2  Online Cataloging

There are three categories of online cataloging activity.

• Copy cataloging. If a matching bibliographic record exists in WorldCat, optionally modify the record for local use, set holdings, optionally create associated Local Holdings Records (LHRs), and optionally export a copy of the record with your modifications. Do not replace the WorldCat record. WorldShare Management Services (WMS) users may instead create Local Bibliographic Data Records (LBDs).
• Original cataloging. If no matching bibliographic record exists in WorldCat, create a new record in accord with current cataloging practices, e.g., Resource Description & Access (RDA). If local edits are needed, then follow the steps listed under copy cataloging. For more information about original cataloging, see section 2.2, Creating Bibliographic Records.
• Transcribing cataloging copy. If no matching bibliographic record exists in WorldCat and there is an existing printed or other non-machine-readable record available, create a new record using data found in the existing non-MARC record. For more information about transcribing cataloging copy, see section 2.3, Transcribing Cataloging Copy.

For information on determining whether an existing bibliographic record matches an item to be cataloged versus the need to create a new bibliographic record, see chapter 4, When to Input a New Record.

For information about correcting existing bibliographic records, see chapter 5, Quality Assurance.

There are three levels of cataloging that indicate the completeness of bibliographic records. These levels are generally defined in terms of cataloging standards such as Resource Description & Access (RDA) or Anglo-American Cataloguing Rules, 2nd edition (AACR2).

• Full-level cataloging. Records that contain all the applicable core-designated elements in RDA (RDA 1.3) or meet the requirements of AACR2 second-level description (AACR2 1.0D2). BIBCO Standard Records and CONSER Standard Records are also considered full-level cataloging.  ELvl is coded blank or 1.
• Minimal-level cataloging. Records that lack one or more applicable core-designated elements in RDA (RDA 1.3) or meet the requirements of AACR2 first-level description (AACR2 1.0D1). ELvl is coded 7.
• Abbreviated-level cataloging. Brief records that do not meet the requirements of Minimal-level cataloging specifications. ELvl is coded 3.

Whether you are creating new bibliographic records as a result of original cataloging or transcribing cataloging copy, use the following guidelines.

Indicate the cataloging instructions used by coding Desc and, if appropriate, field 040 subfield ǂe with codes found in the Description Convention Source Codes . In the cooperative environment of WorldCat, OCLC strongly encourages the use of standard description conventions. If you are using your own description conventions for archival or rare materials, enter the code local followed by a slash and your MARC organization code in field 040 subfield ǂe, e.g., ǂe local/DFo.

Indicate the language of cataloging in field 040 subfield ǂb. For information on inputting the language of cataloging see section 2.6, Language of Cataloging, and chapter 4, When to Input a New Record.

OCLC provides the capability to include all Unicode characters in bibliographic records. For a list of available scripts, see the Unicode Character Code Charts . Include non-Latin scripts in your original cataloging or transcribed cataloging copy if you have the expertise and knowledge to do so. For more information, see section 2.7, Character Sets.

You may provide romanized, transliterated, or Latin-script equivalent data associated with a non-Latin script. For English-language cataloging, follow the ALA-LC Romanization Tables . For cataloging in other languages, follow recognized standards for romanized, transliterated, or Latin-script equivalent data associated with non-Latin scripts, such as e.g., ISO transliteration standards.

For cataloging in English, search the LC/NACO Authority File and use the established forms of names and preferred titles found there. If not found, construct the access point according to RDA or AACR2. For original cataloging in other languages, search applicable authority files and use the established forms of names and preferred titles found there. If not found, construct the access point according to instructions or rules used to create the bibliographic record.

For transcribed cataloging, see section 2.3, Transcribing Cataloging Copy below.

For subject access points, search applicable authority files and use the established subjects found there.

General Material Designations (GMDs) are a controlled list of generic terms describing the medium of the item. Prior to AACR2, they were also known as media qualifiers. Do not include GMDs in bibliographic records. Instead, prefer use of content, media, and carrier terms and codes in fields 336, 337, and 338. For more information, see field 245 subfield ǂh.

Many cataloging communities, whether national, regional, based on language of cataloging, or format of matieral to be cataloged, maintain their own cataloging policies.

For cataloging in English, generally follow LC-PCC practice when creating new records. If OCLC practice conflicts with LC-PCC practice, follow OCLC practice.

For cataloging in other languages, generally follow your community's practice. If OCLC practice conflicts with your community's practice, follow OCLC practice.

The following list provides guidelines to resolve conflicts between LC-PCC practice and OCLC practice for cataloging in English under RDA.

