1. What is RDA? Alan Danskin
I am very pleased to have been invited to Reykjavik to talk to you about RDA.
I am particularly pleased that I can talk about it in English rather than Icelandic. However, if I speak too quickly or there is something you don’t understand or can’t hear, please let me know. This is your seminar and I want you get as much from it as possible.
I am happy to take questions as we go, but I may not answer immediately if it will be covered later or if it needs lengthy explanation. I am also happy to talk to anyone in the scheduled breaks or over lunch.
There will be an FRBR Game later. Don’t PANIC!. It’ll be fine!

2. What is RDA?
RDA provides guidelines and instructions for description of and access to resources
Principles based
IFLA Statement of International Cataloguing Principles (ICP)
RDA is an implementation of IFLA models
Functional Requirements for Bibliographic Records (FRBR)
Functional Requirements for Authority Data (FRAD)
Developed from AACR2 & ISBD traditions
RDA is a principles based standard. RDA was developed in parallel with and to conform to IFLA’s International Cataloguing Principles. Being principles based means that all RDA guidance and structures are founded on principle and are not arbitrary.
RDA is a cataloguing code which gives cataloguer’s guidance and instruction about the metadata required to describe and enable access to information resources.
RDA is an implementation of two IFLA Functional Models: FRBR and FRAD. These models provide the intellectual foundations for RDA and I will say more about them later.
RDA is also built on AACR2 and ISBD traditions. RDA is different from AACR2 , but there are also many similarities. Some of which are more desirable than others.

3. What is RDA? RDA is comprehensive in scope Covers all material types
Description
Access – authority control
General
Subjects
RDA is comprehensive in scope, that means that it covers all material types likely to be encountered in a general library.
RDA, as the title makes clear, covers description of resources and, unlike AACR2, RDA covers the provision of access, including principles of authority control.
RDA is aimed at the generalist rather than the specialist. It may be supplemented by other more specialist codes, such as DCRM or Cartographic Materials.
RDA is not a subject cataloguing standard, but the content of the resource including what it is about is an essential to description and access. RDA make provision for recording the subject of a work and recommends the use of an appropriate subject standard.

4. What is RDA? A product A community RDA Toolkit Print RDA
LC Cataloger’s Desktop
A community
User contributed content
RDA-L
RDA Registry
EURIG
RDA is also a product of course. Your institution purchases a subscription to license access to the Toolkit. There is also a printed RDA, but RDA was not designed to be used in Print and the Print is not useful for cataloguing. I haven’t actually taken my print copy out of the shrink wrap!
RDA can accessed through Cataloger’s desk top if you have an RDA subscription
RDA is also developing in a community. User contributed content, such as policy statements and workflows are contributed to the Toolkit.
The RDA-List hosted by ALA, provides a forum for questions and discussion with other users. The RDA Registry is a linked open data registry for RDA metadata. I can also mention EURIG, the European RDA Interest Group, which is the reason I am here!

5. RDA is not
A schema …e.g. MARC 21, XML A display format…e.g. ISBD RDA can be used independently of any schema or display format.
I’ve said what RDA is, but I should add that RDA is NOT a schema. At present we do not have a good schema for RDA and this is a barrier to full implementation of FRBR. In practice most of us will implement RDA in MARC 21. More on that later.
RDA is not a display format. AACR2 required the use of ISBD punctuation, RDA does not, although most of us continue to provide it in our MARC exchange records.

6. British Library RDA @ Reykjavik, 21-22 May, 2015
Introducing RDA
British Library Reykjavik, May, 2015

7. Contents RDA Governance and Development RDA Primer Vocabulary
Core Elements
Iteration
Sources
Transcription
Authorized Access Points Representing…
In this section, I want to begin to introduce you to RDA. We will start with governance and why and how RDA is developed.
We are then going to look at what I call the RDA Primer, which describes some of the main features of RDA, particulary those that are different from AACR2.

8. Introducing RDA
RDA Governance

9. RDA Governance Structure Rule development and revision
Ownership
Strategy
Rule development and revision
Product Development
Co-publishers and Fund Trustees:
ALA , CILIP, CLA
Committee of Principals (CoP):
ALA, BL, CILIP, CLA, DNB, LAC, LC, NLA
Joint Steering Committee (JSC):
ACOC, ALA, BL, CCC, CILIP, DNB, LC
ALA Publishing
Committee of Principals
Co-Publishers
Fund Trustees
Joint Steering Committee
American Library Association / CC:DA
Australian Committee on Cataloguing
British Library
CILIP
Canadian Committee on Cataloguing
Deutsche Nationalbibliothek
Library of Congress
RDA Examples Editor
JSC Secretary
JSC Places Working Group
JSC Music Joint Working Group
JSC Technical Working Group
JSC RDA/ONIX Framework Working Group
The current governance structure was inherited from AACR2 . AACR2 and RDA are funded from revenues generated from sales/subscriptions. RDA is owned by the original publishers or AACR2: ALA, CILIP and CLA. It is managed by the Committee of Principals: with representatives from each of those National Associations and the National Libraries of the UK, Canadara, USA, Australia and Germany. The Joint Steering Committee for Development of RDA (JSC) has responsibility for the content of the standard. Finally, the product is managed by ALA Publishing on behalf of the other co-publishers.
The present agreement dates back to The governance structure has served us well but is recognised to be in need of change. It has to be made more representative internationally and more responsive. The Committee of Principals met in Chicago last month and agreed a way forward in principle, which will move away from the current “constituency” model to a more representative model. This has to be done in a responsible and measured way, to ensure that expertise is retained and to ensure that JSC is adequately resourced to sustain continued development, over the challenging period ahead.
The way in which JSC operates is already changing, with much more work being delegated to working groups whose members are selected for expertise rather than representation of a particular community. This is a way in which individuals can become involved in the development process, wherever they may work.

