Module 7.1.2: Bibliographic Access

LIS 5043: Organization of Information

Dr. Manika Lamba

Introduction

In Module 7.1.2, we will discuss the concepts and process of Bibliographic Control, MARC records and how they are structured, and also Authority Control.

Bibliographic Control and Authority Control are both central to how we organize information in libraries. This module is going to focus more specifically on libraries than the last two lectures but many of the standards and protocols we will be discussing are also included in other systems for organizing information and knowledge resources.

Module 7.1 and 7.2 Overview

• Introduction to bibliographic control
• Brief introduction to AACR2 cataloging
• Overview of differences between AACR2 and RDA
  • structures of both standards
  • additions to AACR2 in RDA
  • new to RDA
• MARC and RDA
• Access points/authority control

So, we’re going to cover, as I said, a brief introduction to AACR2 and AACR2 cataloging. We’re going to review the differences between AACR2 and RDA. We’ll do this in several ways. We’re going to look first at the structure of both standards. And then we’ll look at some additions to AACR2 that are present in the RDA standard and some that are entirely new to RDA and then also how some of the guidelines for AACR2 have changed, some of the local practices. Then we’ll talk just briefly about MARC and RDA and access points and authority control.

Bibliographic Universe

Just to refresh our memories, when we talk about creating representations or organizing information and knowledge resources we can only represent those that are recorded in some manner. This graphic illustrates some broad categories for information objects that we organize.

In Module 7, we are talking about this abstract idea of the universe of knowledge, which is an abstract way of thinking about all of the knowledge, information, etc. that is generated by the human presence throughout the years. Within that universe of knowledge we can describe and we can create records for items that have been recorded in some manner, and we call those recorded items the bibliographic universe.

And this slide just illustrates that even though we have this larger idea of the universe of knowledge, the item has to be recorded first before you can actually create a representation or surrogate for it. Also, this slide shows that there are many kinds of physical and virtual manifestations of items within the bibliographic universe, which have in some manner have recorded the creators’ representation of their ideas.

Bibliographic Control

We attempt to exert this control over the bibliographic universe

Bibliographic control is the process that the cataloger or other information organizers conduct to create representations of the objects in one’s collection. This slide presents some other views on what Bibliographic control is. How are they different? How are they the same?

Smiraglia discusses or defines bibliographic control as “encompassing the creation, storage, manipulation, and retrieval of bibliographic data.” So, he not only sees it from the perspective of a cataloger or a knowledge creator, but he also sees it from an information retrieval perspective, meaning that we create the representations of the different items so that they can then be retrieved. And so, we manipulate them in different ways.

Svenonius sees bibliographic control as a “skill or art … of organizing knowledge (or information) for retrieval.” So, again, we’re seeing not just the idea of creating a representation, it stands in place of something something–the bibliographic items, those recorded pieces of information. But also we have to think about the ultimate outcome, the retrieval of those items. So, we have to create representations that are useful to our users.

Patrick Wilson, who was a philosopher in our field and who I highly recommend his writings, sees bibliographic control as a form of power. And if knowledge itself is a form of power, as the familiar slogan claims, bibliographic control is, in a certain sense, power over power. And it’s the power to obtain the knowledge recorded in the written form.

Wilson also saw catalogers or those that organize information as possessing an unique power as well in that they have to make some decisions about the systems that we design in order to represent those different items in the bibliographic universe. We had to think about our users and how they might retrieve the information in order to design systems that were actually useful to our users.

Wilson’s definition illustrates the power that representations/bib control embed into bibliographic records. But he also states that we organize things in an arbitrary manner we decided long ago (we have really only codified bibliographic control over the last century) what aspects to represent in bibliographic records. Who is to say that those choices were the best choices to make? Later we will talk a bit about FRBR studies of users and what they uncovered.

Two Primary Challenges

• How to individually represent information objects as concrete entities so they can be found when needed?
  • container-oriented
  • uniqueness-oriented
• How to represent and relate information objects as sources of information on various subjects so they can be found by those in search of the information?
  • content-oriented
  • relationship-oriented

Distilling this all down, there are really two primary challenges when we represent objects for bibliographic access. They are noted on the slide and we discussed them in Module 2 also.

One is how to individually represent those information objects as the concrete entities that they are so they can be found when needed. And when we’re talking about it from this perspective, we’re talking about those container oriented aspects or attributes, such as the name of the work, the author or the creator, things that are describing the size or the color or the shape things like that. They’re not related to content. We’re also at this point concerned with creating records that uniquely represent each item in our collection. So, we’re uniqueness oriented as well.

The other challenge, then, is how to represent and relate information objects as sources of information on various subjects. So, in this case, we’re concerned with the content contained in the objects themselves. And so, we’re representing subject and geographic location if it’s used as a subject or chronological representation if we’re thinking about it in terms of a particular time period. Purpose, function those are other ways we can think about content oriented representation. And then also it comes into the play the different relationships that are related to this particular item, not from the container oriented aspect, but how those content oriented aspects are related to other content oriented aspects in other works within the collection.

So, again, we’re looking at this as a dichotomy. We’re looking at the idea of how to represent that item from a container perspective and how to represent the content oriented aspects of the item as well.

Relations in the Bibliographic Universe

We also have to represent relationships that exist within the bibliographic universe. This slide presents a few.

There are relations related to the same author. Several books are created by the same author, or several works are created by the same author. Or same content in different containers. We might have a book but it’s manifested also in microfilm also in digital. Or we might have the same content itself, the same essay in different objects, such as journals and anthologies. Or we may have original and derivative relationship, where we have an original painting then we have a black and white painting and then we have a digitized photo of that painting itself. They’re all the same item within the bibliographic universe but being represented in different ways. So, again, we have to build these relationships into our bibliographic records, into those representations so that our users can find exactly the item or items they’re looking for.

Can you think of others present in library catalogs? In other information systems you have used? Are there relationships that YOU think should be present that are NOT present in bibliographic records?

