Academic journal article Library Technology Reports

Metadata Elements

Academic journal article Library Technology Reports

Metadata Elements

Article excerpt

Abstract

Chapter 3 discusses metadata elements, the building blocks of any metadata scheme. Metadata elements have been defined using Semantic Web standards, which form a universal pool from which anyone creating metadata can make choices. This chapter describes a range of sources for metadata elements with explanations on their use.

The building blocks for any metadata scheme are the elements that will be used to define the information that is provided. These are often called data elements, although in Semantic Web terminology they are referred to as properties. I will call them elements when discussing them generally in this document because that is the term most familiar to the metadata developers and creators in the library world. I will use the terms properties and classes in the Semantic Web sense when describing particular element sets since that is what they will be called in the documentation where they are defined in RDF or OWL. In addition, groups of Semantic Web elements that have been defined are called either vocabularies or ontologies, and these terms are used imprecisely and interchangeably. To make things worse, the term vocabularies is also applied to controlled lists of terms that are used as data, not as elements. As much as I dislike the term ontology for metadata element sets (an -ology should be a study of something, and ontology in its original definition in philosophy means the study of reality), I will use it here for metadata term sets so that I can reserve the term vocabulary for the controlled lists (see table 3.1).

Elements can be as simple or complex as the metadata task warrants. There can be one single data element for the title of the resource being described, or there can be distinct elements for primary and secondary titles, translated titles, titles of articles, titles of books, and so forth. It all depends on the purpose of the metadata and the anticipated uses.

Finding Linked Data Elements

As stated in chapter 2 in the section on the Semantic Web standards, in this environment the preference is to reuse elements that have already been defined. Metadata elements that have been defined using Semantic Web standards form a universal pool from which anyone creating metadata can make choices. This is different from previous generations of metadata, where each metadata need resulted in a separate definition of data elements that were valid only internal to the local application that used the metadata. The question then becomes, "How can I find elements to use?" There is no one place to go on the Web to learn about the existence of elements; a certain amount of hunting and observing is needed. However, you needn't worry overly about missing an element that you might have used: you will define your own elements whenever you do not find one that you can use, and you can later add to your definition links to any equivalent elements that you discover or that are later defined elsewhere. While it would be ideal if there were a limited number of metadata element sets that everyone could reuse, the Semantic Web standards are designed to work in an imperfect world where the same concept may get defined in more than one environment. Still, reuse is preferred to the creation of new, redundant terms, so looking for such terms is advised.

The following are some places you might look for previously defined elements.

Swoogle

* Name: Swoogle

* Creator: eBiquity (University of Maryland, Baltimore County)

* URL: http://swoogle.umbc.edu

* Created: 2007

* Updated: daily

Swoogle is a Google for the Semantic Web, although it is in early stages of development. It crawls the Web looking for Semantic Web documents and performs keyword indexing on them. It has three search modes:

* ontology searches the full text of ontology documents

* data searches actual instance data

* term searches only terms that have been defined as classes or properties

Swoogle also archives copies of the Semantic Web documents that it finds, so it works as a kind of archive for the Semantic Web. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.