Mapping MARCXML data to Dublin Core
1. About this document
- Author
- Corey Wallis, RUBRIC Technical Officer
- Purpose
- This technical report outlines the way in which the EndNote to Fedora data migration scripts map fields in the MARCXML data to Dublin Core metadata.
- Audience
- RUBRIC Project Partners and other users of EndNote and Fedora
- Requirements
- An understanding of the structure of MARC tags and how they're represented in MARCXML
- An understanding of the structure of the OAI compliant Dublin Core metadata
- References
- RUBRIC Technical Report: Migrating EndNote data to Fedora
- Official EndNote website
http://www.endnote.com/ - Official ARROW Project website
http://www.arrow.edu.au/ - Official MARCXML website at the Library of Congress
http://www.loc.gov/standards/marcxml/ - Official Open Archives Initiative (OAI) Dublin Core Specification
http://www.openarchives.org/OAI/openarchivesprotocol.html#dublincore - Notes
2. Background Information
In preparation for the EndNote to Fedora data migration it was necessary to map the available fields in the MARCXML data to the fields specified by the Open Archives Initiative implementation of the Dublin Core metadata standard. The map also had to take into consideration the requirements of the ARROW project.
The map outlined below was developed using the MARCXML templates provided by the ARROW project, and the “central register” a spreadsheet containing metadata guidelines for a variety of items.
The map deviates slightly from the specifications provided by the central register. Two notable deviations are that:
There is one dc.subject field created for each subject. The existing ARROW guidelines produce one dc.subject field containing all subjects
The order of information in some of the fields is different. For example the dc.publisher field has the publishers name appearing before the place of publication
These differences mean that if the Dublin Core metadata is regenerated by the VITAL manager the resulting metadata will not adhere to the map outlined in this technical report.
3. MARCXML to Dublin Core map
3.1. Data map definitions
The MARCXML to Dublin Core map contains the following columns:
- MARC Field
- The MARC field as specified by the ARROW template
- MARC Subfield
- The MARC subfield as specified by the ARROW template
- Description
- The description of the field
- OAI Dublin Core Field
- The OAI Dublin Core field
3.2. MARCXML to Dublin Core mapping
|
MARC Field |
MARC Subfield |
Description |
OAI DC |
|---|---|---|---|
|
245 |
|||
|
a |
Item Title |
DC:Title |
|
|
100 |
|||
|
a |
Author Name |
DC:Creator |
|
|
u |
Author's organisation |
Concatenation of the $a and $u subfields is used to form the value of the field |
|
|
700 |
|||
|
a |
Author Name |
DC:Creator |
|
|
u |
Author's Organisation |
Concatenation of the $a and $u subfields is used to form the value of the field |
|
|
710 |
|||
|
a |
University Name |
DC:Creator |
|
|
b |
Faculty / School / Dept |
Concatenation of the $a and all $b subfields, if any, is used to form the value of this field |
|
|
520 |
|||
|
a |
Abstract |
DC:Description |
|
|
260 |
|||
|
a |
Place of publication |
DC:Publisher |
|
|
b |
Publisher |
Concatenation of the $b and $a subfields is used to form the value of this field |
|
|
c |
Year of publication |
DC:Date |
|
|
787 |
|||
|
t |
Journal Title |
DC:Relation |
|
|
540 |
|||
|
a |
Copyright statement |
DC:Rights |
|
|
856 |
|||
|
u |
URL to publishers copy |
DC:Identifier |
|
|
653 |
|||
|
a |
Keyword |
DC:Subject |
|
|
655 |
|||
|
a |
Item type |
DC:Type |
|
|
773 |
|||
|
t |
Repository Collection Title |
DC:Relation |
|
|
591 |
|||
|
a |
DEST collection year |
DC:Description |
|
|
592 |
|||
|
a |
DEST category |
DC:Description |
|
|
711 |
|||
|
a |
Conference Title |
DC:Relation |
|
|
013 |
|||
|
a |
Primary Patent Number |
DC:Subject |




