[This local archive copy mirrored from the canonical site: http://www.erin.gov.au/database/metadata/anzmeta/anzmeta-1.1.html; links may not have complete integrity, so use the canonical document at this URL if possible.]

Contents

1 A Guide to Elements in the DTD
  1.1 Overall Structure
  1.2 Text Elements
  1.3 New Leaf Elements
2 XML Conformance
3 General Considerations in Writing Data Records
4 Changes from V1.0
5 Detailed Element Descriptions

1 A Guide to Elements in the DTD

1.1 Overall Structure

The DTD structures the elements of the ANZLIC Metadata Guidelines beyond what is given in the Guidelines. This is necessary to provide a framework in which additional thematic or agency-specific elements can later be added, according to ANZLIC "higher pages" concept.

Table 1 shows the hierachial structure of the DTD, and indicates where the information for each of the named metadata elements is stored. The short names used in the DTD correspond to equivalent element names in the FGDC Metadata Content Standards. Clicking on an element name in Table 1 will display detailed information about that element (links to the Metadata Guidelines are under construction).

1.2 Text Elements

The DTD contains definitions for a number of text-markup elements borrowed from the HTML standard. There are two types of elements - Inline and Block, and are described in Table 2 and Table 3. Although the DTD does not treat these elements differently from metadata elements, they were omitted from Table 1 for simplicity.

While the defined text elements appear the same as HTML, there are a number of restrictions on their use in data records:

• Elements must be specified in lower-case only. This is a result of XML being case-sensitive.
• All elements must have end tags. Paragraphs must be ended with </p>, and list items must end with </li>. This is a requirement for XML documents.
• Element attributes are not permitted, except for the "href" attribute of the <a> tag. The link specified in the href attribute of the <a> tag must be surrounded by quotation marks.

Inline Text Elements can be used within the text of any metadata element, except:

		uniqueid (ANZLIC unique ID)
		long (used in Geographic Extent Polygon)
		lat (used in Geographic Extent Polygon)
		northbc (North Bounding Coordinate)
		southbc (South Bounding Coordinate)
		eastbc (East Bounding Coordinate)
		westbc (West Bounding Coordinate)
		date (used in Beginning Date, Ending Date, Metadata Date)

Block elements can only be used in the following SGML elements:

		abstract (Abstract)
		accconst (Access Constraints)
		lineage (Lineage)
		posacc (Positional Accuracy)
		attracc (Attribute Accuracy)
		logic (Logical Consistency)
		complete (Completeness)
		supplinf (Additional Metadata)

1.3 New Leaf Elements

• keyword
• formname
• long
• lat
• date

Use of these leaf elements allows an application to treat each occurrence of these leaf elements in the same way. For example, although the date element appears in three different contexts,

		timeperd - begdate - date,
		timeperd - enddate - date and
		metainfo - metd - date,

The keyword leaf element strengthens the notion that the text content is one of a finite list of allowable values.

2 XML Conformance

This DTD has been written to conform to the requirements of XML 1.0. To ensure that Metadata documents conform, pay attention to the following points:

• All element names are in lowercase. XML is case sensitive, so documents must use lower-case element tags.
• Element attributes must be in quotes (keyword thesaurus, qualifier). For example:

  	<keyword qualifier="monitoring">MARINE</keyword>
  

• All open tags must have corresponding close tags. This is especially relevant because HTML allows many close tags to be omitted. Example:

  	<p>This is a paragraph.</p>
  	<ol>
  	  <li>First list item.</li>
  	  <li>Note closing tags on list items.</li>
  	</ol>
  	<p>Here is another paragraph.</p>
  

  Line breaks (only) can be specified as <br></br> or <br/>. (This is standard XML markup).
• '<' and '&' cannot occur in as data - they must be written as &lt; and &amp;.

3 General Considerations in Writing Data Records

• All elements must now have content if they appear. However, not all elements are mandatory, so that if information is not available then the entire element should be skipped, rather than inserting empty tags. For example, if the Contact Organisation does not have a fax machine, do not use , but rather leave out the element altogether.
• Whitespace in element content may not be preserved when an application present the metadata content. Use <p>, <br> and list structures where necessary to format the information.
• Paragraphs cannot contain other block elements, only inline text elements. This is the also the case for HTML.

