HOLLIS Administrative Advisory Committee
Meeting 56
February 22, 1993

1. Preliminary Report of the HAAC Task Group on Large Index Searching in HOLLIS (LISH), Continued
2. Access Paths to HOLLIS and VINE
3. HOLLIS Remote User Survey
4. Harvard Automation Committee
5. Updates

6. Second Report of the HAAC Task Group on a Gateway to Information Resources and Library Systems (GIRLS)


Bob Buckwalter, Marianne Burke, Priscilla Caplan, Michael Fitzgerald, Dale Flecker, Lynne Schmelz-Keil (chair), Carrie Kent, Mark Van Baalen, Judy Warnement, Robin Wendler (recording), Charles Willard

Absent: John Collins, Rodney Goins, Ted Pappadopoulos

Guests: Ellen Isenstein, Naomi Ronen, Julie Wetherill

Future Meetings: In the OIS Conference Room,

          March 17, 10:00-12:00    
          April 13, 3:00-5:00
          May 18, 3:00-5:00   
          June 15, 3:00-5:00

1. Preliminary Report of the HAAC Task Group on Large Index Searching in HOLLIS (LISH), Continued

HAAC continued discussion of the first LISH report. Michael Fitzgerald clarified the recommendation to consider providing one or more subset databases should keyword qualification prove unworkable. A subset database would be another option to be selected from the database selection screen, like HU or OW, but would only contain records meeting a particular criterion, such as works in English or works published in the last ten years. It was clear from LISH's discussions that determining the most useful subset would be difficult, and the group agreed that before such a database was implemented, a wide survey should be taken to solicit options and opinions. HAAC charged LISH to poll groups such as SSHUSH and RRIC for their ideas about the best subset and to consider the best way to solicit the views of library patrons.

Carrie Kent asked whether the more complex HELP structure recommended by LISH was justified given the relatively low use of the existing HELP screens. Michael Fitzgerald explained that the group wanted to encourage users to make more sophisticated use of the system by making HELP both comprehensive and easy to use. Julie Wetherill added that the recommended structure allows us to unclutter many existing screens, thereby simplifying and focussing necessary instructions.

Robin Wendler asked why the exact-word symbol was needed in string searches, since the titles retrieved using such a symbol are also retrieved without it, and appear at the beginning of the index. Ellen Isenstein explained that when users retrieve many hits, they may not even realize that the titles they want are on the first screen. Eliminating unwanted items from the screen will make it easier for users to home in on the items they do want. She also noted that this was not a high priority among the recommendations.

Priscilla Caplan noted that one category of error messages to be reviewed could be those generated by the keyword software which are not already being captured and converted. (The most common keyword error messages are captured and converted into more intelligible text, but the less common messages are passed through as is.) HAAC approved the report in principle, and charged LISH with researching and making recommendations on three topics: 1) the best subset options, 2) whether the default Boolean operator in keyword should remain AND or be changed to ADJacent, and 3) whether and how to use the LOC subfield c in call number indexing. OIS will proceed with an analysis of the feasibility and technical impact of the approved recommendations, to be completed within one to six months.

2. Access Paths to HOLLIS and VINE

Priscilla Caplan distributed a paper describing the methods of access to the HOLLIS and VINE systems to HAAC members for their information.

3. HOLLIS Remote User Survey

Dale Flecker presented the results of the HOLLIS remote user survey, an automatic survey screen which was received at random by dial and network HOLLIS users during Fall 1992. One interesting finding was that relatively little network use is by Harvard affiliates. Dale suggested that this may change next year when the freshman dorms are connected to the network. Currently, HOLLIS can accommodate up to 64 simultaneous network users and 24 dial users. At this time, no more than 32 users have accessed HOLLIS at the same time via the network. Dial access is still increasing, and we have hit the limit of available lines. More lines will have to be added.

4. Harvard Automation Committee

The HUL Automation Committee has been reconstituted, and will undertake a formal planning effort over the next year. The committee's first task will be to survey trends in technology and services both on-campus and nationally, followed by a discussion of priorities and strategies. Sessions will be presented by experts both inside and outside the libraries on such topics as networking, electronic resources, and client-server applications.

The group will produce a planning or strategy document including both short-term recommendations and a strategy for addressing longer term goals and directions. Topics to be addressed may include a determination of which functions should be performed centrally and which in a decentralized way, how to coordinate automation activities throughout HUL into a coherent whole, what the priorities are for the next few years, and how long this generation of HOLLIS should remain in place.

Committee members are Richard DeGennaro, Dale Flecker, Steve Hall, Terry Martin, Judith Messerle, Lynne Schmelz-Keil, Sid Verba, Linda West, and Anne Wolpert. Other members may be added from University administration. HAAC will be kept informed of HAC's progress.