10. JSC Working Groups
RDA/ONIX Framework WG JSC Technical WG JSC Places WG JSC Music WG JSC Capitalization Instructions WG JSC Aggregates WG JSC Fictitious Entities WG JSC Relationship Designators WG
Additional working groups have been established this year. Hear is a complete listing. Clearly there is a lot of work going on, reflecting broader changes in technology and internationalization of RDA.

11. JSC Washington, D.C., 2014
L-R: Simon Edwards (Chair of CoP; Judy Kuhagen (JSC Secretary); Alan Danskin (BL); Kathy Glennan (ALA); Susanne Oehlschlaeger (DNB); Ebe Kartus (ACOC); Gordon Dunsire (Chair of JSC) + Galen Jones (CILIP); Bill Leonard (CCC); Dave Reser (LC); Kate James (Examples Editor, LC)
Here is the Joint Steering Committee. We do really exist!
If we look a bit grumpy, it’s because JSC meetings are hard work!
Holding the computer is Gordon Dunsire – the Chair of the JSC. On the extreme left is Simon Edwards, Chair of the Committee of Principals. Next to Simon is Judy Kuhagen, the JSC Secretary.

12. Changing RDA JSC Constituencies Communities Working Groups
Annual cycle
Proposals
Discussion Papers
Responses
Fast Tracks
JSC Meeting

13. Constituency Responses
The development cycle
Discussion Papers
(August)
Proposals
Constituency Responses
(October)
(November)
JSC Meeting
(November – March)
Drafting
RDA Update
April

14. JSC Working Principle COP IFLA Toolkit changes Strategy Review
FR Consolidation:
RDA Development team
Governance
ISBD Strategy and review
Content management system
JSC Membership
RDA Registry
Non MARC carriers
Moratorium on implementation of short term changes
Greater focus on broader range of entities
Prefer relationships to attributes
14th of April 2015
EURIG Members Meeting 2015, Bern

15. Introducing RDA
RDA Development

16. From AACR2 to RDA
[AACR3]
1978
1988
1998
2002
2004
2005
2010
1997 Toronto Conference
FRBR Published
Editor appointed
New Title
Here is the RDA time line. RDA supersedes the Anglo-American Cataloguing Rules, 2nd edition revised. AACR2 was first published in 1978, unifying the separate British and American texts of AACR, and was continuously updated until In 1997 a major conference was held in Toronto to discuss the Principles and future development of AACR. Some of the outcomes from the conference could be accommodated in the routine revision process, but it was recognised that a new edition would be needed. Tom Delsey, a Canadian consultant, who had already prepared the logical structure of the Anglo-American Cataloguing Rules, was appointed as Editor to draft the new edition, which was imaginatively called AACR3.
The response to the initial draft made it clear to JSC that a new edition was not enough and work began on a new cataloguing code, called, Resource Description and Access in 2005, with Delsey as editor. Further drafts were published as RDA developed and comments were received from around the world. The new code was published as an online toolkit and in print in 2010. 16
Tom Delsey

17. AACR seen as no longer fit for purpose Cataloguing environment
Why RDA?
AACR seen as no longer fit for purpose
Cataloguing environment
Communications Technology
Changing expectations
To understand why AACR2 no longer meets our needs and why RDA is necessary, we need to think about the impact of technological change on how we catalogue, what we catalogue and what users expect of us.

18. From Panizzi to Facebook
RDA has essentially the same objectives as AACR2, i.e. to provide a set of general instructions to enable libraries (and other agencies) to describe all kinds of content and all kinds of carrier.
AACR2 is a continuation of a cataloguing tradition dating back to Rules for the compilation of the catalogue of the British Museum, compiled by Antonio Panizzi, illustrated here. Panizzi was a contemporary of Charles Dickens. If you look closely at Panizzi you can see that his right hand is resting on a printed book, his age was an age of print, but print is no longer the dominant medium for communication.
The world has changed in ways that neither Panizzi, nor his successors, the editors of AACR2, working in the 1970s, could have imagined and cataloguing rules need to embrace those changes. 18

19. The Problems with AACR2 Lack of logical structure
Mix of content and carrier data
Hierarchical relationships are missing
Increasingly complex
Anglo-American bias Pre-dates FRBR Not designed for Internet No well-formed metadata or vocabularies
During the 1990’s there were many complaints from users about the difficulties of applying AACR2. These issues provided the focus for the Toronto conference mentioned above.
“It’s increasingly complex”
“There is no logical structure”
“It mixes content and carrier data”
“Hierarchical and other important relationships are not adequately addressed”
“It reflects an Anglo-American viewpoint”
“It pre-dates the FRBR entity-relationship conceptual model”
“There is not enough support for the ‘collocation’ function of cataloging”
“It did not foresee the Internet or the existence of well-formed metadata”
[Based on a slide from Ann Chapman, UKOLN]

20. The Cataloguing Environment -- Internet
Catalogues are no longer in isolation
Global access to data
‘linked data systems’
Integrate bibliographic data with wider Internet environment
Share data beyond institutions
Any user – any place – any time
The evolution of technologies took a major turn with the creation of the Internet. Catalogues are no longer isolated within the walls of an institution. Catalogues and bibliographic data from any source can now be integrated into the wider Internet environment. New kinds of links can be made, and new displays can be generated for users from data packaged in new ways -- all of it on a global scale in multiple languages and scripts. These can be called ‘linked data systems.’
We now have the technology to provide global connection anywhere that computers can operate. That includes the digital connections of cell phones or smart phones with Internet connections to link to any user -- any place -- any time.
[Based on a slide by Library of Congress] 20 20 20 20

