Government of Canada Metadata Implementation Guide for Web Resources - 4th edition, October 2005

Chapter 4: Optional Elements

For department or agency Web resources, optional means that the elements described below are not required for compliance with CLF Standard 6.3. The use of some or all of these, including other Dublin Core elements or non–Dublin Core elements, may, however, be mandatory within a particular department or agency.

All element definitions are cited from DCMI Metadata Terms
(http://www.dublincore.org/documents/dcmi-terms).

4.1  Audience

4.1.1      Definition

A class of entity for whom the resource is intended or useful.

4.1.1.1      Government of Canada constraint

Terms must be selected from an authorized controlled vocabulary.

4.1.2      Usage

There are two possible ways to express this element in HTML: dc.audience and dcterms.audience.

dc.audience

The Audience element is expressed as follows:

where abc is the label for a specific scheme and def is a value selected from the scheme to describe the audience for the resource.

Note: The use of dc.audience is to be considered a temporary method of expression that will change as Dublin Core develops. The following method of expressing the audience element is preferred.

dcterms.audience

The Audience element is expressed as follows:

where abc is the label for a specific scheme and def is a value selected from the scheme to describe the audience for the resource.

The Audience element is defined by the Dublin Core Metadata Initiative in a separate namespace. Departments or agencies using dcterms.audience must include a second instance of the link element to support the HTML code. This additional link element is the following:

4.1.3      Guidance

Recommended practice: It is strongly recommended that the Audience element be applied to all resources to which departments or agencies apply other mandatory and optional metadata elements when those resources are identified as being directed toward one or more particular audiences.

The Audience element should be repeated, with the appropriate scheme label, for each audience type that applies to the resource.

For more guidance on the use of the Audience element, including the use of additional audience schemes, see the Guidelines for the Dublin Core Metadata Element Audience and the Government of Canada Audience Scheme, 2nd Version
(http://www.tbs-sct.gc.ca/im-gi/mwg-gtm/aud-aud/docs/2004/dublin-core/dublin-core00_e.asp).

4.1.4      Values

Recommended practice: It is strongly recommended that metadata developers use values from the Government of Canada Audience Scheme
(http://www.tbs-sct.gc.ca/im-gi/mwg-gtm/aud-aud/docs/2003/schemfinal/schemfinal_e.asp). This scheme includes terms that describe a wide variety of users of Government of Canada Web sites.While it includes a broad range of terms, it is not intended to be an exhaustive list of audiences.

The Government of Canada Audience Scheme is registered as a controlled vocabulary with Library and Archives Canada. The scheme label for the Government of Canada Audience Scheme is gcaudience.

Departments or agencies requiring that new audience terms be added to the vocabulary should address their enquiries to the Secretariat's Information Management Division
(im-gi@tbs-sct.gc.ca).

4.1.5      Examples

4.2  Coverage

4.2.1      Definition

The extent or scope of the content of the resource.

4.2.1.1      Government of Canada constraint

Guidance is provided for spatial or geographic coverage only. The element must be populated with terms found in registered controlled vocabularies.

4.2.2      Usage

The Coverage element is expressed as follows:

where abc is the label for an authorized list of geographic names and def is a specific location chosen from the authorized source or set of geographic co.ordinates.

4.2.3      Guidance

The Coverage element is used to spatially reference Government of Canada Web resources using descriptors from registered controlled vocabularies, authority files, or a spatial reference system like geographic co.ordinates. The Dublin Core Metadata Initiative (DCMI) endorses element refinements that state whether spatial or temporal extent is being expressed. This section provides guidance for spatial coverage only.

Coverage is used when there is a distinct and definable spatial extent to the resource and there is a business requirement to enable the identification of the resource by its unique geographic location. When Coverage is used, the spatial information contained within the element can be used to perform powerful searches that allow a user to search on a specific location, exclude certain areas, sort by nearest location, and perform many other operations.

The following examples illustrate appropriate use of the Coverage element:

Example 1: Web resources on employment opportunities for various regions that contain information organized by area.

The Coverage element will contain geographic descriptors, geocodes, or spatial co.ordinates that define the areas referenced by the resource. This geographic information could enable a user to search on employment opportunities for a specific area.

Example 2: Web resources that identify national historical sites and contain information that specifies a geographic location.

The Coverage element will contain a geographic descriptor, such as a place name, postal code or spatial co.ordinates, that defines the location of the site(s). Such information could enable a user to search for historical sites within a certain radius or along a route of travel.

The Coverage element should not be used if the resource has no distinct and definable geographic extent or if there is no business requirement to use the unique spatial reference(s) specifically or inherently contained in the resource contents. For instance, it is not recommended to use the term Canada to describe federal government on.line resources that are Canadian but not directly related to geographic boundaries within Canada.

