Interoperability and Metadata Quality

The interoperability between metadata schemas and institutional repositories is always a big concern since people want to exchange metadata between repositories. Metadata creation is expensive, in order to reuse metadata, institutional repositories start to harvest metadata from other repositories, this not only dramatically reduces the cost, but also facilitates to share knowledge globally.

However, interoperability is never easy to get, even though people have made efforts on it. I would suggest to control metadata quality. If metadata quality is guaranteed, it might take less time for our IT staff to clear up data and smoothly harvest metadata. Although there is no national standards like AACR2 to control metadata creation, metadata creators could use more controlled vocabularies by consulting FAST and database thesauri, and give enough descriptive metadata for users to search.

Controlling metadata quality offers accurate information for users to search and define the information they need. While for information professionals, this will warrant less potential errors and discrepancy, which would be a huge obstacle when we try to reuse the metadata.

Harmonization of Metadata Standards

Once we move into the digital age, we are excited with the direct access to those great and high quality resources. ETDs, digital collection and digital archives are all available to a variety of researchers. The resource creators employ different metadata standards to facilitate the access and display resource in the most appropriate and descriptive way. Now most resources are created with meatadata schema Dublin Core, MODS, METS, EAD and MARC. Because of these various metadata standards, data transformation is becoming a huge challenge for metadata librarians in the digital age.

When I read the report PROLEARN by Network of Excellence in Professional Learning, I feel like we are in the darkness, it seems to me that the harmonization of metadata standards is never globally actualized. As we all know, now the most promising way to harmonize metadata standards is making crosswalks with XSLT among those various metadata schema. But according to the report, we are just like a starting point, even though we are not sure whether what we have done is correct or not, we still have a long way to go. Probably, what we really need is a standard like MARC21, which could be globally used to describe bibliographic information of a publication. Can we get that far? Before we get that far, I believe that creating crosswalks is still a feasible way to solve metadata interoperability.