Idenitfying Relationships

• Separate containers, same content
  • different formats
• Separate containers, similar content
  • translation, abridged, versions, editions
• Separate containers, same author, different content
  • link together all objects by author
• Separate containers, different authors, similar content
  • link together all objects by subject
• Unique representations of container
• Relationships established through various means
  • name authority control
  • controlled vocabulary

This slide just talks a little bit more about identifying those different relationships.

If we’re looking at separate containers but they have the same content, then we’re dealing with the idea of formats. And how do we represent different types of formats within the same collection so that a user can evaluate our representation and determine if it’s what they want.

We also, as I said, can have separate containers but similar content, and this would be applicable when we’re talking about translations of a work or abridged versions or different editions. That again, we’re talking about similar content, but they’re slightly different in each iteration.

Or separate containers with the same author but different content. We can link all of the objects in our catalog that are written by a particular author or created by a particular creator within our catalog even though they’re in different containers and have different content. So, if we have several different books written by the same author, we would have separate containers, same author, but different content, and we want to be able to link those representations by that author attribute so that when a user does an author search, they can find those representations. One way in which we do this is through the use of authority control.

We might have a relationship within separate containers but different authors but similar content, and we link together then all the items that are related to the same or similar subject. Again, we use mechanism, such as controlled vocabularies, which are considered subject authority control in order to create this relationship within our records and our catalogs.

And then we also have unique representations of different containers. Again, we want to represent each item within our collection as its own unique entity and to have its own unique record even though we also establish those different relationships so that users can find all of the same objects by the same author or related to a particular subject area when they do searches in our catalogs.

And, as I said, some of these relationships are established through what we call ‘authority control.’

Bibliographic Records

• Aggregate of data that are associated with entities described in library catalogs and national bibliographies

• A record which accurately describes an item (book, map, computer file, etc.) both physically and intellectually in such a manner as to distinguish it from all other items in the bibliographic file

• What should be included in bibliographic records?

• Why is certain information contained in bibliographic records?

  • remember the objectives of the catalog
  • bibliographic records must support the objectives

Now let’s talk about bibliographic records. You know them as library catalog records, but really this concept can be extended to any record that is representing an information or knowledge resource. They are basically as the slide notes, “an aggregate of data that are associated with entities”.

A bibliographic record can be defined as an “aggregate of data that are associated with entities describes in library catalogs and national bibliographies.” In other words, a bibliographic record is a catalog record within your OPAC system. What you’re doing is you’re aggregating that container oriented and content oriented data about the objects in your collection, and that record itself is what stands in place of the item.

The record should also accurately describe an item, both physically an intellectually in such a manner as to distinguish it from all other items in the bibliographic files. So, again, we’re looking at the uniqueness oriented aspects, where each item in our collection is represented by one bibliographic record within our file.

A lot of your readings as well as a lot of the more current discussions and some of the more current issues are rethinking what should be included in bibliographic records. And throughout class you’re going to see we’ve had kind of the same model throughout the years. We’ve expanded, obviously, from a paper catalog or paper based records when we moved into card catalogs. We then set some very specific standards for what should be included in bibliographic records and even within our different bibliographic utilities like OCLC and even within Library of Congress standards and cooperative arrangements, we have sets of elements that we consider ‘core’ in bibliographic records, or that need to be included in a bibliographic record for it to be an actual representation.

But what we’re doing now is starting to think forward and to think whether or not what we’ve been doing all these years is actually working for our users and if we need to change the items that are included in our bibliographic records.

When you think about why certain information is contained in bibliographic records, we need to think back about those objects of the catalog to find and identify, to collocate or bring like items together, to be able to evaluate items. The bibliographic record at the core really must support those basic objectives, and now that we’re in an electronic environment (and actually beginning in the card catalog environment), we began to give the additional function of location to those records. But if our records contain no other information, we have to have enough information to allow our users to allow the records to fulfill those objectives of the catalog. And again, now we’re starting with the FRBR studies in 1994 and forward. We’re starting to rethink about what should be included in bibliographic records.

In Module 2 you looked at library and non library contexts for organizing objects. Assignment 2 required you to go out and examine a specific commercial context and think about how it was organized. Then you had to describe it/discuss it within your group.

• What did you learn from that exercise that could be extended to this concept?
• Did you see elements of organization that could be included in bibliographic records that are not?

Think also about the Weinberger readings: WHY do we include certain elements/attributes within records?

In libraries, mainly because it was established in the early 1900’s that records should support Cutter’s objectives of the library catalog. We will discuss these in a few minutes. Other cataloging visionaries made choices about what to include and we have not really altered these choices much, until the past 5-7 years. Your readings will talk more about this issue.

Creating Bibliographic Records

Descriptive Cataloging

• oriented towards intrinsically derived attributes of information objects (e.g., titles from books, copyright date, etc.)
• primarily concerned with describing/representing the information object as an physical entity. Also referred as “bibliographic cataloging”

Subject Analysis/Cataloging

• combination of intrinsically derived attributes (e.g., what is this information object about) and extrinsically defined attributes (e.g., controlled vocabulary that represents the concepts treated by an information object)
• uses words and phrases to represent the intellectual content of the object

Classification

• based on subject analysis, but can use other attributes besides topic to group together items that are related to one another
• bibliographic classification in the library community provides a unique physical location/address to the object

We’re going to focus, as I said, primarily at this point of learning more about how we create these bibliographic records. We’ll talk about these different processes that you see on the slide, but then we’ll talk about different tools and standards that are in place to help us create this piece of the bibliographic record. When we create bibliographic records in libraries, there are three distinct processes:

1. Descriptive cataloging geared more specifically in representing the container of objects
2. Subject cataloging to represent subject, topic, etc. of the object
3. Classification related to subject in most systems. In libraries it provides for physical location of the object also. On the web, it is a way of collocating a product based on specific attributes or what are called “facets”.