Libraries participating in cooperative programs with the Library of Congress (i.e., BIBCO, CONSER, NACO, PCC) must follow appropriate LC-PCC PS for records and access points created as part of those programs.

The following list provides guidelines to resolve conflicts between LC-PCC practice and OCLC practice for cataloging in English under AACR2.

When you create a WorldCat record from a printed or other non-machine-readable record, you are transcribing cataloging copy. Such cataloging copy may include current RDA or AACR2 cataloging such as cataloging-in-publication data that is in the book. Alternatively, it may include printed cataloging copy that is pre-RDA and pre-AACR2 such as data found in the The National Union Catalog, Pre-1956 Imprints.

Since OCLC loads current Library of Congress (LC) and National Library of Medicine (NLM) cataloging as well as cataloging from many other sources, catalogers seldom transcribe current cataloging copy.

Since many libraries have completed projects to convert their non-machine-readable cataloging and such data is represented in WorldCat, catalogers seldom transcribe printed cataloging. For those rare cases when cataloging copy exists and needs to be transcribed, this section provides guidelines for the transcription of both current and retrospective cataloging copy.

Cataloging-in-publication (CIP) data found within a resource, whether it claims to have been provided by a national library or another source, should be treated as pre-publication data. As stated in LC-PCC PS 2.2.2.1, "Do not consider pre-publication cataloging data (foreign or domestic) appearing in the item as a source of information for transcribed elements." Verify all authorized access points appearing in CIP data in the appropriate authority file(s).

You may use existing cataloging for the original item to create a new record for a microform or photocopy reproduction. Treat the new record as your institution's original cataloging. Code Srce appropriately and enter your OCLC symbol in field 040 subfield ǂa. For more information see section 3.1, Provider-Neutral Cataloging, section 3.1.2, Photocopies and Print-on-Demand Reproductions, and section 3.3.4, Microforms.

You must search the appropriate authority file to verify forms of entry in fields x00, x10, x11, and x30. If the forms of name and preferred title entries are not in the authority file, OCLC encourages you to construct the access point or preferred title according to the appropriate instructions, but you are not required to do so.

If an access point appearing on the printed cataloging copy was incorrectly constructed, revise that access point to an appropriate form

For cataloging in English, many pre-AACR2 access points for personal names, corporate bodies, etc., are similar to corresponding RDA or AACR2 forms. However, some pre-AACR2 access points will never have corresponding RDA or AACR2 forms. Often, these access points include obsolete elements. Their inclusion in WorldCat and in local catalogs usually results in split files of access points for similar materials because of different practices under AACR2 and older instructions. If you construct replacement access point(s), update or delete the old access point(s) in favor of corresponding name, title, subject, and/or genre access point(s). Examples:

Example of pre-AACR2 personal name with obsolete elements:

Example of pre-AACR2 corporate name with obsolete elements:

Example of pre-AACR2 access point with obsolete elements:

Historically, in cooperation with OCLC advisory groups, OCLC established input standards for entering bibliographic data into WorldCat. Unless noted otherwise, these standards conform to national standards. For more information on these standards, see the Library of Congress document MARC 21 for Bibliographic Data, National Level Full & Minimal Requirements .

In addition, OCLC adopted the core-level standard that was defined in 1994 by the Cooperative Cataloging Council (now known as the Program for Cooperative Cataloging (PCC)). Core-level records met the requirements of AACR2 first-level description (AACR2 1.0D1) and some of the requirements of AACR2 second-level description (AACR2 1.0D2). ELvl was coded 4.

Note: Core-level record standards are now obsolete. Do not input records with ELvl coded 4.

For historical information about core-level standards, see section 2.5, PCC Standard Records below.

The code in ELvl identifies the standard to which a record conforms. Data required for online cataloging may not be required by cataloging instructions. Requirements for each encoding level have varied over time. Correspondence between data in records and descriptive cataloging instructions is not exact.

Full-level cataloging

Full-level cataloging seeks to ensure inclusion of the essential data elements necessary to meet user needs. It is a solid "floor" description of a resource that can be built upon in a shared environment. Full-level cataloging does not preclude the use of data in a bibliographic description representing more extensive cataloging treatment.

Generally, full-level cataloging meets the requirements of RDA Core or AACR2 second-level description. Full-level cataloging uses ELvl code blank.

