The Battle of RDA: Victors or Victims

Rick J. Block

Columbia University

Rick Block On RDA:

“I think it is a disaster. I'm hoping it is never implemented.”

Library Journal Nov. 15, 2008

Rick Block On MARC:

Unlike some of his colleagues, he believes the MARC record has a future. He points out the example that Columbia has invested a great deal in it, even in its electronic displays. “We have millions of records in MARC,” says Block, “so I don't think it will go away.”

Library Journal Nov. 15, 2008

Rick Block on ?:

“When I was in library school in the early ’80s, the students weren’t as interesting”

New York Times July 8, 2007

A Hipper Crowd of Shushers

“Still I can not help thinking that the golden age of cataloging is over, and that the difficulties and discussions which have furnished an innocent pleasure to so many will interest them no more. Another lost art.”

Charles A. CutterPreface, 4th ed. Rules for a Dictionary

Catalog (1904)

“Several principles direct the construction of cataloguing codes. The highest is the convenience of the user.”

Statement of International Cataloguing Principles (IFLA, 2009)

Why me? My perspective

• I’ve been quoted

• I ignored it as long as I could

• I’m a teacher and a practitioner

• I’m struggling to understand RDA

• I’ve not lived through a code change

• Goal for today: present a balanced view of RDA as I understand it

Deja Vu All Over Again!

• The War of AACR2: Victors or Victims.– Charles Martell. Journal of Academic

Librarianship. Vol. 7. no. 1 (1981)

• The War of AACR2– Michael Gorman. Our Singular Strengths:

Meditations for Librarians

RDA: Wikipedia Disambiguation

• Radioactive Dentin Abrasion

• Redland Railway Station

• Recommended Daily Allowance

• Remote Database Access

• Reader's Digest Association

• Retirement Date Announced

Naming the Code

• RDA – an international standard• Took “Anglo-American” out of title

– Even AACR2 used internationally• Translated into 25 different languages• Used in 45 countries outside the U.S.

• Took “Cataloguing” out of title– “Resource description” better understood by metadata

communities– Will still include basic principles of bibliographic

description

Why New Cataloging Rules?

• Feeling that continued revision of AACR2 not sufficient to address issues– Evolving formats, including items that belong

to more than one class of material– Limitations with existing GMDs and SMDs– Integrating resources– Separation of “content” and “carrier” concepts

• Integrate FRBR principles

RDA Big Picture Concepts

• Designed for the digital world

• Founded on AACR

• Informed by FRBR and FRAR

• Consistent, flexible and extensible framework

• Compatible with international principles, models and standards

• Useable outside the library community

Why Not AACR3?

AACR3

Why Not AACR3?

• Reviewers of AACR3 Part I (2004-05) identified areas for improvement:– Proposed structure of rules – too awkward– More metadata-friendly; less library jargon– More connection to FRBR– Modify the connection of the rules to ISBD– Changes need to be significant enough to

merit a new cataloging code, but records still need to be compatible with AACR2

RDA is …

• “RDA is a content standard, not a display standard and not a metadata schema. RDA is a set of guidelines that indicates how to describe a resource, focusing on the pieces of information (or attributes) that a user is most likely to need to know. It also encourages the description of relationships between related resources and between resources and persons or bodies that contributed to creation of that resource.” (Oliver, 2007, Changing to RDA)

RDA …

• A FRBR-based approach to structuring bibliographic data

• More explicitly machine-friendly linkages (preferably with URIs)

• More emphasis on relationships and roles

• Less reliance on cataloger-created notes and text strings (particularly for identification)

What RDA is intended to be

• A content standard• A set of guidelines• Focused on user tasks (Find, Identify,

Select, Obtain mantra throughout)• An online product (with possible print

“derivatives”)• A more international standard• An effort to make library catalog data play

better in the Web environment

What RDA is intended to be

• Change in view from classes of materials in libraries to elements and relationships for entities in the bibliographic universe

• May be used with many encoding schema such as MODS, MARC, Dublin Core

• An attempt to improve the way we describe and present relationships among resources and bibliographic entities

