Paskin’s paper provides an excellent summary of the development and discussion of the DOI up to this time. Much of his paper focuses on clarifying the definitions of terms and ideas that arose during the development of the DOI. This part seems invaluable for anyone who wishes to be brought up to speed on the DOI or whoever needs a refresher. Beyond the definitions, Paskin examines the major issues in the discussion, both technical and administrative, that I’ve cited in this and my previous report. He devotes a major amount of space and time toward the persistent naming, single or multiple resolution, and local vs. global management problems.

Besides Paskin’s paper, much of the discussion on the mailing list refines or elaborates upon topics cited in my earlier report. These included:

1. Local versus global resolution of the Digital Object Identifier in the routing of queries on a data object
2. Interoperability standards
3. Event notification protocols
4. Intellectual property management
5. “Dumb” versus “Intelligent” DOIs

The first of these issues received a more focused treatment during the most recent six months of discussion. Several contributors identified software programs that work locally to rout queries on data objects. Arthur P. Smith publishers (APS) employs a links manager called IDEAL links to handle queries on its publications. Unlike the central service called for by the DOI proposal, IDEAL links operates on an algorithmic basis to perform reference and citation linking. Other publisher representatives identified similar types of services to accomplish these tasks.

Much of the difficulty in accepting the DOI for general usage among publishers and other interested parties arises from the ongoing debate about whether to use “dumb” or “intelligent” DOIs. Complicating the issue further still, several parties cited the tripartite distinction among manifestation, performance, and the digital object itself. A work requires several DOIs to address the distinction, which seems to obviate the use of “dumb” DOIs with their one-to-one correlation between object and identifier.

The publication of Paskin’s paper stimulated a great deal of discussion that focused on identifying those “kernel" or minimal metadata elements that would facilitate an extended syntax for referencing and managing these objects, much as CORC and other e-mail lists are doing with Dublin Core elements. Several people lamented the DOI’s comparability to the SICI (Serial Item and Contribution Identifier) which they criticized for becoming over-specialized and too narrow for their purposes. Others longed for a “Killer Application” to galvanize the direction and use of the DOI for published materials.

The comments mentioned above prompted several participants, including Stuart Weibel, to deem the redirection issue largely a red herring. The critical issue he feels is not a technological one, but one of determining a policy for naming these items and developing commitment by involved organizations to adhere to whatever policy they adopt. Several parties urged those concerned to adopt a policy and employ the DOI quickly or else lose its application to other identifiers such as PURLs or URNs. IDEAL’s announcement in July that it is using the DOI to identify and link to its new articles underscored majority sentiment that the time for debate has passed and publishers must move forward and commit to the DOI, despite its current imperfections, as their referencing and managing link.

Discussions in the past month over the distinction between level 1 and level 2 DOIs to determine what the object is suggest that publishers have moved beyond the theoretical to the pragmatic stage of their commitment. The announcement of a DOI booth at the Frankfurt Book Fair further indicates ongoing acceptance and commitment to the DOI by the publishing community.

William Fietzer
Chair, Outreach Subcommittee
Networked Resources and Metadata Committee


______________________________

* (Those of you unfamiliar with the origins, concept and structure of DOI, consult the current status of DOI conceptualization and development at http://www.doi.org/system_spec.html . For additional information, contact the DOI web site at http://www.doi.org/index.html)

www.ala.org/alcts/organization/div/nrmc/doi2.html
Last modified 12/16/1999 jca