Create full-level records when possible using the following guidelines as appropriate:

• BIBCO Standard Record (BSR)
• CONSER Standard Record (CSR)
• RDA 1.3
• AACR2 1.0D2
• OCLC Bibliographic Formats and Standards requirements

Minimal-level cataloging

Minimal-level cataloging represents the minimum data necessary to identify an item. Generally, the records meet the requirements of AACR2 first-level description, rule 1.0D1. Minimal level does not prohibit the addition of data that you think is necessary. Minimal-level cataloging uses ELvl code 7.

Users may upgrade minimal-level records. For more information, see chapter 5, Quality Assurance.

Create records according to the minimal-level standard, as appropriate. Constraints on resources, collection development policies, and cataloging priority can also determine the kinds of materials receiving minimal-level cataloging. For example:

• Conversion of older cataloging records that do not contain full bibliographic data
• Materials in foreign languages in which you lack expertise
• Materials such as theses and dissertations, pamphlets, exhibition catalogs, analytics, audiovisual materials, technical reports, and locally produced publications of little interest outside the issuing institution

See the field descriptions in this manual for lists of full-level and minimal-level standards for the OCLC-MARC Format. Use the following guidelines for minimal-level records.

• Generally, use default values to code the fixed field
• Enter series statements that appear on the piece. You are not required to create a series access point.
• For works entered under title, you must include one added access point if applicable or available. For works entered under the name of a person or corporate body, no added access points are required.

Abbreviated-level cataloging

Brief records that do not meet the requirements of minimal-level cataloging specifications. Abbreviated-level cataloging uses ELvl code 3.

Because abbreviated-level records may not meet minimal-level standards, users with full-level cataloging authorization or higher can upgrade these records. Depending on the authorization level, users may upgrade to a higher encoding level. For more information, see chapter 5, Quality Assurance.

Generally, Dublin Core records are treated according to abbreviated-level guidelines. Dublin Core users apply various standards when creating Dublin Core records:

• The Dublin Core Metadata Initiative provides guidance on commonly-adopted best practices in Using Dublin Core
• In the library community, the DCMI-Libraries Working Group publishes the DC-Libraries-Library Application Profile
• Groups of libraries develop their own specific practices based on more general standards, such as those listed above. For example, see the Collaborative Digitization Program's CDP Dublin Core Metadata Best Practices .

When entering data in abbreviated-level records, use the following guidelines:

All other elements not mentioned above are optional.

Guidelines

The Program for Cooperative Cataloging (PCC) provides metadata application profiles for both a BIBCO Standard Record (BSR) for monographic resources and a CONSER Standard Record (CSR) for serial resources. These standards establish a baseline set of elements emphasizing access points over descriptive data, providing catalogers with a set of required elements appropriate to various bibliographic formats. The standard also does not preclude the use of any data in a bibliographic description representing more extensive cataloging treatment.

These elements have been determined to support user needs to find, identify, select, and obtain resources. While PCC records following these standards contain code pcc in field 042 and must contain access points backed up by authority records, all libraries are encouraged to use the BSR and CSR as models of good full-level cataloging practice when creating new records or replacing existing records.

Policy for parallel records

Parallel records are defined as multiple records for the same manifestation of a work that are described in different languages. Only one record per language of descriptive cataloging is allowed within WorldCat for each manifestation of a work.

In the case of multiple scripts representing the same language, only one record per language of cataloging is allowed in WorldCat. For example, in Chinese language cataloging, do not create separate parallel records in both Chinese traditional and simplified scripts. Transcribe the script as it appears on the resource. If a title access point in the alternative script is needed, use field 246.

Previous policy

Previously, records for the same manifestation but cataloged in different languages, such as English, Spanish, and French, were considered duplicate records. In 2003, OCLC policy changed to allow for parallel records within WorldCat by language of cataloging. OCLC no longer considers these records duplicates, but considers them parallel records.

Identifying the language of cataloging

The language of cataloging is characterized by the use of a single language in cataloger supplied elements. Field 040 subfield ǂb indicates the language of the cataloging used to create the record. Subfield ǂb is mandatory and should be explicitly coded.

The language is indicated by a three-letter language code. For a list of these codes, see the MARC Code List for Languages .

The language of descriptive cataloging, recorded in field 040 subfield ǂb, is not to be confused with the language of the item, which is recorded in the fixed field element Lang of the record.

Do not enter codes mul (multiple languages), sgn (sign languages), und (undetermined), or zxx (no linguistic content) in field 040 subfield ǂb.

Example

Some of the more common fields useful for identifying the language of cataloging are:

Note: The language of the subject access points is not a factor when determining the language of cataloging.