When you create a bibliographic record, there’s actually three processes that you’ll go through, each with its own standards and tools. One is descriptive cataloging, which is the piece we’re going to focus on in this topic. But then we’ll talk about subject analysis and cataloging and classification in later lectures.

Descriptive cataloging is that which is oriented more towards the container aspects, or those that are considered intrinsically derived attributes of information objects titles from books, copyright date, publisher information, author or creator information. And descriptive cataloging is primarily concerned with describing and representing the information object as a physical entity. And we’re talking about, again, those container aspects. You’ll also hear descriptive cataloging referred to as ‘bibliographic cataloging.’ You’ll see that in the reading. You’ll see that in different presentations.

The second piece, of course, of creating a bibliographic record is subject analysis and subject cataloging. This is a combination of intrinsic derived attributes, or what is the information object about? And we actually will refer to this as the word ‘ aboutness ’ in the literature as well as you’ll hear that a lot in common parlance. And also in extrinsically defined attributes, those are the attributes, that we have created words if you will, that represent the different concepts that are present within the information object. And one such tool would be a controlled vocabulary, or a subject heading list is an example, such as LCSH. And we would use these different tools to then represent the concepts or the subjects that are present within the information object itself.

The third piece of this, which is related also to subject analysis, is what’s called ‘classification.’ Classification is based on the subject analysis that the cataloger conducts with each item as they create a bibliographic record, but it can also use other attributes besides topic to group items together that are related to one another. Bibliographic classification in the library community provides that unique physical location, or what you might consider an address, for the object. It’s where we create the classification number or the call number on the side of a book or within our catalog so that the user can actually locate the item physically within our collection.

You put these three processes together using the different tools that are unique to each and the standards that are relevant to each to create one bibliographic record.

Recall What Lubetzky Emphasized…

The functions of descriptive cataloging are

• To describe the significant features of the book which will serve (a) to distinguish it from other books and other editions of this book, and (b) characterize its contents, scope, and bibliographical relations

• To present the data in an entry which will (a) fit well with the entries of other books and other editions of this book in the catalog, and (b) respond best to the interests of the majority of users

Also, to consider as we learn about bibliographic control and creating of bibliographic records is what Seymour Lubetzkey emphasized as functions of descriptive cataloging, which are to describe the significant features of the book, and, of course, he’s thinking of book because that’s primarily what they were cataloging at the time, but that can include any kind of information object. But to describe the significant features of the book, which will serve to (a) distinguish it from other books and other editions of this book so again, to make it unique from other items in your collection. And (b) to characterize its contents, scope and bibliographic relations so, to start to create those relations that we mentioned a few slides ago but also to look at the scope or the contents of the work itself.

He also believed one of the functions of descriptive cataloging was to present the data in an entry that will (1) fit well with the other entries of other books and other editions of this book in the catalog, which basically means that you would need to create representations that are similar to or with following the same standards or local practices of the items that are already in your catalog, and you’ll see more of this when we start looking at MARC records where we have structures in place to tell us what kinds of information to include in each of our records. And Lubetzky was very concerned with this standardization. And then (2) to respond best to the interests of the majority of users.

So, Lubetzky was more of a user-centered person where he wanted representations to be useful for the majority of our users so that those standards come into play. By having standard entries within our catalogs, our catalogs work more effectively.

Functions of the Library Catalog

• Represents holdings of particular collection
• Early objectives of the catalog defined by Charles Cutter
  • identification/finding
  • collocation/gathering
  • evaluation/selecting
• Functions of bibliographic instruments (FRBR)
• Catalog adds one additional function:
  • location or acquisition

Let’s go back briefly and review the functions of the library catalog. One is to represent holdings of a particular collection. Library catalogs at a local level are representative of the collection that the library houses itself. And again, now that we’re thinking more in a digital environment and we’re making our OPACs web accessible, we’re still within our own catalog representing items that are the holdings of our particular collection and primarily for the use of our local users. So, a lot of cataloging agencies, as a result of this idea, have their own local practices, or they have their own, for example, set of subject headings they might use to represent different subjects so that they can provide that second objective defined by Cutter, the collocation or gathering function.

So, again, you have to think about that as we have standards within the library community that are used across different collections so that we can then share more effectively the records that we do create representing our own local collections. But then we might also in house have what we might call local practice, where we have our own locally applied standards different from other collections and other records. And we’ll talk more about that when get more into the specifics of the different processes. But again, I want you to remember that we’re representing for particular collections and a particular set of users.

Cutter developed the objectives of the library catalog in the early 1900’s and they still hold today. Representations (bib records) should enable the three functions of 1) identification/finding, 2) collocation/gathering together based on similarity usually subject, but can be author, title, series, etc., 3) evaluation/selecting enough information so user can determine if it is the object they want operating system if CD, version if editions, etc.

FRBR and RDA are two of the hottest topics in the LIS cataloging community today. You will read quite a bit about the FRBR user studies, their resulting models, and how these models are being incorporated into the new cataloging standard, RDA (Resource Description and Access) as you engage with the readings and web explorations.

The FRBR studies added one additional function/objective to the catalog, location or acquisition. This should come as no surprise. Once you find the object you want in the system, you also need to know WHERE/HOW to access it.

The RDA represents a very significant departure from the cataloging standard we have used since the late 1960’s, the AACR2 (Anglo American Cataloging Rules).

International Standard Bibliographic Description (ISBD)

Areas of Bibliographic Description include

1. Title and statement of responsibility area
2. Edition area
3. Materia(or type of publication) specific details area
4. Publication, distribution area
5. Physical description area
6. Series area
7. Note area
8. Standard number and terms of availability area

Another standard that we use in cataloging is the ISBD. This standard spells out specifically which elements should be part of the descriptive aspects of bib records. The eight areas of ISBD description are listed on this slide.

What is not included that you think should be? Or do the elements encompass what you think is necessary for descriptive access (non subject)? or be expanded upon in the new digital environment.

