On April 17, NETSL, the New England Technical Services Library Association, held its annual Spring Conference. The morning keynote speaker was Karen Coyle. Her presentation called “Metadata is a plural noun” can be found at the NETSL conference website at: http://www.nelib.org/netsl/conference/2009/index.htm.
The thrust of Karen’s presentation was that metadata is constructed for a purpose. Metadata is designed with intent in order to accomplish a specific action. In this sense, metadata comes in various flavors and styles. Perhaps it is a map or a series of textual strings. What is certain is that metadata is larger than our library catalogs or data solely bound in bibliographic records. As a result, Karen urged that we step beyond metadata for the sole purpose of bibliographic records to see the enormous possibilities of metadata.
I came away from with a renewed sense of what metadata could do. I’d like to share an idea that came from this question:
- In the ever growing field of metadata, should we be concerned with the encoding and transferring of metadata as two separate and related activities?
It is important to consider how information is encoded and transferred. Both of these activities demand specific skill sets. Both also require thinking about the purpose of metadata in slightly divergent ways.
Encoded information is data that has been structured according to a set of rules. For example, MARC21 is an encoding scheme that uses numbers to differentiate types of information and structure this information within a bibliographic record. Just as Karen discussed, encoding has a purpose. However, it is not necessarily for the sole purpose of displaying information on the Internet or transferring that information. This is why HTML is seen to be less robust than XML because HTML lacks the ability to manipulate data. Encoding asks that the person deciding on the structure and code be aware of the purpose for which the data will be used. Beyond this, it asks that the person know which particular code and structure work in order to fulfill the necessary purpose. Further, this requires a knowledge of what information to encode (which title, from where, in what languages, etc.). In the world of librarianship, this means having knowledge of MARC21, metadata schemas, controlled vocabularies, and content standards.
Though encoding and transferring rely on each other, they are two different activities because they rely on different skills and knowledge sets.
In librarianship, it is not often that one person does both the encoding and transferring. Often the systems engineer (or systems librarian) does the transferring and the librarian (cataloger or metadata librarian) the encoding. In smaller places, the encoding and transferring is done by one person or librarians rely on vendors to solve technical problems of transferring and displaying information.
In Karen’s presentation, encoding and transferring were not distinguished from one another. As a result, her presentation did not answer for example these questions: Can encoding be extended towards different types of data or even fragmented in order to begin to create linked data? Can metadata transferal take advantage of fragmented encoding to search out more related data on the Internet? What would fragmented encoding look like?
I will leave you with these questions. I urge you to read her PowerPoint presentation found on the NETSL website. It provided me with a great deal to think about how information is chosen, encoded, and transferred now and the possibilities for the future.