Metadata

Metadata is more abstract data or names of managed data; for instance, "fruits" for "apples" or call numbers for books used in the library.

Overview
A theory developed to interpret a certain theory, i.e. "a theory about a theory," is called a meta-theory. So, "meta" is a prefix which means "beyond" or "at a higher level," etc. Similar to "meta-theory," metadata means data about data, in other words, abstract data of higher-level about managed data. Specifically, metadata are, for example, "fruits" for "apples" or "strawberries" or call numbers or abstracts for books in the library when a book is treated as a datum. Also, if we treat the value in each cell of spreadsheet software as a datum, names of the columns are metadata. Tags attached to posted contents in picture-sharing or video-sharing sites and labels on products in a supermarket are also metadata. Abstract data attached on concrete data which enable us to understand those data are metadata. This concept is well known and widely used as an effective way of data management in the environment such as Information Search or Library and Information Science where digital contents are distributed.

The term "meta" is often used in the field of software. "Meta-language" refers to XML or SGML etc., and "meta-tags" are used to indicate, regarding web pages, information related to character codes, summary of contents, requests to search engines, authors, creators, software used to create. Metadata about metadata are called metameta-data.

Usage of metadata in the field of information search system
When searching information, it is unpractical and ineffective to judge whether each target datum should be included into the search result by interpreting its property upon each search. Therefore, it is necessary to remake and organize targeted data in advance, and metadata are used as a method for it. By representing targeted data as metadata, search can be performed based on metadata. So, it is not necessary to look into each targeted datum. As a result, search efficiency can be improved.

In current major information search system, words that represent contents of data or ones that frequently occur in the data are attached as metadata of targeted data.

Usage of metadata in the field of Library and Information Science
As catalogs or indices, metadata are used as indispensable elements to organize, manage, and provide materials such as books or journals in libraries. Metadata are utilized mainly for managing, sorting, and searching data. An online library catalog is a concrete example. Here, metadata are used in the same way as they are used in the field of information search. For selecting data that suit characteristics of users, judgment are made based on metadata attached on each datum, and then the data are provided.

Metadata and Computers
In order to use metadata on computers, it is necessary to decide how to express them on computers. Metadata schema is useful for it. Furthermore, standardizing metadata schemas enables interoperation of metadata defined by multiple entities. However, nowadays, various schemas are established by schema-related bodies or organizations in specific industries. On the other hand, metadata schema creators, when developing metadata schemas, often tend to be in dilemma to decide which should be more focused, interoperation of metadata or satisfaction of demands about metadata notation specialized for the community's goal.

Metadata schema is made up of the following elements.


 * 1) Attribute set (attribute vocabulary)
 * Definition of the attributes of information resources notated as metadata. E.g., definition of words representing attributes, such as titles, authors, publishers, etc., and sets of attributes.
 * 1) Attribute value type set (attribute type vocabulary)
 * Definition of notation format of attribute values described as metadata. E.g., definition of words that show types or formats of attribute values, including formats for dates and names, controlled vocabularies, etc., and definition of the attributes.
 * 1) Structural Constraints (abstract syntax)
 * Structural constraints on metadata, such as omissibility and loop conditions of attribute values, etc. Structural constraints specify the syntactic rules of metadata that don't depend on the expressive forms of metadata on a concrete system.
 * 1) Expressive form (concrete syntax)
 * Concrete expressive forms of metadata on systems. Codes of metadata that are expressed on editors.

Related Articles

 * W3C
 * XML
 * Folksonomy