Levels of Description

• First Level

      Gypsy ballads/Federico Garcia Lorca -- Warminster: Aris & Phillips. c1990. -- vii, 161p. -- ISBN 0-85668-490-2.

• Second Level

      Gypsy ballads = Romancero gitano translated and with an introduction and commentary/Federico Garcia Lorca; translated by R. Harvard; illustrated by M. Gollanz. -- Warminster: Aris & Phillips. c1990. -- vii, 161p: ill; 22cm. ----. ISBN 0 85668 490 2.

• Third Level

        Gypsy ballads = Romancero gitano translated and with an introduction and commentary/Federico Garcia Lorca; translated by R. Harvard; illustrated by M. Gollanz. -- Warminster: Aris & Phillips. c1990. -- vii, 161p: ill; 22cm. ---- (Hispanic classics). -- ISBN 0 85668 490 2.

ISBD also codifies three levels that could be present in a bibliographic system. Each are shown on this slide. The first and maybe second level are used in smaller libraries, while the second and third levels are used in larger libraries.

Record in Card Form

(Card info from Taylor 2000, 15)

Putting them to use results in a catalog card. We will look at MARC record examples in the next lecture.

See if you can identify the ISBD elements in the bib record. We would add in subject headings and added entries for additional authors, etc. The classification number is also present on this example.

Same Info in MARC Bibl.

The descriptive cataloging information (plus main entry) from the previous record in MARC:

Here is what a MARC (Machine Readable Cataloging) record looks like. We will learn more about MARC in the next lecture but all of the same elements we saw on the previous slide with the catalog card are still present on this example.

So, moving from a card form to a MARC environment, we can see very little differences. However, when we look at MARC, you’ll notice that there are many additions that we’ve added as records have changed, record structures have changed, and as formats of objects have changed through the years.

Anglo American Cataloging Rules (AACR2)

• AACR2 is the standard sets of rules used for descriptive cataloging until 2013 and later

  • AACR2 was replaced by Resource Description and Access or RDA which was implemented in March 2013

• Identify elements that should be included in a descriptive record

• Provide a sense of the semantics of those elements

Okay, let’s shift now away from the idea of bibliographic control and talk more about descriptive cataloging itself as a process as well as the different tools that we use within descriptive cataloging itself.

The AACR2 as stated earlier, is one of the cataloging standards for bibliographic description and also for choosing and developing access points and creating them in a process called authority control. We will address authority control in the next lecture. RDA (Resource Description and Access) has replaced AACR2 in March 2013, but it may take years for AACR2 to no longer be used in libraries.

First let’s take a brief look at the AACR2 and then we’ll address RDA and compare the two. The reason for this is if you’re working as a cataloger in a library, you’re going to encounter both these standards being applied within your bibliographic records. So, it’s important for you to understand that there was a predecessor to the RDA and that many of the catalog records will still be using both standards or a hybrid of them.

Beginning in 2015 OCLC stopped creating AACR2 records, and so from that time forward, they only used RDA. However, many libraries have not converted to RDA presently, especially smaller, rural libraries. And so, you’re going to need to know both AACR2 as a cataloger if you’re working in a smaller organization that hasn’t made the move to RDA.

Any of you that have worked in libraries or have worked in cataloging departments know that the AACR2 is our bible for instructions on how to create the container related aspects of bibliographic records. The AACR2 itself identifies elements that should be included in a descriptive record, but those elements were defined by what we call the “ISBD” or the International Standard for Bibliographic Description. Each of those elements are included within the AACR2 rules. The rules themselves also provide a sense of semantics or a definition of what those different elements mean within a record. So, for example, what does title mean? And then it would be defined within the AACR2.

You won’t be using AACR2 or RDA rules in this class. If you decide to take Cataloging and Classification, 5403, then you will learn to create records using RDA. For our purposes, you will be using existing bib records and identifying the fields/subfields within the records that hold the information asked for.

Next assignment gives you the opportunity to dissect a MARC bib record and to use a tool called the MARC Bibliographic Format to decode the record.

AACR2 and RDA

• Give the content rules for entering the data into the MARC record structure

  • Chief source of information: where to find the data on the object
  • Capitalization and spelling rules
  • Form of entry (order of name, elements to include, parenthetical qualifiers, etc.)

• Also, provide guidance in determining access points for names, titles

• Does NOT tell us how to enter subject related data for our records

The AACR2 (and now RDA) is primarily your go to place for the content rules so that you can create and enter the data into you bibliographic records and within the MARC record structure itself as most of our catalogs are using today.

Within each of the different kinds of formats within the AACR2, you’ll see what’s called a chief source of information outlined, and what the chief source of information is, is where you would look on the item itself to find the data to put within that area within your bibliographic record. Also within AACR2 are capitalization and spelling rules.

AACR2 has very detailed rules relating to forms of name within spelling, capitalization what to capitalize, what not to capitalize ––, it also has rules for form of entry the order of name or names, the elements to include in each of those different areas, when to include parenthetical qualifiers, when to include bracket information, and things like that.

AACR2 also has a section for providing guidance on how to determine access points and authorized headings for names and for titles. But what AACR2 does not do for us is tell us how to create subject entries within our records.

AACR2 Layout

Part I: Description

Introduction

1. General Rules for Description
2. Books, Pamphlets, and Printed Sheets
3. Cartographic Materials
4. Manuscripts
5. Music
6. Sound Recordings

7. Motion Pictures and Videorecordings
8. Graphic Materials
9. Electronic Resources
10. Three Dimensional Artefacts and Realia
11. Microforms
12. Continuing Resources
13. Analysis

Here is how AACR2 is organized.

First, it is divided into two parts. Part I includes a general chapter with rules that can be applied in general situations. The other chapters are specific to the format of the object you are representing. Chapters 2 12 often will refer you back to Ch. 1 for the general rule. Ch. 13 includes instructions on when to include what is called an analytic entry when to add in additional access points that refer to additional authors, series titles, etc.

