| THE INTERNET | |
|
Mapping Metadata: Crosswalks |
Kevin Butterfield Souther Illinois University kbutterf@siu.edu |
In early October, the Network Development and MARC Standards Office of the Library of Congress released the latest version of the Dublin Core/MARC/GILS crosswalk. As the number, size and complexity of metadata standards continues to grow, supplying metadata for each standard becomes repetitious, time consuming and tedious. In order to minimize the amount of time needed to create and maintain metadata and to maximize usefulness to the widest community of users, there is a need for the metadata created and maintained in one standard to be accessible via related content standards. As methods of mapping one metadata standard to another, crosswalks provide the ability to make contents of elements defined in one standard available to communities using related standards. In this case, the crosswalk would enable users of MARC to translate between it and the Dublin Core or GILS and vice versa.
Examples of Dublin Core/MARC crosswalks
The fifteen elements of the Dublin Core constitute a set of unqualified descriptive metadata. All elements are optional and repeatable. When compared with the large number of elements available in MARC, they seem inadequate. The creators of the Dublin Core have attempted to address this by allowing the elements to be qualified. This narrows the focus of the element and allows for a more specific mapping between standards.
An example of a mapping between MARC and the Dublin Core is the Dublin Core tag Contributor.
The Dublin Core defines Contributor as an entity responsible for contributing to the content of the resource. According to the crosswalk released on October 14, this maps to the following MARC tags.
Unqualified:
720 |a (Added entry—Uncontrolled Name/Name) with |e = contributor (or other term used as value of role qualifier)
Qualified:
Contributor.Personal:
700 1# |a (Added Entry—Personal Name) with |e = collaborator
Contributor.Corporate:
710 2# |a (Added Entry—Corporate Name) with |e = collaborator
Contributor.Conference:
711 2# |a (Added Entry—Conference Name) with |e = collaborator
Contributor.Role:
720 ## |a (Added Entry—Uncontrolled Name/Relator term)
GILS uses the term Contributor.
Another example of a mapping between MARC and the Dublin Core is the Dublin Core tag Title.
The Dublin Core defines Title as the name given to a resource. According to the crosswalk released October 14, this maps to the following MARC tags.
Unqualified:
245 00 |a (Title Statement/Title Proper)
If repeated, all titles after the first: 246 33 |a (Varying Form of Title/Title Proper)
Qualified:
Title.Alternative:
246 33 |a (Varying Form of Title/Title Proper)
Title.Release:
250 ## |a (Edition Statement)
Title.Translated:
242 00 |a (Translation of Title/Title)
Title.Uniform:
130 0# |a (Main Entry—Uniform Title)
GILS uses the term Title.
The crosswalk document lists two uses for mapping the Dublin Core to MARC, but many others could be determined. The enhancement of a simple resource description record is one such item listed. A library may wish to extract the metadata found in a Dublin Core record, convert the data elements to MARC fields, and use the result as a skeletal cataloging record that could then be enhanced as needed. The mappings can also support searching across syntaxes and databases. This would enable users to access data whether it appears in a MARC catalog record, the Dublin Core or a TEI header.
Issues with SGML
Mapping elements between the Text Encoding Initiative (TEI) and Encoded Archival Description (EAD) header systems introduces another item into the mix. Both TEI and EAD can be expressed in Standardized General Markup Language (SGML). In order to map items between systems effectively, the Network Development and MARC Standards Office at the Library of Congress developed the MARC DTD. A document type definition (DTD) defines the markup rules for a given class of documents. It provides a grammar, so to speak, of what tags can be used where and in what combinations and establishes consistency within a class of documents. The primary purpose of the MARC DTD project was to create standard SGML DTDs to support the conversion of cataloging data from MARC to SGML (and back) without loss of data. The driving force behind this project was the desire for a non-proprietary conversion by machine between MARC encoded data and SGML. By agreeing on one set of DTDs for MARC, this conversion becomes much simpler.
Utilities able to handle these conversions automatically are currently available. They and the MARC DTDs are available from the Network Development and MARC Standards Office homepage. These utilities work quite well for small batches of records and can be very useful in insuring consistency between the catalog record and headers for electronic text collections. One of the key things to remember when doing these conversions is that while the elements may be mapped to their equivalents in MARC, the content of the fields may not be represented in a form equivalent to AACR2. The records will then need to be modified to conform. Libraries creating texts using the TEI may wish to consider employing AACR2 when formulating data for the headers. This would ensure a smoother transition. Catalogers could create MARC records for the texts and the headers could be created automatically. This would eliminate duplication of effort.
MARC DTDs
MARC Mappings/Crosswalks
The Text Encoding Initiative |
Studying these crosswalks is a great way to learn more about metadata systems since they relate them to MARC, a more familiar standard. It can also bring to light issues of terminology, for instance, how many different terms can be used to represent the concept of author. Because these standards develop independently of each other, they are often specified differently using specialized terminology, methods and processes. There exists in each of these standards a core set of elements (author, title, publishers, rights, etc.) that, while expressed with different terminology, still represent the same principles. Crosswalks and common DTDs are an effort to address this problem.