A record that contains descriptive cataloging data in more than one language is not standard practice. More specifically, a record where the fields, noted in the table above, contain data in different languages is considered a hybrid record.

These records should be corrected to represent a single language of cataloging. See section, Best practices for editing hybrid records, below.

Basis for creating parallel records

When searching WorldCat, if the matching record is cataloged in a language other than that used by the inputting library, the library may enter a parallel record in its language of descriptive cataloging. For example, a library using Spanish as the language of cataloging that only finds a record for a resource with English as the language of cataloging may create a record with Spanish as the language of cataloging for the same resource.

Using existing records

When using an existing record for copy cataloging, do not change the language of cataloging when upgrading the WorldCat record unless the language of cataloging coded in field 040 subfield ǂb does not accurately reflect the actual language of the cataloging record.

Example of records, representing the same resource, appropriately cataloged in English and Spanish.
Elements in bold identify the language of cataloging in the relevant fields.

Example of records, representing the same resource, appropriately cataloged in French and English.
Elements in bold identify the language of cataloging in the relevant fields.

Example of records, representing the same resource, appropriately cataloged in Chinese and English.
Elements in bold identify the language of cataloging in the relevant fields.

Best practices for editing hybrid records

Take these factors into account when considering a change in coding for the language of cataloging in records with more than one language of descriptive cataloging:

• The intent of the cataloging agency
• The language of cataloging of holding libraries
• The number of descriptive cataloging elements in different languages. As a general rule, a record's language of cataloging should reflect the predominant language used in the descriptive cataloging elements.

Based on these factors, if you determine that the language of the record matches the language used in your institution, then correct the errors and use the record for cataloging.

If you determine that the language of the record does not match the language used in your institution because no record currently exists for the resource in your language, then create a new record. If you are unsure, send record to WorldCat Metadata Quality for staff to review.

However, if you suspect that the record was incorrectly changed from one cataloging language into another then contact WorldCat Metadata Quality. The WorldCat Metadata Quality staff will analyze the data and decide whether to change the record back to its original language of cataloging.

Examples of hybrid records

These examples demonstrate common hybrid record situations. Use judgment in dealing with hybrid records that do not reflect the examples below.

Example of a hybrid record where language of cataloging coding does not match the physical description.

This record does not have subfield ǂb in field 040, so the language of cataloging is assumed to be English. However, field 100 subfield ǂe, field 300, and field 504 are in Spanish. This is a hybrid record. The contributing library catalogs most of its material in Spanish. All fields, except for field 040 subfield ǂb indicate that the language of cataloging is Spanish.

If the language of cataloging used at your library is Spanish, add subfield ǂb coded spa to field 040.

If the language of cataloging used at your library is not Spanish, search for a record in your language of cataloging. If no record is found, create a new record, as appropriate. Report the hybrid record to WorldCat Metadata Quality for correction.

Example of a hybrid record where language of cataloging coding and physical description do not match notes.

This record has the language of cataloging in field 040 subfield ǂb as dan, field 020 subfield ǂq has indb. for indbundet, field 300 has s. for sider, but it also includes field 504 in English, which makes the record a hybrid record. The contributing institution catalogs most of its material in Danish. Only field 504 is inconsistent with the language of cataloging.

If the language of cataloging used at your library is Danish, then change the note in field 504 to Danish: Indeholder bibliografiske referencer og indeks.

If the language of cataloging used at your library is not Danish, search for a record in your language of cataloging. If no record is found, create a new record, as appropriate. Report the hybrid record to WorldCat Metadata Quality for correction.

Example of a hybrid record where language of cataloging coding and fields 33x do not match the extent and notes.

This record either became a hybrid record as a result of editing or was entered originally as a hybrid record. Field 040 subfield ǂb is coded fre along with the 33x fields, the extent could be either French or English, and the unquoted note is in English.

If the language of cataloging used at your library is French, you may convert this record into a French language record by editing field 504 to French: Comprend des références bibliographiques (pages 459-516) et un index.

If the language of cataloging used at your library is not French, search for a record in your language of cataloging. If no record is found, create a new record, as appropriate. Report the hybrid record to WorldCat Metadata Quality for correction.

Example of a hybrid record where language of cataloging coding, the 33x fields, and second unquoted note do not match extent and first unquoted note.

In this example, fields 300 and 500 are both English language of cataloging, while field 040 subfield ǂb, the 33x fields, and field 504 are in Dutch. While the inputting library's intent was to input a Dutch language of cataloging record, most of the libraries who have holdings attached are English language libraries.