21. The Cataloguing Environment -- Current
Web-based
Wide range of information carriers
More types of content and complexity of content
Metadata (bibliographic information)
Created by a wider range of personnel in and outside libraries
Element-based metadata schemas
Dublin Core, ONIX, etc.
The cataloguing environment continues to evolve:
It is increasingly Web-based.
Content is issued on many more carriers than was traditionally the case. It is not feasible to infer the carrier from the content as AACR2 tried to do. A novel like Bleak House, by Charles Dickens, which was originally issued in successive parts in print, is now available as a novel in print or in various electronic formats; as talking books, or dramatizations in sound or vision; in analogue or digital ;hand held or online.
We need to deal with many more types of content and complexity of content in the resources that we catalogue.
Metadata is now created by a wider range of people, who have a wider range of skill levels -- not only by skilled professional catalogers, but by support staff, non-library staff, vendors, wikipedians, and also publishers. Some of us are using structures other than the MARC format for our records (e.g., using Dublin Core for some digital resources).
And we now have access to descriptive data for resources in digital form – even when the resource is in standard book format, the descriptive data is now available from many publishers using ONIX, which is information we can capture for our bibliographic records.
In the digital world we sometimes find that basic bibliographic description is an integral part of a digital object - the software that helps create the digital object or digitizes an analog object, automatically provides a basic set of metadata, that is attributes or data elements. Think of how the software for word processing, like Microsoft Word, suggests a name for your document based on the first words you type (ironically the “titles” for early manuscripts were the first line of text, too!) Or how it can automatically provide the date you created the document. So we can envision the automatic creation of some of the bibliographic information our cataloging systems can capture, saving the cataloguer’s time. RDA builds on this to emphasize transcribing what you see for the basic elements of bibliographic description (‘the representation principle’).
A key aspect of this new “Semantic Web” environment is that it is built on element-based metadata schemas and vocabularies -- and that is exactly what RDA delivers.
[Based on a slide by Library of Congress] 21 21

22. RDA: A Tool for the Digital World
Optimised for use as an online product
RDA Toolkit
Description and access of all resources
All types of content and media
Resulting records usable in the digital environment (Internet, Web OPACs, etc.)
Resulting records readily adaptable to newly emerging database structures
The Joint Steering Committee stated among the goals for RDA that it was to be a tool designed for the digital world. This had several implications:
RDA was to be a Web-based tool optimized for use as an online product. The result is the RDA Toolkit, which continues to be refined with feedback from users.
RDA was to be a tool that addresses cataloguing all types of content and media
RDA was to be a tool that results in records that are intended for use in the digital environment, through the Internet, Web-OPACs, etc.
RDA was intended to result in records with a metadata set of elements intended to be readily adaptable to newly emerging database structures. 22 22

23. International

24. EURIG: European RDA Interest Group
„Any European national library and/or bibliographic centre or other institution which is responsible for the use and implementation of the cataloguing code in its country may be a member of the Group.”
(EURIG Cooperation Agreement, Article 3)
I mentioned EURIG earlier. EURIG is a membership organization that promotes RDA and shares experience and knowledge among its members.

25. We will look at the toolkit in more detail later, but it is sufficient to say for now that since publication, RDA has been translated into several lanagues

26. French

27. Spanish

28. German, which have been incorporated into the Toolkit
German, which have been incorporated into the Toolkit. Italian and Finnish are expected to be added later this year. There are also printed translations into other languages, including Chinese and Arabic.

29. Vocabulary, Conventions, etc.
Introducing RDA
Vocabulary, Conventions, etc.
In this session, I want to introduce some of the conventions and terminology that you will need to master in order to understand RDA. It is important to remember, that for practical reasons you will be implementing RDA in MARC 21 on an Aleph Library Management System. Much of what you do will be very familiar, but the terminology you use and the way you think about things will change.

30. RDA Primer
Element A word, character, or group of words and/or characters representing a distinct unit of bibliographic information. Note An element in which additional information about a specific attribute may be recorded
This Primer is a quick guide to vocabulary and concepts used in RDA, some of which may be different to those used in other rules.
Element: this is the basic unit of bibliographic information in RDA, for example Title Proper is an element, Publication Statement is an element, Date of Birth is an element. We will use this term a lot.
Note: in RDA a note is an element in which we record additional information about another element; for example Note on Title.

31. RDA Primer Access Point
A name, term, code, etc., representing a specific entity.
Authorized Access Point
The standardized access point representing an entity
Variant Access Point
An alternative to the authorized access point representing an entity
Access point in RDA is the general term for name, term, etc. that represents an entity. For example, the name of a person, or the title of a book. In AACR2 they would be called headings.
Authorized access points contain a controlled or standardised form of the access point and are the form used in the description. For example Rowling, J.K is the authorized access point for the author of the Harry Potter novels
In MARC the authorized access point would occur in the 1XX or 7XX in the bibliographic format or 1XX in the authority format.
Variant Access Points record alternatives to the authorised access point. For example: Rowling, Jo is a variant access point for J.K. Rowling.
In MARC variant access points would be encoded 4XX in the authority format.

32. RDA Primer
Relationship designator A designator that indicates the relationship between entities represented by authorized access points, descriptions and/or identifiers Rowling, J.K. author of Harry Potter and the Chamber of Secrets
Relationships designators in RDA are used to describe the relationship that exists between two entities. For example “author” is a relationship designator. J.K Rowling is the author of Harry Potter and the Chamber of Secrets.

33. RDA Primer
Cataloguer judgement Instructions cannot anticipate every possible question. Cataloguers are encouraged to employ their understanding of the underlying principles in the application of RDA Supplied Data A piece of information from outside the resource, provided by the cataloguer
RDA is based on principles. Understanding the underlying principles will help cataloguers to make decisions when situations arise which are not explicitly covered in the instructions.
RDA sometimes instructs you to supply information when none is available from the resource. This can be taken from any source, but should be identified as supplied data within a transcribed element.