Just looking at this structure you can see that the AARCR2 is really quite nicely organized. One of the problems with the AACR2, however, is that the language of the rules is counterintuitive and hard to follow. I added a sample of Chapter one to the Canvas so you can take a look at AACR2.

ISBD, AARC2, RDA, and Catalog Cards

AACR2 and the ISBD were written to accommodate cards. However, some aspects of these rules were changed when the record is moved from card form to MARC record form.

For example, from AACR2 1.0C1: “Precede each area, other than the first area, or each occurrence of a note or standard number, etc., by a full stop, space, dash, space (. –) unless the area begins a new paragraph.”

Here is an example of an AACR2 rule. I have added a few chapters of AACR2 to the Canvas site for your review. Both AACR2 and the ISBD were written originally to accommodate catalog CARDS (remember those?).

So, you’ll see that some of this information may seem a little out of date; however, the elements that were in card catalogs in cards themselves still remain in today’s electronic MARC records.

FRBR and RDA

• Functional Requirements for Bibliographic Description
  • Began research/development in mid 1990’s by an IFLA (International Federation of Library Associations) study group
  • Resulted in a set of models for elements that should be present in bibliographic records based on what we understand about user’s use of MARC catalog records
  • Also resulting in the complete revision of the AACR2 into the RDA

The FRBR models, as you learned about in Module 7.1.1, began development back in the mid 1990s. The study itself resulted in a set of models that would show elements that should be present in bibliographic records based on what we understand about how our users actually use MARC catalog records. And the FRBR studies because they’re conducted by IFLA, the International Federation of Library Associations, are more of a global perspective than say just an American or Canadian or Australian perspective. They’re really looking at more of a global user or MARC catalog records. The study itself resulted in the complete revision of the AACR2 into the Resource Description and Access, or the RDA.

What is RDA?

• RDA is based on the conceptual metadata models, the FRBR (Functional Requirements for Bibliographic Records) and FRAD (Functional Requirements for Authority Data), developed by IFLA and the international cataloging community

• It is also aligned with the principles of the International Cataloging Principles (ICP)

Now that we’ve talked just briefly about AACR2 and you’ve had a bit of a refresher about how the standard was developed and some of the goals behind this standard, let’s talk a little bit more about RDA.

As you already know, RDA is based on the conceptual metadata models from the FRBR, or Functional Requirements for Bibliographic Records, and FRAD, Functional Requirements for Authority Data, that were developed by the IFLA and International Cataloging Community. RDA is also aligned with what we call the ‘International Cataloging Principles’ or the ICP.

RDA

• Efforts have been underway since 2003 to completely revise the AACR2. The Resource Description and Access or RDA began testing in 2010

• In 2011 a decision was made by US national libraries to implement RDA

• Individual libraries will have to determine if and when they will also implement but many academic and larger public libraries are already implementing RDA

• International libraries are already beginning to implement

The effort to develop the RDA really began in 2003-2004, when the AACR2 committee got together and started determining if are we were going to have an AACR3, which was the initial intention of the committee, or develop a new standard. But then they clearly determined at that point that really they needed to go beyond the AACR2’s Band Aid approach of revisions and restructure and reorganize the AACR2 to accommodate the digital environment and a lot of resources that are now available that do not fit well in the AACR2 structure.

In 2011 the decision was made by the US national libraries, and others around the world to implement RDA. Academic libraries followed suit. Now public libraries and school libraries had to decide to adopt RDA or stay with AACR2. At this point the extent of implementation is not known but in Oklahoma, I have heard that many libraries, of all types, have adopted RDA and currently are converting AACR2 records to RDA or creating new records using RDA.

Why was RDA developed?

• The world and the world of data has changed since the AACR, AACR2, and MARC were developed
  • We now have the Internet
  • Many communities are creating metadata that could be used/useful to libraries
  • Libraries share data between themselves but our current structure (MARC records) is not designed to share data easily with other systems (databases, Web)/communities (publishers, vendors, Web and metadata e.g., Dublin Core, ONIX)

The RDA was developed because we realized that the Anglo American Cataloging Rules in its present form was not going to give us enough of a solution for cataloging environments, but also we realized and again this is something catalogers and others that create metadata have been thinking about for a long time is that the whole world of data and the world itself has really changed since the first AACR was developed and even since MARC was developed in the 1960s. We have the internet now, which allows people to share data, create data, and encode data in various different ways. We’ve had many different communities creating metadata and developing metadata schemes that are not MARC but can be used or a very useful to libraries.

So, we want to be able to take advantage of these other metadata communities that are creating metadata and to harvest data from them as well as to share data with them. Libraries have always shared data even when we were in the card catalog days, though it was a little more difficult then. Through our MARC record structure, it made it even more easy to share data internationally, not just locally. And MARC itself is designed to share data as I mentioned, but it’s within only the MARC record structure. It’s not designed to share data easily with other systems, such as databases or web environments, other communities, such as publishers and vendors, and web and metadata communities that are using other non MARC structures. Though, over the last ten years we’ve made great strides in mapping MARC to other metadata schema, such as Dublin Core or ONIX.

Goals of RDA

• A new standard for resource description and access

• Designed for the digital world

  • Optimized for use as an online product
  • Description and access of all resources
  • All types of content and media
  • Resulting records usable in the digital environment (Internet, Web OPACs, etc.)

One of the goals of RDA as a new standard for Resource Description and Access, is that it is designed for the digital world. It is optimized so that it can be used as an online product, and it is designed to be useful for the description and access of all types of resources whether they’re analogue or digital all types of content and media. And these resulting records that we create using the RDA standard will be useable in various different digital environments, from the internet to web to OPACs and within different communities as well.

Goals of RDA

• A consistent, flexible, and extensible framework
• Compatible with internationally established principles, models, and standards
• Primarily for use in libraries, but also adaptable across many information communities worldwide
• Support user tasks of FRBR/FRAD

Other goals of the RDA is that it is a consistent, flexible, and extensible framework, meaning that it can be used in various different contexts and various different formats and within different communities without needing to change. It also, we’re hoping, will not need to be updated as frequently as the AACR2.