If the language of cataloging used at your library is English, change field 040 subfield ǂb from dut to eng and convert the 504 field to an English language field along with the 33x fields.

If the language of cataloging used at your library is not English, search for a record in your language of cataloging. If no record is found, create a new record, as appropriate. Report the hybrid record to WorldCat Metadata Quality for correction.

Example of record where language of cataloging coding, extent, 33x fields, and first unquoted note do not match the second unquoted note.

In this example, field 040 subfield ǂb should remain ger because German language is used in field 300, field 500, and the relationship designator in field 700, while field 504 is the only note in English.

If the language of cataloging used at your library is German, change field 504 to the German: Enthält Literaturverzeichnis und Index.

If the language of cataloging used at your library is not German, search for a record in your language of cataloging. If no record is found, create a new record, as appropriate. Report the hybrid record to WorldCat Metadata Quality for correction.

Reporting hybrid records

If you cannot correct the WorldCat record, you can edit the record locally and also report it so that WorldCat Metadata Quality staff can edit the WorldCat record for all OCLC member libraries. For information about reporting methods, see section 5.5, Requesting Changes to Records.

Bibliographic records in WorldCat were limited to use of Latin characters in the MARC-8 character set until 1986 when Chinese, Japanese, and Korean scripts were implemented. That was followed by the introduction of other scripts, i.e., Arabic, Armenian, Bengali, Cyrillic, Devanagari, Ethiopic, Greek, Hebrew, Syriac, Tamil, and Thai. OCLC implemented use of all UTF-8 Unicode defined characters in 2016.

In some cases, Unicode offers more than one character or method to express the same text. This flexibility, while useful in ordinary written prose and published texts, often results in unnecessary complexity in the creation and maintenance of bibliographic data. Some Unicode characters, such as control characters, have no functional use in bibliographic records. Many classes of characters, e.g., dingbats, emoticons, game symbols, etc., are unlikely to be needed in bibliographic data.

Some characters have been routinely identified as problematic for users of cataloging copy found in WorldCat. When copying and pasting data from internet resources for inclusion into bibliographic records, be aware that some non-preferred or problematic characters may be included in the copied text. Non-MARC-8 characters are output as numeric character references in MARC-8 records.

With exceptions noted below, you may enter whatever characters you need to appropriately transcribe and record bibliographic data. If you are creating or authenticating PCC records, follow PCC character set guidelines.

LC-PCC PS 1.4 provides information on characters that cannot be reproduced by available facilities when creating bibliographic data. LC-PCC PS 1.7.5 provides similar information on symbols that cannot be readily reproduced. You may disregard instructions in both policy statements in cases where needed characters are available in the Unicode character set. Optionally, you may follow provisions of both policy statements.

If a non-PCC record has been created following LC-PCC PS 1.4 or LC-PCC PS 1.7.5, you may update the substitute characters or words to their corresponding Unicode counterparts and replace the WorldCat record if your authorization allows you to do so. Otherwise, you may edit the record for local use as needed.

If a non-PCC record has been created using Unicode characters in place of the substitute characters or words as outlined in the policy statements, do not update the record to follow provisions of LC-PCC PS 1.4 and LC-PCC PS 1.7.5 unless the record is being authenticated for PCC. You may edit the record for local use as needed.

Example of a record with a title that contains substitute Latin characters, indicated by use of the double underscore, created in accord with LC-PCC PS 1.4:

Example of the same record with a title transcribed with the characters as they appear on the resource:

Example of a record with a title that contains two Greek characters created in accord with LC-PCC PS 1.4:

Example of the same record with a title that contains two Greek characters transcribed as they appear along with a linked Latin version that may facilitate searching:

For information on coding and display of non-Latin scripts, see field 066 and field 880. For information on specific Unicode characters, see Unicode Character Code Charts .

For consistency and predictability of bibliographic data, use the following table listing problematic characters with instructions on use, options, and preferred alternative characters.

For information on problematic characters in Uniform Resource Identifiers in subfield ǂu, see Control Subfields.

Angle brackets

In bibliographic records, temporary or uncertain data are enclosed in angle brackets (< >) (e.g., 300  <1-5> volumes ; ǂc 28 cm). When creating bibliographic records do not transcribe angle brackets containing text that can be interpreted as executable computer code. Replace such angle brackets with parentheses and add an explanatory note:

Some bibliographic records in WorldCat include punctuation separating elements of the description while other bibliographic records omit punctuation separating elements of the description. When included, that separating punctuation is often in the form prescribed by the International Standard Bibliographic Description (ISBD), but in other cases separating punctuation can be pre-ISBD.