34. RDA Primer
PFC An abbreviation for Person, Family, and Corporate Body. PPDM An abbreviation for Production, Publication, Distribution, Manufacture Statements Agent Another way of referring to Person, Family, and Corporate Body

35. RDA Primer Instruction types
Alternative: may be used instead of the preceding instruction
Optional addition: of data that supplements that called for in the preceding instruction
Optional omission: of data specified by the preceding instruction
Exception: used instead of the preceding instruction, subject to conditions
RDA has instructions, whereas AACR2 had rules.
In addition to the basic instructions the following special instruction types are important:
These are all labelled in the text.
An alternative may be used instead of the instruction that immediately precedes it. Alternatives may be an indication of conflicting legacy practices which cannot be reconciled. Choice of alternatives is a policy issue.
An optional addition allows more information to be recorded than is required by the preceding instruction. Whether to apply the option may be a matter of cataloguer judgment or local policy.
An optional omission allows less information to be recorded than is required by the preceding instruction. Whether to apply the option may be a matter of cataloguer judgment or local policy.
An exception should be followed instead of the main instruction when certain conditions are satisfied; for example, for a particular type of resource.

36. “Core is the floor, not the ceiling”
Core Elements
“Core is the floor, not the ceiling”
Core elements were selected according to the FRBR/FRAD assessment of their value in supporting user tasks
Core
Core if…
Core elements are required if applicable and readily ascertainable
RDA 0.6 for more information
Core is an important concept in RDA. A record that does not satisfy core requirements cannot be considered to comply with RDA.
The core is the floor not the ceiling. It provides the basics necessary to satisfy the FRBR/FRAD user tasks, in many cases exceeding the core will improve the discoverability of the record.
Some elements are Core, i.e. they are always required
Others are core if specific conditions apply
If the information to record a core element is not from the item or basic reference sources, RDA does not expect you to undertake research to find it.

37. Iteration
If there is more than one instance of an attribute, RDA allows you to:
Record them all
Record some
Record the first
Record none (unless core)
There is no “Rule of Three” in RDA
Sometimes an attribute will occur more than once, for example a resource may list multiple contributors or places of publication. In an ideal world we would record them all, but in the real world we may have to place limits. In AACR2 there was an arbitrary rule of three which stopped cataloguer recording more than three instances of an attribute. RDA allows the latitude to record all, some or none. The only proviso is that if the attribute is recorded as a core element you must record the first instance.

38. Sources: Overview Preferred source: source of title proper
Sources for information
Whole resource, with priority order
Then any other source
Indicate if from outside the resource (e.g. brackets)
Exception: if information is typically supplied
Three categories (RDA – )
Pages, leaves, etc., or images of pages …
Moving images
All other resources
The AACR2 concept of “chief source” has been replaced by the RDA concept of “preferred sources.” This is not only a change of term but also reflects RDA’s expansion from a single source to multiple sources for information. The preferred source of information is still the source where you find the title proper. For most elements, RDA gives you permission to take information from any source, with the choices given in a priority order.
If data is taken from outside the resource, it is shown by some means, e.g. enclosed in square brackets. For some elements, the data can come from “any source”; consult the ‘sources’ instruction for each element, or
 Exception: when cataloging a resource that doesn’t typically have bibliographic information on the resource (e.g., photograph, sculpture), you do not need to indicate that the source is from outside the resource.
The sources given in the AACR2 part I chapters have been condensed into three categories:
A. Resources with pages, leaves, etc., or images of pages
B. Moving images
C. All other resources

39. Sources: Comprehensive Description
: Resources issued as a single unit
The resource is the preferred source
: Resources issued in more that one part
Generally, choose the lowest numbered/earliest part
: Integrating Resource
Choose a source of information identifying the current iteration
When creating a comprehensive description, choose a source of information appropriate to the mode of issuance. The resource is considered as a whole. For monographs the preferred source is the resource itself; for serials and other resources issued in more than one part generally choose the first part, if available. For integrating resources, choose the current iteration as the authoritative source.

40. Sources: Resources with Pages, Leaves, etc
Sources: Resources with Pages, Leaves, etc.(or images of pages, leaves, etc.)
RDA priority order:
Title page, title sheet, etc.
Cover
Caption
Masthead
Colophon
If none of these, source containing title
If no title, source with formal presentation
Exception for Early Printed Resources
RDA provides a priority order for the preferred sources for these materials:
Title page, title sheet, etc. (or image)
Cover (or image)
Caption (or image)
Masthead (or image)
Colophon (or image)
If you have exhausted those sources, you can use the source where the title is located.
If your resource doesn’t have a title, use a source where the information may be formally presented (e.g., perhaps introductory pages). But these are still within the resource.
There is an alternative for digital resources or microform, which says to use an eye-readable label. The BL applies the alternative for the former but not the latter. There is also an exception for early-printed resources, with a different priority order.

41. Sources: Moving Images
RDA
Title frame(s), or title screen(s)
If only individual contents listed, use a formally presented collective title from elsewhere in resource
Alternative to use permanent label on resource
If no title frame or screen, RDA has separate instructions for:
Tangible resources
Online resources
Use the title frame or frames, or title screen or screens as the preferred source of information.
If the title frames or screens only list the titles of the individual contents and another source which is part of the resource has a formally presented collective title, use the first source with the formally presented collective title.

42. Sources: Other Resources
RDA
There are different guidelines for:
Tangible Resources
Online Resources
Apply to resources that are not covered by:
Resources Consisting of One or More Plates, Leaves, Sheets, or Cards
Resources Consisting of Moving Images
RDA provides a priority order for the preferred sources for resources other than those in the first two categories. The instruction differentiates whether the resource is tangible or online – there are specific sources listed for each of these.
Tangible:
Textual source on resource (e.g. label)
Internal source such as a title screen, formally presenting the title textually
Container or accompanying material issued with a resource
Online:
Textual content
Embedded metadata in textual form that contains a title (e.g. metadata embedded in an MP3 audio file)
In either case above, if none of the sources listed is appropriate, use a source where the data is formally presented

43. Sources: Other Sources of Information
RDA 2.2.4
Sources of information outside the resource
Priority order
Accompanying materials
Other published descriptions
Container not issued with the resource itself
Any other source (e.g., a reference source)
RDA provides a priority order for the other sources when you can’t find a title from the resource itself:
Accompanying materials
Other published descriptions of the resource
Container not issued with the resource itself (e.g., a box or case made by the owner)
Any other source (e.g., a reference source)
When the instructions specify transcription, indicate that the information is supplied from a source outside the resource itself (e.g. by means of a note, or some other means such as the use of square brackets).