• Flexible and adaptable

What it is NOT intended to be

• A display or presentation standard

• A metadata schema

• A rigid set of rules

• Structured around ISBD areas and elements

• Instructions on creating and formatting subject headings (yet)

• Instructions on classification numbers

Goals of RDA

• Provide consistent, flexible, and extensible framework for description of all types of resources and all types of content

• Be compatible with internationally established principles, models and standards

• Be usable primarily within the library community, but be capable of adaptation for other communities (e.g. archives and museums)

• Be compatible with descriptions and access points devised using AACR2 in existing catalogs and databases

Goals of RDA

• Written in plain English, and able to be used in other language communities

• Be independent of the format, medium, or system used to store or communicate this data

• Be readily adaptable to newly-emerging database structures

Foundations and Influences

• FRBR (Functional Requirements for Bibliographic Records)

• FRAD (Functional Requirements for Authority Data)

• AACR2

• Paris Principles (“Statement of International Cataloguing Principles” 2009 version)

• ISBD (International Standard Bibliographic Description) But RDA does not follow ISBD order and ISBD punctuation is no longer required.

Stakeholders• Joint Steering Committee for Development of Resource Description

and Access• American Library Association (ALA) • Association for Library Collections and Technical Services (ALCTS)• Cataloging and Classification Section• RDA Implementation Task Force• Australian Committee on Cataloguing (ACOC)• The British Library• Canadian Committee on Cataloguing (CCC)• CILIP: Chartered Institute of Library and Information Professionals• The Library of Congress• International Federation of Library Associations and Institutions

(IFLA) • Dublin Core Metadata Initiative (DCMI) • RDA/MARC Working Group

Stakeholders

• Catalogers – and –• Library administrators• Cataloging educators• Public service librarians• Systems developers• Metadata communities• MARC format developers• National and international programs (PCC, ISSN, etc.)

• You• ………………………………………..to name a few…..

Well, only if the rules actually achieve these lofty, if laudable, goals

2.1.1.1

If the resource does not contain any of the sources listed above, use as the preferred source of information another source within the resource itself, giving preference to formally presented sources

Well, only if the rules actually achieve these lofty, if laudable, goals

Construct the preferred access point representing a libretto or song text, by adding Libretto to the preferred access point representing the work or part(s) of the work if the work or part(s) contain only the text of an opera, operetta, oratorio, or the like, or Text to the preferred access point representing the text of a song. For compilations by a single composer, add Librettos if the compilation contains only texts of operas, operettas, oratorios, or the like; otherwise add Texts.

RDA Structure

• General introduction• Elements• Relationships

Appendices• Capitalization, Abbreviations, Initial articles, etc.• Presentation (ISBD, MARC, etc.)• Relationship designators• Etc.

GlossaryIndex

Structure of RDA

• RDA contains:– 10 sections– with 37 chapters – and 13 appendices

• Table of Contents is 113 pages

0 Introduction (purpose and scope, foundations, objectives, principles, structure, core elements, etc.)

Section Attributes

1Chapters 1-4

Manifestation and item (e.g., title, statement of responsibility, edition statement, publication information, etc.)

2Chapters 5-7

Work and expression (e.g., title of the work, content type, etc.)

3Chapters 8-11

Person, family and corporate body (e.g., name, identifier, associated dates, etc.)

4Chapters 12-16

Concept, object, event, and place

Section Relationships

5Chapter 17

Primary relationships between work, expression, manifestation, and item (hierarchical)

6Chapters 18-22

Relationships to persons, families, and corporate bodies associated with a resource

7Chapter 23

Subject relationships

8Chapters 24-28

Relationships between works, expressions, manifestations, and items (successive, derivative, etc.)