Records with separating punctuation may need some punctuation suppressed in local system displays intended for the catalog user, while records without separating punctuation may need some punctuation supplied in local system displays intended for the catalog user.

Punctuation is not included in most formats for encoding bibliographic information outside of MARC 21, e.g., BIBFRAME, Dublin Core, ONIX, UNIMARC, etc. The use of punctuation in MARC 21 records is often prone to error, and it presents complexities in mapping data between MARC 21 and other formats. Because such mappings are likely to become routine in the future as bibliographic data is created and maintained in a variety of formats, a transition to a MARC 21 environment where most records omit punctuation would be beneficial.

The following examples for the same resource illustrate different cataloging and punctuation practices.

Example of pre-ISBD record with punctuation (including optional terminal periods):

Desc
040			XXX ǂb eng ǂc XXX
100	1		Mitchell, Margaret, ǂd 1900-1949.
245	1	0	Gone with the wind, ǂc by Margaret Mitchell.
260			New York, ǂb The Macmillan Company, ǂc 1937.
300			1087 p. ǂc 22 cm.

Example of pre-ISBD record with punctuation omitted except in the access point:

Example of AACR2 record with punctuation (including optional terminal periods):

Example of AACR2 record with punctuation omitted except in the access point:

Example of RDA record with punctuation (including optional terminal periods):

Example of RDA record with punctuation omitted except in the access point:

You may create records with or without punctuation separating elements of the description which coincide with subfield coding. If you are creating or authenticating PCC records, follow PCC punctuation guidelines.

Records with full punctuation

Records with full punctuation include all punctuation separating elements of the description. Use the following guidelines when creating records with full punctuation:

• Supply punctuation between elements within a variable field per cataloging instructions, i.e., ISBD, RDA, AACR2, etc.
• Supply punctuation at the end of the preceding subfield in cases where punctuation separates data that occur in separate subfields
• Supply punctuation within the text of any subfield as needed per cataloging instructions used to create the record
• Supply punctuation within authorized access points per the practice used in the related authority file
• Do not supply brackets around preferred titles in access points
• Do not supply parentheses around series statements
• Do not supply a space-dash-space that occurs between variable fields, e.g., no space-dash-space at the end of field 245 preceding fields 250, 260, 264, etc.
• Optionally, supply terminal periods following the displayed text in those fields where terminal periods are allowed (Note: Currently, the control heading process adds terminal periods to access points)
  
• Code Desc with blank, a, or i for records with full punctuation whether terminal periods have been included or excluded

Records with no or minimal punctuation

Records with no punctuation exclude all punctuation separating elements of the description. Records with minimal punctuation generally exclude all punctuation separating elements of the description with the possible exception of authorized access points and/or certain problematic fields or subfields. Use the following guidelines when creating records with no or minimal punctuation:

• Omit punctuation at the end of the preceding subfield in cases where punctuation separates data that occur in separate subfields (exceptions are noted on pages for individual fields)
  
• Omit brackets around preferred titles in access points
• Omit parentheses around qualifier subfields in fields other than access points, e.g., field 020 subfield ǂq
• Omit parentheses around geographic coordinates
  
• Omit parentheses around series statements
• Omit a space-dash-space that occurs between variable fields, e.g., no space-dash-space at the end of field 245 preceding fields 250, 260, 264, etc.
• Omit terminal periods following the displayed text unless the field ends with an abbreviation or initial (Note: Currently, the control heading process adds terminal periods to access points)
  
• Supply punctuation within the text of any subfield as needed per cataloging instructions used to create the record
• Supply punctuation within authorized access points per the practice used in the related authority file
• Code Desc with c or n for records with no punctuation or minimal punctuation

Note: If the prescribed punctuation requires a space that would occur ahead of a subfield delimiter (e.g., space-colon-space), enter one space only rather than one in addition to the system-supplied space preceding all subfield delimiters.

Adding or removing punctuation

If you find an existing record that omits punctuation, do not add punctuation and replace the WorldCat record at this time.

If you find an existing record that includes punctuation, do not remove the punctuation and replace the WorldCat record at this time.

In either case, you may edit the record for local use as needed. When performing such a local edit, you should update Desc and field 040 subfield ǂe to reflect the inclusion or omission of punctuation.

