Bib info

This is an ongoing discussion of bibliographic information to recommend every bundle to include for itself and its subelements.

Elements

Collection information

for a single collection.

The same as for any other resource; but the description and title should indicate that it is a collection; and the RELATION field should indicate that it has subelements. Open question: how to link to subelements (presumably by Resource ID).

Resource information

for a single work of code, writing, music, design, &c

RESOURCE-ID     An ID unique to this resource; combined with version ID for a unique identifier for this resource-version.
VERSION         Version ID
VERSION-HISTORY Sequence of Version IDs

CREATOR         Person(s) who created this work. 
CREATION-DATE   YYYY-MM-DD, YYYY-MM, or YYYY   
LAST-MODIFIER   Person who contributed this version should change every time a new individual revises the item         
LAST-MODIFIED-DATE  YYYY-MM-DD, YYYY-MM, or YYYY 
PUBLISHER       Organization that put together/published this item, i.e. University of Bombay, OXFAM, Doctors without Borders 
REPOSITORY-URL  Optional URL to an original or mirror  
REPOSITORY-ID   Optional repository ID, with a description (such as a Course ID; often redundant with part of repository-url)
SOURCES*        Other resources from which this resource is significantly derived.  Should include full resource IDs where posible.
REFERENCES      A bibliography of references and sources for quotes, notes, and information.
RELATION        Plain text description of how this item may relate to other resources or collections  

TITLE           Plain text title of the resource
SUBJECT         Similar to the Category of a bundle info file. 
DESCRIPTION     Plain text description of the item. May include an abstract, manifest of subelements, table of contents, etc.   
COVERAGE        Period of time, geospatial dimensions, intended age group, etc.  

TYPE            ... basically a MIME Type 
FORMAT          Format information specific to the type, such as 400 X 500 pixels
LANGUAGE        ISO 639-1 codes
RIGHTS          URL to license http://creativecommons.org/licenses/by/3.0/ -- Default 
                NB: this is partly redundant with, but more granular than, the Acknowledgements file for each bundle/collection. 

Notes:

• If you translate, revise, or modify someone else's work, you should always update the version, version-history, and last-modifier fields. If the change is dramatic enough, you may want to change the Creator and Creation date, and add the original to the list of Sources.
• COPYRIGHT_TYPE can be designated by a URL to the appropriate license; but should also link to a local description of the license where possible (and especially where required).

• http://creativecommons.org/licenses/by/3.0/ -- Default
• http://creativecommons.org/licenses/by-nc/3.0/
• http://creativecommons.org/licenses/by-nc-sa/3.0/
• http://creativecommons.org/licenses/by-sa/3.0/

bib_info File Format

<?xml version="1.0"?>
<dc xmlns:dc="http://dublincore.org/schemas/xmls/simpledc20021212.xsd">
<dc:title>Coliseum in Rome</dc:title>
<dc:creator>Thornton Staples</dc:creator>
<dc:subject>images</dc:subject>
<dc:description>Image of Coliseum in Rome</dc:description>
<dc:publisher>University of Virginia Library</dc:publisher>
<dc:contributor>Bryan Berry</dc:contributor>
<dc:date>2008-01-30</dc:date>
<dc:type>image</dc:type>
<dc:format>image/jpeg</dc:format>
<dc:identifier>ISBN:0385424728</dc:identifier>
<dc:source>http://www.virginia.edu/pictures/</dc:source>
<dc:language>en</dc:language>
<dc:relation>University of Virginia Antiquities Collection</dc:relation>
<dc:coverage>50-100 AD</dc:coverage>
<dc:rights>http://creativecommons.org/licenses/by/3.0/</dc:rights>
</dc>

XML format implementation

Here is a sample bib_info.xml file

Why use XML? The point of bib_info.xml is to make it easy for large-scale repositories like DSpace, Fedora Repository Server, GLOBE, and ARIADNE to serve up .xo bundles. For this reason we chose a simple standard -- Dublin Core encoded in XML -- that plays well with others rather than create our own.

External links

Here are some good links on Metadata and Controlled Vocabularies

• Wikipedia Controlled Vocabularies
• What is a Controlled Vocabulary?
• Cory Doctorow's MetaCrap
• National Cancer Institute Enterprise Vocabulary Services

Basically, a Controlled Vocabulary is a set of standard values. This can be useful for the Copyright Type, language, and country; less so for the other elements.