4 Changes from V1.0

The major changes from V1.0 to V1.1 have been to make the DTD compatible with recent developments such as XML, to allow higher page elements (thematic or agency specific) to be incorporated simply, and to clarify the document structure thus reducing interpretation problems.

• New structural elements - Adding new structural elements to encapsulate lower level elements, such as to enclose , allows higher page elements to be added under a relevant heading. Thus can be used by applications to extract all information about the metadata record itself. The following structural elements are new:

  	descript - theme
  	descript - spdom
  	descript - spdom - place
  	distinfo - native - nondig
  	distinfo - native - digform
  	distinfo - avlform - digform
  	distinfo - avlform - nondig
  

• Made implied structure explicit - The ANZLIC Guidelines describes rules for filling out the content of fields. In some cases, these rules actually describe document structure. The structure should be in the DTD to ensure that documents conform to the structure, and to make element retrieval easier. Affected elements:

  	distinfo - native (nondig, digform subelements)
  	distinfo - avlform (nondig, digform subelements)
  	timeperd - begdate (date, keyword subelements)
  	timeperd - enddate (date, keyword subelements)
  	status - progress (keyword subelement)
  	status - update (keyword subelement)
  

• Added leaf elements - adding leaf elements for common data types allows applications to act according to the data type. (long and lat were defined in version 1.0).

  	keyword
  	formname
  	date
  

• Date is now ISO8601 - internal structure of dates have been replaced with a single PCDATA field, with the intent that the field will contain information in the ISO8601 date format. See the notes for the date field below.
• Name changes to match FGDC - Some elements have changed names to match corresponding FGDC elements:

  	timeinfo	=>	timeperd
  	accscons	=>	accconst
  

• HTML-like elements - have replaced the <paragraph>, <list> and <item> elements. Allows users to work with tags that are more familiar, and allows a richer set markup commands without dramatically increasing the work for application programs.
• Some elements are optional - Elements which are optional are indicated in the details below. For these elements, it is recommended that if information is not applicable then the entire element should be skipped, rather than inserting empty tags. For example, if the Contact Organisation does not have a fax machine, do not use <cntfax></cntfax>, but rather leave out the <cntfax> element altogether.
• Qualifier is an attribute of keyword - the themekey/qualifier content model of V1.0 was flawed, allowing multiple keywords in a single themekey element, as well as multiple themekey elements. This has been corrected, and qualifier is now an attribute of .
• Thesaurus is an attribute of keyword - Although not appearing in the ANZLIC Guidelines, a word selected from a finite list has an implied thesaurus (that finite list), even if it is not formally named. We have introduced the thesaurus attribute which is optional and takes a default value of "anzlic". The thesaurus "anzlic" can be assumed to refer to the relevant list from the Guidelines.
• Clarified structure of spdom - Spatial domain is now represented as a list of places, and a bounding box. Each place can be specified using a name or a list of polygons. The bounding element will contain the outermost bounding extent of each of the described places.
• XML compatibility - The DTD is written to conform to XML 1.0. It is expected that documents will also be written to conform to XML 1.0.

Changes from Version 1.0 to Version 1.1 by:

Benjamin Hatton
Live Content Engineering & Design
http://www.livecontent.com.au/
under contract to Environmental Resources Information Network (ERIN)

5 Detailed Element Descriptions

citeinfo

Example:

<citeinfo>
	<uniqueid>
		ANZCW00000000
	</uniqueid>
	<title>
		Possible Scenarios for the Spread of Rabbit
		Calicivirus Disease (RCD) to December 1995
	</title>
	<origin>
		<custod>
			Bureau of Resource Sciences (BRS)
		</custod>
		<jurisdic>
			Australia
		</jurisdic>
	</origin>
</citeinfo>

jurisdic

Allowable Jurisdiction keywords under the default thesaurus are (as defined in the ANZLIC Metadata Guidelines):

     Australia
     Australian Captial Territory
     New South Wales
     New Zealand
     Northern Territory
     Queensland
     South Australia
     Tasmania
     Victoria
     Western Australia
     Other

