queried whether MAC might instead consider the use of indicator values to capture some of what CCM is currently encoding in different subfields if this would by helpful.

Adam Schiff (SAC) responded that indicator values are generally not used for the purpose of flagging a $2 code in 3XX fields. A normal approach would be to repeat the field if some terms are controlled and others are not.

Sally McCallum (LC) pointed out that terms from different source vocabularies would also be recorded in separate iterations of the field.

Christopher Carr (CCM) asked for feedback from MAC on Section 3.1 of the paper regarding the suggestion to revise the Field Definition and Scope of the 341 field.

Jay Weitz (OCLC) suggested that the 341 field may now need a tighter definition; he felt that the wording of subfield $a was clear.

Ethan D’Ver (Juilliard) asked whether the 341 field definition change set out in the paper could be the basis of a fast track proposal.

Thurstan Young (BL) responded that, while the definition change could be regarded as the less complex amendment under discussion, treating it as fast track may be a different matter.

Sally McCallum (LC) said that NDMSO was happy to work with CCM on how to take the existing paper forward.

Thurstan Young (BL) suggested that the paper's authors consider the approach taken to recording controlled vocabularies set out in MARC 21 field range 34X and the 382 field as regards precedent. He welcomed the prospect of seeing changes made to MARC 21 documentation which would replace the longstanding placeholders for accessibility vocabulary terms with actual examples.

The paper will return as a proposal.


DISCUSSION PAPER 2023-DP06
:
Adding Subfield $3 to Field 532 in the MARC 21 Bibliographic Format
URL: https://www.loc.gov/marc/mac/2023/2023-dp06.html
Source: National Library of Finland, Canadian Committee on Metadata Exchange (CCM), Online Audiovisual Catalogers (OLAC)
Summary: This paper suggests adding new subfield $3 (Materials specified) in field 532 (Accessibility Note) in the MARC 21 Bibliographic Format to allow for expressing which part of the resource different 532 fields refer to.
Related Documents: 2017-DP03; 2018-03; 2022-DP01; 2022-DP07

Summary of pre-meeting comments:

There was support from NLM, Australia, Spain, LC, OCLC, OLAC, CC:DA, Germany, MLA, PCC, Canada and Britain.  Some of these groups also suggested that this be fast tracked. OCLC commented that specificity be encouraged in the $3 and the use of the $8 should be considered for ease of BIBFRAME sequencing. OLAC and Canada indicated that they would like to see a non-media example and perhaps one that pointed out hazards or deficiencies. LC would encourage the use of the 341 with the 532.

MAC Discussion:
Tatja Pusa (National Library of Finland) introduced the discussion paper.

Adam Schiff (SAC) agreed with the sentiments of others in saying that this paper definitely appears to be a candidate for the MAC Fast-Track process.

Everett Allgood (NYU, Secretary) suggested that best practices might help to determine what labels should be recorded in subfield $3 going forward. This would be helpful in terms of analyzing entities over time.

A straw poll of MAC members also indicated unanimous (17-0) support for Fast-Track treatment.

The paper was referred to the MARC Steering Group for final approval as a Fast-Track proposal.


DISCUSSION PAPER 2023-DP07
:
Recording Numeric Designation of Musical Expressions in the MARC 21 Bibliographic and Authority Formats
URL: https://www.loc.gov/marc/mac/2023/2023-dp07.html
Source: Music Library Association (MLA)
Summary: This paper suggests changing the name of and defining an indicator for Field 383 (Numeric Designation of Musical Work) in the MARC 21 Bibliographic and Authority Formats to allow for recording numeric designations of musical expressions.
Related Documents: [none]

Summary of pre-meeting comments:
Supported by Australia, Spain, OCLC, OLAC, CC:DA, MLA, and PCC. Germany and Britain commented on the need to wait for this element to be added in RDA; however, the RSC representative sent clarification that it is unlikely that any change will be made in the guidance about this, because the current policy is to not type identifiers.  She indicated that all of the "typed" identifiers that currently exist in the new RDA Toolkit were grandfathered in from the original Toolkit. Canada supported this paper but was not convinced an indicator is needed to differentiate work and expression.  Likewise, LC thought the indicator value is not needed since the 1XX/245 or 1XX will match whatever entity is there.  LC would also prefer the word "entity" be used rather than "work" or "expression". Britain would like an example of 1st indicator 0.  They also wondered if there is a need to differentiate between RDA Work and BIBFRAME Work.