44. Transcription ‘Take What You See and Accept What You Get’
ICP Principle of Representation
Generally, do not alter what is on the resource
Accurate representation
Encourage re-use of found data
This is the overriding principle of RDA concerning the transcription of data. It is consistent with the ICP “Principle of Representation” to represent the resource the way it represents itself. This is a fairly significant change from AACR2, which includes extensive rules for abbreviations, capitalization, punctuation, numerals, symbols, etc., and in some cases directs the cataloger to ‘correct’ data which is known to be wrong (e.g., typos). With RDA we generally do not alter what is on the resource when transcribing information for certain elements. This is not only to follow the principle of representation, but also for a more practical reason: to encourage re-use of found data you can copy and paste or scan or download into your description of the resource.

45. Transcription: Transcribed Elements vs. Recorded Elements
Distinction between transcribed elements and recorded elements.
For transcribed elements, generally accept the data as found on the resource.
For recorded elements, the found information is often adjusted (for example, the hyphens in an ISBN are omitted).
RDA distinguishes between transcribed elements and recorded elements.
For transcribed elements, generally accept the data as found on the resource.
For recorded elements, the found information is often adjusted (for example, the hyphens in an ISBN are omitted). 45

46. Transcription: Language and Script
RDA 1.4
“Transcribe … in the language and script found in the resource”
List of elements to be transcribed
The basic instruction for most of the elements for describing a manifestation is to transcribe the data in the language and script found in the resource (“take what you see”). RDA 1.4 contains a list of elements to be transcribed from the resource in the found language and script.

47. This screenshot from RDA shows part of the list of elements to be transcribed from the resource in the found language and script. 47

48. Transcription: Choices
RDA 1.7
1.7.1 General Guidelines on Transcription
Alternative instructions
Follow local policy
Style Manual
Accept without modification:
Derived record
Data derived from digital source
The general guidelines for transcription cover capitalization, punctuation, and spacing, symbols, diacritics, etc. The instructions point to additional guidance in the appendices.
The alternative instructions allow an agency to follow in house guidelines or a published style manual or to accept data without modification when using a description created by another agency or derived from a digital resource
“If the agency creating the data has established in-house guidelines for capitalization, punctuation, numerals, symbols, abbreviations, etc., or has designated a published style manual, etc., (e.g., The Chicago Manual of Style) as its preferred guide, use those guidelines or that style manual in place of the instructions given under 1.7.2–1.7.9 and in the appendices.”
BL policy is to follow the guidance given in RDA at
BL does not follow the first alternative at 1.7.1
BL policy is to follow the second alternative at 1.7.1, which allows you to accept conventions followed in a description created by another agency or when data are automatically scanned or downloaded from a digital source.
“ If a description created by another agency is used or if data are derived from a digital source of information using an automated scanning, copying, or downloading process (e.g., by harvesting embedded metadata or automatically generating metadata), accept the data without modification.”
Note that The LC/PCC PS under has important information about conventions for recording (non-transcribed) data in access points.

49. Transcription: Capitalization
Accepting found capitalization:
$a Cairo : $b THE CITY VICTORIOUS /
$c Max Rodenbeck.
250 ## $a FIRST VINTAGE DEPARTURES EDITION.
Changing found capitalization:
$a Cairo : $b the city victorious / $c Max Rodenbeck.
250 ## $a First Vintage Departures edition.
These are equally acceptable in RDA
Here are two versions of four transcribed elements – title proper, other title information, statement of responsibility and edition statement: The use or not of uppercase letters does not affect searching and retrieval in the vast majority of systems, and many Web applications use uppercase letters.
[Animation for second example on mouse click]

50. Non Transcribed Elements
When recording other elements (e.g., extent, notes), record them in the language and script preferred by the agency creating the data
When adding data within an element, record it in the language and script of the element to which it is added
When supplying an element, generally supply in the language of the agency, e.g. English
Regarding non-Latin scripts, BL policy is to record a Romanized form instead – where possible.

51. Authorized Access Points Representing…
The standardized access point representing an entity
…one of the methods by which an entity can be identified
They are constructed by stringing together attributes that identify the entity
For example:
Eliot, T. S. (Thomas Stearns), 1888–1965
Name of the Person Fuller form of Name Date of Birth-Date of Death
Earlier, we touched on authorized access points. In RDA these are defined as standardized access points representing an entity. So, they are one of the ways that we can identify a person or work, etc. in RDA. For example, here is the authorised access point for T.S. Eliot
The authorised access point is not an element, but it has been composed by stringing elements together. RDA therefore has separate instructions on how to record elements and how to construct authorised access points using elements. This can be confusing at first, so it is important to understand where about in the instructions you are.

52. Here’s an illustration for Chapter 9.
At the top of the screen are the instructions on how to Record the Fuller Form of a Name (at 9.5)
At the bottom of the screen, are the instructions on when to include this information in an authorised access point.

53. Using RDA
RDA Toolkit
In this session, I will introduce you to the RDA Toolkit. The Toolkit is essentially a content management system. It includes RDA of course, but it also includes a lot of other resources. It will be easier to show you directly. So first of all, how to access the Toolkit.