It’s also compatible with internationally established principles, models, and standards, not just within the communities that have developed it.

It will be primarily used in libraries, but it is also written in a way that it can be adaptable across many information communities worldwide.

And we hope that RDA, once implemented in FRBRized systems, will actually support those users in the FRBR and FRAD models, as well as provide additional relationships and the collocation functions that are possible when we encode these new relationships into our systems.

On this slide I wanted to show you the differences between the RDA and AACR2 structures.

The Anglo Anglo-American Cataloging Rules is set out in two separate parts. First is for the use for creating the descriptive aspects in catalog records, and then Part II is for headings, uniform titles, and references references–creating access points, such as authorized headings for names of people, corporate bodies, geographic names, uniform titles.

The Part I description is set out by type of resource. So, in Chapter 1 we have general rules that you’re referred back to continually from the other chapters. And then in Chapter 2-12 we have rules that are specific to a type of format. So, for example, in Chapter 2 is dealing with books or monographs, so that you would have specific rules you would go to Chapter 2 if you’re cataloging a book.

RDA, however, is not set up in the same way. It’s set up in four main sections that have chapters within them. And they’re looking at the item at a work level as opposed to a database record level, which we would see in the Anglo Anglo-American Cataloging Rules. And then in Sections 5-10 we have rules for recording those different relationships we saw in the FRBR models and in the FRAD model as well. And then we have Appendices A-M and Glossary that give us special considerations, and then in the AACR2, we have appendices A-E which talk about punctuation and abbreviations.

Tables extracted from Miksa, S.D. (2009), Resource description and access (RDA) and new research potentials. Bul. Am. Soc. Info. Sci. Tech., 35: 47-51. https://doi.org/10.1002/bult.2009.1720350511

AACR2 vs. RDA: Difference in Proportions

AACR2

• Description of information entities – 13 chapters (Part 1)

• Weak on access points; talks of main and added access points, have to look all over Part II for access point provisions (e.g., title access points are discussed in chapter 21 only and then only as a default provision, not much direction)

• Is not really based on the idea of a “work”, rather it is very much based on the unit record system

AACR2 arranges chapters by the type of information resource and then by type of main or added access points (see Tables 1 and 2). In AACR2’s Part I, chapters 2-12 each focus on a separate format and address only the description of the resources.

It is weak on access points, even though Part II is devoted to choice and formation of personal, corporate body, title access points, and talks of main and added access points.

AACR2 is not really based on of the idea of a work, rather it is much more based on the unit record within the system. So, in AACR2 we’re looking at one bibliographic record for one item within your catalog, where RDA is looking at that work level and then breaking it down into expression, manifestation, and item and all the different elements that are associated with each of those.

AACR2 vs. RDA (Cont.)

RDA

• Description is covered in 4 chapters, everything else is about access points
• Form is no longer the first decision; chapters are not based on form (e.g., no longer have chapters 2-12 as in AACR2)
• Does not focus on the unit record system – it can be, but it doesn’t need to do so – rather it operates on the idea of a “work”
• Does not put the cataloger in the decision of having to decide Main and Added Access points; we don’t need those distinctions any longer although it does use the idea of a “preferred access point”

So, as I mentioned, there’s a big difference in the structure of how AACR2 and RDA are set up. And this has been one of the contentions of the library community, is that why did we have to shift our whole way of thinking about how to describe resources or even what comprises a resource? Within RDA, we’re no longer going to have to go to Chapter 2 whenever we’re looking for rules for describing books, which for catalogers was pretty straightforward. If you had a book in your hand, you knew which chapter in AACR2 to use. It would oftentimes refer you back to Chapter 1 for the general rule. But if you had a more specialized resource, such as cartographic materials, or maps, you would go to that specific chapter, and it would have very specialized instructions, where you don’t see that at play in the RDA.

In the RDA the description is covered in four chapters as I said, and the rest is all about access points and building those relationships into the records so we can serve that collocation function within our systems.

The form of the object is no longer the first decision. Chapters are not based on form, which again, is what a lot of catalogers are having real issues with. It also does not focus on the unit record system. It can, but it doesn’t need to do so. It relates mainly to the idea of “what is a work?” So, it’s really important that you really understand that concept of ‘work’ in the conceptual model.

It also does not put the cataloger in the decision of how to decide between main and added entry points, such as who is the “main” author, therefore the main access point or the main heading? And then who might be additional creators or contributors to that work? In RDA we don’t need those distinctions any longer. Although it does make use of the term “preferred access point,” meaning that you have to have at least one access point to the creator or the contributors of the work, and that would be the preferred access point. There were limits within AACR2 as to how many access points you could create for an individual item, and in RDA, those limits are no longer in play.

RDA and AACR2

• RDA now outlines the first step in creating a catalog record as deciding on the type of description to be represented, and not deciding on format, although format is still integral

• Types of description (rules 1.2)

  • Comprehensive, analytical, or multi-level description

• More emphasis on showing bibliographic relationships (e.g., taxonomy of bibliographic relationships) in order to better allow clustering of records

RDA also now outlines the first step in creating a catalog record as deciding on the type of description to be represented, whether it’s a work, an expression, a manifestation, or an item - but not deciding on its format - whether it’s a book or a CD ROM or a DVD. But format is still integral, and as you’ll see, when you start looking at those different maps from AACR2 to RDA, that we deal with format in new ways.

We have more of a comprehensive, analytical, or multi-level description rather than just one item equals one record in a database.

We also have more of an emphasis on showing those bibliographic relationships that I mentioned earlier and that are present in FRBR models in levels one, two, and three. This, again, is going to allow us to have better clustering, or collocation, of records within our library catalogs once the vendors takes advantage of these new FRBRized systems.