MAC Discussion:
Ethan D'Ver (The Juilliard School) introduced the discussion paper and responeded to some of the pre-meeting comments. He appreciated the British Library's request for an example and more information regarding the proposed 1st indicator "0" (i.e., Work).

Adam Schiff (SAC) pointed out that in the bibliographic record, the use of indicator values appears appropriate because bibliographic descriptions often include both Work and Expression attributes. But these indicators are not necessary or appropriate within the Authority format/context. That is, the 1XX entity in the applicable Name Authority Record (as well as the 075 field) will represent either a Work entity or an Expression entity – the indicator values would be unnecessary.

Thurstan Young (BL) agreed with Adam's point distinguishing between the different uses of this field within bibliographic and authority entity descriptions. In an authority description, field 075 could be used to flag the entity described by the record as a whole.

Cate Gerhart (University of Washington, Chair) suggested that perhaps indicator values could be considered for bibliographic descriptions, but not in the Authority format.

Adam Schiff (SAC) mentioned that when creating/maintaining name authority entity descriptions, it is important to retain the principle of describing only the entity in hand. That is, within Name Authority Records (NARs) most if not all descriptive attributes relate directly to the 1XX entity described.

John Myers (CC:DA) wondered about overlap between this paper and the ways in which Representative Expression NARs are assembled and created.

Adam Schiff (SAC) described the distinction between Work and Expression NARs and the importance of keeping them clear and distinct. He also reminded MAC that the Representative Expression NAR inherits a great deal of data, often including at least the initial base of the 1XX text string, with added attributes as necessary to distinguish/identify (or disambiguate). Representative Expression entities belong in the Work entity pool, not in the Expression entity pool.

Thurstan Young (BL) and others agreed.

Kathy Glennan (University of Maryland/RSC Past Chair) clarified that there was no way that any Representative Expression entity, from an RDA modeling perspective, would represent anything other than a canonical Work entity.

Ethan d’ Ver (Juilliard) responded that he had discussed the matter of Representative Expressions with Keith Knop (MLA) and concluded that the changes sought to field 383 bore little relevance to a canonical Work entity.

Thurstan Young (BL) agreed, adding that music specific representative expression characteristics were limited to medium of performance and key. Field 383 on the other hand was focused on recording a thematic index, serial or opus number.

Ethan d’ Ver (Juilliard) added that he was under the impression that a Bibliographic format option for recording numeric designations of musical expressions was preferred to the Authority format.

Pat Riva (Concordia University/CCM) commented that there was simply a need to find a place for catalogers to record these Expression identifiers in the MARC formats. Most music folks who produce and generate these identifiers are unaware of the nitty-gritty distinctions within the cataloging community between Works, Expressions, and Representative Expressions, etc. From this viewpoint, recording them in the same field makes a great deal of sense.

The paper will return as a proposal.


DISCUSSION PAPER 2023-DP08
:
Adding Subfields $0, $1, and $5 to Linking Fields 76X-78X in the MARC 21 Bibliographic Format
URL: https://www.loc.gov/marc/mac/2023/2023-dp08.html
Source: OCLC
Summary: This paper proposes adding subfields $0 (Authority record control number or standard number), $1 (Real World Object URI), and $5 (Institution to which field applies) in linking entry fields 760 (Main Series Entry), 762 (Subseries Entry), 765 (Original Language Entry), 767 (Translation Entry), 770 (Supplement/Special Issue Entry), 772 (Supplement Parent Entry), 773 (Host Item Entry), 774 (Constitute Unity Entry), 775 (Other edition), 776 (Additional Physical Form), 777 (Issued With Entry), 780 (Preceding Entry), 785 (Succeeding Entry), 786 (Data Source Entry), and 787 (Other Relationship Entry), and 788 (Parallel Description in Another Language of Cataloging) in the MARC 21 Bibliographic Format.
Related Documents: 2010-06; 2016-DP04; 2016-DP17; 2016-DP19; 2017-08; 2019-03; 2020-FT02; 2020-FT03; 2021-04