54. Contents
RDA Toolkit MARC 21 & RDA Cataloguers’ Perspective Managers’ Perspective
I have slightly altered the order of this section from what you may have in the schedule. I will start with a look at the Toolkit, which will give you a break from power point – if not from me.
We’ll then look at MARC and RDA.
I’ll finish by looking at implementation for perspectives of cataloguers and managers – which are not always the same!

55. Using RDA
In this session, I will introduce you to the RDA Toolkit. The Toolkit is essentially a content management system. It includes RDA of course, but it also includes a lot of other resources. It will be easier to show you directly. So first of all, how to access the Toolkit.

56. Accessing the RDA Toolkit
(Toolkit website) (RDA content)
Either of these URLs will get you there. The first URL takes you to the Website for the Toolkit, which is maintained by the publishers and has information about subscriptions, releases, events, etc. You can access RDA from the Toolkit website, or you can go directly to RDA using the second URL.
I am not going to train you in how to use the Toolkit, but I will use it to illustrate the scope of the content and show you some of the features of RDA. For those of you attending tomorrow’s workshop, there will be a hands on session and an opportunity to practice.
<LINK to Toolkit – point out features of publisher site, briefly.
<CLICK access button>
Opening Screen: layout – navigation pane; document pane; scroll bars; tabs; search box, subscription, help, feedback.
Open a document from navigation pane– point out icons – illustrate dual language view; warn about printing
Quick Search 2.10 – policy statements; update flag; glossary entry; cross references
Scroll down – alternative instructions
Switch tabs
Tools: User generated content (Workflows, Mappings)
Resources: AACR2; Policy statements

57. USING RDA
MARC 21 & RDA

58. RDA & MARC RDA designed to implement FRBR
RDA intended to exploit potential of Web
Most implementations in MARC
MARC 46 years old
MARC 21 has adapted to RDA but cannot deliver full benefit
Does not identify separate entities
Does not separate content and carrier
Semantics may be implicit in position or order of information

59. Composite Records Starting point is the item in hand
Using MARC 21 we create a composite record to:
Identify and describe the manifestation
Identify and describe the content (work and expression)
Identify and describe a specific copy (item)
Relate to other resources, persons, families, corporate bodies, etc.
Assign subject indexing and classification as usual
In MARC, the bibliographic record is a composite record. The book or other resource in hand which you are cataloguing is mainly the manifestation in FRBR terms. BUT IN MARC you will have to include attributes of the work, expression, and relationships to other entities, such as persons.
The bibliographic record for the manifestation may also contain elements of copy-specific information, which in RDA are called item attributes

60. Anatomy of an RDA record in MARC
1XX/240 = work & expression [contained] , 490 = manifestation 300 = expression, manifestation 3XX = work, expression, manifestation 5XX = work, expression, manifestation = work & expression [related] = work, expression, manifestation [related] 8XX = work & expression [related] manifestation & item
This slide gives a summary view. You don’t need to memorize this breakdown. Don’t worry if it isn’t easy to understand right now. It’ll get easier! After you’ve created some practice records, you’ll understand more and more about what characteristics you’re including in the MARC fields.

61. New MARC Fields
Several new MARC fields introduced to support RDA List on MARC 21 Website: Most of the new fields were needed to define attributes more explicitly in the Authority Format The RDA Toolkit contains mappings between MARC 21 and RDA. An RDA in MARC 21 looks very similar to an AACR2 record in MARC 21

62. Identifying RDA Records
040 Cataloging Source
$e = rda
Record Leader/18 Descriptive Cataloging form =
“i” if ISBD punctuation used
“ “ if ISBD punctuation is not used
In MARC 21 RDA records can be identified by the code rda in the 040 field.
The Record Leader field in a descriptive cataloguing form can distinguish an RDA record indirectly. Position 18 is coded “i” for records formulated using ISBD punctuation, which is specified by RDA. (Code “a” was used for AACR2 records).

63. Hybrid records
Policies and processes for: Re-use of non-RDA records Editing or updating legacy (pre-RDA) records Serials Multiparts

64. Questions?

65. Cataloguer’s Perspective
Using RDA
Cataloguer’s Perspective
The British Library formally implemented RDA on 1st April This means that we’ve been using RDA for over two years now. I am not someone who catalogues on a day to day basis, but I thought it would be useful to feedback something of

66. Adapting to RDA Continuity Change Aleph MARC Cataloguing Subject
RDA templates, Indexes
New elements, coding
Terminology, sources, transcription, policies, documentation
None
For most cataloguers, adapting to RDA was relatively straightforward. The descriptive cataloguing standard had changed, but much else had stayed the same. We are still using Aleph and MARC 21. Although, RDA is different from AACR2, it comes from the same stable and the cataloguing process has not fundamentally changed. RDA is not a subject standard, although subject is a core relationship (since April). We carried on using LCSH and DDC.
Some things have changed of course.
In Aleph the AACR2 templates were replaced by Aleph templates and macros used as shortcuts were amended or defined. Some new indexes were also introduced to support new MARC fields.
Most of the new MARC fields were in the authority format and are used to specify attributes that would previously have been held in notes or in generic subfields, such as “qualifier”. In the bibliographic format, the main changes to Marc are the replacement of General Material Designation with Content media and Carrier types and the introduction of 264 field to replace 260.
There would be no point in doing this, if cataloguing was exactly the same, so some key changes include terminology, sources of information, transcription. We also changed most of our internal documentation and rewrote our policies.