4.2.4      Values

While each registered encoding scheme has its own method of expressing geographic names, certain general guidelines apply to the construction of these names for use in dc.coverage.spatial.

Sufficient detail must be provided to unambiguously identify the location. A place name must be expressed as a hierarchy, beginning with the largest geographical unit (usually the country name) and ending with the smallest.

Within the hierarchy of location information, geographic units should be separated by a semicolon (;). A semicolon is used as a separator rather than a comma because values in the encoding scheme may contain commas.

For example, the village of Kingston, P.E.I., would be identified using the Canadian Geographical Names Data Base as follows: Canada; Prince Edward Island; Queens; Kingston; Village.

Coverage must be populated using a controlled vocabulary, authority file or other registered scheme. Controlled vocabularies or authority files recommended by these guidelines for usage with Coverage are registered with Library and Archives Canada and can be accessed at
http://www.collectionscanada.ca/8/4/r4-281-e.html. All recommended schemas are able to be translated to unique geographic locations, expressed as spatial co.ordinates. Named geographies from known, controlled sources can be translated to simple geographic co.ordinates (a point) or more complex geometries (bounding box, polygons or lines) through Web services, such as the CGDI Gazetteer Service (http://cgdi-dev.geoconnections.org/prototypes/cgdigaz/index.htm).

Departments and agencies are strongly encouraged to register with Library and Archives Canada other standardized geographic descriptors, geocodes or resources that may better meet their particular requirements. These will be assigned a unique scheme name that must be stated in every instance of Coverage.

The most appropriate choice of vocabulary depends on the resource being documented. More than one vocabulary may be used to describe a single resource.

The Coverage element must be repeated, using the appropriate scheme name each time, for every location described in the resource.

4.2.4.1   Language equivalents in the Coverage element

The language equivalency rule applies for place names. See section 2.3.4 for further information.

4.2.5 Examples

The following are examples employing encoding schemes currently registered for use with Coverage by Library and Archives Canada or the Dublin Core Metadata Initiative.

Canadian Geographical Names Data Base (CGNDB)

http://geonames.nrcan.gc.ca/index_e.php

This database contains the official names of Canada's provinces and territories, names of cities (inhabited places, towns, villages, etc.), topographical features, conservation areas, military areas, and First Nations communities.

To be identified unambiguously, the location must be constructed with the following components, in the following order: country name (i.e. Canada); province or territory name; administrative or geographical location name (e.g. district, county, township, municipality, section town); name of feature; feature type. The CGNDB provides further information on the definition of these components and the allowable values.

Using this encoding scheme, the dc.coverage.spatial metadata for a resource containing information on an employment opportunity in Peterborough, Ontario, would read as follows:

Regions of Canada

http://geodiscover.cgdi.ca/gdp/util/etc/CanadaProvRegions.jsp

This vocabulary consists of Canada's provinces and territories, six large inter-provincial regions and 39 intra-provincial regions. The list provides names for commonly recognized areas within Canada. The boundaries are based upon census divisions or provincial and territorial boundaries and therefore have an associated geometry.

The location is constructed using the complete hierarchy given in the encoding scheme. The hierarchy is composed of the term Canada; then a province or territory name or the term Canadian Regions; then a more precise geographic location name, if applicable.

Using this encoding scheme, Coverage for a resource containing agricultural output statistics for Canadian prairie farms would read as follows:

Getty Thesaurus of Geographic Names (TGN)

http://www.getty.edu/research/conducting_research/vocabularies/tgn/index.html

The TGN can be used as a source for foreign place names. The Dublin Core Metadata Initiative prefers this vocabulary.Since the TGN contains only English terms, use another authoritative source to record international place names in French, such as a good dictionary or the Répertoire de vedettes-matière (http://www.collectionscanada.ca/8/19/index-e.html) to select equivalent French terms. Do not identify a scheme.

To be identified unambiguously, the location must be constructed using the hierarchy given in the encoding scheme, starting at the "nation" level and including all lower levels. The "hierarchy root," "facet" and "continent" levels are omitted. The TGN provides further information on the construction of location hierarchies.

Using this encoding scheme, Coverage for a resource involving plans for a war memorial in Exeter, England, would read as follows:

For the French Web page, the Coverage element would read as follows:

ISO 3166—Country Codes

http://www.iso.ch/iso/en/prods-services/iso3166ma/02iso-3166-code-lists/list-en1.html

ISO 3166 can be used as a source to identify foreign country names. This authority list is registered with the DCMI as a source for country names. Complete place names should be used in preference to abbreviations or symbols.

