First Monday

Getting the Word Out: Making Digital Project Metadata Available to Aggregators by Diane I. Hillmann

Digital projects managed by traditional libraries have tended to follow the “exhibit” model, with an expectation that Web search engines are the best and only way to attract users to available resources. The development of aggregations based on OAI–PMH provides another avenue for libraries and museums to use to market their resources to users who might not discover their resources using individual portals.


Checklist for making metadata aggregator–ready



Most library digitization and collections projects are excellent, providing compelling and useful material for students, researchers, and the generally curious. But, like the printed products that crowd bookstores and libraries, they can easily be overlooked on the increasingly busy Internet Commons. To a great extent, libraries and other cultural institutions have attempted, consciously or unconsciously, to use the “exhibit” as the primary paradigm defining their online presence. They have sought to create better, more graphically interesting and diverse exhibits to attract users, rather than look carefully at the limitations inherent in their “bespoke portal” based marketing efforts.

In the print world, the publishing industry uses a number of techniques to get their products noticed: author tours, advertising, special placement at bookstores, and review media among them. Yet digital projects seem to rely primarily on search engines to get the word out about their materials. Almost under the radar, aggregation services have begun to find their place in this vacuum, providing services for general or specific groups of users — consuming, repackaging, repurposing and highlighting the work of digital library projects of all kinds.

To take advantage of these services, digital library projects will need to learn about and understand the needs of the aggregators, who work differently than search engines and generally do not support their work with advertising. Metadata aggregators are usually neither creators of resources nor collectors of content. They are service providers, bringing together appropriate resource information and providing users with a broader view of what might be available to them in their areas of interest.

This kind of activity would not have developed as quickly as it has (or maybe not at all) without the Open Archives Initiative Protocol for Metadata Harvesting (OAI–PMH). The basic notion underlying OAI — that metadata is useful because it provides the basis for building services to users — is remarkably robust, and extends beyond the simple “discovery” that motivates most initial work. Other services often discussed in the context of aggregation are those of annotation or review, where information about the quality, context or use of a resource might be provided, either by a formal mechanism related to the collection’s home community or stated purpose, or by informal means, similar to the reviews provided by Amazon.




One of the biggest obstacles to widespread use of aggregators is the notion of metadata as an “asset” not to be “given away” without remuneration. This idea is found even within cultural institutions relying heavily on partnerships and collaborations in any number of program areas. Another obstacle, perhaps more subtle, is a reluctance to give up the control of how resources are presented within a bespoke portal. Technical barriers to OAI implementation, particularly within institutions where IT resources are spread thin, is endemic. Because an aggregation strategy is relatively new, lacking guaranteed results, some institutions also worry that important funders won’t approve and might not continue funding when work is shared freely within an aggregation.



Checklist for making metadata aggregator–ready

For those who are ready to make the leap, here are some suggestions for getting ready:

  1. Make your metadata available in a variety of formats. OAI–PMH requires Simple Dublin Core as a base format, but any format with a schema can be distributed in addition. A variety of format options gives aggregators choices that enable richer and more reliable services.

  2. Consider how context affects your metadata choices. When you distribute your information beyond your site, it’s essential that each record contain the complete context necessary for understanding the item the record describes, without relying on any outside information.

  3. Expose your use of standard vocabularies. The use of standard vocabularies enables better integration of metadata records from one source with records from other sources.

  4. Manage your metadata to enforce consistency. All decisions made about application of elements, syntax of metadata values, and usage of controlled vocabularies, should be consistent within an identifiable set of metadata records so those using this metadata can apply any necessary transformation steps consistently.

  5. Document your practices. Provide a page somewhere on your site where you explain how your data is created and managed and link to it from your home page. If you use a controlled vocabulary of your own, make sure that vocabulary is available at your site, with definitions and relationships exposed.




The world of aggregators and service providers offers opportunities for digital projects to break out of the limitations of one–site shopping. Using these opportunities requires a bit of understanding of the digital world outside libraries, but the possibilities for employing this knowledge to expand the reach of library digital projects are inspiring. End of article


About the author

Diane is currently Research Librarian at Cornell University Library (CUL), after five years as Director of Library Services and Operations at the National Science Digital Library (NSDL). During her nearly 29 years at CUL she functioned as a serials and spformats cataloger and technical services administrator for the Law Library and the library system’s authorities librarian and manager of catalog maintenance.

During her years in traditional libraries (before going over to the dark side of digital libraries) she was a liaison to MARBI from the American Association of Law Libraries and a full MARBI member representing LITA.

In 1995 she accepted an invitation to the first Dublin Core conference, and is now (after 10 peripatetic years on the Dublin Core World Tour) a member of the DC Usage Board and Advisory Board, Editor of “Using Dublin Core,” co–chair of the DC Education Working Group, and administrator of AskDCMI.

June 2004 saw the release of Metadata in Practice by ALA Editions — from then on, her advice has consisted of three words: “Read the book!”
E–mail: Dih1 [at] cornell [dot] edu

Editorial history

Paper received 26 May 2006; accepted 20 July 2006.

Contents Index

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 2.5 License.

Getting the Word Out: Making Digital Project Metadata Available to Aggregators by Diane I. Hillmann
First Monday, volume 11, number 8 (August 2006),