ETD Guide/Students/Providing metadata – inside, outside documents

There is nothing new about the concept of metadata. Metadata is resource description; the kind of information found in a library catalogue. What is new in the digital world is the essential role that you, the creator, now play in providing this information. Good quality metadata is easy to provide at the point of creation but usually difficult, expensive or impossible to discover retrospectively.

At one level, this is because all digital resources are in some way dependent on electronic mediation by computers and software and it is only at the point of creation that a record of these dependencies and descriptions can be recorded. At another level, it is the sheer volume of creation that alters the role of the librarian or custodian from cataloguer to metadata repository manager.

In an ideal world, all digital material would be created independent of proprietary hardware and software. In other words, everything would run on commonly available hardware using freely available (public domain) software such as a web browser.

In the real world, many content creators will be producing work on-line or off-line that is either hardware or software dependent (or both). Unfortunately, the costs of emulation, migration and licensing increase if resources are generated in proprietary or platform dependent formats. If possible, try to use commonly available open source formats.

Metadata is information about these applications and formats, which allows for licensed versions to be archived so that the material can be displayed or accessed. In order to be able to provide long-term access to a digital resource, the NDLTD needs the following metadata:


 * Information about the content creator (rights, contributors, publisher,);
 * Information about the content that will help it to be found or discovered (coverage, description, title, subject, relationships);
 * Information about the resource (formats, system requirements, date, identification).

Storing metadata

Metadata can be stored in:

1.  The object or document being described.

There are a growing number of audiovisual formats that allow for metadata to be embedded in the file itself. For example, a text format like HTML allows you to embed metadata in the header of the file and recent versions of image formats such as MPEG include space for metadata. This has the advantage that the information is self-contained and is truly transportable across systems. The major disadvantage is that systems accessing the object will have trouble catering for multiple views or meanings.

2.     A separate file that can be externally accessed but is linked to the object or document.

This has the advantage that different communities can gather the metadata for different purposes. It has the disadvantage of being open to misinterpretation through syntax error or unrecognised schema.

3.  A separate file stored in a database.

The NDLTD model encourages students to submit their metadata to a central repository for indexing in a database. The database will then point to the object/document. This also allows for multiple instances of the metadata for one document. It also provides for enhanced administrative tools (as are normally provided by database systems). Advanced database systems could provide a very sophisticated management system. This is the most expensive method to implement but it has the advantage of being significantly more flexible and provides administrative support from the outset.

See: European Projects such as Metadata Observatory. The aim of the Observatory is to maintain and promote a knowledge base for metadata for multimedia information to continually assess relationships between Dublin Core and other initiatives, especially undertaken in Europe, in order to assist evolution of standardised metadata schemes.

See: The Dublin Core Metadata Initiative. This is an open forum engaged in the development of interoperable online metadata standards that support a broad range of purposes and business models. DCMI's activities include consensus-driven working groups, global workshops, conferences, standards liaison, and educational efforts to promote widespread acceptance of metadata standards and practices. Available [on-line] http://dublincore.org/

Next Section: Protecting intellectual property and how to deal with plagiarism