Bob Buckwalter, Michael Fitzgerald, Carrie Kent, Kathy Klemperer, Eva Moseley, Jon Rothman, Lynne Schmelz (chair), Hinda Sklar, Dorothy Solbrig, Mark Van Baalen, Robin Wendler (recording)
Absent: Marianne Burke, Rodney Goins, Ted Pappadopoulos, Judy Warnement
October 19, 10:00-12:00 OIS Conference Room November 16, 10:00-12:00 OIS Conference Room December 21, 10:00-12:00 OIS Conference Room
Lynne announced several changes in HAAC membership. The committee is sad to lose John Collins and Charles Willard after many years of valuable contributions (many, many years in John's case!), but is pleased to welcome new members Eva Moseley and Hinda Sklar. Charter HAAC member Dale Flecker is also officially leaving the committee, but promises he'll visit from time to time.
Discussions have taken place at ULC and elsewhere in the Harvard library community about the possibility of a free trial of databases offered on OCLC's FirstSearch. The consensus at ULC was that we would not have public trials of databases unless we have both money and intention of purchasing them. We will be getting a trial of INSPEC on FirstSearch from October 1, 1994 through the end of January 1995. HCL and the Division of Applied Sciences will be working out the financing of the database for the period after the trial. There is a perception that the FirstSearch interface is a problem, but it is not clear how bad a problem it is, so we may also be mount a staff trial of a wider array of FirstSearch databases to evaluate the interface formally.
The Course Reserves project as originally approved by HAAC included a reporting component, where staff would be able to download records from HOLLIS technical services mode into a local relational database. With the advent of the End-User Reporting Project, the same functionality can be provided through Sybase. Instead of selecting records within HOLLIS, staff would access the End-User Reporting database and identify records based on the reporting files there. Daniel Bednarek has been checking with Hilles, Cabot, and Lamont to make sure that this approach will meet their needs. The only drawback is that the data will not be as timely as that in HOLLIS. OIS proposes to update the RV reporting database daily or twice daily. HAAC agreed that the original plan can be changed, as long as this new approach is acceptable to the affected units.
Kathy reported that HOLLIS Plus moved to production on September 16. The main problem encountered has been the multiplicity of gopher clients and environments and the ways they don't work together. Send gopher problems to your local network administrator or to the Network Information Center (NIC) helpdesk (496-2001). OIS has rewritten the scripts connecting to some resources to improve keymapping and functionality. The registration of in-library devices is proceeding apace. OIS will soon begin working on the next phase of the project, which includes the search and launch facility.
Michael raised three topics from the HOLLIS Plus working group:
Jon asked whether the COERS document would be kept up-to-date. If so, the document itself can be accessible in HOLLIS Plus.
The group recommended the following actions:
The working group was asked why the weather appeared on the library gopher. HAAC agreed that if the weather was to be in only one place, it wouldn't be HOLLIS Plus, but since it can be pointed to from many Harvard gophers, it does no harm.
The hurdle to the implementation of a HOLLIS User Comment facility has always been administrative rather than technical, and that barrier was overcome when the Harvard College Library offered to coordinate the referral and tracking of questions. Carrie spent much of this year contacting libraries to enlist their support, to discuss possible pitfalls, and to establish the mechanisms for tracking and referral. Carrie asked HAAC to agree that the facility be made public on November 1, 1994, and that it be treated as a pilot project until the end of spring semester 1995. At that time we should have enough experience with the type and volume of questions to evaluate the project and its impact on reference departments. It may be necessary to adapt the facility or to remove the general questions option. Several HAAC members emphasized the importance of setting users' expectations correctly. This includes making it clear that it is an e-mail system, not a real-time communication with a live on-line librarian, that the questions may be referred, that the answers may often be brief, and that policies on answering questions differ from library to library. HAAC endorsed Carrie's proposal.
Several administrators from OCLC visited Harvard in early September to share their vision of a distributed information environment. They met with many groups of librarians here, including the Automation Planning Committee, the recently-formed Search-Engine Evaluation Committee (SEEC -- see below), and OIS staff. Dale and Larry Dowler are setting up a group to design a FirstSearch interface evaluation project. A large-database searching joint research project is also under discussion.
An ad-hoc subcommittee of HAAC has been established to evaluate available database engines. The committee will be visiting MIT to look at their implementation of OCLC's Newton search engine, and will be drawing up criteria for evaluation of search engines. SEEC members are: Cathy Conroy, Rod Goins, Charles Husbands, Kathy Klemperer, Hinda Sklar, and Dorothy Solbrig.
OIS has received PsycInfo records from 1984 to the present, and received programs from the Florida Center for Library Automation which convert the records from the American Psychological Association's two internal formats into NOTIS MARC. OIS is writing a conversion program to turn this NOTIS MARC output into loadable HOLLIS records. OIS expects the database, with journal article records, book records, and book chapter records, to be in HOLLIS by the end of October. PsycInfo Thesaurus records will not be included in this initial implementation, but may be added later.
After the HAAC discussion and approval of a separate database for auction catalogs, the Fine Arts Library decided they want to send the catalogs to HD. This raises technical problems, because circulation and patron-initiated retrieval will not be supported outside the HU database. Discussions are still underway about a final approach.
Since August 31, several modem pools for different online services have been consolidated, so now the public can dial a single number (496-8500) from which they can use the network to access a variety of systems and services. The old numbers will be eliminated later in the year.
Jon Rothman gave an update on the progress in the End-User Reporting project, now renamed HOLLIS Distributed Reporting. The system will be mounted on a networked Sybase server. A test database is up, the working group is evaluating query tools and will soon choose one to recommend, and OIS will soon generate the full databases. The pilot phase with beta test sites will begin in November.