A format can have multiple identifiers, which may be based on entirely separate naming schemes; however, one must be unique within the GDFR and declared as the canonical identifier for the format.
It is the intent of the GDFR to include actionable links to external documents, as well as maintain soft and hard copies of the documents within the registry itself.
All information submitted to the GDFR is subject to technical review for accuracy, completeness, and authoritativeness.
In addition to these properties, the GDFR will investigate the use of format assessment characteristics.
The GDFR defines a set of core registry services in two broad categories: Management Services and Access Services.
Additional administrative services regarding delegation and synchronization between the individual nodes of the distributed registry network will be integrated into the GDFR protocol.
The JSTOR/Harvard JHOVE tool for format-specific object identification, validation, and characterization (Chapman & Abrams, 2004) and the National Library of New Zealand (NLNZ) Metadata Extraction Tool (Searle & Thompson, 2003) are two well-known examples of systems whose implementation and maintenance would be facilitated by the existence of the GDFR to provide sufficiently detailed and authoritative format specifications.
Two criteria for success of the GDFR project are long-term sustainability and trustworthiness.
It remains unclear if the GDFR should operate under the administrative aegis of some existing institution or if an entirely new organization is required.
The initial GDFR data and service models are being tested in a proof-of-concept prototype registry known as Fred (Format Registry Demonstrator) under development at the University of Pennsylvania Library.
The GDFR project will encourage and welcome participation in the project from all appropriate stakeholders, including national, academic, and institutional libraries and archives; standards bodies; commercial interests such as regulated industries with statutory requirements regarding long-term record retention, software vendors as both developers and consumers of formatted information, and content providers; as well as others with an interest in the archival preservation of digital assets.