5. Updates


Carrie Kent reported that SSHUSH spent the fall examining alternatives to the Distributable Union Catalog (DUC). The DUC has grown so large and so expensive to produce that most libraries have discontinued their subscriptions. For the short term, SSHUSH recommended that the next 2 or 3 supplements not be cumulative, and that the subject portion be omitted. They requested that OIS come up with an alternative technology for catalog backup within 2 or 3 years. Dale reminded the group that Linda West had suggested use of RLIN as a backup for short- to medium-term outages. (Long-term outages should be covered by an agreement between the university and a "hot-site" vendor, where HOLLIS could be installed on a remote computer within a few days of a major emergency.) A CD-ROM catalog may be another option. Mark Van Baalen suggested that the first step might be to categorize the kinds of failures which should be compensated for, and that this process could suggest appropriate types of backup.

SSHUSH also approved a proposal to index journal abbreviations in the HOLLIS title index. These abbreviations would particularly help scholars in certain science fields, where standard abbreviations are regularly used in journal citation indexes. This proposal was passed on to the Standing Subcommittee on Series, Serials, and Continuations (SSSSC) for further review.

Carrie was asked whether SSHUSH had discussed indexing untraced series in HOLLIS; she will report back to HAAC about this.

5.b. Loading HU into RLIN

Lynne Schmelz-Keil asked Dale about the timeframe of loading HU into the RLIN database. Dale outlined three things which must take place before the entire HU database can be loaded:

  1. OCLC authority processing must be completed on the existing HU file;
  2. a plan must be in place for loading data prospectively;
  3. RLIN must complete review of its policies on what data from member libraries should be loaded and the extent to which such data should be maintained.

The authority processing of the HU base file is likely to be completed in late Summer of 1993; Fall of 1993 is a possible target date for loading HU into RLIN.

5.c. Bibliographic Standards and Policy

Michael Fitzgerald reviewed Bib Standards' recent agendas. Re-evaluation of the published Bibliographic Standard continues. A Catalogers' Roundtable was recently established under the aegis of BSP, planned as a forum for discussion of cataloging issues among staff in HUL libraries. A workgroup consisting of John Hostage, Michael Kaplan, Peter Lisbon, and Robin Wendler has been set up to fully test and document the global change feature added to HOLLIS last year. Later this year, the group plans to make presentations to the Technical Services Managers' Roundtable, HOLLIS Liaisons, and the Catalogers' Roundtable about the global change function.

Several indexing issues have been approved, including proposals to create new indexes for Physical Form Added Entries (755s) and Hierarchical Place Added Entries (752s).

Peter Lisbon is drafting guidelines for subject tagging. Julie Wetherill is working with SSSSC on the new edition of the Holdings Editing Guide.

5.d. RECON

The Widener and Design projects are underway, with Education and Law scheduled to begin in the coming months. The authority work on the HU database is projected to be completed in July.

5.e. VINE

VINE is up and running; approximately 80 public HOLLIS terminals are also providing access to VINE. Additional data sources are being identified.

5.f. Course Catalog

A steering committee has been formed, and their first charge was to select a technical platform for the catalog. [Recent news: After evaluation of the nature of course catalog data and the available platforms (VINE, HOLLIS, and Gopher), and input from a focus group, the committee selected Gopher.]

5.g. Cooperative Ventures

There is no progress to report on cooperative projects at this time except for the Harvard/Yale/MIT Serials in the Sciences project. The library catalogs of MIT (BARTON) and Yale (ORBIS) will be added to the HOLLIS staff terminal VTAM menu, in order to facilitate cooperative collection development and expedited ILL of those journals within the scope of the project.

6. Second Report of the HAAC Task Group on a Gateway to Information Resources and Library Systems (GIRLS)

Discussion of the second GIRLS report began with Carrie Kent reporting on answers to specific questions the group had been asked to address. GIRLS decided that a context-specific HELP feature was very important, and that it was also very important to be able to "launch" to a resource from any screen within the Gateway. The group also redesigned the first Gateway screen to include the categories Bridge, Library, Within HU, Beyond HU, and Explorer (Explorer would include the previous draft's FINDER and LIST).

Dale Flecker asked why someone should care where a database sits, i.e. within HU, beyond HU? Carrie explained that it might not be necessary in a few years, but for now it is important to give users a sense of where they are. There are many ways of categorizing resources, and the group found none of them to be completely satisfactory. Cilla Caplan suggested the Library/Campus/World breakdown used by some other systems, where Library could include those resources supported or sponsored by the libraries (e.g. Citadel databases). Dale asked whether the group had conceived this menu as the libraries' menu or the university's menu? Carrie felt this question should be answered by HAAC rather than by GIRLS. At this time, the libraries would provide the only centrally available menu to Harvard-based resources, whether or not those resources are library-based.

Charles Willard expressed concern that the Gateway was designed from the libraries' point of view rather than from the user's point of view. Carrie noted that no user studies emerged during the committee's literature review. GIRLS could form users groups to review options. Mark Van Baalen urged the libraries to adopt a strategy which allows users to construct their own menus. Cilla asked how we trade off the needs identified in the report with available technologies such as Gopher. There was acknowledgement that there were no easy choices, and that such trade-offs would be necessary.

HAAC will continue discussion at the next meeting.

7. Minutes of previous meeting

The minutes of the February 2nd meeting were approved with minor changes.


Return to the HAAC Home Page

rev. 6/28/96