67. Separation of content and carrier
There is no General Material Designation in RDA
Replaced by
Content Type e.g. text, sound [Work, Expression]
Media Type [Manifestation]
Carrier Type e.g. volume, audio disc [Manifestation]
RDA is much better than AACR2 for digital resources
Reproductions are described in their own right in RDA
The General

68. Cataloguer’s judgement
Cataloguers like clarity and consistency
AACR2 “Case Law”
provided detailed guidance down to a very granular level.
Rule interpretations
RDA is clear and consistent RDA is principles based
Cataloguer judgement is important. RDA cannot provide guidance for every situation you will encounter. RDA provides a framework for taking defensible decisions. It aims for clarity and consistency of approach. It follow clear principles. C

69. Example
Collaborative work No primary responsibility
12 collaborators
RDA First named must be recorded Any or all others may be recorded
AACR2 Only first three are recorded
BL policy: Generally record all, unless unduly onerous.
“unduly onerous” is not defined, it is left to cataloguer judgement

70. Manager’s Perspective
Using RDA
Manager’s Perspective
The British Library formally implemented RDA on 1st April This means that we’ve been using RDA for over two years now. I am not someone who catalogues on a day to day basis, but I thought it would be useful to feedback something of

71. Preparation
Training
Productivity

72. Preparation Communications Business case Cataloguers Service impacts
Public
Other stakeholders
Infrastructure
Toolkit
Documentation
Training materials
Systems configuration
We had a lot of time to prepare. RDA was a long time in development and even after publication, implementation was delayed. Communicating what you are going to do is important and it is also important that people understand why, particularly if it is going to disrupt their life. From British Library perspective, being so closely identified with the development of RDA there would have needed to be a compelling reason for us not to implement RDA. Nevertheless, I still had to make a business case. This was mainly concerned with

73. RDA Business Case: Costs
RDA Governance & Development
Long term commitment
Planning
Time and opportunity costs
Lost production
(estimated 20,000 items)
Annual Subscription
Concurrent licenses
The costs of RDA for BL have been considerable.
The Library has a commitment under the joint understanding of 1989 to contribute to the governance and development of AACR2/RDA through the Committee of Principals and the Joint Steering Committee for Development of RDA (JSC). Throughout the AACR3/RDA development period, from the present, this has required a much more substantial commitment than was previously the case.
Planning and preparation has involved a large number of managers and staff over a longer period than anticipated because of delays in publication. This was at the expense of other opportunities and activities.
We forecast implementation would result in a loss of at least 20k items from planned production during
The cost of licensing RDA year on year will be more significant than the cost of replacing printed AACR2 every 7 years or so. We have made full use of offers from the publishers to control costs and we are also trying to maximise the value of each license. We have about 140 staff who say they need access, but we anticipated that fewer than 100 concurrent licences will be sufficient. This has been borne out, so far 135 user profiles have been registered and we have 80 licences for concurrent use. Based on new reports available from ALA and anticipated staff losses, we should be able to reduce this significantly in , although savings will be offset by price rises.

74. Business Case: Benefits
Long Term
Short Term
Infrastructure
Discovery
Technology
Linked Data
Strategic Continuity
International commitments
Better Cataloguing
Consistency
Authority Control
Electronic resources
Content / Carrier
LONG TERM
The case for implementing RDA is heavily weighted towards investment in future infrastructure. The FRBR model (we believe) offers a more intuitive basis for discovery of our collection. We believe that RDA will provide the community with the necessary impetus to look beyond MARC and to embrace Web technologies. Prospective users are on the Web and in social media, they are not in our catalogue or on out Website. Our data needs to be where they are.
We are also interested in the potential of linked data to leverage productivity and support discovery
Since the early 1990s our cataloguing strategy has been to follow North American standards as closely as possible to maximise the efficiency of our copy cataloguing. Library of Congresses’ decision on RDA implementation was therefore very important to the British Library.
The Library has invested heavily in RDA through our participation in CoP and JSC and we need to realise that investment.
SHORT TERM
RDA does offer qualitative improvements over AACR2: it is more consistent and there are fewer examples of format specific “case law”; RDA provides comprehensive instructions for authority control. RDA allows much more latitude to identify person, families and corporate bodies than AACR2. AACR2 was concerned with differentiation not identification. Electronic resources are handled much better in RDA than AACR2, because the model separates content and carrier and is no longer based on class of materials.

75. Systems Checklist Aleph ILS: acquisition and cataloguing
PRIMO: search interface
Validation
Researchers’ tools, reference/citation management

76. Systems issues and changes
Import/Export profiles
Batch Upgrade (automatching)
MARC changes implemented on Aleph (95 change requests)
Most changes affect authority records
New indexes
Shared configurations with UK community
Templates / Macros
MARC Report
Very little change to PRIMO
The main technical issues were to reconfigure Aleph and related systems to support RDA. This has involved a lot of table changes to accommodate new MARC fields and content designation. So far we have made about 95 substantive changes; these were mostly very straightforward and mainly affected the authority libraries. We have defined some new indexes, e.g. content, media and carrier type. We have shared these configuration details with other libraries on request.
All of our cataloguers use MARC Report to check their records before releasing them to products. We have worked closely with TMQ, the manufacturers, to developed MARC Report for RDA.
We have also made changes to templates and macros used in Aleph to support record creation. Our public interface does not use Aleph directly and we have made very few changes to PRIMO, which underlies Explore the British Library. There are several reasons for this.

77. Legacy Data LC NACO re-distribution of authority data
Additional clean-up
Long term enhancement:
Identification of works and expression
Enrichment of attributes
Content, media & carrier types
Hybrid records are a fact of life for us
British Museum; AACR; AACR2; “other”;
RDA is just the latest…
During the implementation we made very few changes to legacy data. We implemented the NACO Phase 1 and Phase 2 changes to authority data and applied the results to the bibliographic records.
RDA accounts for a very small proportion of records viewable on Explore the British Library. We have 15 million records created using AACR2 or AACR1 or British Museum and other local rules or even no rules at all. PRIMO also contains 40 million article records. RDA has potential to integrate some of this diverse cataloguing, but it is not the main driver for our metadata enhancement programme.
Before we can effectively FRBRise our data, we need to enrich these legacy records. This is part of our long term strategy for data enhancement to improve identification of works and expressions. As a first step we need to enrich the attributes that enable identification, such as language and content type, media type and carrier type.
We live with hybrid records in our legacy data and RDA adds a new layer.

