Why metadata? Metadata, put simply, is information that describes other information. Metadata allows information to be codified and described in standardised ways. Metadata is useful in that it allows information to be: 

• searched for and retrieved via reference to standard descriptors (ie the metadata that describes it)

• managed within systems of storage, or for purposes of processing or manipulation 

As increasing quantities of information exist in digital form, the need to make information responsive to 'machine processing' has been the key driver behind the development of standards and conventions in the use and application of metadata. Much information these days resides on the web, in a web deliverable format, or in electronic repositories and systems. Metadata is the key to managing electronic information. 

How does metadata work? 

On the web, metadata constitutes a 'hidden' attribute of a page. One specific element of HTML (the web 'mark up language' that expresses how the content of web pages should be understood by an application) is the <META /> tag. This HTML element is used to carry the metadata for a web page. It not visible in the content of the page that is ordinarily viewed in the browser, but is accessible to search engines and other applications that can read HTML, and can be seen when viewing the 'source code' that is behind any web page. 

For example, the following metadata describes provides 'keywords' and a 'description' of the content of the home page of the World Wide Web Consortium (W3C) at www.w3.org:

<meta name="keywords" content="W3C, World Wide Web, Web, WWW, Consortium, computer, access, accessibility, semantic, worldwide, W3, HTML, XML, standard, language, ..." /> 

<meta name="description" content="W3C is over 400 organizations leading the World Wide Web to its full potential. Founded by Tim Berners-Lee, the Web's inventor. The W3C Web site hosts specifications, guidelines, software and tools. Public participation is welcome..." /> 

Search engines can 'find' this web page by examining the metadata contained in it. A user who searches on the keyword "W3C" may have search results returned which lists this page and includes the text of the "description" metadata. The description can help in the user assessing if this is what they want. For web pages, the <META/> element can be used to carry any number of descriptors, as defined by the author. The name of the descriptor is declared by the following formula: 

<META NAME = "name" CONTENT = "content"/> 

The name is the name you give the descriptor type. The content is the descriptor instances. For web pages, some standard meta "names" are widely is use, such as the "keywords" and "description" names in the example shown above, and they form the basis upon which many search engines are configured to look for web information. But there are also formal schemes which provide predefined 'sets' of names which can be used in describing web content. 

What are metadata schemes? 

A metadata scheme is the set of descriptor types available to be applied to information. Numerous standard schemes have been developed to address specific information use and management needs. These standards have emerged from the needs of specific interest groups to standardise how they classify information. 

One of the most prominent general standards is the Dublin Core Metadata Element Set, which looks to define some basic general attributes of information - attributes essentially which can apply to all information, such as Title, Creator, Subject, Description - and to set some rules for using them. Elements of the Dublin Core are frequently incorporated in other standards, which have a particular instead of general focus, for example the Australian Government Locator Service (AGLS) Metadata Standard, which is designed to accommodate descriptors relating to the areas of activity associated with government in Australia. 

Why use metadata? 

Metadata should be seen as an important attribute of any element of information that is in electronic format - or is required to be catalogued electronically. Any content that is designed for web use should be classified and 'tagged' within a metadata scheme. 

Schemes which provide a standard for certain types of information should be employed, and if necessary adapted or extended, and use practices should be adopted that relate to the scheme. The value in this approach is that the potential for information use, where it is properly identifiable within a sound metadata scheme, is increased. Opportunities to share the information developed by others are also increased when it is possible to bring it together with your own information through compatible classification schemes. 

How can you start to apply metadata? 

First you will need to analyse the information you have - to consider its nature, how it is currently developed and used, and how its properties compare to other types of information in broader information domains. 

You can then examine what applicable standards already exist - and even work on creating your own metadata classification scheme or adapting an existing one to suit your specific purposes. 

The best initial approach is to work from a broader knowledge management perspective whereby the whole environment of information products and resources is given consideration. The application and use of metadata is an outcome of a more formalised approach to information. Such an approach looks to develop a strategic perspective on information. 

Some commonplace areas of business endeavour which warrant this degree of formalisation lie in the areas of records management, document management, intranets and internets. The use of metadata properly needs to be considered in the wider and long-term frame as the greatest benefits will accrue where the most opportunities to leverage information value are in view. 

Start with… 

Take a look at the Dublin Core Metadata Initiative website - http://dublincore.org . The Australian Government Locator Service (AGLS) Manual for Users also offers a good general explanation of metadata and its application and provides some additional references. It can be accessed from the Australian Office for Government Online website at http://www.ogo.gov.au/ 

Addendum

In response to the original publication of this piece, Duncan Jamieson, Assistant Director Recordkeeping Standards & Policy at National Archives of Australia, kindly provided an update on developments with the Australian Government Locator Service (AGLS).

"There have been some recent changes to AGLS and the Commonwealth's implementation of AGLS. Standards Australia (SA), in conjunction with the National Archives of Australia (NAA), has now released AS 5044 - AGLS metadata element set (Parts 1 & 2).  This standard is intended to for use by any organisation creating or maintaining information services or resources that are locatable on the Internet, particularly the WWW.

While AS 5044 sets out the element set and qualifiers, the Commonwealth Government (through NAA and National Office for the information economy - NOIE) requires Federal agencies to include additional metadata elements in the Commonwealth's implementation of AGLS to help in the management of a number of Government portals and to help set the foundations for future

Government strategy such as e-Government (details at http://www.noie.gov.au/publications/NOIE/better_services-better_gov/index.htm)

• NAA's information on AGLS can be found at www.agls.gov.au

• An overview of the Commonwealth's implementation of AGLS can be  found at: http://www.naa.gov.au/recordkeeping/rkpubs/advices/advice47.html

• The Commonwealth's AGLS implemation manual  can be found at http://www.naa.gov.au/recordkeeping/gov_online/agls/cim/cim_manual.html

• The two parts to AS 5044 can be found at:    http://www.naa.gov.au/recordkeeping/govonline/agls/metadata_element_set.htm"

Any readers seeking further information are invited to email National Archives of Australia at agls@naa.gov.au.