CAP METADATA
Determining and implementing the appropriate metadata scheme/schema for this digital collection is as critical as determining what will and will not be included in the collection. An appropriate metadata scheme makes it possible to find a collection, manage it, and describe the relationships between objects within a collection. Adopting the six metadata principles recommended by the National Information Standards Organization in its Framework of Guidance for Building a Good Collection affords us the opportunity to assess how effective our metadata scheme will be in making this collection findable, accessible, and preserved into the indefinite future.
We have chosen the Dublin Core Metadata Element Set (DC) to meet the needs of the collection primarily because it is general enough to meet the metadata needs of CAP, which will house a variety of different items. We considered utilizing MARC 21 to organize our metadata, but felt that even with the enhanced descriptors added to manage digital collections, MARC would ultimately prove to be unwieldy to a novice librarian or volunteer who may maintain this collection after it is initially established. Similarly, we felt that DC proved adequate to describe the collection with a level of granularity (down to the page numbers of an article from an issue, creator, etc…) that would meet our users’ needs. We are also swayed by the ease with which a basic understanding of DC metatags could be learned and applied — as opposed to infinitely more complex MARC scheme.
Using the DC scheme for CAP also establishes solid footing for making sure the collection meets the interoperability needs for this collection. While initially established as an internal collection for the organization, the goal of CAP is to create a model for preserving the history of small press magazines that will be accessible “across domains and institutions,” according to the NISO report. We plan to increase interoperability by supporting Open Archives Initiative Protocol for Metadata Harvesting — allowing the metadata (including information about rights and terms of use) from our collection to be harvested for inclusion in other institutional databases. While it currently exceeds the personnel and financial resources we have available for this collection, we may also choose to employ DC/MARC crosswalks to facilitate inclusion of this collection in even more databases.
Adopting a controlled vocabulary for CAP is essential. As a project with limited resources, we do not have the luxury of creating a collection-specific thesaurus, nor would it be desirable if this collection should prove to be a collection that will ultimately include numerous small-press magazines. Because ProQuest has shown some initiative in collecting and making accessible some small press titles as full text, we will be utilizing the UMI ProQuest Thesaurus to describe CAP. The thesaurus is specific enough to meet the collections needs, and will provide future possible future interoperability with UMI ProQuest databases.
In employing DC for our metadata scheme, we will provide rights and terms of use information at the collection level. All work published by Clamor Magazine has been approved by the contributors to be archived for preservation (though not republication) purposes. Reprinting rights may be granted in some cases and information about acquiring those rights will also be included via collection-level metadata.
Metadata ensuring the long-term preservation of CAP is where the project certainly falls short. At this point, no considerations have been made for including preservation metadata in the collection. To do so would add a significant amount of time at the outset of the project (employing PREMIS Data Dictionary for Preservation Metadata), but it may be more beneficial than attempting to add this metadata at a future date.
Another shortcoming of the CAP metadata strategy is the lack of any administrative metadata — or “meta-metadata” as described in the NISO report. DC's suggested Administrative Core may prove useful for this project, primarily if or when more publications are included in the collection. At the outset, administrative metadata is less critical to the collection, which is being created and maintained by a small staff. As more titles are included, administrative metadata will be more essential. Given that, inclusion of A-Core metadata at the outset will save time and money in the long run and all attempts will be made to include this metadata for each resource.
We feel confident that CAP will achieve a “goodness” threshold meeting at least four of the six metadata principles. Reaching that threshold with the remaining two principles will require additional resources and expertise to the project that will need to be factored into any decisions about whether to include preservation metadata and meta-metadata in the CAP metadata strategy.