78. Phased implementation of RDA
Expert Groups Established
2008
Aleph Implementation Environment
2009
Re-distribution of Derived RDA Records
2012 June
Authority Control (NACO)
2012
RDA Cataloguing in Publication
2013 January
Descriptive Cataloguing (Boston Spa)
2013 Q1
Descriptive Cataloguing (London)
The British Library project was to implement RDA, not to implement FRBR. FRBR implementation was not feasible within the constraints of current systems.
The implementation was planned as a phased project.
An group of experts was created very early in the project with members in London and Boston Spa. This group was involved in the project planning and development of documentation, etc.
We planned to implement RDA in stages, starting by accepting derived records from third parties. Our Authority Control Manager was an enthusiastic convert to RDA and we started contributing RDA records to NACO in early 2012.

79. External Change to CIP contract with BDS
switch to RDA in advance of BL implementation date
Contractural changes with other record suppliers
as agreements renewed
Legal Deposit Libraries Shared Cataloguing Programme (LDLSCP)
Shared documentation

80. Training
Mix activities Classroom and Hands on Check work and give feedback Start with easy stuff to build confidence

81. PRODUCTIVITY Individual productivity targets suspended during 2013-14
Production hit was estimated at 20k records
Mitigation strategies kept output on target at end of year (31/3/2013)

82. Productivity
RDA has been neutral No evidence for any change in productivity Despite implementing RDA 3 days before legal deposit extended to non-print.

83. Production Targets Comparative snapshot % Items processed
(to Feb)
All Print
104%
100%
103%
All Digital
90%
84%
107%
Comparative snapshot
Measure
October 2013
October 2012
Cataloguing Daily Rate
10.41
10.66
Copy Cataloguing daily rate
39.09
39.02
Team Authority Control %
33%
20%
The following figures illustrate the relative neutrality of RDA with regard to productivity. The first table shows the productivity targets for processing intake. The target is to process 100% of intake. In 2012 we exceeded the targets for print and digital. In 2013 the target for print was achieved, despite implementation. The target for digital was not achieved, but RDA was not the issue. The same factors apply this year, but it will be noted that the print target has again been exceeded.
The second table shows comparative figures for October 2012 (before RDA training really started) and for October 2013, by which point we had been using RDA for about 6 months. Three figures are provided. The first row is the daily rate for original cataloguing. The second is for copy cataloguing. There is very little variance between the two periods.
The third figure, illustrates the one area where RDA has had a negative impact on productivity and needs some explanation.
Not all cataloguers are NACO trained, therefore within each team those who are not NACO trained, pass their work to colleagues who are. We measure the amount of time these cataloguers spend on team authority control. We noticed that the time spent on authority control within the cataloguing teams had substantially increased; from 1 day/week to almost 2 days/week. Further analysis showed that this increase is attributed to more names being controlled under RDA. This is a good thing in principle, but we have to balance the value of the additional access against the number of items processed. We found that without the rule of three, cataloguers were proving many more access points for compilations and conferences. We have reluctantly, amended our policy to limit this. We have also tried to limit how much information is included in authority records.

84. Cataloguing issues High Level
Change to well established policies and procedures
Realising the benefits/deferral of gratification
Quality Assurance
Cataloguing
Compilations / Collaborations
Cataloguer Judgement – “unduly onerous”
Authority Control productivity/demands

85. Future Developments Linked Data
In this section, I am going to talk briefly about linked data. It is a big topic and we don’t have time to cover it in depth. It is a technology, but also an aspiration and has become closely linked with the open data movement.

86. Linked Data
Internet = web of computers WWW = web of documents Next? = web of data
Fundamentally, linked data refers to the next stage of development of the Web.
First we had the internet, which linked computers together across the world.
That was useful, but not really of interest to the public.
Next came the World Wide Web, growing out of the need Tim Berners-Lee perceived at CERN to shared information seamlessly. The World Wide Web is a web of documents. It links documents together.
However, Berners-Lee has always seen the Web as much more than that and has espoused the idea of the semantic web in which data is linked and can be used by applications to deliver services.

87. Linked data
RDF = Resource Description Framework Model for the web of data
Subject
Object
Predicate
The underlying model for the web of data is RDF- Resource Description Framework
It is a relatively simple model…

88. Linked data
RDF = Resource Description Framework Model for the web of data
This item
The British Library
Is owned by/owns
The underlying model for the web of data is RDF- Resource Description Framework
It is a relatively simple model…to say anything about anything
If each component is represented by a persistent identifier, such as a Cool URI, the data can be interpreted by a computer using the Web.
This is also looking very similar to the entity model for FRBR that we looked at this morning

89. Linked data
Cataloguing is really about identifying entities and explaining the relationships between them Linked data has the potential to remove many of limitations that have constrained cataloguing in the past. RDA identifies entities RDA defines lots of relationships RDA and linked data are a good match

90. However…
Implementations of RDA are almost all in MARC 21 MARC 21 is not a good match for linked data The full benefits of RDA are dependent on a replacement for MARC and a new generation of systems. This is a slow process…

91. Initiatives
RDA Registry
BIBFRAME
RIMMF

92. RDA Registry

93. BIBFRAME
Bibliographic Framework Initiative Being developed as a replacement for MARC 21 Still in early stages Does not explicitly support FRBR

94. RIMMF: RDA in Many Metadata Formats
Tool for visualisation and experimentation

95. Links: IFLA
Statement of international cataloguing principles FRBR FRAD ISBD description

96. Links: RDA
Joint Steering Committee for Development of RDA RDA Toolkit (website) (toolkit) RDA Registry RIMMF ?id=rimmf

97. Links: linked data
Resource Description Framework (RDF) Bibliographic Framework (BIBFRAME) (official site) (background)

98. Links: other
RDA in MARC