Linking IEEE LOM Metadata with HTML Documents
Document Information
Status: Draft
Version: V1.00.20050102
Revision Date: 2005-01-02
Type: Recommended Practice
CORDRA ID:
H
[2000.01/70084F7903D34169BF475033C518F84B]
Metadata ID:
H
[2000.01/AD745C34DA8B4940A00B6FB26D60B32C]
Abstract
This document describes an approach to associate sets of IEEE LOM Metadata with an HTML document using the HTML <link> element.
The IEEE Standard for Learning Technology -- Extensible Markup Language (XML) Schema Definition Language Binding for Learning Object Metadata describes how to produce a machine-processible instance of IEEE LOM Metadata. The general.identifier item in the LOM Base Schema provides the mechanism to associate the content object with the metadata instance. However, the LOM model does not provide the inverse mechanism to associate a metadata instance with a content object. While structured data representations, such as IMS Content Packaging, provide a way to create structured collections of content and their associated metadata instances, there is no simple mechanism to link an HTML document to an associated LOM XML instance. This document describes an approach using the HTML <link> element to associate IEEE LOM XML instances with HTML documents, and also outlines how a harvester or robot may use the element to harvest LOM instances from HTML documents.
As RECOMMENDED practice, these HTML (and XHTML) document conventions MAY be used to associate IEEE LOM XML instances with any HTML or XHTML document.
These XHTML document conventions SHALL apply to all CORDRA documents, i.e., all documents that define the CORDRA System, the CORDRA System instance or implementation of the CORDRA system, or a definition, instance or implementation of Federated CORDRA.
The HTML <link> element "conveys relationship information", i.e., how the document specified in the link element is related to the document containing the link element. The link element may also be used to provide information about the document to search engines.
For example, the link element is used to define the relationship between a document and its stylesheet, an RSS feed, or a FOAF (Friend of a Friend) element. Three examples of link elements are shown below, one for a stylesheet, one for RSS and one for FOAF relationships.
<link rel="alternate" type="application/rss+xml" title="RSS" href="rss.xml" />
<link rel="stylesheet" type="text/css" href="stylesheet.css" />
<link rel="meta" type="application/rdf+xml" title="FOAF" href="foaf.rdf" />
This same approach may be used to associate an IEEE LOM XML instance with HTML documents. An example of such a link element is shown below.
<link rel="ieeelommetadata" type="application/ieeelom+xml" title="LOM Metadata" href="lom.xml" />
A user agent (web browser, harvest agent, web spider) may use the link element to obtain the associated LOM metadata from the document containing the element. This Autodiscovery approach is used to find RSS feeds and build FOAF networks.
The remainder of this document specifies the constraints on the attributes of the link element and specifies how a user agent should process the element when harvesting and indexing metadata.
Specifying Metadata Relationships with the link Element
The link element SHALL conform to the latest W3C HTML Recommendation.
The link element SHOULD conform to the latest W3C XHTML Recommendation. For CORDRA Documents, the element SHALL conform to the XHTML Recommendation.
The link element SHALL only appear in the head section of an HTML document.
Multiple link elements MAY be used. Each link specifies a unique relationship between the HTML document and the associated LOM XML instance.
The link element SHALL include the following four attributes:
Any other attribute of the link element may be present but conveys no special meaning in defining the relationship between the document and the LOM XML instance.
Use of the type attribute is consistent with the approach used for RSS and FOAF relationships. The MIME type follows that described in RFC 3023.
Harvesting and Indexing Metadata
A harvesting or indexing user agent MAY use the rel and type attributes to identify that there is LOM Metadata associated with the document.
The user agent SHALL use the href attribute as the URI of the LOM XML instance.
The user agent MAY ignore the title attribute. There is currently no special meaning associated with the title element.
The user agent SHOULD obey all indexing directives specified in a robots.txt file or in meta elements specifying indexing directives (see XHTML Appendix B.4). The directives applied to the document SHALL apply to the related LOM metadata document specified by the href attribute in the link element. For example, if the document is not indexed, the metadata should not be indexed.
A CORDRA-conformant user agent SHALL accept and process an href attribute where the value of the URI is a handle encoded in the URI (see H Encoding CORDRA Identifiers in URI Syntax). The URI-encoded Handle SHALL be resolved to obtain the LOM XML Instance.
| Version | ID | Date | Change Summary |
|---|---|---|---|
| 1.00 | H | 20050102 | Initial release |