Summary of pre-meeting comments:
There was general agreement that at least for some of the 76X-78X fields the $0 and $1 were needed.  The 773, 774 & 776 were the most preferred fields for the inclusion of the $0 and $1. Many were not convinced the $5 was needed in all of the fields in the range, but all agreed that if it came back either as a proposal or as another discussion paper it should be separate from the paper about the $0 and $1. NLM and Spain had difficulty understanding the paper, especially the examples.  NLM also did not like the linking of both Work and Manifestation in one field and wondered if it would be possible to repeat the field if you need both. Britain suggested that the new $l (el) for data provenance might be used instead of the $5.  They also pointed out that the options seem to point to best practices rather than MARC coding.  They found Option 1 and 3 problematic because URIs are not used for portions of access points.  Also, the 76X-78X block is used to link Bibs not Authorities, so Work is out of scope.  Lastly, they said the URIs in the examples are wrong. LC thought that we should wait to see if this is really needed.  It is possible that the $w and the ISSN/ISBN will be enough to do the linking. Everett Allgood (NYU, Secretary) pointed out that in the example at section 2.2, the 500 should also have a $5. Option 2 was preferred by OLAC, Germany and some of CC:DA. Option 3 was preferred by MLA. Multiple people pointed out that the first example in 2.2 should be field 767 not 762.

MAC Discussion:
Hayley Moreno (OCLC) introduced the discussion paper. Regarding the subfield $5 component OCLC agreed to the suggestion that this may need breaking out into a separate MARC paper. OCLC would like additional guidance on whether this should come back as a Discussion Paper or as a Proposal.

Everett Allgood (NYU, Secretary) and Regina Reynolds (ISSN/LC) both thought that a Discussion Paper may be appropriate because there appears to be uncertainty as to whether subfield $5 is necessary in all of the 760-788 bibliographic linking fields.

Jackie Parascandola (RBMS) expressed support for subfield $5 and would welcome the opportunity to work with OCLC on the follow-up paper.

Everett Allgood (NYU, Secretary) and others encouraged OCLC and RBMS to think through this carefully. If/when the paper returns as a proposal, OCLC should enumerate specific, practical use cases wherein subfield $5 is needed for specific bibliographic linking fields. The Serials Cataloging Community encodes these 76X-78X Bib Linking fields extensively to describe very specific relationships; tweaking these MARC Linking fields will need to take such community practices into account.

Hayley Moreno (OCLC), moving on to the subfield $0/$1 components, said that OCLC would like additional guidance from MAC and from the MARC cataloging community. They needed feedback regarding where URIs belong in MARC.

Thurstan Young (BL) agreed and added that since the same issue had come up in several recent MARC paper discussions, MAC needed to consider and carefully think it through.

Regina Reynolds (LC/ISSN)–Added that Deanna White (ISSN International Centre) and many others at the ISSN Centre are also really interested in additional MAC feedback and guidance re: subfields $0 and $1. Also as a Serialist who is aware of how MARC bibliographic descriptions often include numerous WEMI entity attributes within a single Manifestation description, she was uncomfortable with this proposal overall. Within the Serials Cataloging Community at least, the 760-787 bib linking fields are used almost exclusively to link specific manifestations to related  manifestations. For Serial & Continuing Resource bibliographic descriptions, including subfield $0 and/or $1 URI links might be counterproductive.

Everett Allgood (NYU, Secretary) said that, as a Serials cataloger, he completely concurred with Regina's comments, and shared in her uneasiness. Having spent many years creating and maintaining Serial bibliographic descriptions and the necessary, granular bibliographic relationship links among them, he had difficulty envisioning how incorporating subfield $0 or $1 URIs into these relationship fields might work alongside existing practices. CONSER and the Serials cataloging community have very specific legacy linking practices regarding bibliographic descriptions in order to optimize descriptive accuracy and user displays. He felt that the Serials community would need additional information describing OCLC's plans for this as well as the desired results and outcome.

