Table of Contents
Ross Atkinson suggests there are two kinds of metadata, iterative and explicative. In AACR terms, iterative metadata is the ‘description’. it is not clear to me from this account whether he sees the ‘headings’ as iterative or explicative.
Iterative metadata repeat and thus privilege language or other signs that occur in the referent (as, e.g., in the case of a citation). Explicative metadata, on the other hand, provide information about the referent that is not directly contained or expressed in it. Iterative metadata play therefore primarily the role of the surrogate, accelerating access to the referent. In the case of electronic data, some of its iterative metadata can be produced automatically, assuming a standardized mark-up system – and it is essential that all traditional formats (e.g., books), for as long as they continue to be produced, be designed in such a way that iterative metadata can in future be automatically derived from them. Explicative metadata, on the other hand, add new information about the referent. This is not to imply that explicative metadata can never be automatically generated. Some explicative metadata for information in online form can certainly be created through calculation or reference to other sources or scales. Most explicative metadata, including especially evaluation and analysis relating to the referent, must necessarily be created by people. One of the major administrative and political challenges of information services in the future will be to agree upon how the responsibility and authority for creating explicative metadata will be distributed and shared. [SPO: Scholarly Monograph Series]