9Chapters 29-32

Relationships between persons, families, and corporate bodies

10Chapters 33-37

Relationships between concepts, objects, events, and places (such as broader or narrower terms)

New Terminology

Current term RDA Term

Heading Access point

Authorized heading Preferred access point

See references Variant access point

Authority control Access point control

Uniform title Preferred title

Chief source Preferred source of information

No real equivalent—some possible parallels in MARC relator codes and relator terms

Relationship designator (indicates the nature of a relationship/role, e.g. author, arranger of music; abridgement of [work], contained in [work]; alternate identity, family member, employee)

RDA Appendices

• Capitalization• Abbreviations• Initial articles• Record syntaxes for descriptive data• Record syntaxes for access point control data• Additional instructions on names of persons• Titles of nobility, terms of rank, etc.• Dates in the Christian calendar• Relationship designators (4 appendices)• Complete examples

Transcription – Principle of Representation in RDA

• “Take what you see” – Correction of inaccuracies elsewhere– No more abbreviating (but take abbreviations

found on the resource)

• Accept what you get– Facilitating automated data capture

Sample Changes from AACR2

• Transcribed data – Option to keep rule of 3

• e.g., [and five others] – no more “… et. al.”

– First place of publication is “core”– Place of publication not identified – not “s.l.”– Publisher not identified – not “s.n.”– Date of publication not identified

Sample Changes from AACR2

• General Material Designator ONIX/RDA (icons?)– Content type

• e.g., notated music, performed music, sounds, spoken word, text, still image, two-dimensional moving image (MARC 336)

– Media type• e.g., audio, computer, microform, projected, unmediated,

video (MARC 337)

– Carrier type• e.g., audio disc, online resource, microfiche, volume, object,

videodisc (MARC 338)

Sample Changes from AACR2

• Access points – Bible– Treaties– No more “Polyglot”– Birth/death dates (no more b. or d.)– More data in authority records

Reaction to RDA drafts

• Rhetoric is at times heated• Mostly taking place on email lists and the

blogosphere, rather than in the published literature

• Falls into two camps:– Too extreme– Not extreme enough

• Both sides have some valid points; both miss the point entirely at times

Jenn Riley. “RDA and FRBR: An Update.”http://www.dlib.indiana.edu/~jenlrile/presentations/ilf2007/rdafrbr.pdf

Reaction to RDA drafts

• The JSC claims RDA will make shifts in the theoretical framework without invalidating previous cataloging work

• So, we must both change the standard and not change the standard

• This is why JSC’s work has been criticized for being both too dramatic a change, and not a sufficient change

The “too extreme” argument goes something like:

• Abandonment of ISBD as a guiding structure is a step backwards

• FRBR is just theory, we shouldn’t be basing a cataloging code on it

• Language is incomprehensible• Planned changes don’t give enough

benefit to warrant the costs of implementation

Adapted from Jenn Riley. “RDA and FRBR: An Update.”

“Too Extreme”

• No other communities are going to use this thing anyways

• Any simplification of rules might reduce record quality and granularity

• Trying to cater to multiple audiences pollutes a library cataloging standard.

• Retraining staff will be expensive for libraries and confusing to catalogers – the bigger the change, the more the cost and confusion.

“Too Extreme”

• See Gorman paper for an example

“The RDA seeks to find a third way between standard cataloguing (abandoning a slew of international agreements and understandings) on the one hand and the metadata crowd and boogie-woogie Google boys on the other.”

The “not extreme enough” argument goes something like:

• Too much data relegated to textual description• Length and specificity make it unlikely to be applied

outside of libraries• Plans to remain backwards-compatible prohibit

needed fundamental changes• FRBR integration only a surface attempt• RDA is a “legacy standard” mired in past thinking. It

will never catch on outside of libraries if it remains so complicated (example: 2 chapters = 120 pages of info.).

Adapted from Jenn Riley. “RDA and FRBR: An Update.”

“Not Extreme Enough”

• RDA is too bottom heavy. JSC should create broad rules for most scenarios and let specialized groups produce details.

• JSC cannot create a robust standard for both digital and analog records. It must choose digital or risk losing forward thinking supporters.

• A less structured approach would allow for more sophisticated computer mediation, which would create superior search results and better serve patron demands.

