Tools and Services Working Group
September 24, 2002
Minutes

Attending: Laine Farley (CDL), Catherine Soehner (UCSC), Michael Yonezawa (UCR), Roy Tennant (CDL); guests from CDL: Trisha Cruse, Rosalie Lack, Ellen Meltzer, Brian Tingle.

1. Announcements

a. CDL's public web site: Farley described a project to create a site for non-UC users to make it easier for them to discover resources that are freely available. This site is in response to a request from the Digital California Project, a networking initiative for K-12 schools, which wants to offer more content. In addition, public users have difficulty identifying freely available sources amongst the licensed sources. The project, led by Trisha Cruse, will also attempt to provide access to digital objects at the object level even if they are accessible via some other mechanism (e.g, OAC digital objects). The time frame currently calls for an early 2003 debut which will be a "soft" rollout with publicity mainly via DCP. Eventually CDL hopes that partners with K-12 expertise and/or public libraries or museums may be able to take the content and provide context and other tools for working with these materials.

b. Transitions: Yonezawa reported that in spite of efforts to engage faculty in preparing for the transitions, UCR has had little interest/response thus far. Farley summarized the types of comments CDL has received directly or via other campuses. Responses to the reoccurring complaints have been shared with TSC. Yonezawa reported a series of complaints about PsycINFO which he sent to the Resource Liaison.

Action: Farley will remind RLs to copy CDL on these issues and any exchange they have with the vendors.

2. Academic Research Task Models/Usage Models

Because S. Toub is out on a medical leave, he has not yet digested the research articles members have provided. Several new general user studies have appeared recently and C. Soehner shared one other article she found. The group also reviewed the analysis of queries from UCLA's ERDB shared by Kawakami.

Action: All members will review user studies (both those sent to Steve and new ones) to identify 1) common functions, 2) user expectations, 3) potential pilot projects for services (and building blocks-see below), and 4) principles that we might develop to make systems more successful, both those we develop and for influencing vendors. Share these as they occur to you via the listserv-don't wait for the next meeting!

Some examples: Expectation: users expect search results to be relevancy ranked because of their experience using web search engines. Common function: common workspace for saving, sorting, ranking items. It would be desirable for this function to operate across many resources rather than within individual databases. Possible project: develop a tool to help undergraduates find a few good things (e.g, the Panic button approach to SearchLight discussed at the last meeting). In addition to a few basic databases with good access to full text, it might include a search of Google, the top ten books from Melvyl based on number of owning libraries, and a filter based on online availability. Note that this could become a project once the SearchLight strategy becomes clearer (see item 5).

3. CDL Building Blocks

Members of CDL's Services group joined to discuss work on defining and categorizing "building blocks" for campus libraries to use in creating their own services. The group is still refining what these are or should be. One example is the Content Management system developed to manage CDL's digital objects by harvesting and storing metadata. It is being used for eScholarship's implementation of UC Press books and for objects in OAC. Tingle and Tennant demonstrated how the system can easily present search results with different interfaces based on a simple parameter and thus could be easily adapted for campus based projects.

Discussion ensued about other possible building blocks and their characteristics, with focus on the continuing problems of managing access to electronic resources ( aka the "Directory" problem). Soehner commented that a building block would have to be easier than what they are doing now (e.g., Yuengling's database of electronic journals). Yonezawa mentioned that a tool for creating online pathfinders which could pull authoritative descriptive and access information from a central source would be useful.

There is a need for a "backend" or staff function that CDL's Directory and MIS databases partially fill for providing information on licensing issues. Either a central database accessible by staff or the ability to pull information from a central source for inclusion in a local database would be desirable. Other behind the scenes possibilities include statistics, information on customizable options in vendor resources

The recent Digital Library Forum identified some areas where campuses are working on solving the same problems, the "Directory" and electronic resource management issues being approached from different angles.

To be successful, building blocks might need to be very customizable and flexible, capable of interoperating with many systems at the campus level. Another approach is to focus on things that are smaller in scale and focused on a specific problem. The group agreed it might be best to simply solicit (perhaps via campus visits) and listen to problems and issues from campuses and then try to deduce what building blocks might be appropriate and possible.

4. UC-eLinks

With SOPAG's endorsement, CDL is pursuing licensing SFX for campus libraries. Campus staff will have direct control over turning on targets and services for their local view from an administrative interface.

• Because good metadata is the foundation of the service, we should work toward improving its quality and the understanding of its importance.
• CDL defines minimum standards for Request and communicates to source and target vendors.
• Resource Liaison tests UC-eLinks for all known document types in the resource to see if the metadata is sufficient for Request.
• CDL and Request liaisons review test results to see where problems occur for Request.
• CDL may be able to work with the vendor and/or the data producer to improve metadata.
• Providing one-stop convenient service to users for as many resources as possible is paramount, even if it introduces some additional staff workload.
• If problems still exist for Request, the RL can recommend turning it on by providing a justification (e.g, it works for most of the doc types in the database, the chances for improvement are slim, etc.)
• Alternative access to ILL will be offered (e.g., links to fill-in-the-blank forms).
• CDL and advisory groups will review the alternative mechanisms for Request (e.g., the blank form) for user satisfaction and impact on ILL staff and could revisit the decision on whether and how to offer Request for a given source.

Action: Farley will see if it is possible to populate the blank form with citation information from the vendor's web site so that the user doesn't have to cut and paste.

Update: The only information available to pass to the form is what is in the UC-eLinks window since the link is from that window. Also, the VDX form and most campus forms do not have a general space where the information can be dumped.

5. SearchLight

6.Next steps