Using this encoding scheme, Coverage for a resource on the spread of severe acute respiratory syndrome (SARS) in Southeast Asia (China, the Philippines and Singapore) would read as follows:

DCMI Point

http://www.dublincore.org/documents/dcmi-point

DCMI Point is a scheme for uniquely identifying a point on the Earth's surface using its geographic co.ordinates. DCMI guidance provides further details on how to use this scheme.

Using this encoding scheme, Coverage for a resource on Grand Valley, Ontario, Canada, would read as follows:

DCMI Bounding Box

http://www.dublincore.org/documents/dcmi-box/

DCMI Bounding Box is a scheme for uniquely identifying an area on the Earth's surface by defining its geographic limits as a bounding box. DCMI guidance provides further details on how to use this scheme.

With this encoding scheme, Coverage for a resource on Prince Edward Island, Canada, would read as follows:

4.3  Description

4.3.1      Definition

An account of the content of the resource.

4.3.2      Usage

The Description element is expressed as follows:

where abc is the description of the resource.

4.3.3 Guidance

An accurate, clear, concise and objective description helps users determine whether a resource will be useful to them.

The words in a description can be indexed by search engines. The text is often displayed in search results. As a general rule, descriptions should be no more than 100-120 words in length. Because only the first few lines of a description may be displayed in a listing of Web resources, it is important that the most significant information be contained in the first few sentences.

Word or character length of description may be dependent on department or agency policies or search engine requirements. Some search engines will index all words in a description but only display the first portion of the text.

4.3.4 Values

4.3.4.1 Tips on preparing a description

• Use the headings, table of contents, and the introduction or summary as a guide for information about the resource.
• Try to identify the type of product or service in the description, e.g. "This report (document, CD-ROM, publication, diskette) . . . ."
• Use simple, direct language.
• Use words that clearly describe the product or service.
• Make the description understandable to a wide audience.
• Use the objective third person point of view (avoid using "you" or "we").
• Be brief but as complete as possible.
• Ensure that the description is coherent.
• Use abbreviations, acronyms, and initialisms if they are widely known (e.g. NATO, UNESCO) but define unfamiliar terms and symbols the first time they occur.
• Avoid jargon.
• Ensure that the description informs the reader about the subject of the resource and presents each distinguishing characteristic of the product or service.

4.3.4.2      HTML description and dc.description

Departments or agencies using HTML may wish to use both HTML description and dc.description as some search engines ignore dc.description. In such cases, the content values for both must be identical.

4.3.5      Other considerations

Departments or agencies are urged to develop their own organizational guidance for writing descriptions based on Guidelines for Writing Descriptions for Government of Canada Web Sites (http://www.tbs-sct.gc.ca/im-gi/references/meta-descrip/meta-descrip00_e.asp).

4.3.6      Examples

4.3.6.1      English-only Web resource

http://www.collectionscanada.ca/trains/index-e.html

4.3.6.2      French-only Web resource

http://www.collectionscanada.ca/trains/index-f.html

4.3.6.3      Bilingual resource

http://bsa.cbsc.org

4.4  Format

4.4.1      Definition

The physical or digital manifestation of the resource.

4.4.1.1      Government of Canada constraint

Terms must be selected from a registered controlled vocabulary.

4.4.1.2      Government of Canada interpretation

Format is indicated by the file extension of the program or application used to create the resource (<dc.format>: Guidelines,
http://www.tbs-sct.gc.ca/im-gi/mwg-gtm/fmt-fmt/docs/2003/guidelines-lignes-directrices_e.asp).

4.4.2      Usage

The use of the Format element is essential in the creation of metadata for multimedia digital collections that contain a multitude of file formats and applications. The metadata can be embedded in the resource itself or stored externally in a metadata repository.

The form of expression for the Format element will depend on the technical environment of local implementations.

The Format element is not repeatable.

4.4.3      Guidance

The Format element should not be used unless an organization has the ability to manage metadata in a database or a repository.

Every distinct resource should have distinct metadata. Metadata contained in the source code of an HTML resource describes the HTML resource itself. It does not describe resources linked to, or displayed on, that resource. As a result, metadata that does describe a sound, video, other textual non-HTML format, etc., linked to an HTML resource should not be stored in the source code of that HTML resource.

This situation may be confusing. A photo or image (using the .gif or .jpeg file format) appears to be part of the HTML resource in the Web browser; indexers may wish to describe that picture in the metadata for the HTML resource. It is important, however, to separate the metadata for these resources. Metadata for non-HTML resources will likely have to be stored and managed in a database or a repository.