“Not Extreme Enough”

• See Coyle/Hillmann paper for an example

“Particularly problematic is the insistence that notions of "primary" and "secondary," designed to use effectively the space on a 3 x 5 inch card, must still be a part of RDA. Preferences about identification of materials continue to focus on transcription in concert with rules for creating textual "uniform" titles by which related resources can be gathered together for display to users. Similarly, relationships between works or derivations have been expressed using textual citation-like forms in notes. “

Working Group on the Future of Bibliographic Control

• Develop a More Flexible, Extensible Metadata Carrier

• Integrate Library Standards into Web Environment • Extend Use of Standard Identifiers • Develop a Coherent Framework for the Greater

Bibliographic Apparatus • Improve the Standards Development Process,

including return on investment and greater focus on lessons from user studies

• Suspend Work on RDA

WG Recommendation 4.2• Presented their preliminary recommendations Nov. 13, 2007 at the

Library of Congress, recommendation 4.2 directed at RDA. The working group expressed their concerns about the new guidelines:– RDA is being written on a framework that is not yet tested--FRBR

concepts need to be tested on real cataloging data – "Temporarily suspend all further new work on RDA" – need thorough exploration of FRBR and implications on

bibliographic control – WG needs assurance that RDA is based on practical reality as

well as on theoretical construct, that this would improve the support for the new code

– need more info on cost of implementation – need identification of the real benefits of implementation – need info on hospitality of systems to be able to handle the new

rules – urge the JSC to go back and address these outstanding issues, as

well as language issues, organization, and usability

• “We want to make clear that NAL and NLM have not yet reached a conclusion regarding the adoption of RDA. We are mindful that the sponsoring organizations have economic limitations and revenue projections tied to the publication of RDA. However, the decision to adopt a new code must be based on the content of that code and not the economic needs of the sponsoring organizations.”– Statement posted to Autocat and other listservs. July

11, 2007

Draft Review Process: Positive Features of RDA

• Re-organization of the instructions around a clearly-defined element set

• Effort to support both current and forward-looking implementation scenarios

• Application of the FRBR/FRAD data models, including the attributes, relationships, and user tasks

• Emphasis on relationships among resources and entities

• Greater emphasis on describing entities, as opposed to creating access points

Draft Review Process: Positive Features of RDA

• Consistent specification of resource identifiers as an alternative to text strings for identifying entities

• Effort to support international application of RDA outside of an English-language environment

• Decision to define a place for subject entities and relationships in the RDA structure

• Collaborations with the ONIX and DCMI communities have already yielded what may turn out to be some of the most significant products of the RDA project

Draft Review Process: Not So Positive Features of RDA

• Constituency review of the RDA draft was deeply flawed and a difficult and unpleasant experience. – Calls into question whatever credibility the

RDA project has left – The PDF files in which the full draft was finally

issued were flawed documents, characterized by abundant typographical errors, faulty references, and a layout that obscured rather than supported the content

Draft Review Process: Not So Positive Features of RDA

• Frustrating combination of a forward-looking structure with the retention of vast amounts of case law and arbitrary decisions from the past.– Instructions retain many of the arbitrary

decisions inherited from AACR2, and the current reorganization now highlights how arbitrary many of those inherited decisions are.

Draft Review Process: Not So Positive Features of RDA

• Catalogers of special types of resources, such as cartographic, archival and moving-image resources, have become convinced that they have nothing to gain from RDA and much to lose

• RDA fails to meet many of its objectives, but none more fatally than the objective of clarity: RDA is not “clear and written in plain English.”

Will RDA Ever be Implemented?

• Heidi Hoerman's presentation on RDA from the 2008 OLAC/MOUG/NOTSL Conference. She reviews RDA and predicts:

• "RDA will die a quiet death.”

• “AACR2r2010 will be published.”

• “RDA's aims will be realized in due time."

Will RDA Ever be Implemented?

• Even if RDA proves to be as bad as detractors suggest, it may still have some important things to say about cataloging

• Perhaps is RDA proves to be insufficient, its shortcomings will be addressed and the next standard will be the dramatic change

• Or, maybe RDA will be just as dramatically wonderful as it has been suggested it will be

Cooperative Cataloging Rules

• The site has two primary purposes– 1) to offer a serious alternative to RDA – 2) to offer a place for sharing bibliographic

concepts within the general metadata community.

