In the paper The Seven Levels of Identification: An overview of the current state of identifying objects within digital libraries, Juha Hakala reviews the state of bibliographic identifiers being used within digital libraries in 2005 with a goal of arguing that more needs to be done in this area in order to fulfill the particular needs of digital publishing. Previously used methods are not sufficient to incorporate all-digital scenarios and publishers and may conflict with traditional ways of identifying objects within a collection.
Using the Book Industry Study Group’s (BISG) Roadmap of Identifiers as a starting point, she groups identifiers into seven categories:
• Collections and Services
• Works and Expressions
• Component Parts
• Search Attributes
Hakala excludes combination systems which utilize both traditional identifiers and resolution services, such as DOI or URN because it is her belief that such identifier must ALWAYS include a traditional identifier such as an ISBN and cannot be used independently. She also excludes such combinations as they do not require any kind of formal registration method which makes international standardization impossible.
In looking at the problem continually on an international scale, Hakala shows the difficulties in reconciling multiple naming conventions, lack of organizational standardization (in some case on the local and national level, but most crucially – on the international level), and lack of widespread access to portal metadata which would facilitate service element creation. One of the goals of a unified system would be enable mutual recognition between clients and servers.
Hakala recognizes that in meeting the needs of all parties involved, component part identifiers would become to cumbersome and complex to have any real use. The answer to such a quandary would be to create identifiers that are “…self-sustainable, in the sense that it can be derived directly from the object itself.” Automatic generation would need to be the standard, and thus resources would have to be structured with data elements from which the identifier could be built.
The author recognizes that identifier proliferation is a massive problem. Standardization bodies such as the NISO and ISO rely on voluntary assistance and lack sufficient resources to create and maintain multiple systems at one time. Also at play is the very real issue of prioritization – local institutions are concerned with maintaining their own collections – to help their local population. The integration of a global standardization system is not a concern in their focus on day to day operations, as quite frankly, some are struggling to manage their current workloads. Trying to get all participants to work in common, creating a shared “vision” is quite frankly, fantasy. The practicalitiy of such an arrangement is not very strong. While the author’s intentions are good, I think the reality of the situation demands something closer to NASA’s Unified Metadata Model, where translators help bridge the gap between independent systems, allowing users on the local level to meet their immediate needs and still allowing cross-collaborative research and access.
Just as in real life humanity has failed to utilize a single common language, I think it would be folly to expect that level of simplification and perfection on a technical level, when really, what we generate as content, is merely a reflection of ourselves, our priorities, our concerns, and our cultures at any given moment.