Adam Schiff (SAC) pointed out that one platform MAC could look at and consider is Wikidata. The Wikidata platform contains clearly articulated Work, Expression and Manifestation entities containing Wikidata Real World Object (RWO) links (i.e., subfield $1 in MARC encoding). With the caveat that many OCLC bibliographic descriptions contain piecemeal Work and Expression attributes, for the most part, Bibliographic descriptions in OCLC represent Manifestation entities.

Hayley Moreno (OCLC) felt that MAC was leaning towards choosing Option 2 as a basis for making changes to the Bibliographic format linking fields.

Kevin Ford (LC) suggested that Option 2 would be preferable in terms of providing an unambiguous link.

Thurstan Young (BL) commented that it was important to align community practices when including URIs within bibliographic linking fields; the included URIs need to describe the content of the entire MARC field in question, not individual components of it. Fields 76-78X are used to record controlled values for manifestations, but not authorized ones because they do not identify Work or Expression entities. This is in contrast to fields 700, 710, 711 and 730. If the current paper is followed up by OCLC, then any subsequent study should focus on the unique use case for identifying manifestations which is supported by bibliographic linking fields.

Adam Schiff (SAC) spoke to the importance of using URIs based on the entity described and of distinguishing between whether  the URI represented an authority identifier or a Real World Object (RWO).

Reinhold Heuvelmann (DNB) suggested the possibility of extending the scope of $w to contain a URI. MAC has done a similar thing with $0, where it initially contained only an authority record control number but was then expanded for a standard identifier and then a URI. Something similar could be done with $w in the case of linking fields.

Kirk Hess (OCLC) expressed the view that bibliographic linking fields are a bit of a hodgepodge, possibly warranting some compromises. Responding to Reinhold Heuvelmann's (DNB) $w solution, it sounded good to him. However, if MAC did pursue the use of subfield $w for recording URIs, then the result would be four MARC separate subfields to hold URIs. Hence OCLC's desire for additional MAC discussion and feedback.

Everett Allgood (NYU, Secretary) picked up on comments from Kirk Hess (OCLC) and others regarding the tensions in MARC bibliographic linking fields that sometimes try to perform multiple duties. Theoretically, the MARC 21 standard allows these 76X-78X links to point to related works. However, in practice the cataloging community has applied the encoding standard in various ways: 1) When the necessary specific bibliographic description (i.e., OCLC Manifestations) exists, this precise link is provided; 2) When the necessary Bib description is not yet cataloged and available to create a Manifestation-to-Manifestation link, the 76X-78X field often provides a broader Work/Expression entity link containing only MARC subfields $a and $t, etc. to represent the related Entity. As a result, different 76X-78X linking fields sometimes reference different WEM entity descriptions. For the CONSER and serials cataloging community these distinctions and legacy practices are vitally important and must remain clear.  He encouraged OCLC to include or consult CONSER and possibly some serials catalogers as this paper moved forward.

Sally McCallum (LC) agreed with a great deal of what had been said. She was not clear why MARC had ended up with so many 76X-78X field subfields other than to cram in more and more data attributes from the related record. As Everett Allgood (NYU, Secretary) and others mentioned, theoretically the MARC Standard was not necessarily intended to link to a specific entity (i.e., Manifestation). She suspected that in many 76X-78X field occurrences most of their subfields were not used.

Kirk Hess (OCLC) agreed with Sally McCallum (LC) that some subfields were not used much, but OCLC could look more closely to see which cataloging communities were using them.

Cate Gerhart (University of Washington, Chair) wondered whether, considering the discussion thus far and in particular the uncertainty around $0/$1 application, a follow up Discussion Paper would be preferable to a Proposal.

Hayley Moreno (OCLC) thanked MAC for its feedback and added that OCLC would likely bring this matter back as another Discussion Paper in order to continue the discussion.