– James Weinheimer post to Autocat, Oct. 15, 2009

MARC

“The electronic embalming of the catalog card.”

--Michael Gorman

“MARC has always been an arcane standard. No other profession uses MARC or anything like it.”

--Roy Tennant

MARC

• “There are only two kinds of people who believe themselves able to read a MARC record without referring to a stack of manuals: a handful of our top catalogers and those on serious drugs.”

• Roy Tennant. MARC Must Die

Rocky [realia] : beloved pet / raised and loved by Rick Block and Bill Vosburg. -- Shih tzu. -- Missouri : Farm, 1999-

1 dog : male, black and white, 18 lbs. ; 51 x 33 cm. (Block/Vosburg dog series ; no. 1)

ISDN: 101-01-0101 : priceless.

MARC: WoGroFuBiCo

• 3.1.1.1 LC: Recognizing that Z39.2/MARC are no longer fit for the purpose, work with the library and other interested communities to specify and implement a carrier for bibliographic information that is capable of representing the full range of data of interest to libraries, and of facilitating the exchange of such data both within the library community and with related communities.”

What about MARC? How will RDA change this standard?

• RDA/MARC Working Group is to propose changes to MARC21 to accommodate encoding of RDA data

• MARC is only one possible encoding schema for RDA data

• RDA online product will include mappings to MARC (current PDF draft has mappings to MARC21 in Appendix D)

• “JSC has gradually backed away from their original stance that RDA could be expressed easily in MARC21”—Diane Hillmann

• Well supported rumors indicate that LC is considering discontinuing update of MARC21 sometime in 2010

What about MARC? How will RDA change this standard?

• We don’t have complete answers about how MARC will change with the adoption of RDA.

• The RDA/MARC Working Group has formed to address these questions:– Under the auspices of the British Library, the Library and Archives

Canada, and the Library of Congress, an RDA/MARC Working Group has been established to collaborate on the development of proposals for changes to the MARC 21 formats to accommodate the encoding of RDA data. With the implementation of RDA anticipated for late 2009, the Working Group will be drafting proposals for review and discussion by the MARC community in June 2008.

– Although the MARC 21 formats support the encoding of descriptions created according to a wide range of content standards, the close relationship between AACR and MARC 21 has contributed to the efficient exchange of information among libraries for decades. The RDA/MARC Working Group will identify what changes are required to MARC 21to support compatibility with RDA and ensure effective data exchange into the future.

• (Taken from an email posted by Marjorie Blossto RDA-L on April 13, 2008.

Future of MARC

• Discussion of the future of MARC is only partially about MARC– The broader digital information landscape– Technologies– Cataloging practices– The diminishing market share of:

• Libraries in the information marketplace• Library catalogs as a resource discovery tool

MARC’s Richness

• Metadata record with approximately 2,000 elements available– Approximately 200 fields– Approximately 1800 subfields or other

structures

• To what extent is the richness/complexity exploited

MARC: My Thoughts

• Rumors of MARC’s death have been greatly exaggerated.

• Nevertheless, the “cult of MARC” could keep us from seeing or moving ahead

• It’s not MARC that’s killing us, it’s the record

• The pursuit of the perfect record must end

MARC: My Thoughts

• Librarians have had greatest success with data sharing

• Don’t sweat over MARC• Can re-package MARC data• ILS systems need to gather and display

records: not a lot needs to be done to MARC records

• Not convinced MARC will die either by murder or natural causes … but

MARC: My Thoughts

• MARC does limit our ability to share and exchange data outside of libraries while the creation of metadata outside of libraries is undergoing exponential growth