For more guidance on the use of dc.format, see dc.format: Guidelines at
http://www.tbs-sct.gc.ca/im-gi/mwg-gtm/fmt-fmt/docs/2003/guidelines-lignes-directrices_e.asp.

4.4.4      Values

Recommended practice: It is strongly recommended that metadata developers use values from the Government of Canada Format Scheme
(http://www.tbs-sct.gc.ca/im-gi/mwg-gtm/fmt-fmt/docs/2003/schem_e.asp). This scheme includes terms that describe a wide variety of Web site formats.

The Government of Canada Format Scheme is registered as a controlled vocabulary with Library and Archives Canada. The scheme label for the Government of Canada Format Scheme is gcformat.

The Government of Canada Format Scheme is intended to encompass all formats found on federal Web sites. Departments or agencies requiring that new formats be added to the vocabulary should address enquiries to the Secretariat's Information Management Division
(im-gi@tbs-sct.gc.ca).

4.5  Type

4.5.1      Definition

The nature or genre of the content of the resource.

4.5.1.1      Government of Canada constraint

Terms must be selected from a registered controlled vocabulary.

4.5.1.2      Government of Canada interpretation

The Type element may describe the purpose or functions of content; it does not describe the subject (i.e. what the resource is about) or the format (physical or digital manifestation) of the resource (GOL Metadata Working Group, <dc.type> Sub-group, Final Report, November 6, 2002,
http://www.tbs-sct.gc.ca/im-gi/mwg-gtm/typ-typ/docs/2003/finalrep-final/finalrep-final_e.asp).

4.5.2      Usage

The Type element is expressed as follows:

where abc is the label for a specific scheme and def is a value selected from the scheme to describe the type of resource.

4.5.3      Guidance

The Type element is used to describe general categories, functions, or genres of Web resource content. It is not to be confused with the Format element, which describes the physical or digital manifestation of a resource, or with the Subject element, which describes the topic of a resource.

Including a value for dc.type helps users find the information by giving them an additional way to narrow their search queries. When users select a resource type along with other terms (such as title, subject, or date) in a search query, they will be able to narrow their search for an event, presentation, report, policy, etc. For example, a user can reduce the results of a subject search for information management by also selecting the type event or policy.

The Type element must be repeated, with the appropriate scheme label, for each type that applies to the resource.

4.5.4      Values

Recommended practice: It is strongly recommended that metadata developers use values from the Government of Canada Type Scheme
(http://www.tbs-sct.gc.ca/im-gi/mwg-gtm/typ-typ/docs/2003/schem/schem_e.asp). This scheme includes terms that describe a wide variety of resource types found on Government of Canada Web sites.

The Government of Canada Type Scheme is registered as a controlled vocabulary with Library and Archives Canada. The scheme label for the Government of Canada Type Scheme is gctype.

While the Government of Canada Type Scheme includes a broad range of terms, it is not intended to be an exhaustive list of types.For information on adding terms to the gctype scheme, see the Process for Updating the GoC Type and Aggregation Level Schemes Draft (http://www.tbs-sct.gc.ca/im-gi/mwg-gtm/typ-typ/docs/2003/type/type_e.asp).

Organizations may also request the inclusion of new resource types by contacting the Secretariat's Information Management Strategies Division (im-gi@tbs-sct.gc.ca).

For more guidance on the use of the Type element, see dc.type Sub-Group: Usage Guidelines (http://www.tbs-sct.gc.ca/im-gi/mwg-gtm/typ-typ/docs/2003/usage-util/usage-util_e.asp).

4.5.5      Examples

4.5.5.1   English resource

The National Flag of Canada

http://www.canadianheritage.gc.ca/progs/cpsc-ccsp/sc-cs/df1_e.cfm

4.5.5.2      French resource

Le drapeau canadien

http://www.canadianheritage.gc.ca/progs/cpsc-ccsp/sc-cs/df1_f.cfm

4.5.5.3      Bilingual resource

Canadian Heritage welcome page

http://www.canadianheritage.gc.ca

4.5.5.4      Resource with multiple dc.type element terms

The Type element is repeated for each individual term.

4.6  Keywords

HTML keywords metadata contains additional words or phrases to serve as access points for search engines. This metadata may be useful in the context of local information management systems. Keyword values can be natural language terms that are not part of a controlled vocabulary. Whether and how keywords are used depends on local metadata access policies and search engine configuration.

The keywords element is not a Dublin Core element.

For guidance on the use of keywords, please see CLF Best Practices
(http://www.tbs-sct.gc.ca/clf-nsi/inter/inter-06-03_e.asp).