Marianne Burke, Bob Buckwalter, John Collins, Michael Fitzgerald, Dale Flecker, Rodney Goins, Carrie Kent, Kathy Klemperer, Ted Pappadopoulos, Lynne Schmelz (chair), Jon Rothman, Dorothy Solbrig, Mark Van Baalen, Judy Warnement, Robin Wendler (recording), Charles Willard
Guests: Karen Carlson-Young, Jeffrey Horrell
April 27, 1994 -- 10-12 a.m. OIS Conference Room May 18, 1994 -- 10-12 a.m. OIS Conference Room June 15, 1994 -- 10-12 a.m. Countway Library
Dale Flecker introduced the Automation Planning Committee paper From HOLLIS to HOLLIS II by emphasizing three main points:
The APC paper is intended to give a taste of the difficulties we will face in moving to a new system and the criteria to be used in evaluating and selecting a new system. The environment has changed tremendously since the initial development of HOLLIS, and this time the libraries will need lots of input from information technology experts on campus. Any development the libraries do will be much more visible this time around, and we will need to lay the groundwork for change in the community well in advance of concrete recommendations. Harvard is not alone among research libraries in looking toward the next generation system. A number of automation directors from peer institutions have been meeting informally to discuss common needs and to survey vendors. This group hopes to issue a white paper for discussion at the CNI meeting in the Fall. The timeframe Harvard has defined is consistent with that of most other comparable institutions.
Charles Willard was pleased that the large research libraries are defining common needs in time to influence the development of vendor systems. Dale reiterated that given the small number of choices for systems this size, it is critical that we make our needs known early. The worldview of large libraries differs significantly from that of vendors. Vendors generally prefer to work on non-ILS development, whereas large libraries need a better ILS from the vendor and anticipate providing the extra functionality with a variety of compatible, peer systems.
Mark Van Baalen noted that a crucial but unstated assumption in the paper is that there will be a campus network. To help shape university planning, documents like this need to make that assumption explicit.
Marianne Burke was concerned about the paper's focus on traditional functionality. She felt that image servers and other newer functions should be basic. Dale explained that image servers profit from having an engine optimized for that specific task, and that it is important to bear in mind that there will not be ONE image server in our future, but many. You could plan to integrate one, but that would leave dozens of others. Better to look for flexibility in the ILS system you buy in working with a variety of non-bibliographic servers.
Dale pointed out that the level of customization of ILS functions in the new system is likely to be very low, which makes the need to define our requirements up front that much more critical. Charles asked whether the premise that we won't modify our next system is based on the assumption that we won't have to (that is, we'll be able to buy what we want) or that we can't afford to because it would be a poor stewardship of programming resources. Dale said that both are true up to a point. NOTIS asked why every site needed to modify the system, and is trying to make those particular areas easily customizable as it develops its new HORIZON system.
Discussion of the final list of large HOLLIS projects will begin with the APC, the ULC, HAAC, and general community input.
Jeffrey Horrell outlined his proposal to establish a separate database in HOLLIS for auction catalog records. The art auction sales catalog is a very important research tool for museum curators, art historians, and students of material culture. The Fine Arts Library currently has approximately 40,000 titles and add 800 each year. Houghton is also interested in adding book auction catalogs to a HOLLIS database.
RLIN has a special database for auction catalog records, SCIPIO. These records, although they are bibliographic, often have common, generic titles, a small number of unique issuing bodies, and have special access points such as date of sale. Jeff anticipates a conversion project to import SCIPIO records, along with current cataloging of new records into the SCIPIO database. SCIPIO is not at this time available through RLG's public Eureka interface.
Exporting records from SCIPIO for loading into a local system has not been done before. It is not clear what modifications to the MARC format would be necessary to accommodate the date of sale information and any other non-standard data. Duplicate detection would not be feasible on such records, so staff would be responsible for keeping duplication to a minimum.
HAAC unanimously approved the proposal in principle. Karen Carlson-Young, Jeff Horrell, and Robin Wendler will clarify the technical issues and submit a revised proposal.
Kathy Klemperer, the new Assistant Director of OIS for Applications Development and Support reviewed active ADS projects: