A metadata schema is the underlying organizational pattern or framework for the metadata, which consists of pre-defined elements representing the specific characteristics of an information resource. Examples of pre-defined metadata elements include title, creator, creation date, and other related bibliographic features. It is important to recognize that there is no single, comprehensive schema for metadata. Instead, there are different types of schemas, each created for specific types of information according to their own set of standardized guidelines.
Although individual schemas are controlled and standardized, there can be a significant amount of variation among the different schemas created by different institutions. Flexibility refers to the ability of the metadata creators to determine the level of detail contained within a record. Consequently, not all schemas possess the same levels of detail. As a result, schemas can vary in the number and types of metadata elements used, in the use of controlled vocabularies, and in encoding into machine-readable form (Taylor, 2004, p. 142). Additionally, it has become increasingly difficult to distinguish the bibliographic features of resources of different types of media and formats when compared to traditional print materials (Kim, 2003, p. 103).
Most schemas, however, do tend to exhibit three common traits: structure, semantics, and syntax. Structure refers to the model that coordinates the data, ultimately arranging how the data is presented. Semantics refers to the meaning associated with the pre-defined metadata elements that compose the schema. For example, does the meaning of the term “author” used in one schema correspond to the meaning of the term “creator” used in another? Syntax refers to how the metadata elements are to be encoded into machine-readable form. The encoding allows the metadata to be processed by a computer program. Unless the encoding scheme understands the semantics of the metadata schema, the data will be unusable (Taylor, 2003). Discrepancies among element meanings and incompatibility among encoding formats of different schemas usually result in interoperability issues.
References
Kim, K. (2003). Recent work in cataloging and classification, 2000-2002. Library Resources and Technical Services, 47(3), 96-109. Retrieved May 20, 2008, from InfoTrac OneFile database.
Taylor, A.G. (2004). The organization of information (2nd ed.). Westport, CN: Libraries Unlimited.
Taylor, C. (2003). An introduction to metadata. Retrieved May 20, 2008, from http://www.library.uq.edu.au/iad/ctmeta4.html
Subscribe to:
Post Comments (Atom)
5 comments:
"Although individual schemas are controlled and standardized, there can be a significant amount of variation among the different schemas created by different institutions. Flexibility refers to the ability of the metadata creators to determine the level of detail contained within a record."
That is a very good point. Often you can have some difficulty retrieving information because of several factors pertaining to the individual who created the Metadata. There could be any number of issues ranging from the original cataloging that had to take place to the control vocabulary that was created.
I guess that's where the whole idea of crosswalks come in, but even that can get dicey, especially as you say when the semantics don't match with what should be equivalent elements between schemas. While crosswalks do help in converting from schema to schema, one must give consideration to minimizing the loss of data as some crosswalks are not bidirectional--you can't always recapture all the original elements from schema 1 once you use a crosswalk to change it to schema 2 and then try to go back to schema 1.
This is a common issue when attempting to merge data from two disparate databases. Schemas are flexible, which is one of their benefits, but that also means you have to identify each property of an object before you can assess their values.
It seems logical that some properties will be universal, say the field 'Title' or 'Author' for example, but there is always the chance that this assumption will result in problems when crossing schemas when merging data.
It seems to me to be a seesaw - one one end you have the rigidity and cross-compatibility of a controlled system and on the other you have a more flexible system that can be used more easily but with the loss of compatibility.
Would you consider "tagging" as part of Metadata?
That is a great question, and I hope my understanding of this concept is correct since I am new to metadata and tagging.
According to Taylor, a Tag is “a number, set of letters, certain set of punctuation marks, and so forth, that designates the kind of field in an encoding standard” (2004, p. 380). Based on that definition, I would say that tagging helps permit the search and retrieval of metadata records. Tags designate the kind of field on the record, and can help create more logical arrangements of records. Thus, tagging definitely plays an important role in using metadata to retrieve information.
Taylor, A.G. (2004). The organization of information (2nd ed.). Westport, CN: Libraries Unlimited.
Post a Comment