Brackets

Practices related to the inclusion of brackets within a bibliographic record have varied over time. Older cataloging instructions would have catalogers record multiple adjacent elements to be bracketed within a single set of brackets. In the MARC record, a left bracket could occur in field 245 with the corresponding right bracket appearing in field 260. Current ISBD practice requires use of separate sets of brackets for different elements within a single ISBD area. You may change existing records to reflect current ISBD practice regarding use of brackets and replace the WorldCat record if your authorization allows you to do so.

In 2000, OCLC first experimented with controlling bibliographic access points to authority records in the former Cooperative Online Resource Catalog (CORC). Later that same year, controlling was introduced in Connexion. Since then, OCLC has continued to add new authority files. OCLC's controlling functionality is a type of linked data that makes an association between bibliographic and authority records and supports the mechanism to automatically update authorized access points when changes are made to the associated authority records. Recognizing the desire of catalogers for increased accuracy in name and subject access points, OCLC has since created additional projects to control more access points in WorldCat.

The controlling functionality currently available in Connexion and Record Manager enables users to link authority records to authorized access points in WorldCat bibliographic records. Cataloger-initiated controlling functions facilitate automatic updating of access points in bibliographic records. Connexion allows controlling to Library of Congress authority records from the LC/NACO Name Authority File (LCNAF), Library of Congress Subject Headings (LCSH), and Library of Congress Genre-Form Terms (LCGFT). Record Manager allows controlling to those files, as well as Art & Architecture Thesaurus (AAT), Biblioteca Nacional de España (BNE) Authority File, Canadiana authorities (Canadiana: Autorités de noms en français and Canadian Subject Headings), Faceted Application of Subject Terminology (FAST), Integrated Authority File (GND), Medical Subject Headings (MeSH), Māori Subject Headings, Dutch Names Authority file (NTA Names), and Répertoire de vedettes-matière (RVM).

The system normalizes the authorized access points in bibliographic records, matches them with fields 1xx or 4xx in authority records, inserts the authorized form of the heading, and creates a link in the bibliographic record. Controlling does not necessarily correct coding or typographical errors. The fact that an access point can be controlled does not mean that it has been constructed correctly nor does it mean that it is appropriate for a given resource.

In Connexion, controlled Library of Congress (LC) name and subject access points display as hyperlinks. Controlled access points from thesauri other than LC display with a subfield ǂ0 containing the authority control number associated with that authorized access point.

In Record Manager, controlled access points from each controllable thesaurus display as hyperlinks.

Do not edit controlled access points containing subfield ǂ0 in Connexion. These access points are controlled to a thesaurus that cannot be controlled in Connexion. If a controlled access point is inappropriate to the described resource and you have sufficient permissions, you may delete the entire field using Connexion.

For additional information about controlling headings in Connexion client, see Summary of authority control functions. For additional information about controlling headings in Record Manager, see Control headings. For information about applying authority data from each individual supported authority file to a bibliographic record in Record Manager, see Work with authority records.

In 2012, OCLC developed an offline batch controlling service that runs independently. The software evaluates and controls access points on different schedules for each the following categories of bibliographic records in WorldCat:

• Records newly added to WorldCat
• Records changed and replaced in WorldCat
• Existing records in WorldCat

In other words, all WorldCat bibliographic records are considered for controlling, but there are records with access points that cannot be controlled to one of the existing authority files.

The software:

• Automatically expands dates for candidate personal name access points from LCNAF and LCSH only
• Partially controls fields and automatically appends subfields that are not controlled
• Maintains the links between non-Latin script access points and their Latin-script equivalents

As with the Control All Headings function in Connexion and Record Manager, the offline controlling software does not:

• Attempt to control unqualified personal names (personal names that do not include any subfields ǂb, ǂc, ǂd, and ǂq)
• Control access points to an undifferentiated personal name authority record (008/32 or Name (Undifferentiated personal name) is coded b)
• Cause automated change of field tag of any type
• Control access points to an LCSH authority record containing field 667 with text Record generated for validation purposes
• Control an LC geographic subdivision to an LC authority record that does not contain field 781
• Control access points that have no exact match
• Control access points that match only a portion of a subfield
• Control access points that contain coding errors, typographical errors, or other errors that require manual correction
• Control access points that contain square brackets, with the exception of access points controllable to BNE
• Control access points to a reference authority record (008/09 or Auth/ref (Kind of record) is coded b, c, or g)
• Control names to authority records coded as not valid for name use (008/14 or Name use (Heading use-main or added entry) is coded b)
• Control subjects to authority records coded as not valid for subject use (008/15 or Subj use (Heading use-subject added entry) is coded b)
• Control series to authority records coded as not valid for series use (008/16 or Ser use (Heading use-series added entry) is coded b)

The software also makes the following adjustments as needed:

• Applies any appropriate punctuation changes
• Attempts to fully control previously controlled access points that were only partially controlled
• Uncontrols descriptive access points in fields 1xx, 7xx, and 8xx, unless those descriptive access point fields are controlled to a thesaurus appropriate to the language given in the subfield ǂb of field 040
• Deletes a geographic subdivision from a subject access point if the access point can be completely controlled and no element of the access point can be subdivided geographically (008/06 or Geo subd (Direct or indirect geographic subdivision) is not coded d or i)

Symbol OCLCO is appended in field 040 subfield ǂd in bibliographic records that are modified by the software.