The subfield $5 content will be broken off into a separate paper and will likely return as a proposal. The $0/$1 content will likely return as another discussion paper.


DISCUSSION PAPER 2023-DP09
:
Adding Subfields $0 and $1 to Fields 082 and 083 in the MARC 21 Formats
URL: https://www.loc.gov/marc/mac/2023/2023-dp09.html
Source: Dewey Editorial Team, OCLC
Summary: This paper proposes adding subfields $0 (Authority record control number or standard number) and $1 (Real World Object URI) to fields 082 (Dewey Decimal Classification Number) and 083 (Additional Dewey Decimal Classification Number) in MARC 21 Authority, Bibliographic, and Community Information formats.
Related Documents: 2017-04; 2021-04; 2020-FT03; 2020-FT02

Summary of pre-meeting comments:
There was general support for this paper from Australia, NLM, Spain, OLAC, CC:DA, PCC, Germany, MLA, Canada, and LC. Most preferred the $0 be used including NLM, OLAC, CC:DA , and LC. However, Spain thinks the $1 is the only acceptable choice. Germany wasn’t sure if it should be $0 or $1. Britain commented that the choice of $0 or $1 is a best practice decision and depends on how DDC would be modeled as linked data. From the perspective of the SKOS Primer a classification scheme is modeled as a concept scheme. This being the case, $0 may be preferable to $1, since the latter subfield is only intended to code Real World Object URIs. The paper does not describe how a DDC URI would relate to the preceding content in field 082 and 083. It would be useful to know whether it expresses the built number recorded in subfield $a alone or the built number, edition and timestamp.

MAC Discussion:
Alex Kyrios (OCLC) introduced the discussion paper.

Hong Cui (LAC) asked for clarification regarding whether this paper included consideration of URIs for both the 23rd Edition of Dewey and the Abridged Edition.

Alex Kyrios (OCLC) recommended against creating or using URIs for the Abridged Edition Dewey numbers.

Reinhold Heuvelmann (DNB) asked whether Alex Kyrios (OCLC) has a rough idea of how many URIs would be needed if URIs are created for the entire Dewey Classification scheme.

Alex Kyrios (OCLC) responded that approximately 44,000 would be required on the basis that only non-built numbers would be in scope.

Cate Gerhart (University of Washington, Chair) asked Alex if he had what he needed to follow up this Discussion Paper.

Alex Kyrios (OCLC) replied yes, subject to other possible considerations arising from the upcoming discussion concerning subfields $0 and $1 in MARC 21 overall.

The paper will return as a proposal.

 

MAC Discussion: URIs in Subfields $0 and $1 of the MARC 21 Formats

 

BACKGROUND:

As an addendum to its original agenda, MAC agreed to include a discussion about encoding URIs in the following subfields as part of its MARC Advisory Committee Annual 2023 meetings:

$0 (Authority record control number or standard number)
$1 (Real World Object URI)    

(cf. https://www.loc.gov/marc/bibliographic/ecbdcntf.html)

This was on the basis of observations shared by Thurstan Young (BL) as part of the BIC Library Metadata Group's pre-meeting comments on the MARC listserv: https://listserv.loc.gov/cgi-bin/wa?A2=ind2306&L=MARC&P=6987

Many of the MAC papers currently under discussion include both subfields $0 and $1 amongst their proposed lists of additions to the formats:

The approach of routinely proposing both $0 and $1 to encode URIs in MARC 21 may indicate that a degree of uncertainty exists within the community as to which subfield is required on a case by case basis. If this is correct, then MAC itself is not necessarily the appropriate forum for establishing best practices of application going forward. However, wider consultation with the linked data community may offer a way of making more informed choices on whether to record either of the following to meet a specific need:

1) URIs that identify a "Record" or "Authority" entity describing a Thing (e.g., madsrdf:Authorities, SKOS Concepts for terms in controlled or standard vocabulary lists);

2) URIs that directly identify a Thing itself (sometimes referred to as a Real World Object or RWO, whether actual or conceptual).

DISCUSSION:

Thurstan Young (BL) provided a brief introduction to the discussion based on the comments set out above.

