Section D - Guidance for Encoding Metadata

Target audience

Metadata authors who are encoding metadata into HTML or XML, and developers.

Who needs this information?

This section summarises the DC approach to syntax to provide guidance for people who are already proficient in using HTML or XML. It is not necessary for metadata authors who are solely using Metalogue or are storing the metadata in other systems not using the provisions of HTML or XML.

Syntax in NZGLS follows the standards set for Dublin Core. These Dublin Core guidelines follow international standards such as those from W3C. References to the key documents are given below.

A complication is that case is sometimes used in HTML, but XML (which appears to be the emerging standard) is case sensitive and the guidance is to use only lower case. But, so long as you consistently follow current conventions regarding the use of case in HTML documents then the conversion should not be too difficult.

When encoding directly into HTML/XML:

• The format for element refinements is: element name, full stop, element refinement.
• Reference DC for all DC terms, otherwise reference NZGLS.
• Elements can be repeated when there are multiple values that are important for resource discovery.
• Element refinements, including component labels, consisting of more than one word are concatenated as a text string, with no initial capital, but subsequent words in the string being capitalised.
• Where a controlled vocabulary/encoding scheme is used to select the element value, the scheme name should be included in the HTML metatag for the element, e.g.
  <meta name="abc" scheme="def" content="xyz">
• The HTML <title> tag and the NZGLS metatag <DC.title> should have the same content
• NZGLS is an evolving standard, so it makes sense for agencies to record which version of the standard their metadata complies with. This can be shown by use of a validation statement (in the Head portion of an HTML document) e.g. <link rel="schema.nzgls" href="http://nzgls.govt.nz/.....">
• The format that qualifiers are written in is called "lower camel case". Single word qualifiers are written all in lower case. Qualifiers consisting of more than one word have no spaces in them between words, but have the initial letter of the second and subsequent words in uppercase. For example, the qualifier "act" is all in lower case. The qualifier "isBasedOn" has a lower case initial letter but uppercase letters for subsequent words.

References

Using Dublin Core http://dublincore.org/documents/usageguide/

Encoding Guidelines http://dublincore.org/resources/expressions/

Encoding Dublin Core in HTML (IETF RFC 2731) http://www.ietf.org/rfc/rfc2731.txt

Expressing Simple Dublin Core in RDF/XML http://dublincore.org/documents/dcmes-xml/

Guidelines for implementing Dublin Core in XML http://dublincore.org/documents/dc-xml-guidelines/

Expressing Qualified Dublin Core in RDF/XML http://dublincore.org/documents/dcq-rdf-xml/

Expressing Qualified Dublin Core in HTML/XHTML meta elements http://dublincore.org/documents/dcq-html/