If a bibliographic record is locked, the software skips and reevaluates it when the lock has been removed.

For more information about the offline batch controlling service, see How does OCLC control headings in WorldCat?

You are encouraged to continue using the Control Headings functions in Connexion and Record Manager when adding or upgrading records as you share your cataloging expertise with the cooperative.

When an authority record is updated with changes that impact controlling, a controlled heading update service updates any bibliographic records with authorized access points linked to that authority record. Such changes include:

• Any change to field 1xx
• Certain changes to the following elements:
  • 008/06 or Geo subd (Direct or indirect geographic subdivision)
  • 008/09 or Auth/ref (Kind of record)
  • 008/12 or Series (Type of series)
  • 008/13 or Ser num (Numbered or unnumbered series)
  • 008/14 or Name use (Heading use-main or added entry)
  • 008/15 or Subj use (Heading use-subject added entry)
  • 008/16 or Ser use (Heading use-series added entry)
  • 008/32 or Name (Undifferentiated personal name)
• Addition of field 667 with text Record generated for validation purposes in LC records
• Deletion of field 781

Symbol OCLCA is appended in field 040 subfield ǂd in bibliographic records that are modified by the controlled heading update service.

OCLC makes use of the fields 7xx in Répertoire de vedettes-matière (RVM) authority records to add equivalent subject access points to and from Library of Congress authority records (LC/NACO Name Authority File (LCNAF), Library of Congress Subject Headings (LCSH), and Library of Congress Genre-Form Terms (LCGFT)), Medical Subject Headings (MeSH), and Art & Architecture Thesaurus (AAT). Equivalencies may be added only if there is a one-to-one relationship between the RVM heading and an equivalent thesaurus.

If equivalent subject access points are found, they are added and controlled to their respective authority file.

• If a bibliographic record contains fully controlled RVM access points, the system searches for equivalent access points found in AAT, LCNAF, LCSH, LCGFT, or MeSH
• If a bibliographic record contains fully controlled AAT, LCNAF, LCSH, LCGFT, or MeSH access points, the system searches for equivalent access points found in RVM

Equivalent subject access points may be added using the Add Equivalent Subject Headings command in Record Manager or via the offline controlling process.

Based on RVM 201-0000897:

If this is the subject access point in the bibliographic record:

The following subject access points will be present in the bibliographic record after adding equivalencies:

For additional information about adding equivalent subject access points in Record Manager, see Add equivalent subject headings.

After the decision was made to proceed with RDA implementation, PCC investigated whether it would be possible to apply any of the new conventions that RDA would introduce to non-RDA records. There was also concern about whether the implementation of RDA would require significant recataloging.

It was determined that the legacy databases that PCC member libraries create and maintain already contain many records created under different cataloging practices. For example, AACR2 practices have been applied to pre-AACR2 records.

It was decided that adding or editing data elements in legacy records that reflect RDA conventions would cause no harm, and could result in positive changes, making the bibliographic records easier for end users to read and interpret.

OCLC has updated records in WorldCat to incorporate RDA elements such as adding 33x fields, spelling out non-transcribed abbreviations, converting Latin abbreviations to English equivalents, converting dissertation notes in field 502 to multiple subfields, and removing the general material designation (GMD).

Do not remove valid AACR2 elements or valid RDA elements when enhancing records. Do not recode the descriptive rules for minor differences.

Retain elements that are correct, even if you would not normally supply them yourself. If it's not wrong, leave it alone.

Avoid "editing wars" that are merely stylistic, such as style of note.

Ask first if you are not sure. Requests for advice on cataloging policies, standards, and practices should be sent to askqc@oclc.org.

Libraries may add new unique records to the WorldCat database using any cataloging code that they wish. Libraries may move to using RDA on their own timetable. Libraries may upgrade AACR2 records to RDA but may not change RDA records to AACR2 or add new AACR2 records if a matching RDA record already exists.