See also the general discussion of keywords.

descript

Example:

<descript>
	<abstract>
		<p>This is the abstract.  This data was
		collected to:
		</p>
		<ul>
		<li>reason 1</li>
		<li>another point</li>
		</ul>
	</abstract>
	<theme>
		<keyword qualifier="monitoring">SOIL</keyword>
		<keyword>AQUACULTURE</keyword>
	</theme>
	<spdom>
		<place>
			<dsgpolyo>
				<long>134.0</long>
				<lat>-11.0</lat>
				<long>133.5</long>
				<lat>-11.0</lat>
				<long>133.5</long>
				<lat>-12.0</lat>
				<long>134.0</long>
				<lat>-12.0</lat>
			</dsgpolyo>
		</place>
		<place>
			<keyword thesaurus="auslig_topo250k_names">
				Wessel Islands
			</keyword>
		</place>
		<bounding>
			<northbc>-11.0</northbc>
			<southbc>-12.0</southbc>
			<eastbc>136.5.0</eastbc>
			<westbc>133.5</westbc>
		</bounding>
	</spdom>
</descript>

abstract

Can make use of block text elements (<p>, <ul>, etc).</dd>

theme

The theme of the dataset may be described using a number of keywords. In addition, data and/or attributes may be simultaneously collected for a number of different themes, hence multiple <theme> elements are possible. Refer to the ANZLIC Metadata Guidelines for the list of possible keywords.

spdom

The spatial domain consists of a detailed specification of places, plus an encompassing bounding box. This element is optional, since some datasets and/or projects may not be related to a specific spatial area. For example, a laboratory study may collect information generally relevent to a theme, but not specifically relevent to a spatial area.

place

A place is defined by a keyword, or by one or more polygons (eg. "Australia", or polygons covering the mainland and Tasmania). A polygon should not be given in addition to a name which describes that polygon, since this repeats the same information. The <place> element has been moved to within the <spdom> element to indicate its relevance to the spatial domain. A dataset which covers multiple distinct places or areas will have multiple <place> elements defined (eg. A study of 5 beaches in scattered locations). The place element is optional, because data may be available for a large area, eg. the southern hemisphere, for which the <bounding> element is a complete description of its spatial domain. Refer to the ANZLIC Metadata Guidelines for a list of possible keywords.

long,lat

The <dsgpolyo> element contains structured information, being a stream of points, but this structure was not specified in the previous DTD. Including the <long> and <lat> tags has made the structure available. Combining the pairs of <long> and <lat> elements into "coordinate" elements is not necessary, since the alternation between long and lat is fixed, and hence the structure is fully specified.

bounding

The <bounding> element contains an overall bounding box for the dataset, which indicate the outermost extent covering each listed <place>.

keyword

Keywords can have a qualifier and/or thesaurus defined as attributes. Qualifiers are additional attributes which could apply to any valid keyword. The thesaurus attribute is included to allow future definitions of keyword lists specific to themes, applications or agencies. It is envisaged that ANZLIC will maintain a list of valid thesauri. Thesaurus content can be delegated to the relevant agencies. For example, AUSLIG maintain a list of names and codes used to specify mapsheets in the Topographic Map Series.

There is an implied thesaurus with the name "anzlic" which applies to any keyword element that does not specify a thesaurus. This default thesaurus contains the keyword lists which appear in the ANZLIC Metadata Guidelines.

place - keyword

The default keyword thesaurus includes each of the place names listed in the ANZLIC Metadata Guidelines under the section "Geographic Extent Name". See also the general discussion of keywords.

theme - keyword

The default keyword thesaurus includes each of the theme Search Words listed in the ANZLIC Metadata Guidelines under the section "Search Word(s)". Any of the qualifiers listed in that section may also be used with the default thesaurus. See also the general discussion of keywords.

timeperd

Example:

<timeperd>
	<begdate>
		<date>
			1770-01-01
		</date>
	</begdate>
	<enddate>
		<keyword>
			current
		</keyword>
	</enddate>
</timeperd>

begdate, enddate

Prior to this DTD, these elements were declared as PCDATA, making it difficult for indexing programs to handle both numeric dates and keywords. These elements now must use one of two subelements, date or keyword, to clarify how the time information is being represented.