Cate Gerhart (University of Washington, Chair) asked whether MAC should continue to make the distinction between whether to record URIs in subfield $0 or subfield $1. Is MAC the best, most appropriate venue for making this decision, especially considering that these decisions now come up routinely, often proposed within multiple papers at each MAC meeting?

Adam Schiff (SAC) responded that it was important to continue to distinguish between subfield $0 Authority identifier/entity description URIs and subfield $1 Real World Object URIs. This is especially important because of the manner in which these URIs can be acted upon within a Semantic Web environment.

Regina Reynolds (LC/ISSN) said that when the ISSN Centre was writing its most recent paper (Proposal 2023-06), the authors struggled with this distinction between Authority identifiers and Real World Objects. They questioned what a Real World Object was in this context, what its parameters were and how catalogers should define such entities or objects.

John Myers (CC:DA) said that discussions around these issues and trying to understand these distinctions continued to be the thing which he struggles with. In talking with others, he believed that many of his colleagues felt similarly.

Adam Schiff (SAC) pointed out that Authority identifiers are not actionable in the Semantic Web context; though they are routinely maintained, these URIs are considered "static"; They are internally cohesive and rigorous, but not outside of their own context or platform. Real World Object (RWO) URIs in subfield $1 represent actionable URIs that are cohesive both within the Semantic Web and among wider, different platforms. Real World Object URIs lack provenance data though.

John Myers (CC:DA) noted that Real World Objects (RWO) may be equivalent, while authority identifier URIs may not.

Adam Schiff (SAC) said that it is the subfield $1 URI that really makes the Linked Data functionality operate, not the subfield $0 containing authority work.

Corine Deliot (British Library) noted that it is important that we include SKOS (Simple Knowledge Organization System) within this discussion; the subfield $0 & $1 discussion is really restricted to the library world domain. In principle the context and the modeling are critical, especially in relation to the role of a particular URI. She referenced one of the examples in the MARC 022 documentation and how the formulation of that URI contradicts some ISSN examples in the most recent 022/023 field Proposal. Referring to the ISSN Centre linked data model and application profile [https://www.issn.org/wp-content/uploads/2020/09/ISSN-LinkedDataApplicationProfile-v2_0.pdf], she pointed out that ISSN URIs with the pattern "http://issn.org/resource/ISSN/{ISSN}" identify the ISSN resource, i.e., a continuing resource while ISSN URIs with the pattern "http://issn.org/resource/ISSN/{ISSN}#ISSN" represent the ISSN identifier itself. Hence http://issn.org/resource/ISSN/{ISSN} is recorded in subfield $1 while http://issn.org/resource/ISSN/{ISSN}#ISSN is recorded in subfield $0 (as per the example in the MARC 022 documentation).

Thurstan Young (BL) said that, going forward, it will be really important for MARC paper authors to stipulate whether they need subfield $0 or subfield $1 based on the context of the paper and the modeling in question. If MAC proposes one particular approach based on specific modeling, it will be critical that MAC and the Library Community remain consistent with that approach.

Regina Reynolds (LC/ISSN) commented that Corine's observations were helpful. It made her think that with her input and assistance, the ISSN Centres would be able to model the ISSN platform context correctly. They believe that they will need both the subfield $0 in order to point to ISSN identifiers, and also subfield $1 to point to the ISSN resource Real World Object. In reference to Thurstan's introductory comments about recent MAC papers and the presence of both subfield $0 & $1, Regina suspected authors are likely modeling based on precedents. Libraries and librarians want to jump on the Linked Data bandwagon, but many/most do not know enough about it to model such concepts accurately. Hence the "easy answer" has been to follow preceding or earlier MARC proposal papers.

Thurstan Young (BL) said that, based on MAC's approval of the ISSN Paper at this meeting (Proposal No. 2023-06), how MAC and the ISSN community put this into practice and deploy the MARC 023 field will be very important.

John Myers (CC:DA) agreed and added that the current discussion demonstrated the need for MARC paper authors to continue to distinguish between these two types of URIs, as Thurstan Young (BL) said, based on the context and modeling of each MARC paper.