RDA Online Product: Planned Features

• Browse and Search text (chapters and appendices)• RDA-AACR2 Mappings• Mappings to Dublin Core, ISBD, MARC• Full or Core View options• Workflows and examples for different formats and

types of resources• Links to external resources• Customizable views and settings• Demo from the IFLA Satellite Meeting, August 2008:

http://www.collectionscanada.gc.ca/jsc/docs/iflasatellite-20080808-demo.pdf

Testing

• Six months

• Coordinated by U.S. national libraries: LC, NAL, NLM

• Also includes PCC libraries of varying sizes, some archives, ILS vendors, OCLC

• RDA itself and compared to AACR2

Testing

• Feasibility of creating bibliographic data and populating MARC record

• Workflow and time comparison to AACR2• Determination of possible changes to MARC to

accommodate data created using RDA• Financial impact of training, workflow, and workflow

adjustments• Usability: for catalogers, by systems, ability of users

to locate desired information• Co-existence of RDA and AACR2 records• Integration between online product and other tools• System development needed for implementation

Testing

• Initial release of RDA Online will be tested• All methodology, results and data will be shared

and available• Core set of 25 resources including text, AV,

serials and integrating resources• Each institution will create both an RDA record

and a record using their current rules– Different staff members will create the RDA record

and the current rules record• Each institution will produce at least an

additional 25 RDA records

• “The goal of the test is to assure the operational, technical and economic feasibility of RDA … At the very least, the testing may simply reveal that the rules don’t work and thus show us how not to develop cataloging guidelines, which is always a valuable lesson.”

• Shawne Miksa. Resource Description and Access (RDA) and New Research Potentials.

Current Timeline Version ??• Full draft released in PDF November 17, 2008• Comment period on full draft ended February 2, 2009• RDA Online release January 2010?• Testing will begin only after RDA is available• Test Days 1-90

– Training period• Test Days 91-180

– Records creation period• Post-Test Days 1-90

– Steering Committee analyzes results• After Post-Test Day 91

– Report is shared with US library community• Implementation?

Controversies, questions, considerations …

• Cost and accessibility of online product– It is unlikely that RDA in its entirety will be available

through open access. • Too radical or not radical enough?• Drafts have been difficult to understand and

inconsistent• Has FRBR been tested enough?• FRBR model doesn’t apply equally well to all

types of materials• WoGroFuBiCo’s recommendation to suspend

work on RDA

Controversies, questions, considerations …

• Internationalization vs. Anglo-American membership on JSC

• Flexibility and adaptability vs. specificity and detail

• Break with the past vs. compatibility with legacy data

• Simplicity and ease of use vs. length and FRBR jargon

• Must MARC die?• What is OCLC going to do?• … and others

Final Thoughts

• The road to RDA has been extremely frustrating

• I’ve become even more convinced that despite its flaws we need to have it out and used (or not!)

• Releasing an imperfect code is better than another 15 years of discussion

• Release early, release often!

Too much change?!

“In cataloging, all changes cost money. The larger the catalog in which the changes are introduced, the more they cost. That is why there is always a powerful conservative lobby among administrators of the largest and richest libraries when the revision of cataloging rules is under consideration.”

Lewis, P.R. (1980). “The Politics of Catalog Code Revision and Future Considerations.”In The Making of a Code: the Issues Underlying AACR2. held March 11-14, 1979, Tallahassee, Florida. Edited by Doris HargrettClack. Sponsored by the School of Library Science, Florida State University. Chicago: ALA

Consider this past observation…

“…failure to keep cataloging practice in line with changes in the characteristics in the documents in our libraries, and with the expectation and needs of document users in those libraries, leads to increasing inefficiencies; and so long-term costs of avoiding catalog changes may be as high as those of accepting them, although this is not easy to demonstrate in library budgets. Either way, the longer the changes are deferred, the more they cost...the proper method is to carry out revisions promptly.”

Lewis. P.R. (1980)

What Should I Be Doing Right Now?

• Get familiar with FRBR and RDA terminology• Explore the RDA website and other resources—

official and unofficial• Watch discussion lists and blogs for discussions

and updates• Ask questions, talk with colleagues, participate

in the online discussions• Keep an open mind• Be prepared for change, even if RDA dies• And, most importantly…