Allowable keywords for the begdate under the default thesaurus are:

	not known

Allowable keywords for the enddate under the default thesaurus are:

	current
	not known

See also the general discussion of dates and keywords.

date

Dates must be presented according to a subset of ISO 8601:

	Full date:	1993-02-14
	Month only:	1993-02
	Year only:	1993

status

Example:

<status>
	<progress>
		<keyword>
			complete
		</keyword>
	</progress>
	<update>
		<keyword>
			not planned
		</keyword>
	</update>
</status>

progress

Progress keywords for the default thesaurus:

	complete
	in progress
	planned
	not known

See also the general discussion of keywords.

update

Update / Maintenance Frequency keywords for the default thesaurus:

	continual
	daily
	weekly
	monthly
	quarterly
	bi-annually
	annually
	as required
	irregular
	not planned
	not known

See also the general discussion of keywords.

distinfo

Example:

<distinfo>
	<native>
		<nondig>
			<formname>
				Hardcopy maps drawn by ...
			</formname>
		</nondig>
	</native>
	<avlform>
		<digform>
			<formname>
				Arc/Info Export, Double
				Precision.  Single precision
				available on request.
			</formname>
		</digform>
		<digform>
			<formname>
				Raster Image in GIF or TIF
				format.
			</formname>
		</digform>
		<nondig>
			<formname>
				Assorted printed maps - Visit
				service center for samples.
			</formname>
		</nondig>
	</avlform>
	<accconst>
		Internal use in government at cost of transfer
		and restricted access licence for non
		government at negotiable price.
	</accconst>
</distinfo>

nondig,digform

These elements have been created to appear under <native> and <avlform> to allow nondigital and digital forms to have different structures at higher page levels.

avlform

Optional, as there may not be standard available forms for the information. For example, the data may be stored in a proprietary database system, or the dataset may not currently be available due to loss, damage or contractual obligations.

accconst

Can make use of block text elements (<p>, <ul>, etc - see the discussion at the top of this file).

dataqual

Example:

<dataqual>
	<lineage>
		<p>These are the basic datasets received from
		Marine laboratories.</p>
	</lineage>
	<posacc>
		<p>Positional error should not exceed 1km for
		the vast majority of pixel centres.</p>
	</posacc>
	<attracc>
		<p>As determined by CSIRO processing of
		spectral response and computation of NDVI, see
		'AVHRR DOCUMENTATION' held by Neil Freeman
		(ERIN).</p>
	</attracc>
	<logic>
		<p>The method of aggregation employed selects
		the maximum fortnightly value for output to
		each monthly pixel, so as to minimize
		atmospheric contamination - thus the monthly
		pixel values are cloud free.</p>
	 </logic>
	<complete>
		<p>All datasets are spatially complete - there
		are no missing sections. The data is
		temporally incomplete - March to November 1994
		are missing.</p>
	</complete>
</dataqual>

All subelements of dataqual can use block text elements (<p>, <ul>, etc - see the discussion at the top of this file).

cntinfo

Example:

<cntinfo>
	<cntorg>
		Environmental Resources Information Network
		(ERIN), Environment Australia 
	</cntorg>
	<cntpos>
		Scientific Coordinator - Remote Sensing
	</cntpos>
	<address>GPO Box 787</address>
	<city>Canberra</city>
	<state>ACT</state>
	<country>Australia</country>
	<postal>2620</postal>
	<cntvoice>+61 2 6274 1203</cntvoice>
	<cntfax>+61 2 6274 1333</cntfax>
	<cntemail>shane@erin.gov.au</cntemail>
</cntinfo>

Voice, fax and email address are optional depending on the how the agency prefers to be contacted.

metainfo

Example:

<metainfo>
	<metd>
		<date>
			1998-01-01
		</date>
	</metd>
</metainfo>

supplinf

Example:

<supplinf>
	<p>This dataset is available through any or our <a
	href="http://mydomain.com.au/resellers.html">resellers</a>.
	</p>
</supplinf>

This field has no metadata subelements, and can make use of block text elements (<p>, <ul>, etc).