Charlene Chou (New York University/PCC) said that during the PCC OpCo Meetings in May, they considered $1 in several fields. The major issue is including identifiers in $0, but not as a URI. She mentioned that a new PCC Group would be considering/proposing URI guidelines. It would look at these Authority identifier URIs and Real World Object URIs in order to provide clear guidance and Best Practices throughout the PCC and cataloging community.

Adam Schiff (SAC) reminded people about the Formulating and Obtaining URIs in MARC PDF document that Adam has now converted into a Wikidata table at: https://www.wikidata.org/wiki/Wikidata:WikiProject_URIs_in_MARC. He emphasized the importance of how the individual URIs are modeled within each specific context. For instance, one might look at the ULAN example recommending that both types of URIs be provided, along with instructions for creating them.

Corine Deliot (British Library) expressed concern about the suggestion of possibly providing URIs in subfield $w (as suggested during deliberations over Discussion Paper No. 2023-DP08). The MARC community already has confusion regarding whether and when to use subfield $0 or $1; she questioned whether the community really wanted to include yet another encoding option for catalogers. MAC and the MARC Community are not experts in this Linked Data/Semantic Web environment. But in the five years since the PCC URIs Best Practices document was issued, there has been additional development within the Semantic Web/Linked Data community. It would be both instructive and wonderful for MAC to hear from some of the experts in that community who can inform and guide it. Perhaps the library community should not be recording and encoding these URIs in separate subfields if it is not fully understanding the need, the output, the eventual repercussions, etc.

Regina Reynolds (LC/ISSN) commented that, regarding the 023 paper (Proposal No. 2023-06), the ISSN Centre authors definitely had a great deal to think about. The suggestion during discussion for that paper that we may not include or implement subfield $0 or subfield $1 with the 023 MARC field right away was certainly something to consider. That implementation might be delayed based on this and further discussion, though she was not certain. As an international organization with many active, inputting constituents, the ISSN Centre cannot necessarily always follow either PCC or the RDA Steering Committee (RSC). Different ISSN Center members sometimes engage in different practices.

Charlene Chou (New York University/PCC) asked if Kevin Ford (LC) or Sally McCallum (LC) could provide additional information concerning why the current MARC-to-BIBFRAME crosswalk guidelines say that subfield $1 is not being included when cross-walking entity descriptions from MARC to BIBFRAME; subfield $0 is included, but not subfield $1.

Kevin Ford (LC) responded that LC's current thinking and modeling indicates that the library community primarily depends upon identifiers in subfield $0.

Adam Schiff (SAC) countered that the PCC Wikidata Community encodes URIs in subfield $1. Also there have been recent PCC discussions concerning NACO-Lite, SACO-Lite, etc. and creating some authority entity descriptions outside of MARC. If some of those descriptions are encoded in Wikidata, MARC will need subfield $1 in order to reference them, following PCC practice.

Cate Gerhart (University of Washington, Chair) asked whether it was possible to define rules (or consistent modeling) to help MARC paper authors know which subfield to use in the future.

John Myers (CC:DA) responded that MAC provides guidance on the encoding structure only. The MARC Communications Formats are intended to be content-agnostic. The MAC group members can provide guidance on what it means to model one way or the other and might recommend one subfield or the other, but implementation decisions are then left to user communities.

Cate Gerhart (University of Washington, Chair) expressed the hope that the Discussion Paper authors have a better idea as regards the handling of $0 and $1 going forward.

Regina Reynolds (LC/ISSN) commented that the ISSN Centre authors had a lot to consider. They may not immediately implement subfields $0/$1 in the new 023 field at the outset.

Hayley Moreno (OCLC) commented that the discussion had been helpful. OCLC was okay with subfield $1, but based on today's discussion, OCLC and the authors will also explore additional use case modeling when submitting a recommended subfield in any follow-up papers.



Respectfully submitted,
Everett Allgood


MARC 21 HOME >> MAC

The Library of Congress >> Librarians, Archivists >> Standards
( 11/16/2023 )
Legal | External Link Disclaimer

Contact Us