Comparing RDA to AACR2

We can also take a table to table comparison from AACR2 to RDA. The points that are in red are those that have been added or are specific to RDA as opposed to AACR2. You can see that quite a bit has carried over from AACR2 to RDA. You can also see that the names of some of these items have changed as well, though they’re still very similar to the actual element being described that we have in the AACR2.

So, for example, under “Description” we use the International Standard Bibliographic Description elements, those eight areas of ISBD description. Within RDA we’re no longer using those eight elements. We’re using the attributes of FRBR entities, the WEMI model.

And we are no longer using classes of materials, such as we saw in AACR2; we are using types of content and carrier.

Mode of issuance is still the same. Type of description is still the same.

Now, in terms of access points in AACR2 we had guidance for how to choose access points, and they were very prescriptive. We had name, title, and subject access points, though subject access points were not part of AACR2 unless it was using an individual or corporate name as a subject. In RDA we have those FRBR relationships built in as access points. So, for example, we would point from the work was created by an individual individual–that would be on relationship.

That then becomes an access point. In AACR2 we had rules for how to create these headings, or the form headings for the names.

AACR2 Structure

Okay, so just to go back to the idea of the RDA versus AACR2 structure. In AACR2 we have Parts I and II. Part I is where we have all the rules about how to create the descriptive aspects in our records using those eight areas of ISBD description, such as author, title, statement of responsibility, editions statement, etc.

You can see Part I is set out by chapters, which are specific to the form of the work you actually have in your hand, or the form of the item you have in your hand, such as books, cartographic materials, manuscripts, music, sound recordings, etc. Each of these different forms have their own individual chapter. Now, in the case of books, you’re often referred back to Chapter for General Rules for Description, and this again can happen for other forms as well.

In Part II that’s where we have the guidelines and rules for choosing access points, how to create those actual headings, how to deal with headings for people, geographic names, corporate bodies, uniform titles, and references.

RDA Structure

RDA, as you can see here, is structured very differently. Instead of the descriptive aspects, RDA is structure according to the FRBR and FRAD attributes. RDA includes the attributes of manifestation and item, work and expression; attributes of person, family and corporate body; attributes of concept, object, event, and place all of which are derived from the FRBR and FRAD models that you looked at in the last topic. So, it’s really important to understand those different conceptual models and the role that they will play in the new RDA framework.

Also in Sections 5 10 the FRBR and FRAD relationships set out as well as how to establish each of those different relationships within our records.

Categorization of Resources

General Material Designations

On this slide we’re looking primarily at categorization of resources, or how RDA puts different resources into categories. For example, we would have electronic resources in AACR2 on microform or sound recording or video recording. With RDA we have broken this out into different types of categories.

So, for example, in AACR2 we had what were called “General Material Designations,” or GMD, which in your catalog, if you’re sorting by say book or CD or DVD, the system itself is reading that GMD designation, which is 245 subfield H in a MARC record. And it would allow you and your users to actually filter through all of the different works on a specific title by format.

So, if they only wanted books, they would only get books in their search. If they wanted CDs, they would get CDs, etc. Now, with RDA we’re no longer going to be creating GMDs, which is again is a point of contention among librarians, especially in public and school libraries, whose users like to filter or like to sort by different format of object.

So, we’re no longer going to be creating GMDs; however, some people have decided they’re going to do it anyways. As long as their system will still read that field and sort, they’re going to include GMDs in their records, even in their RDA record structure.

In RDA we’ve broken those General Material Designations into three different categories, we have media type, carrier type, and content type. And this is where RDA gets a little bit more confusing than AACR2. But just keep in mind that there are now three, what we would have called before, categorization of resources. We no longer have the GMDs. We now have media type, carrier type, and content type.

Sources of Information

There are also some changes into what is considered a source of information. In AACR2 we referred to this as the chief source of information. And the chief source of information is specified for each class of material. Chief source of information basically means that where do you look on the item itself in order to find information to put into your bibliographic records. So, where do you look for title? Where do you look for statement of responsibility or who created this resource? And in AACR2 there’s a different chief source of information for each of those different forms of items. So, for books it would be the title page or the verso of the title page, meaning the back of the title page. For manuscripts or maps you would have a different chief source of information. For sound recordings, for example, you might look at the title screen or you might look at the liner notes if it’s a CD of music. So, there would be various different places you would look on the item, and again, we would call that ‘chief source.’

In RDA we now call it preferred source of information, and it’s specified for three categories of resources. So, things that are print based would be in one group. You would look for one or more pages, leaves, sheets, or cards, for example. So, you would look at pages, leaves, sheets, or cards to find the information to enter for the title. For moving images, we would have a different chief source of information. Generally, if it’s a moving image, it’s going to be the title screen or even the credits. And then we would have other resources. All these other formats would have specific chief sources that are identified within that other resources category.

So, AACR2, again, is set out specific to each class of material. RDA is specified for three categories of resources. And then there’s a lot of differences in the language here, and that’s something that people that have been using AACR2 for years now have to change their mindset. They now have to think about instead of ‘chief source of information,’ ‘preferred source of information.’ So, any new catalogers that are coming in under RDA, at least in the probably first five years of its implementation, are probably still going to have to learn both sets of vocabulary and how AACR2 versus RDA works. But anyone that’s coming in towards the end of this transition period will probably just learn the new language of RDA and not have to learn some of the other vocabulary that was present in the AACR2.

Transcription

• Elements transcribed from source
  • title, statement of responsibility, edition statement, etc.
• Modification of transcribed data
  • capitalization, diacritics, symbols, spacing of initials and acronyms
• Abbreviation
  • AACR allows abbreviations to be used in certain transcribed elements (e.g., edition statement, numbering, place of publication, distribution, etc., series)
  • RDA permits abbreviations in transcribed elements only if the data appears in an abbreviated form in the source
• Inaccuracies
  • AACR allows inaccuracies to be corrected within transcribed elements
  • RDA requires inaccuracies to be recorded as they appear in the source

Okay, let’s look at some other changes in terms of transcribing information from the resource into the bibliographic record. Using RDA we now are going to transcribe elements directly from the source. Before we followed AACR2 conventions. Now we have this kind of “see it as it is” viewpoint, that even if there is some type of an error, we’re supposed to transcribe it in the same way using RDA. In AACR2 we could indicate that there was an error, and we could put things in brackets, indicating some issue with the title, for example. So, elements that are transcribed exactly from the source are the title, statement of responsibility, edition statement, etc.

