PCC Participants’ Meeting Summary
ALA 2007 Mid-Winter Conference

Hotel W, Great Room 1
Seattle, Washington
January 21, 2007
4:00–6:00 p.m.

PCC Policy Committee Chair Mechael Charbonneau (Indiana University) opened the meeting with thanks to all those involved in PCC activities for their many contributions, especially the PCC Policy Committee members, PCC standing committee chairs, members of various PCC committees and task groups, and the many individuals involved in PCC training initiatives. Charbonneau then presented an overview of PCC statistics, commenting that everyone's hard work has resulted in strong growth in all areas of PCC activity.

Charbonneau summarized the November 2006 PCC Policy Committee meetings held at the Library of Congress:

• The PCC governance document was being updated to include PoCo SACO representation
• Continuation of optional series authority control by PCC members was strongly endorsed
• A new task group will be appointed to consider simplifications in the area of series authority record creation and maintenance
• The Policy Committee approved the new CONSER standard record
• Work continues on finalizing the PCC strategic plan's tactical action items

Introduction of Guest Speaker

PCC Policy Committee Chair Emeritus Mark Watson (University of Oregon) introduced the evening's guest speaker, Lorcan Dempsey, Vice President and Chief Strategist, OCLC. Watson noted that Dempsey’s presentation would be the first in a lecture series on professional development topics of interest generated by the tactical objectives identified by the PCC Policy Committee. Lorcan Dempsey is well known for writing and speaking on leadership, and Watson acknowledged that Mr. Dempsey’s weblog should be mandatory reading for information professionals.

Keynote Address

Discovery, Disclosure, and Inventory Management: A New Landscape for Catalogs and Cataloging

Lorcan Dempsey likened library catalogs and databases such as OCLC WorldCat to a lake, with sediment collecting on the bottom while other sources of information are more like a river containing constant flowing and ever-changing data. Dempsey observed that the catalog is becoming more river-like in today’s world. The key is to make cataloging data work harder by removing it from the “lake” and manipulating it for other results.

Dempsey pointed to FictionFinder, a FRBR-inspired model for searching and browsing bibliographic records tagged as fiction in OCLC WorldCat, as an example of how data can be made to work harder. The FictionFinder prototype creates a super record at the work level and then collects manifestations into groups based on author and title information from bibliographic and authority records. By re-purposing existing cataloging data, search results in FictionFinder offer richer and more complete descriptions that are more readily understood by library users. Mining information from the existing database and processing the data in new ways gives entry into the full range of WorldCat holdings, and therefore to the world of literature. A new OCLC prototype in development, WorldCat Identities, is yet another example of how processing existing information in WorldCat can be used to make data work harder. This new OCLC product will build collective summaries of the life and work of personal authors and corporate entities combining information available from other sources such as authority records, FictionFinder, RedLightGreen, and Wikipedia. By doing different things with our databases we can provide users with a more engaging bibliographic experience.

Using a grid with the top axis of "stewardship" and the side axis of "uniqueness," Dempsey discussed how description of information objects has been changing over time. Stewardship refers to the level of attention we give things. Books and journals have typically represented an area of high stewardship within libraries despite the fact that most of what is collected has been published in multiple and widely available copies. Special collections, because they are more unique, require the highest level of attention in the area of stewardship. Research and learning materials constitute a variety of community-based materials, such as e-prints and learning objects, which exist in many places. Typically generated by staff and students, this type of material is increasingly becoming digital but is not receiving much curatorial attention at the moment.

The digital record is becoming more important in today's environment as libraries struggle with storage and access issues surrounding digitized resources. This is resulting in new behaviors and support for research and learning models and the development of new services for teaching faculty. Dempsey noted that electronic resources management systems (ERMS) and knowledge bases will likely replace the current integrated library systems. Increasingly libraries will focus on digital materials, archival practices, and providing access to their hidden collections. In the future, "classic cataloging" will likely be applied to a smaller part of a library's collection with element sets being created that may be more or less detailed in MARC, Dublin Core, and other metadata schema.M

The discovery experience for library users must be improved and it must be more attractive. We need to provide disclosure by putting the discovery experience where the users are. How do we get users aware of additional resources in our catalogs? Is there some way we can reveal more to them along the discovery way? Dempsey suggested four ways libraries can help make this happen:

1. Local catalogs must be improved as a discovery tool and we need to develop a richer "skin" on top of our catalogs. Professionals should be discussing the next generation of the catalog in this context, such as Endeca, Primo from ExLibris, and WorldCat. It will be increasingly more important to combine access to holdings information in local collections to internet sources via one interface.
2. Continue to develop more shared catalog discovery environments such as those currently provided by OhioLink and WorldCat. By aggregating information offered to users in these combined resources, collections will be better able to meet the supply and demand needs of a larger audience.
3. Create syndicated catalog discovery environments by projecting our discovery experiences into other environments. This will enable individuals to find data and services in a different stream, but bring them to the library to obtain the resources. It is important that we expose our information into other places, such as Google Book Search and Live Search Books. Course management systems, RSS feeds, and the development of online educational collaboration and learning settings, like the Sakai Project, are other areas where we can insert the catalog discovery process.
4. Leverage the discovery environment by bringing users into our local library environment. For example, using a Firefox Extension/Greasemonkey plug-in script, users of Amazon.co.uk can take advantage of a library holdings lookup feature.

Dempsey predicted that the future will bring more acquisition of catalog data from upstream and more automated creation of metadata from digital materials. We should expect to see an emergence of more common metadata creation environments across resource types. There will be an increase in the creation of structured data coming from automated sources that can be further processed programmatically resulting in less manual intervention needed on the part of staff. He concluded his presentation by stating that we need to make the data work harder in order to serve our users better.

Questions and Answers following Keynote Address

Q: In the connection between Amazon and a library's catalog, can the system determine if one item is a US publication and another one is from the UK?
A: Yes, the retrieval is based on ISBNs, so it can be distinguished.

Q: Publishers often use incorrect ISBNs--how does the system deal with that?
A: In that instance, we just fail. Data must support more automated processes but it is not infallible.

Q: How does Amazon benefit from this collaboration?
A: It becomes more attractive and maybe users will spend more time there, although they don't need to participate in your discovery process.

Q: Please talk about the relationship of PCC work with its rigorous standards and how it fits with the future where the "lake" has currents from different rivers and streams.
A: We are in danger of misinterpreting the relationship between traditional and new environments and seeing them as competitors. The Web environment desires data with more structure and is able to develop lists of names and topics for consistent data indexing. However, Google and Amazon are also interested in the traditional consistent data that libraries are able to provide. However, we need to be certain that the relationships in our data are both good and usable by machines. For example, Google Book Search throws up a website for a book telling of entities with relationships to the book. By contrast, there is a bit more constraint in the relational structure we currently provide with our cataloging data.

Q: What is an example of traditional data that is not structured enough?
A: (Eric Childress, OCLC): 7xx access points are not clear in giving relationships, for example, in the 7xx fields for Festschrifts--which person is the honoree?
A: (Ed O'Neill, OCLC): Uniform Title Authority records are very valuable to bring clusters of records together properly. For even better control of the database, records need more editor, publisher, and illustrator data.