Minutes for Oct. 26, 2006

Start time: 2 pm. Location: SML 410. Present: Tom Bolze (recording), Daniel Lovins (chair) Youn Noh, Britta Santamauro, Matthew Beacom (guest)

Discussion of DPIP Report

According to the agenda, this meeting was intended primarily to discuss DPIP's Production and Content Integration Working Group (PCIWG) Final Report of Aug. 2006, and presentation to Catalog Librarians on Oct. 12, 2006. Yale Metadata Librarian and DPIP member Matthew Beacom joined the discussion.

For Matthew's benefit, Daniel reviewed the VTF charge and history. In the summer of 2005 Joan and the team leaders assembled a list of process improvement recommendations for consideration by our library colleagues and Meg Bellinger. The idea was that by rationalizing workflows, the department could free up scarce resources to pursue non-MARC metadata projects and collaborate more on digital library initiatives. Wider discussion and implementation were postponed indefinitely because, at least in part, the services to be reduced were spelled out clearly, while details on new services were still relatively vague. There was concern that asking staff members to scale back long-standing practices without providing details on the resulting benefits could be demoralizing and even counterproductive.

The Vision and Directions Task force was formed shortly after these discussions, in September 2005. The group's principal charge has been to help Joan formulate a new vision, mission, and goals statement for the department, with the understanding that the process improvement recommendations would be easier to implement once expectations for the department's future were more clearly understood. Similarly, it seemed that a new name for the department should be recommended only after the new mission, vision, and goals had been more fully articulated.

The recent release of the DPIP report was of great interest to the Task Force because it provided examples of the kind of non-traditional metadata that staff will increasingly be called upon to provide. One such example is the WW-I Experience project, where catalogers, archivists, and others are co-developing an application profile and metadata element set to support a virtual special collection. Other opportunities for collaboration mentioned in the report include the FEDORA digital repository, Classes*v2 server, and the YaleInfo portal.

Matthew mentioned that Emily Horning and Kalee Sprague will be discussing Sakai (customized by Yale as Classes*v2) at the Catalog Librarians meeting on November 2. The wiki features of Sakai could provide the Department with a new platform for documentation. Tom expressed concern that this would add yet more complexity to a system that already spans various servers and applications. Matthew and Daniel raised the possibility that some of the existing documentation could be consolidated in Sakai, thereby streamlining resources and actually improving the findability of needed documents.

Daniel suggested that learning how to use Sakai fits into a broader training mission, namely, to apply concepts, distinctions, and best practices to metadata projects irrespective of the specific hardware or software in which they happen to be embodied at a given moment. Due to the accelerated rate of format obsolescence and data migration, focus should be on optimizing metadata for resource discovery and interoperability. The same trend is driving ILTS to require new web pages to be encoded in XHTML rather than plain HTML. XHTML invokes XML namespaces, allowing data to integrate and migrate more easily within 'Library 2.0' web services. [For similar reasons, the forthcoming RDA tries to distinguish rules on data structure from those regarding data presentation].

Matthew mentioned the Catalog Dept. has already taken on some of the kinds of activities outlined in the DPIP report, including the World War I metadata project (see above). The Metadata Services Team, and eventually Catalog Management, may soon be taking on some responsibility for the MetaLib knowledge database, specifically, keeping track of resource availability and broken links, and communicating with patrons about access.

Daniel suggested that another contribution to MetaLib would be for the Catalog Department to maintain a registry of required data elements that would apply to all record submissions. The absence of standard data elements poses a growing problem as the number of non-Orbis records (i.e., those beyond the reach and care of catalog librarians) continues to expand. For example, the MetaLib Find Database page currently offers the option to search by name of vendor. The problem is that those creating IRDs (Information Resource Descriptions) in MetaLib may not have considered the vendor name (or other data element) important, and therefore left it out. This kind of inconsistency can compromise our users' research experience. Perhaps something like the Yale Elements Set (Y.E.S.) could be made to include such things as MetaLib; IRD submissions could then be machine validated to assure compliance with the proposed standard. In any case, the Catalog/Metadata department would need to be involved. Moreover, the Department needs to stay engaged with public interface developers, to make sure that the search options we offer patrons correspond to what the underlying metadata can support.

In light of the numerous portals available at Yale (e.g., Orbis, MetaLib, subject-based pathfinders), a discussion ensued regarding the ones that users most prefer. Tom and Britta noted that subject-based portals seem especially highly valued. Matthew pointed out that there is often a disconnect between various resources at Yale: resources that differ in format (e.g., digital, paper, sound recording) and ownership status (e.g., leased vs. owned digital resources) are often discovered and used in isolation rather than as integrated by their subject and contents.

Britta noted that it is often unclear what drives a given digitization project, e.g., classroom instruction needs, or curator initiative, [reader requests], etc. She suggested that it is difficult for catalogers to stay motivated about new workflows and technologies if they are largely unaware of the projects already taking place around them. Daniel agreed that the catalog dept. should be more aggressive about publicizing projects internally (as well as externally); moreover, while metadata training should in general be applied to real, immediate projects, certain skills, such as familiarity with XML, may be so fundamental as to be worth cultivating even without immediate application.

Matthew noted that several initiatives may be nearing fruition in a way that would help create more project-based metadata opportunities. FEDORA/VITAL, for example, will probably be implemented soon, and this could create a dynamic metadata production environment. This is a system that has strong backing at the highest levels of the library administration (rather than being specific to any one division), and it will include the need for high-quality supporting metadata. Also, IAC (Integrated Access Council) is considering ways to address Yale's digital preservation needs, any of which will require the involvment of skilled metadata practitioners.

Tom expressed concern that the language of the DPIP report could be interpreted in such a way as to undercut Joan's objectives. Specifically, Joan's plan is to distribute metadata services across various teams rather than consolidating them in a single unit. Matthew noted that DPIP has not taken a position on this matter. Daniel suggested that the expert user program, which has increased the availability of workstation support across the library, might productively be applied as a model to metadata services. Britta was concerned, though, that this might produce the kind of segregated expertise (albeit by seemingly voluntary means) that Joan and the VTF have been trying to avoid.

Discussion on the suitability of the expert user model ended with an apparent consensus that, if the model were modified to include a general call for volunteers (including both C&Ts and M&Ps) and some provision to include those interested in production but not necessarily in support work, it could be a useful way to distribute training and responsibilities across the department. Both Matthew and Daniel noted that this model would require a significant initial investment in training. Mathew suggested that staff outside the department who do related work should be included as well.

Meeting adjourned: 3:00 pm.