We also can modify transcribed data in term of capitalization, diacritics, symbols, spacing of initials, and acronyms. There are some differences between how we’re doing this in AACR2 and in RDA. In AACR2 we used sentence style capitalization. In RDA we’re supposed to keep the same capitalization as on the item itself, which again is a contention that librarians are not very happy about right now, especially as it’s translated into systems where we might have records that have one form of capitalization and other records that have a different form.

Also in AACR2 we allow abbreviations to be used in certain transcribed elements, and those are listed here. We even have specific appendices in AACR2 that tell you how to abbreviate certain words. In RDA they permit abbreviations in transcribed elements only if they appear as an abbreviated form in the source itself. So, we’re no longer using these abbreviations that catalogers have learned through the years. Things like “illustrated” will now be spelled out as opposed to “Ill.” The idea behind this is that our users do not understand our abbreviations, nor do other systems and metadata schemes that we want to share our data with. So, we’re now going to get rid of abbreviations and use them only when prescribed that way on the actual item itself, which again for me is a real issue because it deals with the consistency of how things are displayed in records, how they’re searched or how a system reads them and understands what to do with them.

In terms of inaccuracies, AACR2 allows inaccuracies, as I mentioned earlier, to be corrected within those different transcribed elements. RDA requires inaccuracies to be recorded as they appear within the source. So, if something is misspelled, you have to leave it according to RDA. You should no longer be putting it in the correct form and then in brackets and indicating that it was in some way inaccurate.

Rules of Three

• Collaborative words
  • AACR2: entry under title if more than three persons or corporate bodies responsible
  • RDA: first named person, family, or corporate body with principal responsibility (or first named if principal responsibility not indicated)
• Compilations of works by different persons or bodies
  • AACR2: entry under heading for first work if no collective title (with added entries if no more than three works in the compilation)
  • RDA: separate access points for each work (and/or devised title for compilation)
• Treaties, etc.
  • AACR2: entry under title if more than three parties
  • RDA: party named first (exception for single party on one side) ; title if first named party cannot be determined

Okay, another thing that was changed from AACR2 to RDA is what is called ‘the rule of three.’ And this is the rule that states that no more than three access points should really be created for an individual item. Now, RDA has changed this obviously because we see in the FRBR and the FRAD models that access points are very important to our users to fulfill those user tasks. So, let’s talk a little bit about how these are addressed within AACR2 and RDA.

In collaborative works, meaning that the work was created by more than one individual, in AACR2 we would have a main entry under the title. The entry under the title if more than three persons or corporate bodies was responsible. So, we would have a main entry for the main author, the main creator. Then we would have additional access points in the 700 field of our records. We could also under the statement of responsibility in the 245 subfield C in a MARC record include that addition of those persons’ names. In RDA we treat this a little bit differently. We have the first named person, the family or corporate body, with the principle responsibility. So, they’re the first named if principle responsibility is not indicated in collaborative works. But we have the ability to go beyond that rule of three to provide additional access points.

Now, if we had a compilation of works by persons or bodies say we have a conference proceeding or we have an anthology ––, under AACR2 the entry under the heading is for the first work if no collective title with added entries if there are no more than three works in the compilation. So, if we have an anthology and we have no more than three separate works in that item, then we would be able to add those entries as access points. If we had a conference proceedings, however, we might have 400 different contributors, that really wasn’t very helpful with AACR2. With RDA we can create a separate access point for each work and/or devised title for that compilation. So, if we had 400 authors, I’m not sure we would want to put all these in an RDA record, but you have the possibility of doing that now.

Under treaties under AACR2 we have an entry under title if more than three parties. In RDA we have a party named ‘first,’ again, with an exception for single party on one side. And then in title if first named party cannot be determined, we would also give additional information in the title area.

So, there are some differences. And some of these things are very minor, and some of these to catalogers are actually good changes that they’ve really wanted for a long time, and this rule of three is an example of one of those. They have wanted for years to be able to provide access to not just the overall work itself, but in the case of the conference proceeding, the individual authors or titles within that work.

RDA and MARC

• The MARBI working group (comprised of members from the Library of Congress, British Library, Library and Archives Canada, and other entities) has been working to revise the MARC structure to work with RDA

• MARBI disbanded in 2012/13 joint LITA/ALCTS group on metadata and new bib framework forming

• Documentation (www.loc.gov/marc)

  • Draft documentation of approved changes
  • Links to documents describing possible changes under discussion

RDA changes are of course related to changes that were made to the MARC record structure. The MARBI working group was tasked with this effort early in the development of RDA. However, MARBI was disbanded in 2012/13 and a new working group comprised of members from the Library Information Technology Association (LITA) division of ALA and ALCTS, division of ALA are now tasked with suggesting and making changes to MARC and emerging bibliographic frameworks, like BIBFRAME, that will be used to create bibliographic records.

You can learn more about the changes to MARC and the joint working group following the link on the slide.

Putting it all together

• Descriptive cataloging using AACR2, RDA
• Combined with subject analysis and selection of controlled terms from controlled vocabularies like: LCSH, MESH, Sears, ERIC
• Classification added, based on subject of object (LCC, DDC, faceted schemes)
• Implemented in different formats/tools
  • In past: manual card catalog
  • Presently: MARC records (OCLC), metadata schemes (Dublin Core), XML MARC

All in all, in this module, you learned more about how bibliographic records are structured, the standards that guide cataloger/record creators, and the tools we use to create the descriptive aspects of the records.

In the next lecture, we will look at the MARC record standard and discuss the concept of authority control and how it is used in bibliographic records.